GABBR2 supplies the G-protein-coupling subunit of the GABBR1–GABBR2 GABA-B receptor. The review separates that receptor role from ligand recognition by GABBR1, auxiliary-protein interactions and downstream ion-channel conductance. The annotation reasons and reference assessments are in the review YAML. The companion source-evidence file preserves original annotation contexts, sequence, source locators and phylogenetic evidence; it does not repeat the current judgments.
The selected original Results and Methods of PMID:22660477(https://pmc.ncbi.nlm.nih.gov/articles/PMC3374333/) and PMID:24305054(https://pmc.ncbi.nlm.nih.gov/articles/PMC3865065/) support human receptor-subunit association and the division between ligand-binding GABBR1 and coupling GABBR2. Ectodomain structures and solution measurements establish the heterodimeric interface. Functional experiments separately test receptor constructs and signaling. A structural interface, a cellular association reporter and a native G-protein readout are different measurements; none is described as all three. The original contributes_to qualifier on the GABA receptor activity annotation is retained, and the single core uses the same scope. Other literal upstream relationship fields remain unchanged. The five refined or accepted receptor-activity reasons explicitly interpret the biological claim as contribution by the coupling subunit, without asserting an independently verified qualifier defect.
The early papers are retained with their actual claims. PMID:10328880(https://pubmed.ncbi.nlm.nih.gov/10328880/) reports an R2-alone adenylyl-cyclase response, whereas PMID:9872744(https://pubmed.ncbi.nlm.nih.gov/9872744/) reports different responses for R2 alone and enhanced signaling on coexpression. Later human heterodimer experiments support the physiological receptor model without rewriting those historical observations. PMID:10075644(https://pubmed.ncbi.nlm.nih.gov/10075644/) and PMID:10773016(https://pubmed.ncbi.nlm.nih.gov/10773016/) independently support inhibitory adenylyl-cyclase signaling. Abstract-only access does not establish every construct or assay detail in those papers.
Three incoming ComplexPortal contexts cite PMID:987231(https://pubmed.ncbi.nlm.nih.gov/987231/), an unrelated chemistry paper. Their machine identifiers remain literal, with separate CHECK source-defect records. The correct receptor paper PMID:9872316(https://pubmed.ncbi.nlm.nih.gov/9872316/) is independently present elsewhere and supports the receptor core, but identifier similarity does not prove it was the intended replacement for all three bad citations.
Six R1-specific interaction contexts support the more informative heterodimerization term. Their assay scopes remain distinct. The PMID:16724110(https://pubmed.ncbi.nlm.nih.gov/16724110/) abstract identifies NSF-dependent GABA-B signaling work in rat preparations and CHO cells; the exact human receptor construct behind its curated interaction was not independently recovered. Independent human structures support receptor heterodimerization. The NSF association itself remains non-core. The PMID:30814734(https://pmc.ncbi.nlm.nih.gov/articles/PMC6405316/) GABBR2-tail/KCTD16 structure uses human residues 876–913. Its KCTD16 pentamer and the separately studied KCTD12–G-protein-subunit mechanism are not interchangeable claims.
For PMID:24658140(https://doi.org/10.1038/nmeth.2895), the normal cache is abstract-only, but the authentic publisher supplement was read. Supplementary Figure 3e, PDF page 10, names GABBR1 reporter baits and positive GABBR2-Nub prey; Figure 6b, page 13, confirms the pair in HEK293T reporter cells. These are split-ubiquitin mammalian reporter measurements, not purified affinity measurements or endogenous neuronal co-immunoprecipitation. The supplement does not independently establish an exact natural GABBR1 splice product for the cloned pair.
Two exact interaction contexts remain unresolved. The PMID:23864651(https://pubmed.ncbi.nlm.nih.gov/23864651/) cache is marked as containing full text, but its retained body supplies the abstract and Discussion rather than the target Results/Methods/table. The GLP-1R–GABBR2 row and tested GABBR2 construct were not recovered. The broad MYTH and CHO follow-up design does not identify that particular positive pair. For PMID:37207277(https://pubmed.ncbi.nlm.nih.gov/37207277/), selected neuronal IP-MS Results and Methods and the GABBR2 genetic discussion were read, but the incoming Q86UR5–GABBR2 pair table and pair-specific validation were not recovered. Genetic prioritization is not direct binding evidence. These are access gaps, not evidence that either curator chose the wrong gene.
The current UniProt, direct FASTA and actual family alignment agree on canonical human O75899, 941 residues. No alternative GABBR2 product is declared in those inputs. GABBR1b accession Q9UBS5-2 belongs to the partner, not to GABBR2. Reactome residues 42–941 describe processed canonical GABBR2, not a new isoform.
