The IKK complex — the unit the eviction experiment actually resolves, because the chemistry that produced the result cannot see the subunits
also called IKK complex · IKK signalosome · IKKa/IKKb/NEMO · IkB kinase complex
The three IKK proteins work as one machine, and the drugs used to show that MLL leukaemia needs them block the machine rather than any one part. So when a paper says a subunit is required, check whether the experiment could have told the parts apart.
The same thing, for a scientist
This entry exists for one reason and it is a measurement reason, not a roster reason. **The result that makes this beat matter — IKK inhibition evicting wild-type and fusion MLL from Hoxa9 and Meis1 — was produced with IKK inhibitors III, IV and VII (EMD 401480, 401481, 401486), compounds directed at the complex and not at a subunit** [E1]. The genetic arm of the same paper, which did address subunits with shRNAs against Ikka, Ikkb and Ikkg, measured only growth and clonogenicity and never occupancy [E1]. So the correct resolving unit for the chromatin claim is the complex; the correct resolving unit for the growth claim is the subunit; and the two are not interchangeable. The complex is the canonical IKK signalosome: catalytic **IKKa (CHUK, O15111)** and **IKKb (IKBKB, O14920)** with the non-catalytic ubiquitin-binding scaffold **NEMO (IKBKG, Q9Y6K9)**, which phosphorylates IkBa to trigger its degradation and release RELA:p50 [E1]. **The asymmetry that matters:** IKKa also functions outside this complex, as the NIK substrate that processes p100 to p52 in the non-canonical arm, an arm whose activation *suppresses* MLL-AF9 leukaemia [E2]. A pan-IKK inhibitor is therefore not a canonical-pathway drug — it is a drug that blocks a permissive arm and a protective arm simultaneously, and nothing in this literature separates them. **This is the same failure the brief's Rule 63 records for CK2**, where CX-4945, TBB and DMAT are all pan-catalytic-subunit, so a CSNK2A1-versus-CSNK2A2 attribution is naming rather than measurement.
Subunits, and the one experiment each of them has
| subunit | id | UniProt | length | arm | disease evidence |
|---|---|---|---|---|---|
| IKKa | gene-chuk | O15111 | 745 aa | canonical and non-canonical | shRNA, growth, 'lesser extent' than Ikkb 1 |
| IKKb | gene-ikbkb | O14920 | 756 aa | canonical | shRNA, growth, correlated with knockdown level 1; KMT2A::MLLT3 in vivo via a dual degrader |
| NEMO | gene-ikbkg | Q9Y6K9 | 419 aa | canonical only | shRNA, growth, 'lesser extent' 1; zero symbol-keyed records |
All three accessions were resolved this session by gene_exact queries with gene_primary requested, every row read, and only the primary match kept.
What a complex-level entry is allowed to claim
It may claim what the complex-level perturbation showed: IKK activity is required for MLL-fusion and wild-type MLL promoter occupancy, for H3K4me3 and H3K79me2 at those promoters, for proliferation, for survival, for differentiation arrest, and for LSC potential, with at least a ten-fold drop in LSC frequency after two days of pretreatment 1.
It may not claim which subunit does any of that, and no entry in this folder should. The one thing subunit chemistry could in principle establish — that an IKKb-selective agent reproduces the eviction — has not been tested; the only subunit-selective in vivo agent in this disease is a dual degrader of IKKb and NR4A1 whose effect cannot be attributed to either target alone 3.
| stoichiometry note | Canonical signalosome; exact stoichiometry not asserted here because no structure of the intact human complex was checked for this entry. |
|---|---|
| resolving unit | complex for the chromatin-occupancy claim; subunit for the growth claim only |
complex members
[ "gene-chuk", "gene-ikbkb", "gene-ikbkg" ]
uniprot members
[ "O15111", "O14920", "Q9Y6K9" ]
perturbing agents used in disease
[ "IKK inhibitor III (EMD 401480)", "IKK inhibitor IV (EMD 401481)", "IKK inhibitor VII (EMD 401486)" ]
- A complex-level intervention cannot avoid the non-canonical arm, because IKKa belongs to both. Any therapeutic proposal at this node inherits an unmeasured opposing effect.
- Three research-grade inhibitors agreeing is not the same as target engagement being demonstrated. If the shared activity is an off-target one, every downstream conclusion in this beat moves with it.
- Writing subunit entries around a complex-level result risks manufacturing three claims from one measurement. That is why this entry exists and why each subunit entry states what its own evidence covers.
- Does a subunit-selective agent reproduce the eviction of MLL from chromatin, or does eviction require loss of complex integrity? A NEMO-directed and an IKKb-selective agent read out by ChIP would settle it.
- Are IKK inhibitors III, IV and VII engaging IKK in these cells at the concentrations used? No target-engagement assay is reported, and the working concentrations differ by nearly three orders of magnitude across experiments.
