MYD88 (human, Q99836) review notes

Status of inputs

• UniProt, GOA (236 rows, 229 seeded annotations after de-duplication) and all 63 GOA PMIDs were
  cached before review. Two extra PMIDs were fetched to trace mouse-sourced IEA rows
  (PMID:18319258, PMID:31164412).
• Falcon deep research had not been produced when the review was drafted (MYD88 was queued behind
  the TLR batch); see the end of this file for its status.

Biology summary (with provenance)

• MYD88 is a cytosolic TIR-domain adaptor with an N-terminal death domain (DD), an intermediate
  domain and a C-terminal TIR domain. It couples IL-1 receptor family and TLR receptors to the IRAK
  kinases. PMID:9430229 PMID:9430229
• General adaptor for the Toll/IL-1R family. PMID:9734363
• Recruitment to TLR4 (and TLR2) needs the bridging adaptor TIRAP, which is targeted to PIP2 at the
  plasma membrane. PMID:16751103 PMID:17258210
• DD-mediated oligomerisation builds the Myddosome with IRAK4 and IRAK1/2. PMID:36865541 PMID:36865541 Endogenous MyD88 forms barrel-like scaffolds
  PMID:38961291
• IRAK4 variants in the DD lose MyD88 binding PMID:24316379
• Endosomal TLR7/8/9: MyD88–IRF7 complex drives IFN-alpha PMID:15361868; TLR9 phosphorylation is needed for MyD88 recruitment
  PMID:40980882; TLR8 in human macrophages PMID:33718825.
• IL-1 family receptors: IL-1R, IL-18R, IL-33R all use MyD88 PMID:36865541; IL-33-driven human/mouse T cell
  polarisation needs MyD88 PMID:18802081.
• Non-TIR receptor: TACI binds MyD88 PMID:20676093.
• Regulation by ubiquitin: OTUD5 PMID:38605168; USP3 PMID:37971847.
• Human genetics: MyD88 deficiency phenocopies IRAK-4 deficiency PMID:21057262.

Curation decisions and problems found

1. GO:0008063 Toll signaling pathway (IBA and rat-sourced IEA). GO:0008063 is defined for
   ligand binding to "the receptor Toll" and is NOT an ancestor of GO:0002224 toll-like receptor
   signaling pathway (checked in OLS). The IBA node PTN000386853 spans fly Myd88 and vertebrate
   MYD88; the ancestral adaptor role is sound, but for a vertebrate the pathway is the TLR/IL-1R
   pathway. MODIFY to GO:0002755. This is the project's Toll-vs-TLR conflation issue on an adaptor.
2. GO:0005121 Toll binding (rat IEA) — same issue; MODIFY to GO:0035325 Toll-like receptor
   binding.
3. ATP-dependent histone chaperone activity / chromatin organization / chromatin remodeling (IEA)
   — traced to mouse Myd88 IMP from PMID:18319258, which shows only that MyD88 signaling is required
   for nucleosome remodeling at secondary-response promoters ("MyD88 is required for nucleosome
   remodeling and histone H3K4 trimethylation at secondary response promoters"). This is an indirect
   downstream effect; MYD88 has no histone chaperone activity. REMOVE all three. (Source defect is
   on the mouse Myd88 MGI annotation; no mouse review exists in this repo.)
4. GO:0005123 death receptor binding (TAS PMID:9374458) — MYD88 binds IL-1R, not TNFR-family
   death receptors. MODIFY to GO:0005149 interleukin-1 receptor binding.
5. Phagocytosis (IMP PMID:30883606) — the paper states "MyD88 silencing did not significantly
   reduce phagocytosis of either E. coli or S. aureus" in human macrophages; only mouse Myd88-/-
   iBMDMs showed reduced E. coli uptake. MARK_AS_OVER_ANNOTATED.
6. Cellular response to mechanical stimulus (IEP PMID:19593445) — the cited full text (BAD in
   prostate cancer) contains no mention of MyD88 or mechanical stimulation. Likely a wrong PMID;
   UNDECIDED rather than REMOVE, flagged in reference_review.
   6b. Nucleus (EXP PMID:21057262) — no localisation data found in the cached clinical paper;
   kept as non-core (deferring to the curator) for consistency with the other nucleus rows.
7. Cell surface (IDA PMID:22851693) — MYD88 is a cytoplasmic adaptor; the paper studies
   membrane-associated IL-1R complex I. MODIFY to GO:0031234.
8. protein binding IPI rows — MODIFY to TIR domain binding (TIRAP, SARM1, TLR partners in MAPPIT),
   Toll-like receptor binding (TLR2/TLR4 co-IP), death domain binding (FADD, IRAK4 DD–DD), protein
   kinase binding (IRAK1/IRAK4), TNFR superfamily binding (TACI); REMOVE for high-throughput screens
   with incidental partners, pathogen proteins and regulators (SPOP, STAP2, HDAC6, DOCK8, DHX9/36,
   TcpB, HCV NS5A).
9. Type I interferon-mediated signaling pathway (IMP PMID:23633945) — the miR-21 paper measures
   IFN effector genes downstream of HCV-induced type I IFN production; MYD88 acts in the TLR arm
   that produces IFN, not in IFNAR signaling. MARK_AS_OVER_ANNOTATED (keep the sibling
   GO:0032481 positive regulation of type I interferon production as non-core).
10. NLRP3 inflammasome complex assembly (NAS, IRAK review) — MYD88 does not assemble NLRP3; it
    primes/licenses via TLR signaling. MODIFY to GO:1900227 positive regulation of NLRP3
    inflammasome complex assembly (which is also separately supported by IMP PMID:33718825).

Deep research

• Falcon deep research completed after the first draft (MYD88-deep-research-falcon.md). It is
  consistent with the review: direct TIR recruitment by TLR5/7/8/9, IL-1R and IL-18R; progressive
  MyD88 oligomerisation to a threshold before stable IRAK4 engagement; OTUD5 K11-chain cleavage;
  S257 phosphorylation and the oncogenic L265P TIR mutation. It reports nothing supporting
  histone-chaperone, chromatin or phagocytosis roles. Added as a reference and cited in the IL-1R
  family core function.

NEW annotations considered

• GO:0035655 interleukin-18-mediated signaling pathway: UniProt states involvement; supported in
  review PMID:36865541 (IL-18R uses MyD88). MYD88 performs the adaptor step in IL-18R signaling
  (participation test passes: it is the adaptor that bridges IL-18R1 TIR to IRAK4). Comparator:
  QuickGO (2026-09-30) shows human GO:0035655 annotations only on IL18, IL18R1, IL18RAP, AKT1,
  PIK3R1, PDGFB — not on IRAK4, TRAF6 or mouse Myd88. The adaptor layer is systematically absent,
  so per CLAUDE.md this was NOT added as NEW; raised as a suggested question instead.