NSL complex - the MOF assembly that five NUP98 fusions bind alongside MLL1, and that competes with MLL/COMPASS for WDR5
also called NSL · NSL complex · non-specific lethal complex · KAT8 regulatory NSL complex · MOF-NSL
NSL is a nine-protein machine built around the histone-acetylating enzyme MOF. It matters to this project twice over: five leukaemia fusion proteins that contain no KMT2A at all grab both NSL and the normal MLL1 complex to cause leukaemia, and NSL competes with MLL-type complexes for a shared component called WDR5.
The same thing, for a scientist
NSL (non-specific lethal) is a nine-subunit KAT8/MOF acetyltransferase whose roster includes KANSL1, KANSL2, KANSL3, KAT8, MCRS1, PHF20, OGT, WDR5 and HCF1; WDR5 and HCF1 are also found in MLL/SET complexes and MCRS1 in INO80 [E1]. **Complex identity sets the enzyme's output**: MSL-associated MOF acetylates nucleosomal H4 almost exclusively at K16, whereas NSL-associated MOF has relaxed specificity and also acetylates H4K5 and H4K8 [E1]. Two results connect NSL to this project. **(1) The cross-family result.** NUP98-HOXA9, NUP98-HOXD13, NUP98-NSD1, NUP98-PHF23 and NUP98-TOP1 - none of which contains KMT2A - physically interact with MLL1 and with NSL; NUP98-HOXA9 and MLL1 co-localise at HOX promoters; and **MLL1 is required for NUP98-HOXA9 proliferation in vitro and in vivo**, with MLL1 inactivation reversing a NUP98-rearranged human leukaemia signature [E2]. **(2) The WDR5 partition.** The KANSL1/WDR5/KANSL2 structure shows WDR5's interactions with NSL and with MLL/COMPASS are **mutually exclusive** [E3]. Against that, an earlier proteomic study treats WDR5 and HCF1 as shared subunits and reports NSL acting **upstream** of MLL/SET to promote H3K4me2 [E4]. Genetically, MOF's catalytic activity is required for MLL-AF9 leukaemogenesis, and Mof inactivation also suppresses NUP98-HOXA9-driven AML [E5] - a requirement that spans both fusion families and therefore predicts, correctly, that no NSL subunit shows KMT2A selectivity in DepMap: seven of the eight NSL-associated genes scanned fall on the anti-selective side of the median [E6].
What NSL does to the menin-interface hypothesis
It supports the general form and narrows the specific form, and the two need separating.
The general form - the wild-type MLL1 complex, and the menin interface within it, is the druggable node - gets meaningful support. Five structurally unrelated fusion proteins, built from a nucleoporin rather than from KMT2A, run their leukaemogenic programme through MLL1, and removing MLL1 stops them growing 2. A node that two independent fusion families converge on is a better therapeutic target than a node one family happens to use.
The specific form - this is a KMT2A-fusion-specific vulnerability - is weakened by the same evidence, and more so by the MOF result. MOF's HAT activity is required for MLL-AF9 leukaemogenesis and Mof inactivation suppresses NUP98-HOXA9-driven AML 5. A requirement that holds across fusion families is a requirement of myeloid leukaemogenesis, not of a fusion.
The testable consequence is clinical rather than mechanistic: if NUP98-rearranged leukaemia runs through MLL1, it should respond to menin inhibition. That is a checkable prediction the graph can hold, and it is the most useful thing this complex contributes.
The WDR5 partition, and the disagreement it sits inside
Two papers make incompatible-sounding statements about the same protein:
| WDR5's status | Consequence claimed | |
|---|---|---|
| Cai et al. 2010 1 | a shared subunit of NSL and MLL/SET complexes | shared membership implies functional coupling |
| Zhao et al. 2013 4 | shared, following Cai | NSL acts upstream of MLL/SET to promote H3K4me2 |
| Dias et al. 2014 3 | not shared; NSL and MLL/COMPASS binding is mutually exclusive | WDR5 assembles distinct complexes with different roles |
The reconciliation is available and is stated in none of them: a protein can be a component of two complexes without occupying both simultaneously, and 'crosstalk' produced by competition for a limiting component looks, in a knockdown, much like crosstalk produced by sharing.
