ADGRA3 (GPR125) is a 1321-residue adhesion G protein-coupled receptor of the ADGRA subfamily, built from a cleaved signal peptide, a large extracellular region containing four leucine-rich repeats, two immunoglobulin domains and a hormone-binding domain followed by a GAIN domain, a canonical seven-pass transmembrane bundle, and a cytoplasmic tail ending in the class I PDZ-binding motif Glu-Thr-Thr-Val. Despite an atypical GPCR proteolysis site (Ser-Leu-Ser rather than the canonical His-Leu/Ile-Ser/Thr), the receptor self-cleaves within the GAIN domain early in biosynthesis into N- and C-terminal fragments that remain associated; this exposes a stachel tethered agonist at the new N-terminus of the C-terminal fragment. Through it the receptor activates both Gs and Gi, constitutively and without any identified external ligand, and abolishing the first three residues of the tethered agonist abolishes signalling. The receptor reaches the cell surface with its ectodomain exposed and is continuously retrieved by beta-arrestin-independent, clathrin-dependent endocytosis. In polarised epithelium the C-terminal motif binds the PDZ domains of DLG1 and confines the receptor to the basolateral membrane, where it is required for correct mitotic spindle orientation and single-lumen morphogenesis; the same motif binds the PDZ domains of many other scaffolds in quantitative affinity assays. The zebrafish orthologue recruits Dishevelled and modulates planar cell polarity during gastrulation, although human ADGRA3 does not engage beta-catenin-dependent Wnt signalling and its G protein output is Dishevelled-independent. ADGRA3 is a marker of spermatogonial progenitor cells, and mice lacking it develop obstructive azoospermia or imperforate vagina; in adipocytes its constitutive Gs-PKA-CREB activity drives a thermogenic programme, and it also promotes osteoclast differentiation. Its endogenous ligand is unknown.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0005886 plasma membrane | IBA GO_REF:0000033 | ACCEPT | Summary: Correct, well founded on both sides of the propagation, and corroborated directly in human cells. The WITH/FROM field resolves to five gene products and one node: MGI:MGI:1917943 is mouse Adgra3 (Q7TT36, Swiss-Prot) - the 1:1 orthologue; MGI:MGI:1925810 is mouse Adgra2 (Q91ZV8, Swiss-Prot); UniProtKB:Q96PE1 is human ADGRA2; ZFIN:ZDB-GENE-131003-2 is zebrafish adgra3 (S4X0Q8, Swiss-Prot); and ZFIN:ZDB-GENE-081104-363 is zebrafish adgra2 (E7FBY6 / AGRA2_DANRE, Swiss-Prot, 1367 aa). All five are reviewed entries. Worth recording as a resolver limit: UniProt's xref:zfin- index returns only the unreviewed A0A0U2ULT4 / A0A8M1P7B9 for that ZFIN gene id and misses E7FBY6 entirely, so a lookup through that index alone would wrongly report the source as TrEMBL-only. Asking what those donors themselves carry for this term (QuickGO, goUsage=descendants) shows the propagation is not circular: human ADGRA2 has three EXP annotations plus an IDA, mouse Adgra2 has an EXP and an IDA, and zebrafish adgra3 has its own IDA from PMID:23821037. So real experimental plasma-membrane evidence exists in the clade. Independently of the propagation, human ADGRA3 itself is demonstrably a cell-surface protein: an N-terminally FLAG-tagged construct is detected at the surface of non-permeabilised HEK293 cells and internalises constitutively from there. The term is core: everything ADGRA3 is known to do - G protein coupling, PDZ-motif scaffolding, autoproteolysis-dependent maturation - happens at this membrane. Reason: Every non-node donor is a reviewed ADGRA family member and four of them carry their own EXP/IDA evidence for this term; human ADGRA3 is independently shown at the cell surface in HEK293 cells. Propagation Review Root cause: NO FAILURE CORE Sources checked: MGI:MGI:1917943 · mouse Adgra3 (UniProtKB Q7TT36, reviewed/Swiss-Prot) - the 1:1 orthologue; carries IDA to the descendant term GO:0009897 from PMID:17882221 SUPPORTS TRANSFER MGI:MGI:1925810 · mouse Adgra2 (UniProtKB Q91ZV8, reviewed/Swiss-Prot) - a paralogue, not the orthologue; carries its own EXP (PMID:25558062) and IDA (PMID:28803732) for this term SUPPORTS TRANSFER PANTHER:PTN001738137 · PANTHER ancestral tree node, not a gene product, so it carries no evidence of its own; its reach across the eight reviewed (Swiss-Prot) members of PTHR45930 is the whole set - ADGRA1, ADGRA2 and ADGRA3 in human and mouse, plus adgra2 and adgra3 in zebrafish, which has no adgra1 in the family - i.e. the family root NOT RELEVANT UniProtKB:Q96PE1 · human ADGRA2/GPR124 (reviewed/Swiss-Prot) - a paralogue, not the orthologue; carries three EXP annotations and an IDA for this term SUPPORTS TRANSFER ZFIN:ZDB-GENE-081104-363 · zebrafish adgra2 (UniProtKB E7FBY6 / AGRA2_DANRE, reviewed/Swiss-Prot, 1367 aa) - a paralogue; carries EXP (PMID:26051822) and IDA (PMID:27979830) for this term. Note UniProt's xref:zfin- index does not return this reviewed entry for the ZFIN gene id, only the unreviewed A0A0U2ULT4 / A0A8M1P7B9; the reviewed accession comes from the cached PTHR45930 member table SUPPORTS TRANSFER ZFIN:ZDB-GENE-131003-2 · zebrafish adgra3 (UniProtKB S4X0Q8, reviewed/Swiss-Prot) - the orthologue; carries IDA for this term from PMID:23821037 SUPPORTS TRANSFER Supporting Evidence: PMID:31659746 In summary, GPR125 is mainly expressed on the cell surface and undergoes constitutive internalization under physiological temperatures PMID:36089063 Consistent with the association, in polarized MDCK cells, FLAG–GPR125 colocalized with endogenous Dlg1 at the basolateral membrane |
