CHUK (IKK-alpha) — the only kinase that sits in BOTH NF-kB arms, which makes it the one subunit whose inhibition has an ambiguous sign in this disease

This is a prediction, not a measurement. No one has seen this protein in this shape. The colours say how much the software trusts itself, and that is worth reading before anything else — the orange stretches are places the model is effectively saying this part has no fixed shape.

• Inhibiting IKKa blocks the arm that suppresses this leukaemia at the same time as the arm that supports it. A therapeutic proposal built on 'inhibit IKK' has to say what it expects the non-canonical arm to do and has no data with which to say it.
• IKKa loss impairs skin and immune-organ development in mouse genetics. In an infant, targeting a kinase with developmental roles is a different risk from targeting one in an adult.
• The dependency evidence for this specific subunit is a qualitative comparison inside one screen. That is thin ground for a subunit-level therapeutic claim.

1. Which arm dominates when IKKa is lost in an MLL-rearranged cell? A single Ikka knockdown with p100 processing, RelB localisation and an MLL-fusion ChIP read together would answer it.
2. Are the anti-leukaemic effects reported for IKK inhibitors III/IV/VII achieved despite non-canonical blockade, or partly because of blocking something else those compounds hit?
3. Does IKKa have a chromatin role here independent of NF-kB? IKKa has documented nuclear and histone H3S10-kinase activity in other systems, none of it tested in KMT2A-rearranged leukaemia.
4. Would a NIK-arm-sparing IKK inhibitor - IKKb-selective or a degrader - be more effective than the pan compounds used in 2013?

1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

   IKKa was among 33 phosphoregulator candidates whose knockdown impaired growth of MLL-oncogene-transformed myeloid progenitors by 50% or more; genetic reduction of Ikka reduced growth and clonogenicity in MLL-oncogene-transduced and MLL AML cells but not in E2A-PBX1-transduced cells or normal haematopoietic progenitors, with Ikka and Ikkg contributing to a lesser extent than Ikkb. The paper describes the canonical pathway as one in which the IKKa/b/g complex phosphorylates IkBa and triggers its degradation, and its pharmacology used non-subunit-selective IKK inhibitors III, IV and VII.

   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 2013
   PMID 24054986DOI
   ▸What this evidence cannot show

   Mouse myeloid only. The Ikka result is qualitative relative to Ikkb ('to a lesser extent') and no effect size is given in the text. No occupancy, protein-level or non-canonical readout was taken alongside the Ikka knockdown. Full text read this session (PMC3816582).

   ▸Evidence that argues against this (1)

   • The same kinase drives the arm that suppresses this leukaemia. In NIK-stabilised MLL-AF9 cells, increased IKKa phosphorylation accompanies p100-to-p52 processing, RelB upregulation, and delayed leukaemogenesis - so inhibiting IKKa would remove a protective output as well as a permissive one. Xiu Y et al. 2018
2. 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 in turn phosphorylates p100 and induces its processing to p52, facilitating formation and nuclear translocation of RelB/p52 complexes. In NIK-stabilised MLL-AF9 leukaemia cells, NIK protein rose about two-fold with increased IKKa phosphorylation, RelB upregulation and activated p100-to-p52 processing, increased nuclear RelB and p52, and markedly reduced RelA expression and nuclear localisation; the net effect was suppression of MLL-AF9-induced AML.

   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 2018
   PMID 29320732DOI
   ▸What this evidence cannot show

   Mouse MLL-AF9 model throughout, using a non-degradable NIK mutant lacking the TRAF3-binding domain rather than a drug. IKKa is measured as a phospho-western readout of NIK activity, not manipulated; no IKKa loss-of-function experiment appears in either paper. Full text read this session via Europe PMC fullTextXML (PMC5810947).

   ▸Evidence that argues against this (1)

   • That NIK phosphorylates IKKa and drives p100 processing is canonical pathway biology established outside this disease. Its therapeutic direction in MLL-rearranged leukaemia is the opposite of what an IKKa-inhibition rationale would predict: STABILISING NIK, and therefore increasing this very signalling, SUPPRESSES MLL-AF9 AML. Xiu Y et al. 2018
3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database

   Counted absence. (KMT2A OR MLL OR "mixed-lineage leukemia") AND (CHUK OR "IKK-alpha"[tiab] OR "IKKalpha"[tiab] OR "IKK alpha"[tiab]) returns 4 PubMed records; title-checking removes all four (a trained-immunity review, a circRNA network paper in MLL/AF4 ALL, a hybrid grouper nutrition study, and a monozygotic-twin infant ALL concordance study). Title-checked count: 0. CHUK has never been the subject of a KMT2A or MLL paper.