This entry therefore holds the following position in its own voice: NSL and MLL/SET are functionally coupled through WDR5, and the coupling is at least partly competitive. It does not call WDR5 a shared subunit, and the graph should not draw WDR5 inside both complexes at once.
That position generates the hypothesis hyp-wdr5-partition-nsl-versus-compass, which is the one genuinely new, falsifiable idea this beat produced.
Cell type, cell state, localisation, epigenetics at complex level
Cell type. Murine MLL-AF9 and NUP98-HOXA9 leukaemia models, human MLL-AF9 lines, and human NUP98-rearranged patient expression signatures 25; human cell lines for the biochemistry 14; transgenic Drosophila for the WDR5 requirement 3. No infant ALL, no KMT2A::AFF1 system anywhere.
Cell state. The leukaemia phenotypes are transformation and initiation phenotypes - conditional Mof deletion reduced tumour burden and prolonged survival in a mouse model, and MLL1 inactivation blocked NUP98-HOXA9 proliferation 25. There is no differentiation-block or leukaemia-stem-cell-state readout specific to NSL.
Subcellular and subnuclear. Nucleus; three subunits (KANSL1, KANSL2, KANSL3) are additionally annotated to the mitochondrion, and KANSL1 to the kinetochore and spindle pole. Subnuclear: promoter-proximal, with NSL recruitment to target promoters requiring the KANSL1-WDR5 interaction 3. No nuclear-body or compartment localisation is reported and none is claimed.
Epigenetics - the practically important part. The mark depends on which complex MOF is in: NSL-MOF gives H4K5ac and H4K8ac as well as H4K16ac; MSL-MOF gives essentially only H4K16ac 1. This graph reasons about MOF largely through H4K16ac, which is the MSL-specific output. An H4K16ac readout is therefore not a clean assay for NSL activity, and H4K5/K8ac is the better NSL-specific measurement. Separately, NSL subunit depletion lowers H3K4me2 at a shared target promoter 4, which is the only epigenetic link between this complex and the H3K4 axis the fusion disease runs on.
Why the dependency data is quiet and what that does and does not mean
Across the eight NSL-associated genes in the scan, seven are anti-selective: KANSL2 +0.2265 (rank 17,631), KANSL1 +0.0685 (15,913), PHF20 +0.0339 (13,286), KANSL3 +0.0371 (13,630), MCRS1 +0.0139 (10,971), KAT8 +0.0041 (9,683), and only OGT is negative at -0.0829 (rank 1,634) 6. WDR5, the shared component, is +0.1984 at rank 17,587 - among the most anti-selective genes in the genome by this statistic.
That uniformity is consistent with the published biology rather than contradicting it. The MOF requirement demonstrated genetically is a requirement for leukaemogenesis in two fusion families 5; a statistic built to find what KMT2A-rearranged lines need more than other leukaemia lines is blind to a requirement both arms share.
The honest summary: NSL is required by leukaemia cells and is not differentially required by KMT2A-rearranged ones. Both halves of that sentence matter.
