Oncofusion

start / graph / mechanism-il1-irak-axis-in-leukaemia-stem-cells

Mechanismmoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q38-nfkb-signalling

IL-1/IRAK1/4 — the receptor-proximal input the MLL screen actually pointed at, with an Il1r1 knockout in MLL-AF9 mice and an inhibitor that spares normal stem cells

also called IL-1R/IRAK axis · IRAK1/4 in LSCs · the TLR/IL-1R input to NF-kB in MLL leukaemia

The kinase screen that first linked NF-kB to MLL leukaemia did not find a receptor; it found IRAK3, a relay in the interleukin-1 and Toll-like receptor branch. Later work in AML stem cells shows that deleting the interleukin-1 receptor in MLL-AF9 mice reduces the leukaemia stem cell pool and extends survival, and that a drug against the IRAK kinases kills leukaemia stem cells while leaving normal ones alone.

The same thing, for a scientist

**This is the only upstream receptor layer in this beat with a disease-matched genetic experiment**, which is why it gets an entry and TNFRSF1A, MYD88 and TRAF6 do not. **What the original screen pointed at.** Among the 33 phosphoregulator hits in MLL leukaemia were **IRAK3** and **PPP4C** alongside the three IKK subunits, PANTHER analysis showed enrichment of the candidate kinases and phosphatases in the **Toll-like receptor signalling pathway**, and Toll-like pathway transcripts 'that mediate p105-dependent NF-kB activation' were enriched in MLL-AF1p and MLL-AF10 leukaemia cells relative to normal bone marrow — though not significantly in MLL-AF9 [E1]. The paper then demonstrated pathway responsiveness with **LPS**, a TLR4 ligand, not with TNF [E1]. So the input arm this literature implicates is **TLR/IL-1R -> MyD88/IRAK -> p105/IKK**, and the screen hit is an IRAK. **What the receptor evidence is.** IL1R1 and IL1RAP are upregulated in leukaemia stem and progenitor cells at diagnosis and at relapse; knockdown of either reduces clonogenicity or engraftment of primary human AML cells; and **in leukaemic MLL-AF9 mice, Il1r1 knockout reduced LSC frequency and extended survival** [E2]. A new IRAK1/4 inhibitor, **UR241-2**, suppressed IL-1/IRAK1/4 signalling including NF-kB activation and p65 phosphorylation, selectively inhibited LSC clonogenicity in primary human AML at diagnosis and relapse **while sparing normal HSPCs**, and reduced engraftment in mice [E2]. **The caveat that keeps this honest:** the receptor work is AML-wide with one MLL-AF9 experiment inside it, and IRAK3 — the actual screen hit — is a *negative* regulator of IRAK signalling, so the screen's direction and the inhibitor's direction are not obviously the same.

Go deeper
The IRAK3 direction problem, stated rather than smoothed over

IRAK3 (IL-1 receptor-associated kinase 3, UniProt Q9Y616, 596 aa; also known as IRAK-M) is a pseudokinase and a negative regulator of TLR/IL-1R signalling in most settings. The founding MLL paper places it 'upstream on the pathway' and notes that its over-expression has been shown to activate NF-kB-mediated transcription, citing 1999 work 1. Knocking IRAK3 down impaired MLL leukaemia growth 1, which is consistent with the paper's reading and inconsistent with the textbook negative-regulator role.

This is not resolved anywhere in this literature, and it matters: if IRAK3 in these cells behaves as a negative regulator, then an IRAK1/4 inhibitor and an IRAK3 knockdown should have opposite effects, and both are reported as anti-leukaemic. The graph should carry that as an unexplained inconsistency, not fold it away.

Why MYD88, TRAF6, TNFRSF1A and IL1R1-as-an-entry were not written

Applying the brief's rule — a receptor with no KMT2A-rearranged evidence is a bounded absence, not an entry:

  • MYD88: 7 records against KMT2A/MLL, every one CLL, lymphoplasmacytic lymphoma or a mutation panel. Zero functional work in this disease 3. Bounded absence.
  • TRAF6: 3 records, none TRAF6-focused; the only on-topic one reaches TRAF6 as part of a TLR-TRAF6-NF-kB axis overactivated by IMPDH inhibition, which suppresses MLL-fusion AML 4. That is a real and interesting result — and it says more TLR-NF-kB signalling is anti-leukaemic, which is the opposite direction to the IKK-dependence story and belongs in the tension, not in a TRAF6 dossier.
  • TNFRSF1A: 2 records, neither KMT2A-specific.
  • IL1R1: covered here rather than as its own protein entry, because the single MLL-specific experiment is one knockout inside an AML-wide paper.