| GO:0007166 cell surface receptor signaling pathway | IBA GO_REF:0000033 | ACCEPT | Summary: The general term is the right call here, and I checked the case for refining it and rejected it. Node PANTHER:PTN001738137 reaches all eight reviewed members of PTHR45930 - ADGRA1, ADGRA2 and ADGRA3 in human and mouse, plus adgra2 and adgra3 in zebrafish, which has no adgra1 in the family - so it is the family-root node. Its three named seeds all carry experimental evidence for descendants of this term, but for *different* descendants: human ADGRA2 and zebrafish adgra2 have GO:0060070 canonical Wnt signalling by IDA, while zebrafish adgra3 has GO:0060071 Wnt/planar-cell-polarity by IDA. The least common ancestor of just those two is GO:0016055 Wnt signaling pathway, one level below the term actually used, so a granularity objection is available. It does not hold. The node's reach includes ADGRA1, whose only experimental biology is postsynaptic (GO:0014069, GO:0098978 at the ADGRA1-only node PTN002914505) with no Wnt evidence at all, so the clade is heterogeneous and GO:0016055 is not its LCA. More decisively, refining to a Wnt term would assert of human ADGRA3 something its own data declines to support: in HEK293 cells ADGRA3 produced no beta-catenin/TCF-LEF transcriptional response, and its G protein signalling persists in DVL1-3-null cells. GO:0007166 is therefore the correct conservative ancestor, not lazy curation. It is also now true for a reason PAINT could not have used in 2018: ADGRA3 has since been shown to be a genuine G protein-coupled receptor, and GO:0007186 - already on this gene from InterPro and GDB - is the descendant that its own experiments establish. Reason: Donor set is functionally heterogeneous (ADGRA1 synaptic, ADGRA2 canonical Wnt, zebrafish adgra3 Wnt/PCP), so GO:0007166 is the genuine LCA; and the target's own data argue against the Wnt refinement. Propagation Review Root cause: NO FAILURE CORE Sources checked: PANTHER:PTN001738137 · PANTHER ancestral tree node, not a gene product; family-root node reaching all eight reviewed members (ADGRA1/2/3 in human and mouse, adgra2 and adgra3 in zebrafish), which is why the term sits at this generality NOT RELEVANT UniProtKB:Q96PE1 · human ADGRA2/GPR124 (reviewed/Swiss-Prot) - a paralogue; carries IDA to the descendant term GO:0060070 canonical Wnt signalling from PMID:30026314 SUPPORTS TRANSFER ZFIN:ZDB-GENE-081104-363 · zebrafish adgra2 (UniProtKB E7FBY6 / AGRA2_DANRE, reviewed/Swiss-Prot) - a paralogue; carries IDA to the descendant term GO:0060070 from PMID:30026314 SUPPORTS TRANSFER ZFIN:ZDB-GENE-131003-2 · zebrafish adgra3 (UniProtKB S4X0Q8, reviewed/Swiss-Prot) - the orthologue; carries IDA to the descendant term GO:0060071 Wnt/planar cell polarity from PMID:23821037 SUPPORTS TRANSFER Supporting Evidence: PMID:40127866 No transcriptional activation was observed in an assay of downstream β-catenin activity. PMID:40127866 Collectively, this suggests that ADGRA3 is capable of activating Gai signaling axes independently of DVL presence. PMID:40127866 Collectively, this establishes classical G protein-mediated signaling for ADGRA3. |
| GO:0009897 external side of plasma membrane | IBA GO_REF:0000033 | ACCEPT | Summary: A clean single-orthologue transfer that lands on exactly the term its donor holds, with independent human corroboration. This row comes from a different node than the other two IBAs. PANTHER:PTN002914494 reaches exactly the three ADGRA3 orthologues - human ADGRA3, mouse Adgra3, zebrafish adgra3 - and carries only this one term. Its sole non-node donor, MGI:MGI:1917943, is mouse Adgra3 (Q7TT36), the true orthologue, and QuickGO shows that donor holds GO:0009897 by IDA from PMID:17882221. The propagation therefore neither gains nor loses precision relative to its source, which is the ACRV1 failure mode checked for and not found here. I also ran the reference-projection test on the donor's paper. PMID:17882221 annotates five entities, which is the superficial shape of a complex-projection artefact, but the five are Adgra3, Cd34, Pou5f1, Vim and Acta2 and each receives a *different* compartment (GO:0009897 for Adgra3 and Cd34, nucleus for Pou5f1, intermediate filament for Vim, actin cytoskeleton for Acta2). No phenotype term spreads across the set. This is a marker-panel paper localising five distinct markers, not one finding projected onto five genes, so the donor IDA is its own. Human corroboration is independent: an N-terminally FLAG-tagged human GPR125 is detected on non-permeabilised HEK293 cells, i.e. the large N-terminal ectodomain faces the extracellular space, which is what this term asserts. Reason: Sole protein donor is the mouse orthologue and holds the identical term by IDA; the donor's paper passes the projection test (five entities, five different compartments, no phenotype spread); and human surface exposure of the ectodomain is independently shown. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: MGI:MGI:1917943 · mouse Adgra3 (UniProtKB Q7TT36, reviewed/Swiss-Prot) - the 1:1 orthologue and the only protein donor on this row; carries IDA to the identical term from PMID:17882221, so no precision is lost in transfer SUPPORTS TRANSFER PANTHER:PTN002914494 · PANTHER ancestral tree node, not a gene product; its reach is exactly the three ADGRA3 orthologues (human ADGRA3, mouse Adgra3, zebrafish adgra3) and this is the only term it propagates to them - see suggested_questions on the terms it does not NOT RELEVANT Supporting Evidence: PMID:17882221 Furthermore, both SPCs and MASCs express GPR125, an orphan adhesion-type G-protein-coupled receptor. PMID:31659746 In summary, GPR125 is mainly expressed on the cell surface and undergoes constitutive internalization under physiological temperatures |
| GO:0004888 transmembrane signaling receptor activity | IEA GO_REF:0000002 | ACCEPT | Summary: Correct, and correctly cautious for the signature it comes from. IPR017981 is 'GPCR, family 2-like, 7TM' and InterPro maps it to GO:0004888, GO:0007166 and GO:0016020. GO:0004888 transmembrane signaling receptor activity is a direct ancestor of GO:0004930 G protein-coupled receptor activity, which is separately annotated to ADGRA3 from IPR000832/IPR001879/IPR036445 and from a TAS row - and which has since been demonstrated experimentally, with ADGRA3 activating both Gi and Gs in human cell lines and losing that activity when the first three residues of the stachel tethered agonist are removed. GO:0004888 transmembrane signaling receptor activity is a direct ancestor of GO:0004930 G protein-coupled receptor activity, which this gene also holds - so the row is redundant in the GO record. Redundancy is not error, though, and it is worth being precise about which of the two is the defect. Why ACCEPT and not MODIFY-for-redundancy. IPR017981 is a **7TM-domain** signature, and a seven-transmembrane bundle on its own cannot tell you the receptor couples to a heterotrimeric G protein - many 7TM proteins do not. InterPro's mapping to the non-committal parents GO:0004888 and GO:0007166 is therefore the most this signature can honestly support, not a curator stopping short. The specific children GO:0004930 and GO:0007186 arrive from a different and more specific signature, IPR000832 (GPCR family 2, secretin-like), which does support them. Modifying a correctly-conservative signature mapping downward would be attributing to it evidence it does not carry. This is the discriminator applied throughout this review: a general term is ACCEPTed when it is the most its own source can support, and MODIFIED when the specific term is established for this gene and the general row merely restates a compartment or activity now known precisely. The three GO:0016020 rows fall on the other side of that line - they are superseded statements about where this protein is, and the ISS one additionally sits above the very donor it was drawn from. Reason: Correct and appropriately scoped to IPR017981, a 7TM-domain signature that cannot by itself support G protein coupling; the specific child GO:0004930 comes from the family-level IPR000832 and is separately annotated. Redundant in the record but not a curation defect. Resolved identically to the IEA GO:0007166 row, the other parent this same signature supplies. Supporting Evidence: PMID:40127866 We found low-level activation of Gi and Gs by ADGRA3 and slightly more by its CTF. PMID:40127866 This resulted in abrogated G protein-mediated signaling, as observed for other aGPCRs. |
| GO:0004930 G protein-coupled receptor activity | IEA GO_REF:0000002 | ACCEPT | Summary: A domain-derived molecular function that has since been confirmed by direct measurement - the case where the fold-propagation suspicion is simply wrong. The three signatures are IPR000832 (GPCR family 2, secretin-like), IPR001879 (family 2 extracellular hormone receptor domain) and IPR036445 (its structural superfamily). Read on their own these are architecture, and until 2025 the gene was formally an orphan whose G protein coupling had never been tested. It has now been tested: in human cell-line signalling assays ADGRA3 activates Gi and Gs at low level, its GPS-truncated C-terminal fragment does so slightly more strongly, and deleting the first three residues of that fragment - the class-defined stachel tethered agonist - abolishes the signalling. That is the canonical demonstration of GPCR activity, and it is reinforced by an independent report of constitutive Gs-PKA-CREB signalling in adipocytes. Core function, and the evidence code understates it: this should be re-evidenced as IDA against PMID:40127866. Reason: The InterPro inference is now independently confirmed: ADGRA3 activates Gi and Gs in human cells in a stachel-dependent manner. Supporting Evidence: PMID:40127866 We found low-level activation of Gi and Gs by ADGRA3 and slightly more by its CTF. PMID:40127866 This resulted in abrogated G protein-mediated signaling, as observed for other aGPCRs. PMID:40127866 Collectively, this establishes classical G protein-mediated signaling for ADGRA3. |