| complex name | NSL (non-specific lethal) complex, KAT8/MOF-containing |
|---|---|
| subunit count reported | 9 |
| classification | a MOF-containing acetyltransferase functionally coupled to MLL1; bound by five NUP98 fusions; NOT shown to interact with any KMT2A fusion |
| shown for fusion | NUP98-HOXA9, NUP98-HOXD13, NUP98-NSD1, NUP98-PHF23, NUP98-TOP1 (complex level). No KMT2A fusion. |
| position in mechanism | acetylates nucleosomal H4 at K5, K8 and K16 through MOF; competes with MLL/COMPASS for WDR5; depletion lowers H3K4me2 as well as H4K16ac at shared promoters |
| cell type evidence | murine MLL-AF9 and NUP98-HOXA9 models; human MLL-AF9 lines; human NUP98-rearranged expression signatures; human cell lines for biochemistry; transgenic Drosophila for the WDR5 requirement. No infant ALL, no KMT2A::AFF1 system. |
| cell state evidence | transformation and leukaemia-initiation phenotypes; no differentiation-block or LSC-state readout specific to NSL |
| epigenetic readout | NSL-MOF writes H4K5ac, H4K8ac and H4K16ac; MSL-MOF writes essentially only H4K16ac. NSL subunit depletion also lowers H3K4me2 at a shared promoter. H4K16ac is therefore NOT a clean assay for NSL activity. |
| granularity exception | Complex-level entry written alongside, not instead of, separate subunit entries (gene-kansl1, gene-kansl2, plus the pre-existing gene-kat8, gene-wdr5, gene-hcfc1). It carries only what cannot be said at subunit level: the cross-family NUP98/MLL1 convergence, the WDR5 partition and the disagreement around it, the complex-dependent mark specificity, and the roster-wide dependency pattern. |
subunits
[
{
"gene_symbol": "KANSL1",
"uniprot": "Q7Z3B3",
"length_aa": 1105,
"role": "scaffold; binds WDR5 and activates KAT8",
"separate_entry": true,
"node": "gene-kansl1",
"depmap": {
"difference": 0.0685,
"rank": 15913,
"pan": -0.7402
}
},
{
"gene_symbol": "KANSL2",
"uniprot": "Q9H9L4",
"length_aa": 492,
"role": "WDR5-binding motif; in PDB 4CY2",
"separate_entry": true,
"node": "gene-kansl2",
"depmap": {
"difference": 0.2265,
"rank": 17631,
"pan": -0.9807
}
},
{
"gene_symbol": "KANSL3",
"uniprot": "Q9P2N6",
"length_aa": 904,
"role": "NSL subunit",
"separate_entry": false,
"reason_no_entry": "0 PubMed records against MLL/KMT2A; no structure; dependency unremarkable (rank 13,630)",
"depmap": {
"difference": 0.0371,
"rank": 13630,
"pan": -1.1827
}
},
{
"gene_symbol": "KAT8",
"uniprot": "Q9H7Z6",
"role": "catalytic HAT subunit, shared with MSL",
"separate_entry": true,
"node": "gene-kat8",
"depmap": {
"difference": 0.0041,
"rank": 9683,
"pan": -1.4515
}
},
{
"gene_symbol": "MCRS1",
"role": "shared with INO80",
"separate_entry": false,
"reason_no_entry": "1 PubMed record against MLL/KMT2A; dependency unremarkable (rank 10,971)",
"depmap": {
"difference": 0.0139,
"rank": 10971,
"pan": -0.6226
}
},
{
"gene_symbol": "PHF20",
"role": "NSL subunit",
"separate_entry": false,
"reason_no_entry": "1 PubMed record against MLL/KMT2A; essentially not required anywhere (pan -0.0802)",
"depmap": {
"difference": 0.0339,
"rank": 13286,
"pan": -0.0802
}
},
{
"gene_symbol": "OGT",
"role": "O-GlcNAc transferase, NSL-associated",
"separate_entry": false,
"reason_no_entry": "3 PubMed records against MLL/KMT2A; the only negative difference in the roster but well inside noise; a large independent O-GlcNAc literature that this beat could not read",
"depmap": {
"difference": -0.0829,
"rank": 1634,
"pan": -1.4669
}
},
{
"gene_symbol": "WDR5",
"role": "shared component; NSL and MLL/COMPASS binding mutually exclusive",
"separate_entry": true,
"node": "gene-wdr5",
"depmap": {
"difference": 0.1984,
"rank": 17587,
"pan": -1.159
}
},
{
"gene_symbol": "HCFC1",
"role": "reported in both NSL and MLL/SET complexes",
"separate_entry": true,
"node": "gene-hcfc1"
}
]localization
{
"subcellular": "nucleus; KANSL1/2/3 also mitochondrion, KANSL1 also kinetochore and spindle pole",
"subnuclear": "promoter-proximal; recruitment requires the KANSL1-WDR5 interaction. No nuclear-body localisation reported."