Engineered constructs remain separate from natural products. Selected Methods in PMID:32555460(https://pmc.ncbi.nlm.nih.gov/articles/PMC8020835/) specify GABBR2 42–821; PMID:32581365(https://pmc.ncbi.nlm.nih.gov/articles/PMC7725281/) uses 1–819 for the structural receptor. PMID:32580208(https://pmc.ncbi.nlm.nih.gov/articles/PMC7429364/) describes a signal-peptide substitution and reports a 41–941 construct while also describing removal of the first 41 residues; the retained original wording is not silently harmonized. Chimeric-G-protein readouts, receptor truncations and interface mutations do not establish activity of every possible natural product.
Original rat PMID:14657159(https://pmc.ncbi.nlm.nih.gov/articles/PMC6741037/) Methods and Results directly show GABBR2 at pre- and postsynaptic membranes, including presynaptic specializations in Figures 3 and 5, against a predominantly extrasynaptic distribution. The Discussion connects presynaptic GABA-B receptors with transmitter-release inhibition through calcium-channel suppression by citing earlier physiology; the microscopy study is not represented as a new channel-function assay. The core now names both verified membrane compartments. Numerical distance analysis for R1 is not assigned to R2. Mouse PMID:15013631(https://pubmed.ncbi.nlm.nih.gov/15013631/) reports receptor–ATF4 association and neuronal/glial colocalization, but its exact cytoplasmic localization panel remains unrecovered. The cytoplasm annotation is therefore UNDECIDED; membrane trafficking in general is not positive evidence for that particular assertion. Rat PMID:10196584(https://pubmed.ncbi.nlm.nih.gov/10196584/) supports a source-scoped neuron–glial signaling context from pharmacology and astrocyte calcium responses, rather than an isolated R2 assay. Its original has_start_location CL:0011005 and has_end_location CL:0000127 remain one conjunction.
All seven incoming TAS Reactome events were followed through official participant relationships to the O75899 entity R-HSA-420747, with mature residues 42–941 at the plasma membrane. The events describe receptor/G-protein assembly, activation and downstream signaling. Membership in a GIRK-channel signaling event does not make GABBR2 the channel pore. The broader source-plan pathways are retained as source records without manufacturing additional process annotations.
The actual PTHR10519 book places the 941-residue human target at AN139/PTN002471337. The receptor-activity and GPCR-heterodimeric-complex IBDs are at AN0/PTN000828857; the GABA-signaling IBD is at AN1/PTN000054983. All are on the target lineage. Literal IBD rows, donor identifiers and selected donor-node joins are preserved in the evidence file. The target's own experimental contribution is valid descendant evidence, not circularity. No retained family loss/NOT record contradicts these assertions.
Official typed paths support the proposed refinements: GO:0004965 through GO:0004930 to GO:0004888, GO:0004965 to GO:0016917, GO:0046982 through GO:0046983 to GO:0005515, GO:0007193 through GO:0007188 to GO:0007186, GO:0051932 to GO:0007268, GO:0005886 to GO:0016020, and GO:1902712 to GO:1902710. Each listed edge is is_a. GO:0038039 uses its current official GPCR-heterodimeric-complex label; an unverified extra ancestry edge is not asserted.
The source-evidence file distinguishes selected original full-text reading, abstract-only reading and unavailable exact target tables. A cache's full_text_available flag is recorded separately from the sections actually read. Failed service responses and reCAPTCHA pages are excluded from scientific evidence. The two short core excerpts are exact substrings of the genuine normal publication caches as well as the retained authentic PubMed abstracts. The evidence file records their source hashes and byte offsets.
Genuine native acquisition created 49 pending annotations. All 49 contexts and all 16 TSV columns match the actual parser; the incoming contributes_to qualifier remains intact. Raw26 maps to native46, where both original cell-type extension atoms are restored together. The 46-source union comprises 16 accepted-base files reused unchanged and 30 genuinely acquired files. Current cache titles and access headers are recorded separately from selected reading scopes. Native CREATE history and targeted validation are complete. Target validation passed with three advisories: two generic protein-binding annotations retained under the standing project policy and one optional direct-excerpt advisory for PMID:24305054. The computed status is DRAFT. Strict GOA comparison passed for all 49 annotations. The approved receptor refinement now includes structured term-granularity metadata for the verified ancestral node; no annotation action or source changed.
The follow-up changes only the cytoplasm action to UNDECIDED, clarifies the five receptor-activity reasons and the rat localization scope, and adds presynaptic membrane to the existing core. The single new 10-word excerpt from PMID:14657159 is verified against the unchanged normal cache. All 49 machine objects, original references, 46 source files, both cell-type extension atoms and the two earlier excerpts remain exact. The six heterodimerization refinements remain supported source-specific judgments. Native EDIT history and targeted follow-up checks passed. Default validation reports four advisories: two generic-binding annotations, one optional PMID:24305054 excerpt, and the presynaptic core location without a separate incoming annotation row. The approved scope adds no NEW row. Status remains DRAFT; strict GOA comparison matches all 49 annotations. History validation, private rendering, relative links and scoped whitespace checks passed, and all 46 normal sources remain byte-identical.