| GO:0007166 cell surface receptor signaling pathway | IEA GO_REF:0000002 | ACCEPT | Summary: The biological-process twin of the GO:0004888 row, and resolved the same way for the same reason. IPR017981 supplies this gene three terms - GO:0004888, GO:0007166 and GO:0016020 - and two of them, this one and GO:0004888, are ancestors of terms the gene also holds from IPR000832. An earlier draft of this review modified GO:0004888 downward while accepting this row, which is opposite reasoning applied to one situation; that inconsistency was raised in review and is resolved here by accepting both. Why ACCEPT and not MODIFY-for-redundancy. IPR017981 is a **7TM-domain** signature, and a seven-transmembrane bundle on its own cannot tell you the receptor couples to a heterotrimeric G protein - many 7TM proteins do not. InterPro's mapping to the non-committal parents GO:0004888 and GO:0007166 is therefore the most this signature can honestly support, not a curator stopping short. The specific children GO:0004930 and GO:0007186 arrive from a different and more specific signature, IPR000832 (GPCR family 2, secretin-like), which does support them. Modifying a correctly-conservative signature mapping downward would be attributing to it evidence it does not carry. This is the discriminator applied throughout this review: a general term is ACCEPTed when it is the most its own source can support, and MODIFIED when the specific term is established for this gene and the general row merely restates a compartment or activity now known precisely. The three GO:0016020 rows fall on the other side of that line - they are superseded statements about where this protein is, and the ISS one additionally sits above the very donor it was drawn from. Accepting this row also keeps it consistent with the IBA row carrying the same term, where GO:0007166 is the genuine least common ancestor of a clade spanning ADGRA1's synaptic biology, ADGRA2's Wnt biology and ADGRA3's G protein coupling. Same term, same action, two independently sufficient reasons. Reason: Correct and appropriately scoped to IPR017981, a 7TM-domain signature that cannot support G protein coupling on its own; GO:0007186 arrives separately from IPR000832 and is now experimentally demonstrated. Redundant in the record but not a curation defect, and consistent with both the GO:0004888 row from the same signature and the IBA row carrying the same term. Supporting Evidence: PMID:40127866 We found low-level activation of Gi and Gs by ADGRA3 and slightly more by its CTF. PMID:40127866 This resulted in abrogated G protein-mediated signaling, as observed for other aGPCRs. PMID:40127866 Collectively, this establishes classical G protein-mediated signaling for ADGRA3. |
| GO:0007186 G protein-coupled receptor signaling pathway | IEA GO_REF:0000002 | ACCEPT | Summary: The biological-process counterpart of the GO:0004930 row, and likewise now experimentally established rather than merely inferred from the secretin-like family signature. ADGRA3 signals through heterotrimeric G proteins in human cells: both Gi and Gs are activated, the response is stronger for the GPS-cleaved C-terminal fragment, and it depends on the stachel peptide. It is also DVL-independent, so it is a genuine G protein pathway and not a repackaged Wnt/PCP readout. Core process; the evidence code should be upgraded to IDA against PMID:40127866. Reason: Now demonstrated directly: stachel-dependent Gi and Gs activation by human ADGRA3, sustained in DVL1-3-null cells. Supporting Evidence: PMID:40127866 We found low-level activation of Gi and Gs by ADGRA3 and slightly more by its CTF. PMID:40127866 Collectively, this suggests that ADGRA3 is capable of activating Gai signaling axes independently of DVL presence. PMID:40127866 Collectively, this establishes classical G protein-mediated signaling for ADGRA3. |
| GO:0016020 membrane | IEA GO_REF:0000120 | MODIFY | Summary: True, but two steps less precise than what is known, and traceable to a UniProt subcellular-location line that is itself only a by-similarity inference. The WITH/FROM includes UniProtKB-SubCell:SL-0162, i.e. this row is the machine reflection of UniProt's own 'SUBCELLULAR LOCATION: Membrane {ECO:0000250|UniProtKB:S4X0Q8}; Multi-pass membrane protein'. That line dates from before the human protein was localised, and the human localisation is now settled: N-terminally tagged GPR125 is on the surface of non-permeabilised HEK293 cells, and in polarised MDCK cells it is confined to the basolateral plasma membrane domain - GO:0016323, which sits two is_a/part_of steps below GO:0016020. GO:0005886, one step below, is already on this gene by IBA, so the generic parent is redundant as well as vague. MODIFY rather than REMOVE: the statement is true, and the constructive fix is to say which membrane. The corresponding UniProt line should also be upgraded to Cell membrane - see suggested_questions. Reason: Generic ancestor of GO:0005886, which is already annotated and is directly established for the human protein; the row's own source token is UniProt's by-similarity Membrane line. Proposed replacements: plasma membrane Supporting Evidence: file:human/ADGRA3/ADGRA3-uniprot.txt CC -!- SUBCELLULAR LOCATION: Membrane {ECO:0000250|UniProtKB:S4X0Q8}; Multi- PMID:31659746 In summary, GPR125 is mainly expressed on the cell surface and undergoes constitutive internalization under physiological temperatures PMID:36089063 Consistent with the association, in polarized MDCK cells, FLAG–GPR125 colocalized with endogenous Dlg1 at the basolateral membrane |