}wdr5 partition
{
"claim": "WDR5's interactions with NSL and with MLL/COMPASS are mutually exclusive",
"source_evidence": "E3",
"contested_by": "E1 and E4, which describe WDR5 as a shared subunit",
"graph_position": "NSL and MLL/SET are functionally coupled through WDR5, at least partly competitively; WDR5 must not be drawn inside both complexes simultaneously",
"hypothesis": "hyp-wdr5-partition-nsl-versus-compass"
}enumeration record counts — the numbers behind the chart
[
{
"query": "(MLL OR KMT2A) AND (\"NSL complex\" OR \"non-specific lethal\")",
"records": 4,
"records_after_title_check": 4
},
{
"query": "(MLL OR KMT2A) AND (KANSL1 OR KIAA1267 OR MSL1v1 OR NSL1)",
"records": 4,
"records_after_title_check": 2
},
{
"query": "(MLL OR KMT2A) AND (KANSL2 OR C12orf41 OR NSL2)",
"records": 1
},
{
"query": "(MLL OR KMT2A) AND (KANSL3 OR KIAA1310 OR NSL3)",
"records": 0
},
{
"query": "(MLL OR KMT2A) AND (KAT8 OR MOF OR MYST1 OR hMOF)",
"records": 11
},
{
"query": "(MLL OR KMT2A) AND (MCRS1 OR MSP58 OR INO80Q OR MCRS2)",
"records": 1
},
{
"query": "(MLL OR KMT2A) AND (PHF20 OR GLEA2 OR TZP OR HCA58)",
"records": 1
},
{
"query": "(MLL OR KMT2A) AND (OGT OR \"O-GlcNAc transferase\")",
"records": 3
},
{
"query": "NUP98 AND (\"NSL complex\" OR KANSL1 OR MOF OR KAT8)",
"records": 1
},
{
"query": "(\"NSL complex\" OR KANSL1 OR KANSL2 OR KANSL3) AND (leukemia OR leukaemia)",
"records": 8,
"records_after_title_check": 1
},
{
"date": "2026-09-19"
}
]- The MOF requirement spans MLL-AF9 and NUP98-HOXA9 leukaemogenesis, so an NSL- or MOF-directed agent is a broad myeloid-leukaemia agent rather than a KMT2A-selective one, and should be evaluated against that wider comparator set from the start.
- If WDR5 is partitioned between NSL and MLL/COMPASS, inhibiting NSL could RELEASE WDR5 to SET1/MLL complexes - the wrong direction in a disease dependent on a wild-type KMT2A complex. This compensation risk follows directly from the structural result and has never been measured.
- KAT8/MOF is common-essential (-1.45 across 1,208 lines) and its loss increases gamma-H2AX nuclear foci and downregulates DNA-damage-repair genes [E5]. A MOF-directed agent is a genotoxic agent in every proliferating tissue, which is a poor profile for a paediatric survivor population.
- Three NSL subunits are annotated to the mitochondrion, so NSL perturbation is not purely a chromatin perturbation and an observed viability effect may be metabolic.
- Because MOF sits in both MSL and NSL, every genetic result in this literature is ambiguous between the two complexes, and a programme targeting 'NSL' on the strength of Mof deletion data would be targeting an assembly that has never been separately tested.
- Does NUP98-rearranged leukaemia respond to menin inhibition? The published MLL1 dependency of NUP98-HOXA9 cells predicts it should, and this is a checkable clinical claim rather than a mechanistic guess.
- Which NSL subunit contacts NUP98 fusion proteins? The result is at complex level, and the answer decides whether a subunit-selective intervention is even conceivable.
- Does KMT2A::AFF1 interact with NSL? Five NUP98 fusions do; no report was found either way for a KMT2A fusion, making this an untested absence rather than a negative.
- Does depleting a KANSL subunit - rather than MOF, which sits in both MSL and NSL - phenocopy the Mof leukaemia phenotype? Until that is done, no leukaemia phenotype can be assigned to NSL as opposed to MSL.
- Is H4K5/K8 acetylation, the NSL-specific output, altered at HOXA9 and MEIS1 in KMT2A-rearranged cells? The graph currently reasons about MOF through H4K16ac, which is the MSL-specific mark.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
The NSL complex comprises nine subunits; WDR5 and HCF1 are shared with members of the MLL/SET family of H3K4 methyltransferase complexes and MCRS1 with the human INO80 remodelling complex. Assembly of MOF into MSL versus NSL controls substrate specificity: MSL-associated MOF acetylates nucleosomal H4 almost exclusively on K16, while NSL-associated MOF has relaxed specificity and also acetylates H4 on K5 and K8.