- Is there a kinase-independent scaffolding role for the complex in MLL retention? All the pharmacology is catalytic; a kinase-dead knock-in would separate the two.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
MLL leukaemia cells are preferentially sensitive to three different IKK inhibitors (III, IV and VII; EMD 401480, 401481, 401486) relative to non-MLL leukaemia cells; these compounds target the IKK complex, a major upstream regulator of NF-kB, in which the IKKa/b/g complex phosphorylates IkBa and triggers its degradation. IKK inhibition reduced proliferation, induced apoptosis, reversed the differentiation block, reduced colony-forming cells and gave at least a ten-fold reduction in LSC frequency on limit-dilution secondary transplantation, and reduced promoter occupancy of MLL-AF10, wild-type MLL, RELA, H3K79me2 and H3K4me3 at Hoxa9 and Meis1. Subunit-level perturbation in the same paper was by shRNA against Ikka, Ikkb and Ikkg, with growth and clonogenicity as the only read-outs.
Epigenetic roles of MLL oncoproteins are dependent on NF-κB. · Kuo HP, Wang Z, Lee DF, Iwasaki M, Duque-Afonso J, Wong SHK, Lin CH, Figueroa ME, Su J, Lemischka IR, Cleary ML · Cancer Cell 2013What this evidence cannot show
IKK inhibitors III, IV and VII are research reagents with published off-target liabilities that this entry has not enumerated; three compounds agreeing constrains but does not eliminate an off-target explanation. Concentrations used span a wide range, including 2 nM for inhibitor IV in the differentiation experiment and 1 uM for inhibitor VII in phosphoflow, and no on-target engagement assay is reported. All systems are mouse myeloid or human AML lines. Full text read this session (PMC3816582).
Evidence that argues against this (1)
- Pan-IKK inhibition also blocks the non-canonical arm's kinase, and that arm suppresses this leukaemia: NIK stabilisation raises IKKa phosphorylation, p100 processing and RelB, and delays MLL-AF9 AML. So a complex-level inhibitor's net effect mixes two opposing outputs. Xiu Y et al. 2018
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
NIK activates non-canonical NF-kB signalling by directly phosphorylating IKKa, which phosphorylates p100 and induces processing to p52 and nuclear translocation of RelB/p52 - a route that does not require NEMO and that, when activated in MLL-AF9 AML, suppresses the leukaemia.
Stabilization of NF-κB-Inducing Kinase Suppresses MLL-AF9-Induced Acute Myeloid Leukemia. · Xiu Y, Dong Q, Li Q, Li F, Borcherding N, Zhang W, Boyce B, Xue HH, Zhao C · Cell Reports 2018What this evidence cannot show
The NEMO-independence of the non-canonical route is standard pathway biology stated in this paper's introduction rather than a result it measures. The measured part is that NIK stabilisation raises phospho-IKKa, RelB and p52 and suppresses MLL-AF9 AML in mice. Full text read this session (PMC5810947).
Evidence that argues against this (1)
- A NEMO-independent route is a statement about the non-canonical arm's wiring in general signalling biology, established outside this disease. Its therapeutic direction here is also inverted from what an IKK-inhibition rationale predicts, since driving this very route by stabilising NIK suppresses MLL-AF9 AML. Zhao J et al. 2021
- E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
The only subunit-directed agent tested in vivo in a KMT2A-rearranged model is compound A9, a celastrol-based cereblon PROTAC that degrades both IKKb and NR4A1 and suppresses leukaemia progression in a KMT2A::MLLT3 AML mouse model without inducing neutrophilia; the authors describe the two targets as playing non-redundant oncogenic roles, so the effect is not attributable to IKKb degradation alone.
PROTAC-mediated targeting of IKKβ and NR4A1 for AML therapy. · Maharjan CK, Liu Y, ... Zhang W · Oncogene 2026What this evidence cannot show
Abstract read; full text not retrieved. Celastrol is a promiscuous electrophile and the PROTAC is built on it, so proteome-wide effects are not excluded. No chromatin or MLL-occupancy read-out is reported in the abstract, so this does not test whether subunit-selective degradation reproduces the eviction phenotype.
Evidence that argues against this (1)
- Compound A9 degrades IKKb AND NR4A1, so no effect it produces can be attributed to IKKb. NR4A1 is an orphan nuclear receptor with its own substantial role in myeloid leukaemia, and a dual degrader is the weakest possible tool for a subunit-attribution claim - the same naming-versus-measurement problem this graph logged for CX-4945 and GSK-J4. Xiu Y et al. 2018
catalytic beta subunit of the canonical IKK complex
catalytic alpha subunit; also the NIK substrate of the non-canonical arm
NEMO, the non-catalytic ubiquitin-binding scaffold; canonical arm only
phosphorylates IkB-alpha and triggers its proteasomal degradation
IKK inhibitors III/IV/VII are the perturbation that produced the eviction result
IKKa belongs to both arms, so pan-IKK chemistry cannot separate them
canonical ligand input; but the MLL paper demonstrated pathway responsiveness with LPS, not TNF
TLR/IL-1R to MyD88/IRAK is the input layer the screen pointed at, upstream of IKK
the enumeration's method finding: subunit-keyed absence counts under-report complex-level literature
IDENTIFIER AND MECHANISM TRAP: ELP1 was discovered as IKAP, reported as the IKK scaffold (PMID 9751059), and UniProt still carries that clause at ECO:0000269 and the name in its alternative-names field