   PubMed E-utilities esearch, db=pubmed, run 2026-09-20 · NCBI · PubMed 2026
   ▸What this evidence cannot show

   querytranslation was read; PubMed expanded CHUK to 'chuk protein human' and the IKK-alpha variants to a title/abstract phrase, so the count covers the current symbol and the common spellings. The retired alias IKK1 was not searched separately and could add older records. The founding 2013 paper does not appear in this query because it names IKKa only in its body text, which is itself a reason to distrust a symbol-keyed absence count for a subunit named only inside a figure.

   ▸Evidence that argues against this (1)

   • A zero at the symbol level is not a zero at the pathway level, and on this beat that is demonstrable rather than hypothetical: PMID 29320732 studies the non-canonical arm directly in MLL-AF9 leukaemia and is unreachable from any of these subunit symbol queries, because it names the kinase and the pathway rather than the subunit. The count is honest about what was searched and should not be read as evidence that the biology is unstudied - only that it is not indexed under this symbol. Xiu Y et al. 2018
4. E_q93_chuk_loc1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

   UniProt O15111 carries Cytoplasm (ECO:0000269 PubMed:12789342, PubMed:30341167) and Nucleus (ECO:0000269 PubMed:12789342) with the note 'Shuttles between the cytoplasm and the nucleus'; both values are experimentally coded, which makes it the best-evidenced compartment record among this graph's three NF-kB kinase and subunit entries.

   UniProtKB O15111 (IKKA_HUMAN), SUBCELLULAR LOCATION comment with evidence codes, sources and note, plus TISSUE SPECIFICITY, fetched live 2026-09-21 · UniProt Consortium · Databases 2026
   ▸What this evidence cannot show

   A statement about the record. The ranking claim is comparative within this graph and is exactly what it says: CHUK has two coded values from a paper about its nuclear function; NFKB1's p50 values are coded but rest on a crystal structure; MAP3K14's single value has an empty evidence list. That is a ranking of citation quality, not of how well any of the three compartments is actually quantified - none of the three has a nuclear-to-cytoplasmic ratio reported in a haematopoietic cell.

   ▸Evidence that argues against this (1)

   • The second cytoplasmic attribution weakens the record more than the first strengthens it. PubMed:30341167 is 'DDX3 directly facilitates IKKalpha activation and regulates downstream signalling pathways' - an innate-immune signalling study whose subject is a helicase enhancing kinase activation in the TLR7/9 interferon pathway, not where IKK-alpha is. So one of O15111's two cytoplasmic citations is a rule-91 instance of the ordinary kind: a signalling paper cited for a compartment. The annotation survives on its first citation alone. Fullam A et al. 2018
5. E_q93_chuk_loc2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

   IKK-alpha functions in the NUCLEUS in a manner independent of its cytoplasmic role in IkappaBalpha phosphorylation: it is recruited to NF-kB-responsive promoters and phosphorylates histone H3 at serine 10, and this nuclear function is critical for cytokine-induced gene expression - the question being raised because IKK-alpha-deficient mouse embryo fibroblasts are defective in NF-kB-dependent transcription while IKK-beta rather than IKK-alpha is the kinase critical for cytokine-induced IkappaB degradation.

   Histone H3 phosphorylation by IKK-alpha is critical for cytokine-induced gene expression. · Yamamoto Y, Verma UN, Prajapati S, Kwak YT, Gaynor RB · Nature 2003
   PMID 12789342DOI
   ▸What this evidence cannot show

   Mouse embryo fibroblasts and cytokine-stimulated cell lines, not haematopoietic and not fusion-bearing (brief rule 73). Promoter recruitment measured by chromatin immunoprecipitation is locus OCCUPANCY, which is a genomic address and not a subnuclear compartment - this entry says which it is rather than converting one into the other. Bounded by the abstract.