Subunit composition and substrate specificity of a MOF-containing histone acetyltransferase distinct from the male-specific lethal (MSL) complex · Cai Y, Jin J, Swanson SK, Cole MD, Choi SH, Florens L, Washburn MP, Conaway JW, Conaway RC · Journal of Biological Chemistry 285(7):4268-72 2010What this evidence cannot show
Abstract fetched from PubMed and read on 2026-09-19; full text not read, identifiers copied from that fetched record. Proteomic purification through multiple candidate subunits in human cells; substrate specificity assayed on nucleosomal substrates in vitro, so the in vivo mark distribution may differ. **The nine-subunit roster is from one study and NSL rosters differ between preparations.** The word 'shared' for WDR5 is this paper's and is contradicted as a statement about simultaneous engagement by the later structure (E3); this entry does not adopt it.
Evidence that argues against this (1)
- A structural study of the KANSL1/WDR5/KANSL2 subcomplex concludes that WDR5's NSL and MLL/COMPASS interactions are mutually exclusive and proposes that WDR5 is not a shared subunit but a component assembling distinct complexes. Proteomics cannot distinguish 'found in both complexes' from 'in both at once'. Dias J et al. 2014
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
NUP98-HOXA9, NUP98-HOXD13, NUP98-NSD1, NUP98-PHF23 and NUP98-TOP1 physically interact with MLL1 and with the non-specific lethal (NSL) histone-modifying complex; NUP98-HOXA9 and MLL1 co-localise on chromatin at HOX gene promoter regions; MLL1 is required for the proliferation of NUP98-HOXA9 cells in vitro and in vivo; and MLL1 inactivation decreases expression of genes bound by NUP98-HOXA9 and MLL1 and reverses a gene-expression signature found in NUP98-rearranged human leukaemias.
NUP98 Fusion Proteins Interact with the NSL and MLL1 Complexes to Drive Leukemogenesis · Xu H, Valerio DG, Eisold ME, Sinha A, Koche RP, Hu W, Chen CW, Chu SH, Brien GL, Park CY, Hsieh JJ, Ernst P, Armstrong SA · Cancer Cell 30(6):863-878 2016What this evidence cannot show
Abstract fetched from PubMed and read on 2026-09-19; full text not read, every identifier copied from that fetched record. **None of these fusions contains KMT2A**, so this is evidence about the wild-type MLL1 complex as a convergence node, not about KMT2A::AFF1. Murine transformation and xenograft models plus human expression signatures; no primary paediatric material described in the abstract. The NSL interaction is reported at complex level without naming the contacting subunit.
Evidence that argues against this (1)
- The same group's follow-up shows MOF catalytic activity is required for MLL-AF9 leukaemogenesis AND that Mof inactivation suppresses NUP98-HOXA9-driven AML, which means the NSL/MOF requirement is shared across fusion families rather than specific to either - the direct reason no NSL subunit shows KMT2A selectivity in DepMap. Valerio DG et al. 2017
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
The KANSL1/WDR5/KANSL2 structure shows WDR5 is recruited into NSL via conserved linear motifs of KANSL1 and KANSL2; structure-based KANSL1 mutants in transgenic flies show the KANSL1-WDR5 interaction is required for NSL assembly, efficient recruitment to target promoters and fly viability; and the interactions of WDR5 with the MOF-containing NSL complex and with MLL/COMPASS histone methyltransferase complexes are mutually exclusive.
Structural analysis of the KANSL1/WDR5/KANSL2 complex reveals that WDR5 is required for efficient assembly and chromatin targeting of the NSL complex · Dias J, Van Nguyen N, Georgiev P, Gaub A, Brettschneider J, Cusack S, Kadlec J, Akhtar A · Genes & Development 28(9):929-42 2014What this evidence cannot show
Abstract fetched from PubMed and read on 2026-09-19; full text not read, identifiers copied from that fetched record. Crystallography plus **Drosophila** genetics - the in vivo requirement is established in flies, not in mammalian cells and not in leukaemia. 'Mutually exclusive' rests on overlapping binding sites plus binding data; the competition experiments themselves were not read here. The consequence for WDR5 availability in a KMT2A-rearranged cell is a prediction of this graph, not a finding of this paper.