| GO:0005515 protein binding | IPI PMID:24550280 Large-scale interaction profiling of PDZ domains through pro... | MODIFY | Summary: A real, replicated, functionally consequential interaction recorded under an uninformative term. PMID:24550280 is proteomic peptide-phage display: libraries of all human C-terminal peptides were screened against nine PDZ domains including those of DLG1. The ADGRA3 hit is therefore, by construction, a PDZ-domain/PDZ-binding-motif interaction - ADGRA3's C-terminus is Glu-Thr-Thr-Val (1318-1321), the class I PBM that UniProt annotates as MOTIF 1319..1321. Unlike the other seventeen partners, DLG1 is corroborated well beyond peptide affinity. It was found by direct binding to the bacterially expressed GPR125 C-terminal tail in 2004 (the paper's 'TEM5-like protein' is ADGRA3, alongside TEM5 = ADGRA2), it co-immunoprecipitates with full-length GPR125 from HEK293 cells, endogenous Dlg1 co-precipitates with GPR125 in MDCK cells, and both are lost when the ETTV tetrapeptide is deleted. IntAct records five evidences over three publications and four methods for this pair, against a single method for every other partner. Bare 'protein binding' is uninformative; the informative term is GO:0030165, and it is exactly what was measured. Retain the partner machine-readably via the has_input extension. Reason: Bare protein binding replaced by the specific, measured activity: this is a PDZ-domain/PBM interaction, and DLG1 is the one partner with independent cellular corroboration. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:24550280 With these libraries we screened the nine PSD-95/Dlg/ZO-1 (PDZ) domains of human Densin-180, Erbin, Scribble, and Disks large homolog 1 for peptide ligands. PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) PMID:36089063 These findings indicate that GPR125 associates with Dlg1 in a PBM-dependent manner. PMID:36089063 Similarly, endogenous Dlg1 in MDCK cells was coprecipitated with FLAG–GRP125-FL but not with FLAG–GRP125-ΔETTV PMID:15021905 hDlg furthermore bound a novel seven-pass transmembrane protein, which was homologous to TEM5, and was named here a TEM5-like protein (TEM5-like). file:human/ADGRA3/ADGRA3-uniprot.txt CC -!- SUBUNIT: Interacts (via PDZ-binding motif) with DLG1. |
| GO:0005515 protein binding | IPI PMID:36115835 Quantitative fragmentomics allow affinity mapping of interac... | MODIFY | Summary: One of eighteen partners from a single quantitative PDZ-domain affinity screen; modified from bare protein binding to the activity the assay actually measures. PMID:36115835 is a holdup assay measuring 65,000 PDZ-domain/PBM affinities. Its ADGRA3 bait is the C-terminal peptide, whose last four residues are Glu-Thr-Thr-Val - the class I PDZ-binding motif UniProt annotates as MOTIF 1319..1321. Every one of the eighteen partners GOA imported from this study is a reviewed Swiss-Prot canonical entry carrying between 2 and 13 PDZ domains, with no TrEMBL clone or partial ORFeome construct among them, so there is no partner-identity defect of the kind found on ACRV1. What the screen establishes is a property of the motif, not eighteen independent discoveries: IntAct holds 117 holdup evidences for ADGRA3 spanning 80 distinct PDZ proteins, of which GOA imported these 18. GRID2IP carries 2 PDZ domains and is a postsynaptic/neuronal scaffold. Worth flagging: the postsynaptic compartment is the characterised territory of the paralogue ADGRA1, which is the family member PAINT gives GO:0014069 postsynaptic density and GO:0098978 glutamatergic synapse at the ADGRA1-only node PTN002914505. There is no evidence placing ADGRA3 at a synapse. The most economical reading is that a class I PBM is recognised by class I PDZ domains regardless of where those scaffolds actually reside, which is what a domain-array affinity screen is designed to detect. Real biochemistry, unestablished cell biology. GO:0030165 PDZ domain binding states what was measured and is true of every row in this block; 'protein binding' states nothing. The partner is preserved machine-readably in the has_input extension rather than being discarded by the term change. Reason: Bare protein binding replaced by GO:0030165: the assay measured the affinity of the ADGRA3 class I PBM for the PDZ domains of GRID2IP, which is the specific activity supported. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:36115835 Here, we measure the affinities of 65,000 interactions involving PDZ domains and their target PDZ-binding motifs (PBM) within a human interactome PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) file:human/ADGRA3/ADGRA3-uniprot.txt FT MOTIF 1319..1321 |