   ▸Evidence that argues against this (1)

   • The relevance of this result to this graph rests on an inference the paper does not make, and it should be stated as an inference. H3S10 phosphorylation recruiting BRD4 and P-TEFb is the mechanism by which a KINASE reaches the same elongation machinery the KMT2A fusion reaches through the SEC - but a shared downstream factor is a shared address, not a shared mechanism, and the same logic was already used and correctly hedged for SGF29 in tranche 4 ('both act at H3K4me3-marked promoters, which is a shared address rather than a shared mechanism'). Nothing here shows IKK-alpha acts at a fusion target locus, and nothing shows the fusion needs it. Thorne JL et al. 2012
6. E_q93_chuk_loc3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   In ACTIVATED MACROPHAGES, IKK-alpha travels with the ELONGATING form of RNA polymerase II together with heterochromatin protein 1 gamma at NF-kB-dependent genes; IKK-alpha binds and phosphorylates HP1gamma, HP1gamma controls IKK-alpha's binding to chromatin and its phosphorylation of histone variant H3.3 at serine 31 within transcribing regions, and downstream of transcription end sites IKK-alpha accumulates with its inhibitor CUEDC2, which the authors read as linking IKK-alpha inactivation to transcription termination; the same work records that promoter H3 serine-10 phosphorylation by IKK-alpha triggers recruitment of BRD4 and P-TEFb.

   Heterochromatin protein 1 gamma and IκB kinase alpha interdependence during tumour necrosis factor gene transcription elongation in activated macrophages. · Thorne JL, Ouboussad L, Lefevre PF · Nucleic Acids Res 2012
   PMID 22649058DOI
   ▸What this evidence cannot show

   Activated macrophages - MYELOID and inflammatory, not lymphoid and not leukaemic (brief rule 73, and the claim names the system first). This is a gene-body occupancy and travelling-complex result, so it is a genomic address along transcription units rather than a subnuclear compartment. The BRD4 and P-TEFb recruitment statement is background in this paper carried from the 2003 work rather than a new measurement here. Bounded by the abstract.

   ▸Evidence that argues against this (1)

   • The HP1gamma interdependence cuts against reading this as a clean 'IKK-alpha is an elongation factor' result, and the direction is the interesting part. HP1 proteins are canonical heterochromatin readers, and this paper has one of them travelling with elongating Pol II and CONTROLLING a kinase's chromatin access at active genes - a role opposite to the silencing function the protein is named for. A second, weaker report in liver cancer stem cells also couples IKK subunits to an HP1-HOTAIR axis, so the pairing recurs, but neither paper is in a lymphoid or leukaemic cell and the mechanism has not been reproduced outside inflammatory gene induction. For this graph the honest status is a well-specified hypothesis about a second route to P-TEFb, not an established parallel mechanism. Yamamoto Y et al. 2003
7. E_q93_chuk_loc4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database

   Bounded 2026-09-21: the CHUK gene clause over CHUK, IKKalpha, IKK-alpha and IKK1 returns 3,045 records; against a twelve-term subnuclear clause it returns 4, of which 2 are the HP1gamma papers and 2 are off topic; against a nine-term KMT2A/MLL fusion clause it returns 2, neither a localisation measurement; and against a localisation clause AND the fusion clause it returns 0.

   NCBI E-utilities esearch db=pubmed; querytranslation read back and stored beside every count · NCBI E-utilities · Databases 2026
   ▸What this evidence cannot show

   Titles and abstracts only; supporting figures not searched (brief rule 80). The zero is a LOCALISATION absence and not an absence of IKK-alpha in this disease: brief rule 64's own worked example is this gene, where PMID 24054986 knocks down Ikka and is unreachable from the symbol because its abstract frames everything at IKK/NF-kB, and this graph already cites that paper elsewhere.

   ▸Evidence that argues against this (1)

   • Rule 64 was written from this exact gene, so the counted absence here has to carry its own disclaimer, and the disclaimer is load-bearing rather than formal. The complex-level literature contains the single most relevant result available: IKK/NF-kB signalling is required for retention of both wild-type and fusion MLL protein at target promoters with the associated histone modifications, in murine MLL-AF9 and human MLL-rearranged cells - a chromatin-retention phenotype that this graph holds under gene-rela. So IKK activity already has a measured consequence for where the FUSION sits, while IKK-alpha's own position has never been measured. That is the inverse of the usual gap in this beat and it is the more tractable half. Thorne JL et al. 2012

• ← supportsAbsence claims: 93% are unbounded in prose, but the evidence claims are 71% bounded - and the query that tested them was itself broken

  TESTED, NOT REFUTED: 86 free-text co-occurrences, none naming CHUK in title or abstract alongside KMT2A, on the first page examined 2026-09-20.

