LPAR2 literature notes
Scope and research provenance
This pass is restricted to reference discovery, cached-source review, exact-quote extraction, and evidence boundaries. It does not review GO annotation actions or write description/core-function synthesis.
The project deep-research wrapper was attempted with just deep-research-perplexity human LPAR2. It failed on 2026-08-11 with HTTP 401 insufficient_quota and exited without producing LPAR2-deep-research-perplexity.md. No provider-named substitute was created. The literature assessment below was performed manually from project-cached PubMed/PMC, Reactome, GO reference, GOA/IntAct provenance, and UniProt records.
All 21 seeded references were audited: six GO method references, eight seeded PMIDs, and seven Reactome records. Four decisive primary papers were added and fetched through ai-gene-review fetch-pmid: PMID:10729222, PMID:10922489, PMID:15143197, and PMID:16904289. PMID:16904289 was already present in the shared cache. The reviewed UniProt record was also added as a file reference.
Decisive functional evidence
Ligand binding and receptor identity
- The founding human EDG4 study is the clearest direct receptor-identity source. Recombinant EDG4 produced concentration-dependent LPA signaling: PMID:9525886 It also increased radioligand-binding sites: PMID:9525886
- EDG4/LPAR2 has broad LPA molecular-species responsiveness in recombinant Sf9 cells, not absolute selectivity for a single acyl species: PMID:10922489
Gi, Gq, PLC, calcium, and Rho signaling
- Founding human-receptor evidence supports pertussis-toxin-sensitive Gi input and a Rho-dependent arm: PMID:9525886
- Recombinant human EDG4 mobilizes calcium through PLC-generated IP3: PMID:9804623
- The same study distinguishes EDG4 from EDG2 and supports mixed Gi/Gq input for LPAR2 calcium signaling: PMID:9804623
- In polarized intestinal epithelial models, the Gi arm inhibits CFTR: PMID:16203867
- UniProt integrates Gi/Go, G12/G13, and Gq coupling, but line-prefix formatting in the cached flat file requires quoting the individual continuations separately: [file:human/LPAR2/LPAR2-uniprot.txt "G(i)/G(o),"] [file:human/LPAR2/LPAR2-uniprot.txt "G(12)/G(13), and G(q) families of heteromeric G proteins. Plays a key"] The cached primary abstracts do not provide equally direct G-protein-selective assays for every arm. The G12/G13-to-Rho interpretation is consistent with the C3-sensitive founding response and MAGI3/RhoA work, but should not be described as a direct heterotrimeric-coupling measurement from those abstracts.
Receptor-specific scaffolds and interaction boundaries
NHERF2 and PLCB3
- NHERF2 specificity is direct and motif-resolved: PMID:15143197
- NHERF2 links LPAR2 selectively to PLC-beta3: PMID:15143197
- In intestinal epithelial cells, NHERF2 scaffolds LPAR2 and CFTR into an apical complex: PMID:16203867 CFTR is therefore an NHERF2-mediated complex partner in this epithelial context, not evidence that LPAR2 directly binds CFTR without a scaffold.
TRIP6 and MAGI3
- TRIP6 binds the receptor tail through LIM domains rather than a PDZ domain: PMID:14688263 LPA-dependent TRIP6 recruitment promotes focal-adhesion signaling, and migration perturbation was demonstrated in SKOV3 ovarian cancer cells: PMID:14688263
- MAGI3 is another direct, C-terminal PDZ interaction: PMID:16904289 Its ERK/RhoA effects were tested in SW480 colon cancer cells and should remain context-bounded.
RalA, GRK2, and desensitization
- PMID:19306925 has an LPA1-focused title but explicitly assays both receptors. RalA associates with LPAR2, while agonist-triggered RalA activation in the accessible abstract is assigned only to LPAR1: PMID:19306925 Do not transfer the paper's LPA1 Ral-dependent PLC conclusion to LPAR2.
- GRK2-dependent desensitization does include LPAR2: PMID:19306925
- UniProt records the before/after localization in flat-file continuations: [file:human/LPAR2/LPAR2-uniprot.txt "Prior to LPA treatment found"] [file:human/LPAR2/LPAR2-uniprot.txt "predominantly at the cell surface but in the presence of LPA"] [file:human/LPAR2/LPAR2-uniprot.txt "colocalizes with RALA in the endocytic vesicles."] This supports trafficking to endocytic vesicles, but the accessible PMID:19306925 abstract states that RalA was not found to control LPA-receptor endocytosis; colocalization must not be rewritten as a causal RalA trafficking mechanism.