| GO:0005515 protein binding | IPI PMID:36115835 Quantitative fragmentomics allow affinity mapping of interac... | MODIFY | Summary: One of eighteen partners from a single quantitative PDZ-domain affinity screen; modified from bare protein binding to the activity the assay actually measures. PMID:36115835 is a holdup assay measuring 65,000 PDZ-domain/PBM affinities. Its ADGRA3 bait is the C-terminal peptide, whose last four residues are Glu-Thr-Thr-Val - the class I PDZ-binding motif UniProt annotates as MOTIF 1319..1321. Every one of the eighteen partners GOA imported from this study is a reviewed Swiss-Prot canonical entry carrying between 2 and 13 PDZ domains, with no TrEMBL clone or partial ORFeome construct among them, so there is no partner-identity defect of the kind found on ACRV1. What the screen establishes is a property of the motif, not eighteen independent discoveries: IntAct holds 117 holdup evidences for ADGRA3 spanning 80 distinct PDZ proteins, of which GOA imported these 18. MPDZ carries 13 PDZ domains and is a basolateral/junctional polarity or scaffolding protein, so it is topologically compatible with GPR125, which is itself confined to the basolateral membrane domain in polarised epithelium and is retained there through this very motif. The affinity is therefore plausible in cells as well as in vitro. It is nonetheless a single measurement in a single study with no cellular assay for this pair, so the interaction is recorded at the level the data support - PDZ-domain binding by the ETTV motif - and is not claimed as a core function. GO:0030165 PDZ domain binding states what was measured and is true of every row in this block; 'protein binding' states nothing. The partner is preserved machine-readably in the has_input extension rather than being discarded by the term change. Reason: Bare protein binding replaced by GO:0030165: the assay measured the affinity of the ADGRA3 class I PBM for the PDZ domains of MPDZ, which is the specific activity supported. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:36115835 Here, we measure the affinities of 65,000 interactions involving PDZ domains and their target PDZ-binding motifs (PBM) within a human interactome PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) file:human/ADGRA3/ADGRA3-uniprot.txt FT MOTIF 1319..1321 PMID:36089063 Consistent with the association, in polarized MDCK cells, FLAG–GPR125 colocalized with endogenous Dlg1 at the basolateral membrane |
| GO:0005515 protein binding | IPI PMID:36115835 Quantitative fragmentomics allow affinity mapping of interac... | MODIFY | Summary: One of eighteen partners from a single quantitative PDZ-domain affinity screen; modified from bare protein binding to the activity the assay actually measures. PMID:36115835 is a holdup assay measuring 65,000 PDZ-domain/PBM affinities. Its ADGRA3 bait is the C-terminal peptide, whose last four residues are Glu-Thr-Thr-Val - the class I PDZ-binding motif UniProt annotates as MOTIF 1319..1321. Every one of the eighteen partners GOA imported from this study is a reviewed Swiss-Prot canonical entry carrying between 2 and 13 PDZ domains, with no TrEMBL clone or partial ORFeome construct among them, so there is no partner-identity defect of the kind found on ACRV1. What the screen establishes is a property of the motif, not eighteen independent discoveries: IntAct holds 117 holdup evidences for ADGRA3 spanning 80 distinct PDZ proteins, of which GOA imported these 18. DLG4 carries 3 PDZ domains and is a postsynaptic/neuronal scaffold. Worth flagging: the postsynaptic compartment is the characterised territory of the paralogue ADGRA1, which is the family member PAINT gives GO:0014069 postsynaptic density and GO:0098978 glutamatergic synapse at the ADGRA1-only node PTN002914505. There is no evidence placing ADGRA3 at a synapse. The most economical reading is that a class I PBM is recognised by class I PDZ domains regardless of where those scaffolds actually reside, which is what a domain-array affinity screen is designed to detect. Real biochemistry, unestablished cell biology. GO:0030165 PDZ domain binding states what was measured and is true of every row in this block; 'protein binding' states nothing. The partner is preserved machine-readably in the has_input extension rather than being discarded by the term change. Reason: Bare protein binding replaced by GO:0030165: the assay measured the affinity of the ADGRA3 class I PBM for the PDZ domains of DLG4, which is the specific activity supported. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:36115835 Here, we measure the affinities of 65,000 interactions involving PDZ domains and their target PDZ-binding motifs (PBM) within a human interactome PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) file:human/ADGRA3/ADGRA3-uniprot.txt FT MOTIF 1319..1321 |
