TLR2 (human, O60603) review notes

Sources used

• UniProt O60603 record (TLR2-uniprot.txt)
• 44 GOA-cited publications cached in publications/ (most abstract-only; full text for PMID:16893894, 18765807, 19509286, 19509287, 20067962, 20406817, 20802527, 23155421, 23376921, 23626692, 26649771, 29990310, 31980649, 33576548)
• Falcon deep research: see status note at the end.

Core biology

• Recognition of Gram-positive cell-wall components: PMID:10384090
• Heterodimers set ligand specificity: PMID:19931471
• TLR1-TLR2 structure: PMID:17889651
• TLR2-TLR6-diacyl lipopeptide and TLR2-LTA structures: PMID:19931471
• Preformed heterodimers, rafts, Golgi trafficking: PMID:16880211; PMID:16880211
• Phagosome recruitment: PMID:11095740
• TIR adaptors: PMID:19509286
• LTA: PMID:12594207
• Amyloid-beta: PMID:22198949

LPS question

TLR2 is not an LPS receptor:
- PMID:11521063
- PMID:11986301
- PMID:12443841
- The TAS review cited for GO:0001875 attributes LPS recognition to TLR4 PMID:11518816, so that row was removed. The two IDA rows (GO:0001530 PMID:11274165; GO:0001875 contributes_to PMID:16880211) are UNDECIDED because the full texts are not cached and the abstracts do not describe TLR2 binding LPS.

Decisions (summary)

• Core: pattern recognition receptor activity; triacyl lipopeptide binding; Toll-like receptor binding; TLR1:TLR2 and TLR6:TLR2 pathways; detection of di/triacyl lipopeptide; plasma membrane, rafts, phagosome membrane; TLR1-TLR2 and TLR2-TLR6 complexes.
• NEW: GO:0042498 diacyl lipopeptide binding (IDA, PMID:19931471 structure). An MF from direct structural evidence; no NEW process terms.
• protein binding (24 rows): MODIFY to Toll-like receptor binding where the partner is TLR1/6/10 (PMID:15728506, 20067962, 23155421 x3); MODIFY to TIR domain binding for TIRAP (PMID:19509286) and MyD88 (PMID:29990310); REMOVE the rest (microbial ligands, EGFR, CXCR4, Tollip, TRIP6, ATG16L1, PAUF).
• Downstream cytokine/NF-kappaB outputs kept as non-core; indirect terms (Wnt regulation, Pol II transcription, MMP secretion, responses to IFN-gamma/M-CSF where TLR2 is the regulated target) marked over-annotated; mouse ISS neural terms (learning, microglia development, synapse assembly) marked over-annotated.
• Defense response to virus (IEP, PMID:32634389) removed: the paper states PMID:32634389.
• PMID:23626692 has a 2024 expression of concern; its TLR6:TLR2 row accepted because other papers support it.

Project questions

• GO:0034134 / 0038123 / 0038124 are correctly on human TLR2; IBA GO:0002224 fits.
• No TLR-specific MF exists; GO:0038187 is used as the core MF.

Deep-research cross-check (2026-09-30)

Falcon deep research finished after the review was drafted (TLR2-deep-research-falcon.md;
the runner reported exit code 1 but the file was written). Cross-checked against the
completed review. No annotation actions, descriptions or core functions were changed.

Agreement:
- TLR2 is a signalling pattern recognition receptor, not an enzyme; TLR2:TLR1 for triacyl
and TLR2:TLR6 for diacyl lipopeptides, with TLR6 lacking the TLR1 channel for the third
acyl chain. Matches the two heterodimer core functions and the detection/binding rows.
- Plasma membrane as the principal functional location; cell-type-wide expression on
myeloid and non-immune cells.
- MyD88-dependent signalling through TIRAP/MAL, myddosome, TRAF6, TAK1 to NF-kappaB and
MAPK; no TRIF branch. Matches the accepted NF-kappaB row and the TIR-domain-binding
MODIFY decisions for TIRAP and MyD88.
- TLR2 homodimer signalling is called controversial, supporting identical protein binding
being kept as non-core rather than made a core function.
- Responses to purified peptidoglycan and lipoteichoic acid may reflect contaminating
lipoproteins, which is exactly the open peptidoglycan question already recorded.
- TLR2/TLR4 recognition of LPS is described as context-dependent and incompletely
established, consistent with UNDECIDED on the two LPS rows (GO:0001530 IDA
PMID:11274165; GO:0001875 IDA PMID:16880211) and with the REMOVE of the TAS row.
- Amyloid-beta and alpha-synuclein as DAMPs, consistent with keeping amyloid-beta binding
as non-core.

Additions not acted on (report cites only secondary reviews; no primary paper verified, so
per the "never cite deep research as sole support" rule nothing was asserted):
- TLR2/TLR10 and TLR2/CLEC2D heterodimers; CLEC2D/TLR2 is described as negatively
regulating IRF5-mediated antifungal immunity. No Toll-like receptor binding or
process row was added for these.
- Endosomal TLR2 pools with regulatory (IL-10) output, and soluble/vesicular TLR2 acting
as a decoy. Not annotated; the review's Golgi and cytoplasm IDA rows stay non-core. The
report asserts TLR2 is absent from the Golgi, which conflicts with the IDA Golgi row
(PMID:16880211); the experimental row was left in place per the rule against overruling
curators from incomplete evidence.
- TLR2 promoting monocyte chemotaxis, adhesion, transendothelial migration, tissue-factor
expression and platelet interactions. These are downstream cellular consequences of
signalling and would not pass the participation test for a NEW process term, so none
was added.

Conflict: none that changes a decision.