All queries and counts are in ana-nfkb-upstream-enumeration-and-bounded-absences.

Facts and figures
mechanism classreceptor-proximal-input
implicated layerTLR / IL-1R -> MyD88 / IRAK -> p105 / IKK
receptor evidenceIl1r1 knockout reduces LSC frequency and extends survival in MLL-AF9 mice; IL1R1 and IL1RAP upregulated in human AML LSCs at diagnosis and relapse
chemical matterUR241-2, an IRAK1/4 inhibitor sparing normal HSPCs
contradicting directionIMPDH inhibition OVERACTIVATES TLR-TRAF6-NF-kB and suppresses MLL-fusion AML
screen hit
{
  "symbol": "IRAK3",
  "uniprot": "Q9Y616",
  "length_aa": 596,
  "direction_note": "conventionally a negative regulator; the MLL paper places it as an upstream activator"
}
bounded absences here
[
  "MYD88 (0 title-checked)",
  "TRAF6 (0 TRAF6-focused)",
  "MAP3K7/TAK1 (0)",
  "TNFRSF1A (0 KMT2A-specific)"
]
How this could bite us back
  • IL-1 and TLR signalling are the innate immune system's alarm circuitry. Blocking IRAK1/4 in an infant on intensive chemotherapy is blocking the response to bacterial and fungal infection, the commonest cause of treatment-related death in this disease.
  • IRAK4 deficiency in humans causes recurrent invasive pyogenic infection in early childhood, which is the closest natural experiment available and points the wrong way for paediatric use.
  • The one MLL-specific genetic result is a germline Il1r1 knockout, which models lifelong absence rather than pharmacological blockade in an established leukaemia.
  • If the relationship between TLR-NF-kB activity and leukaemia is non-monotonic - as the IMPDH result suggests - then partial inhibition could land on the wrong side of the curve and accelerate disease.
Still open
  1. Does an IRAK1/4 inhibitor evict the MLL fusion from chromatin? It suppresses NF-kB activation and p65 phosphorylation, so it is the most tractable available agent for testing the retention mechanism upstream of IKK - and nobody has run that read-out.
  2. Is IRAK3 acting as an activator or a repressor in MLL leukaemia cells? The screen's direction and the inhibitor literature's direction cannot both be right without an explanation.
  3. Is IL1R1 or IL1RAP expressed on KMT2A::AFF1 B-ALL blasts? IL1RAP is already a CAR and antibody target in myeloid disease and nobody has checked the index lineage in this context.
  4. How can TLR-TRAF6-NF-kB overactivation be anti-leukaemic while IKK/NF-kB is required for fusion retention? A dose-response on the same axis, with occupancy as the read-out, would test whether the relationship is non-monotonic.
Every source behind this entry (4)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    Among the high-scoring candidates of the MLL leukaemia phosphoregulator screen, several have implicated roles in mediating NF-kB signalling, including the IKK complex, PLK1, PPP4C and IL-1R-associated kinase 3 (IRAK3); PANTHER analysis showed enrichment of the identified candidate kinases and phosphatases in the Toll-like receptor signalling pathway; and transcripts of Toll-like signalling pathway proteins that mediate p105-dependent NF-kB activation are enriched in MLL-AF1p and MLL-AF10 leukaemia cells compared to normal bone marrow, with the MLL-AF9 enrichment not achieving statistical significance. PPP4C and IRAK3 are described as functioning upstream on the pathway, their over-expression having been shown to activate NF-kB-mediated transcription. The demonstration of pathway responsiveness used LPS stimulation.

    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

    The PANTHER enrichment is reported as 'data not shown'. The transcript enrichment failed significance in MLL-AF9, the most widely used model in this field. IRAK3 is conventionally a negative regulator of IRAK signalling and the paper's placement of it as an upstream activator rests on a 1999 citation not re-fetched here. Mouse myeloid throughout. Full text read this session (PMC3816582).