| GO:0005515 protein binding | IPI PMID:36115835 Quantitative fragmentomics allow affinity mapping of interac... | MODIFY | Summary: One of eighteen partners from a single quantitative PDZ-domain affinity screen; modified from bare protein binding to the activity the assay actually measures. PMID:36115835 is a holdup assay measuring 65,000 PDZ-domain/PBM affinities. Its ADGRA3 bait is the C-terminal peptide, whose last four residues are Glu-Thr-Thr-Val - the class I PDZ-binding motif UniProt annotates as MOTIF 1319..1321. Every one of the eighteen partners GOA imported from this study is a reviewed Swiss-Prot canonical entry carrying between 2 and 13 PDZ domains, with no TrEMBL clone or partial ORFeome construct among them, so there is no partner-identity defect of the kind found on ACRV1. What the screen establishes is a property of the motif, not eighteen independent discoveries: IntAct holds 117 holdup evidences for ADGRA3 spanning 80 distinct PDZ proteins, of which GOA imported these 18. TJP1 carries 3 PDZ domains and is a basolateral/junctional polarity or scaffolding protein, so it is topologically compatible with GPR125, which is itself confined to the basolateral membrane domain in polarised epithelium and is retained there through this very motif. The affinity is therefore plausible in cells as well as in vitro. It is nonetheless a single measurement in a single study with no cellular assay for this pair, so the interaction is recorded at the level the data support - PDZ-domain binding by the ETTV motif - and is not claimed as a core function. GO:0030165 PDZ domain binding states what was measured and is true of every row in this block; 'protein binding' states nothing. The partner is preserved machine-readably in the has_input extension rather than being discarded by the term change. Reason: Bare protein binding replaced by GO:0030165: the assay measured the affinity of the ADGRA3 class I PBM for the PDZ domains of TJP1, which is the specific activity supported. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:36115835 Here, we measure the affinities of 65,000 interactions involving PDZ domains and their target PDZ-binding motifs (PBM) within a human interactome PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) file:human/ADGRA3/ADGRA3-uniprot.txt FT MOTIF 1319..1321 PMID:36089063 Consistent with the association, in polarized MDCK cells, FLAG–GPR125 colocalized with endogenous Dlg1 at the basolateral membrane |
| GO:0005515 protein binding | IPI PMID:36115835 Quantitative fragmentomics allow affinity mapping of interac... | MODIFY | Summary: One of eighteen partners from a single quantitative PDZ-domain affinity screen; modified from bare protein binding to the activity the assay actually measures. PMID:36115835 is a holdup assay measuring 65,000 PDZ-domain/PBM affinities. Its ADGRA3 bait is the C-terminal peptide, whose last four residues are Glu-Thr-Thr-Val - the class I PDZ-binding motif UniProt annotates as MOTIF 1319..1321. Every one of the eighteen partners GOA imported from this study is a reviewed Swiss-Prot canonical entry carrying between 2 and 13 PDZ domains, with no TrEMBL clone or partial ORFeome construct among them, so there is no partner-identity defect of the kind found on ACRV1. What the screen establishes is a property of the motif, not eighteen independent discoveries: IntAct holds 117 holdup evidences for ADGRA3 spanning 80 distinct PDZ proteins, of which GOA imported these 18. DLG1 is the anchor of the whole set and the only partner corroborated outside peptide affinity: direct binding to the bacterially expressed GPR125 C-terminal tail, co-IP of full-length GPR125 with FLAG-Dlg1 in HEK293 cells, co-precipitation of endogenous Dlg1 in MDCK cells, and colocalisation at the basolateral membrane - all abolished by deleting the ETTV tetrapeptide. This is the interaction that targets the receptor basolaterally and is required for correct epithelial polarity. GO:0030165 PDZ domain binding states what was measured and is true of every row in this block; 'protein binding' states nothing. The partner is preserved machine-readably in the has_input extension rather than being discarded by the term change. Reason: Bare protein binding replaced by GO:0030165: the assay measured the affinity of the ADGRA3 class I PBM for the PDZ domains of DLG1, which is the specific activity supported. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:36115835 Here, we measure the affinities of 65,000 interactions involving PDZ domains and their target PDZ-binding motifs (PBM) within a human interactome PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) file:human/ADGRA3/ADGRA3-uniprot.txt FT MOTIF 1319..1321 PMID:36089063 These findings indicate that GPR125 associates with Dlg1 in a PBM-dependent manner. PMID:36089063 Similarly, endogenous Dlg1 in MDCK cells was coprecipitated with FLAG–GRP125-FL but not with FLAG–GRP125-ΔETTV PMID:36089063 Consistent with the association, in polarized MDCK cells, FLAG–GPR125 colocalized with endogenous Dlg1 at the basolateral membrane |