Evidence that argues against this (1)
- Two earlier studies describe WDR5 and HCF1 as shared subunits of NSL and MLL/SET and build a functional-crosstalk argument on that sharing. The accounts are reconcilable but neither paper performs the reconciliation, and this entry states the coupling as at least partly competitive rather than adopting either wording. Zhao X et al. 2013
- E4moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
Depletion of the NSL subunits hMOF or NSL1/KANSL1 significantly reduced both H4K16ac and H3K4me2 near the ANKRD2 transcription start site, whereas depletion of RbBP5, a core MLL/SET component, reduced H3K4me2 only - consistent with NSL acting upstream of MLL/SET to regulate H3K4me2 at certain promoters; a set of genes including ANKRD2 was affected by knockdown of both NSL and MLL/SET subunits.
Crosstalk between NSL histone acetyltransferase and MLL/SET complexes: NSL complex functions in promoting histone H3K4 di-methylation activity by MLL/SET complexes · Zhao X, Su J, Wang F, Liu D, Ding J, Yang Y, Conaway JW, Conaway RC, Cao L, Wu D, Wu M, Cai Y, Jin J · PLoS Genetics 9(11):e1003940 2013What this evidence cannot show
Abstract fetched from PubMed and read on 2026-09-19; full text not read, identifiers copied from that fetched record. **The exemplar locus is ANKRD2, a skeletal-muscle gene, in human cell lines - not a HOXA locus and not a haematopoietic system.** Knockdown rather than acute degradation. The epistasis claim rests on an asymmetry between two knockdowns at one promoter.
Evidence that argues against this (1)
- 'NSL upstream of MLL/SET' and 'NSL and MLL/SET compete for a limiting shared component' predict very similar knockdown asymmetries, and this experiment does not separate them. The structural result favours competition; the wording of this paper favours sharing; the graph holds both and asserts neither. Dias J et al. 2014
- E5strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
An RNAi screen for druggable chromatin regulators in MLL-rearranged leukaemia identified the H4K16 acetyltransferase MOF; conditional deletion of Mof in a mouse model of MLL-AF9-driven leukaemogenesis reduced tumour burden and prolonged survival; rescue with catalytically inactive MOF mutants showed enzymatic activity is required; a small-molecule MYST inhibitor blocked growth of murine and human MLL-AF9 lines; and Mof inactivation also suppressed leukaemia development in an NUP98-HOXA9-driven AML model.
Histone Acetyltransferase Activity of MOF Is Required for MLL-AF9 Leukemogenesis · Valerio DG, Xu H, Chen CW, Hoshii T, Eisold ME, Delaney C, Cusan M, Deshpande AJ, Huang CH, Lujambio A, Zheng YG, Zuber J, Pandita TK, Lowe SW, Armstrong SA · Cancer Research 77(7):1753-1762 2017What this evidence cannot show
Abstract fetched from PubMed and read on 2026-09-19; full text not read, identifiers copied from that fetched record. MLL-AF9, not KMT2A::AFF1, and murine models throughout. **MOF is the catalytic subunit of both MSL and NSL**, so a Mof deletion phenotype is not attributable to NSL specifically - the experiment removes both complexes' enzyme at once. The MYST inhibitor is a tool compound of unstated selectivity in what was read.
Evidence that argues against this (1)
- Because MOF sits in both MSL and NSL, and because MSL-MOF and NSL-MOF write different marks, this result cannot be assigned to the NSL complex. An NSL-specific test would require deleting a KANSL subunit rather than MOF, and that experiment does not appear in this literature. Cai Y et al. 2010
- E6moderate confidence 3 of 5: One strong study, or several smaller consistent ones.database
In the genome-wide KMT2A-rearranged selectivity scan of 17,728 genes, seven of the eight NSL-associated genes are anti-selective: KANSL2 +0.2265 (rank 17,631), KANSL1 +0.0685 (15,913), KANSL3 +0.0371 (13,630), PHF20 +0.0339 (13,286), MCRS1 +0.0139 (10,971), KAT8 +0.0041 (9,683), and WDR5 +0.1984 (17,587); only OGT is negative at -0.0829 (rank 1,634). Pan-essentialities: KAT8 -1.4515, OGT -1.4669, KANSL3 -1.1827, KANSL2 -0.9807, KANSL1 -0.7402, MCRS1 -0.6226, PHF20 -0.0802. On the lineage split none is a KMT2A dependency in either half.