    Evidence that argues against this (1)
    • A phosphoregulator screen returns what its library contains and what its readout scores; several high-scoring candidates having 'implicated roles' in NF-kB is a post-hoc annotation of a hit list, not evidence that the axis is the mechanism. IRAK3 in particular is a NEGATIVE regulator of IL-1R signalling, so its presence does not indicate axis activation. Liu X et al. 2023
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Transcriptome analyses of functionally defined leukaemia stem cell populations at diagnosis and relapse revealed significant alterations in IL-1 signalling, with IL1R1 and IL1RAP notably upregulated in leukaemia stem and progenitor cells at both time points; knockdown of IL1R1 and IL1RAP reduced clonogenicity and/or engraftment of primary human AML cells; and in leukaemic MLL-AF9 mice, Il1r1 knockout reduced LSC frequency and extended survival. The IRAK1/4 inhibitor UR241-2 suppressed IL-1/IRAK1/4 signalling including NF-kB activation and phosphorylation of p65 and p38 following IL-1 stimulation, selectively inhibited LSC clonogenicity in primary human AML cells at diagnosis and relapse while sparing normal haematopoietic stem and progenitor cells, and reduced AML engraftment in leukaemic mice.

    PMID 41419607DOI
    What this evidence cannot show

    Abstract read; full text not retrieved. This is an AML-wide study; the KMT2A-relevant content is one Il1r1 knockout experiment in MLL-AF9 mice, and the primary human AML samples are not reported by karyotype in the abstract. UR241-2 is a novel compound from the authors' own programme with no independent characterisation. A bioRxiv preprint of the same work exists (PMID 39605740, doi 10.1101/2024.11.09.622796). No lymphoid or KMT2A::AFF1 data.

    Evidence that argues against this (1)
    • The direction conflicts with the screen that opened this field. There the hit was IRAK3, conventionally a NEGATIVE regulator of IRAK signalling, and knocking it down impaired MLL leukaemia growth. An IRAK1/4 inhibitor and an IRAK3 knockdown should not both be anti-leukaemic, and no paper reconciles them. Kuo HP et al. 2013
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database

    Counted absence for MYD88. (KMT2A OR MLL OR "mixed-lineage leukemia") AND (MYD88 OR "MyD88"[tiab]) returns 7 PubMed records; title-checking removes all seven, which are chronic lymphocytic leukaemia prognostic and mutation-panel studies, a lymphoplasmacytic lymphoma comparison, and a histone-methylation TLR4 wound-healing study. Title-checked count in KMT2A-rearranged leukaemia: 0. Counted absence for MAP3K7/TAK1: the corresponding query returns 3 records and title-checking removes all three.

    What this evidence cannot show

    querytranslation read on both queries. MYD88's alias MYD88L265P was not searched; it would only add lymphoma records. TAK1's alias TGF-beta activated kinase 1 was not searched as a phrase, which is a real recall gap for older literature.

    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. E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    Acute myeloid leukaemias with MLL fusions are susceptible to IMPDH inhibitors in vitro, and alternate-day administration suppressed MLL-AF9-driven AML in vivo; mechanistically, IMPDH inhibition induced OVERACTIVATION of Toll-like receptor-TRAF6-NF-kB signalling and upregulation of VCAM1, which contribute to the antileukaemia effect, and combining an IMPDH inhibitor with a TLR1/2 agonist effectively inhibited MLL-fusion AML development.

    PMID 36453131DOI
    What this evidence cannot show

    Abstract read; full text not retrieved. This is the single most direct challenge to the simple 'NF-kB supports MLL leukaemia' reading in this folder: here, driving TLR-TRAF6-NF-kB signalling HARDER is anti-leukaemic, and a TLR agonist is used as a therapeutic partner. It is a different perturbation (overactivation versus baseline requirement) and a different downstream output (VCAM1 adhesion), so it is not a formal contradiction - but any entry asserting that NF-kB activity is monotonically pro-leukaemic in this disease has to answer it. Mouse MLL-AF9 and human MLL-fusion AML; no lymphoid data.

    Evidence that argues against this (1)
    • The IMPDH result reaches TLR signalling through nucleotide depletion, so it is at least as much a metabolic result as an innate-immune one, and reading it into an IL-1/IRAK axis selects one downstream branch of a pleiotropic perturbation. Mycophenolate's effects on lymphocytes are also well documented and independent of this axis. Kuo HP et al. 2013
Connected to