Broad interaction screens
- BioPlex 2.0 and 3.0 are proteome-scale AP-MS resources. GOA/IntAct assigns LGALS3 as an LPAR2 partner for PMID:28514442 and PMID:33961781, but neither cached narrative establishes an LPAR2-specific signaling function. This remains a screen-level candidate rather than a core receptor mechanism.
- The RAMP study directly names LPAR2 among nine GPCRs with evidence for all three RAMPs across all capture-detection methods: PMID:39083597 The result comes from systematic pairwise coexpression and does not establish an LPAR2-specific trafficking, ligand, or pharmacological consequence in native tissue.
- PMID:9525886 observed two EDG4 RNA sizes: PMID:9525886 Transcript size alone does not establish distinct translated isoforms or isoform-specific function.
- The initially reported ovarian-tumor EDG4 cDNA contained a frameshift artifact/variant: PMID:10729222 It must not be represented as a normal LPAR2 protein isoform.
- The current reviewed UniProt record contains no
ALTERNATIVE PRODUCTS section. Its historical sequence-conflict records and the differently sized transcripts above are not evidence for experimentally validated functional protein isoforms.
Species and experimental-context boundaries
- PMID:9525886 used human EDG4 overexpressed in Jurkat cells; PMID:9804623 used human cDNA in TAg-Jurkat cells and rat HTC4 hepatoma transfectants; PMID:10922489 used recombinant receptors in Sf9 insect cells. These establish receptor capability, not native coupling weights in every human tissue.
- PMID:15143197 is decisive for the NHERF2-PLCB3 complex, while PMID:16203867 anchors that scaffold to polarized intestinal epithelial biology and includes mouse knockout physiology. The in vivo diarrhea phenotype is mouse/context-specific even though the human epithelial complex is directly observed.
- TRIP6 migration and MAGI3 ERK/RhoA effects were established in cancer-cell contexts. They are receptor-regulatory mechanisms with context-specific phenotypic outputs.
- Reactome records are curated pathway context, not independent primary evidence. GO_REF records describe propagation methods and do not themselves validate an LPAR2-specific biological claim.
Reference prioritization
The strongest direct sources for later synthesis are PMID:9525886 (human receptor identity, ligand response, Gi/Rho signaling), PMID:9804623 (PLC/IP3/calcium and Gi/Gq), PMID:15143197 (LPAR2-NHERF2-PLCB3 specificity), PMID:16203867 (apical NHERF2-CFTR complex and Gi function), PMID:14688263 (TRIP6), PMID:16904289 (MAGI3), and PMID:19306925 (GRK2-dependent desensitization with strict LPA1/LPAR2 separation).
Full IBA re-review, 2026-09-20
This assessment supersedes earlier universal coupling and membrane-exclusion arguments. All original source rows and qualifiers are preserved. Actual PTHR22750 ancestry places the target below PTN002733616; the target appearing as an IBD source is legitimate experimental grounding. GO cytoplasm includes internal membrane structures, and primary PMID:26473723 demonstrates internalization of human receptor constructs. Conditional cAMP activation from full PMID:10488122 Methods/Results/Fig.7 is retained non-core alongside cell-specific inhibitory responses; a shared focused report is pending. PMID:10727522 provides contrasting assays and human forebrain expression rather than a universal brain absence.
Detailed primary-source access limits, ortholog chains, protein-binding decisions, NEW comparator/ancestor checks and pending questions are in the shared primary evidence record. The companion JSON records live ontology, annotation and tree responses. No additional NEW terms were added.
Recovery PR review: generic binding policy (2026-09-22)
Applied the repository policy to the re-reviewed GO:0005515 rows. Removal concerns
the uninformative function label and does not refute the source interaction.
Rows whose target-specific assays remain inaccessible are UNDECIDED. Source
assertions and supporting evidence are preserved.
- PMID:16203867: MARK_AS_OVER_ANNOTATED -> REMOVE
- PMID:28514442: KEEP_AS_NON_CORE -> REMOVE
- PMID:33961781: KEEP_AS_NON_CORE -> REMOVE
- PMID:39083597: KEEP_AS_NON_CORE -> REMOVE
- PMID:19306925: KEEP_AS_NON_CORE -> REMOVE
Recovery PR signaling follow-up (2026-09-22)
Restore full forskolin context in the quotation, distinguish external full-text access from the abstract-only cache, and retain branch-specific LPAR2 core terms without redundant NEW assertions beneath existing GPCR signaling.