Genome-wide KMT2A-rearranged selectivity scan, 17,728 genes (DepMap Chronos via Open Targets) · Broad Institute DepMap / Open Targets Platform; analyses/depmap-kmt2a-selective-scan-full 2026What this evidence cannot show
Rows read out of analyses/depmap-kmt2a-selective-scan-full/results.json and analyses/depmap-interactome5-lineage-split/results.json on 2026-09-19 in scripted passes; no number typed from recall. Eleven KMT2A-rearranged against 43 other leukaemia lines, AML-heavy, none infant-derived. **Several of these genes are near the essentiality floor where the difference statistic compresses.** HCF1/HCFC1, the eighth roster member, is covered by its own existing entry and is likewise common-essential.
Evidence that argues against this (1)
- A selectivity statistic is structurally incapable of detecting a requirement that KMT2A-rearranged and other leukaemia lines share, and the published MOF result says the requirement IS shared across fusion families. A null here is therefore not evidence that NSL is irrelevant to leukaemia - only that it is not differentially relevant to KMT2A-rearranged leukaemia. Valerio DG et al. 2017
- E7moderate confidence 3 of 5: One strong study, or several smaller consistent ones.database
As of 2026-09-19 PubMed returns 4 records for '(MLL OR KMT2A) AND ("NSL complex" OR "non-specific lethal")' - all four relevant on title-check: the NSL-MLL/SET crosstalk study, the NUP98 fusion study, the KANSL1/WDR5/KANSL2 structure, and the NSL composition study - 4 for the KANSL1 alias query of which 2 survive title-check, 1 for KANSL2, 0 for KANSL3, 11 for '(MLL OR KMT2A) AND (KAT8 OR MOF OR MYST1 OR hMOF)', and 1 for 'NUP98 AND ("NSL complex" OR KANSL1 OR MOF OR KAT8)'.
PubMed E-utilities esearch record counts for the NSL complex, its subunits and NUP98 against KMT2A, MLL and leukaemia, run for interactome round 5 · NCBI PubMed, queried 2026-09-19 2026What this evidence cannot show
Counts from the esearch `count` field, queried 2026-09-19, with every record in each counted query title-checked individually. **The decisive methodological point: the NUP98 paper does not appear in any subunit-symbol query and is reachable only through the complex-name phrases.** Symbol and phrase set: KANSL1, KANSL2, KANSL3, KIAA1267, NSL1, NSL2, NSL3, MSL1v1, C12orf41, KIAA1310, KAT8, MOF, MYST1, hMOF, 'NSL complex', 'non-specific lethal'.
Evidence that argues against this (1)
- The alias MOF is an ordinary word fragment and NSL1 collides with an unrelated kinetochore gene, so the 11-record MOF count and the 4-record KANSL1 count are both upper bounds on relevance; conversely a subunit-only enumeration would have reported this complex as having no leukaemia literature, which would have been badly wrong. Both directions of the alias rule bit on this complex. Xu H et al. 2016
Scaffold subunit of the nine-subunit NSL complex.
NSL complex subunit.
KANSL1 is the NSL scaffold.
KANSL2 is an NSL subunit contributing a WDR5-binding motif.
KAT8/MOF is the catalytic subunit; complex identity sets whether it writes H4K16ac alone or also H4K5ac and H4K8ac.
WDR5 is required for NSL assembly and promoter targeting, and its NSL and MLL/COMPASS interactions are mutually exclusive.
- part of →HCFC1 (HCF-1) — a founding member of the wild-type KMT2A complex that the fusion cannot bind
HCF1 is reported as a component of both NSL and MLL/SET complexes.
Five NUP98 fusions interact with both MLL1 and NSL, and MLL1 is required for NUP98-HOXA9 proliferation.
NUP98-NSD1 is one of five NUP98 fusions shown to interact physically with MLL1 and the NSL complex.
Supports the general menin-MLL1 thesis - a second fusion family converges on the same wild-type MLL1 node - while weakening the KMT2A-specific form of it.
One side of the proposed partition.
The opposite finding: four directly relevant papers, reachable only through complex-name phrase queries.
NSL-bound KAT8 sits at proximal promoters and TSS and is the essential pool
listed NSL-complex subunit