• part of →The IKK complex — the unit the eviction experiment actually resolves, because the chemistry that produced the result cannot see the subunits

  catalytic alpha subunit; also the NIK substrate of the non-canonical arm

• depends on →KMT2A-rearranged acute myeloid leukaemia in children and infants

  Ikka shRNA reduces growth of MLL-transformed progenitors, to a lesser extent than Ikkb

• ← regulatesMAP3K14 (NIK) — the one node in this pathway where ACTIVATION, not inhibition, is the proposed therapy, and the one that also damages normal stem cells

  NIK phosphorylates IKKa to drive p100 processing

• part of →Canonical versus non-canonical NF-kB in KMT2A-rearranged leukaemia — the two arms have opposite signs, and the two papers usually read as contradicting each other actually agree

  the shared component whose dual membership creates the sign problem

• ← related toNFKB1 (p105/p50) — the only menin-NF-kB subunit with a co-immunoprecipitation, and the arm the MLL leukaemia data points at

  Same pathway, opposite compartment problem, and the three NF-kB entries must not disagree: p50's nucleus rests on a crystal structure (well-founded biology, wrong instrument cited), IKK-alpha's nucleus rests on a real chromatin paper (PMID 12789342), and NIK's cytoplasm rests on nothing at all (Q89). Ranked by evidence quality that is CHUK > NFKB1 > MAP3K14 (Q93 2026-09-21)

• related to →Transcription factories and Pol II clusters - no KMT2A::AFF1 imaging, condensate evidence for SEC and MLL-ENL, and 3D-FISH that actively contradicts the factory story for t(4;11)

  IKK-alpha TRAVELS WITH ELONGATING POL II together with HP1gamma at NF-kB-dependent genes in activated macrophages, phosphorylating H3.3S31 within transcribing regions, and accumulates with its inhibitor CUEDC2 downstream of transcription end sites (PMID 22649058). A genomic address with a coordinate along the gene, not a nuclear body (Q93 2026-09-21)

• related to →BRD4 — the best-drugged protein in this beat, and the one whose clinical result is already in and disappointing

  A SECOND, KINASE-DRIVEN ROUTE TO THE SAME FACTORS. IKK-alpha phosphorylates H3 serine 10 at NF-kB-responsive promoters and that mark triggers recruitment of BRD4 and P-TEFb (PMIDs 12789342, 22649058) - the two factors the KMT2A fusion reaches through the SEC. Nobody has put IKK-alpha and a KMT2A fusion in the same chromatin readout (Q93 2026-09-21)

• related to →P-TEFb (CDK9 + cyclin T1) — the deepest dependency and the clearest therapeutic-index wall in the interactome

  H3S10ph written by IKK-alpha recruits P-TEFb at induced promoters (PMID 12789342), which is an NF-kB-pathway route to the elongation machinery independent of the AFF1/AFF4 scaffold this graph is built on. Shown in fibroblasts and macrophages; never tested in a fusion cell (Q93 2026-09-21)

• related to →MAP3K14 (NIK) — the one node in this pathway where ACTIVATION, not inhibition, is the proposed therapy, and the one that also damages normal stem cells

  The two-arm membership has a compartment consequence the graph should not blur: NIK's own cytoplasmic call is UNSOURCED and the nuclear step of the noncanonical arm belongs to RelB:p52, not to NIK (Q89) - whereas IKK-alpha genuinely shuttles, with a coded citation that is about its nuclear function (PMID 12789342) (Q93 2026-09-21)

• related to →The soluble-nucleoplasm versus chromatin-bound axis - the one localisation readout that keeps a size dimension, and nobody has run it on a KMT2A fusion

  IKK activity already has a measured consequence for where the FUSION sits - IKK/NF-kB signalling is required for retention of wild-type and fusion MLL at target promoters - while IKK-alpha's own chromatin position has never been measured in a leukaemic cell. The inverse of this beat's usual gap (Q93 2026-09-21)