| GO:0005515 protein binding | IPI PMID:36115835 Quantitative fragmentomics allow affinity mapping of interac... | MODIFY | Summary: One of eighteen partners from a single quantitative PDZ-domain affinity screen; modified from bare protein binding to the activity the assay actually measures. PMID:36115835 is a holdup assay measuring 65,000 PDZ-domain/PBM affinities. Its ADGRA3 bait is the C-terminal peptide, whose last four residues are Glu-Thr-Thr-Val - the class I PDZ-binding motif UniProt annotates as MOTIF 1319..1321. Every one of the eighteen partners GOA imported from this study is a reviewed Swiss-Prot canonical entry carrying between 2 and 13 PDZ domains, with no TrEMBL clone or partial ORFeome construct among them, so there is no partner-identity defect of the kind found on ACRV1. What the screen establishes is a property of the motif, not eighteen independent discoveries: IntAct holds 117 holdup evidences for ADGRA3 spanning 80 distinct PDZ proteins, of which GOA imported these 18. IL16 carries 4 PDZ domains but is neither a junctional nor a postsynaptic polarity scaffold (Pro-interleukin-16), so no cellular context connects it to ADGRA3. Recorded as measured PDZ-domain affinity with no claim beyond that. GO:0030165 PDZ domain binding states what was measured and is true of every row in this block; 'protein binding' states nothing. The partner is preserved machine-readably in the has_input extension rather than being discarded by the term change. Reason: Bare protein binding replaced by GO:0030165: the assay measured the affinity of the ADGRA3 class I PBM for the PDZ domains of IL16, which is the specific activity supported. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:36115835 Here, we measure the affinities of 65,000 interactions involving PDZ domains and their target PDZ-binding motifs (PBM) within a human interactome PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) file:human/ADGRA3/ADGRA3-uniprot.txt FT MOTIF 1319..1321 |
| GO:0005515 protein binding | IPI PMID:36115835 Quantitative fragmentomics allow affinity mapping of interac... | MODIFY | Summary: One of eighteen partners from a single quantitative PDZ-domain affinity screen; modified from bare protein binding to the activity the assay actually measures. PMID:36115835 is a holdup assay measuring 65,000 PDZ-domain/PBM affinities. Its ADGRA3 bait is the C-terminal peptide, whose last four residues are Glu-Thr-Thr-Val - the class I PDZ-binding motif UniProt annotates as MOTIF 1319..1321. Every one of the eighteen partners GOA imported from this study is a reviewed Swiss-Prot canonical entry carrying between 2 and 13 PDZ domains, with no TrEMBL clone or partial ORFeome construct among them, so there is no partner-identity defect of the kind found on ACRV1. What the screen establishes is a property of the motif, not eighteen independent discoveries: IntAct holds 117 holdup evidences for ADGRA3 spanning 80 distinct PDZ proteins, of which GOA imported these 18. SCRIB carries 4 PDZ domains and is a basolateral/junctional polarity or scaffolding protein, so it is topologically compatible with GPR125, which is itself confined to the basolateral membrane domain in polarised epithelium and is retained there through this very motif. The affinity is therefore plausible in cells as well as in vitro. It is nonetheless a single measurement in a single study with no cellular assay for this pair, so the interaction is recorded at the level the data support - PDZ-domain binding by the ETTV motif - and is not claimed as a core function. GO:0030165 PDZ domain binding states what was measured and is true of every row in this block; 'protein binding' states nothing. The partner is preserved machine-readably in the has_input extension rather than being discarded by the term change. Reason: Bare protein binding replaced by GO:0030165: the assay measured the affinity of the ADGRA3 class I PBM for the PDZ domains of SCRIB, which is the specific activity supported. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:36115835 Here, we measure the affinities of 65,000 interactions involving PDZ domains and their target PDZ-binding motifs (PBM) within a human interactome PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) file:human/ADGRA3/ADGRA3-uniprot.txt FT MOTIF 1319..1321 PMID:36089063 Consistent with the association, in polarized MDCK cells, FLAG–GPR125 colocalized with endogenous Dlg1 at the basolateral membrane |
| GO:0005515 protein binding | IPI PMID:36115835 Quantitative fragmentomics allow affinity mapping of interac... | MODIFY | Summary: One of eighteen partners from a single quantitative PDZ-domain affinity screen; modified from bare protein binding to the activity the assay actually measures. PMID:36115835 is a holdup assay measuring 65,000 PDZ-domain/PBM affinities. Its ADGRA3 bait is the C-terminal peptide, whose last four residues are Glu-Thr-Thr-Val - the class I PDZ-binding motif UniProt annotates as MOTIF 1319..1321. Every one of the eighteen partners GOA imported from this study is a reviewed Swiss-Prot canonical entry carrying between 2 and 13 PDZ domains, with no TrEMBL clone or partial ORFeome construct among them, so there is no partner-identity defect of the kind found on ACRV1. What the screen establishes is a property of the motif, not eighteen independent discoveries: IntAct holds 117 holdup evidences for ADGRA3 spanning 80 distinct PDZ proteins, of which GOA imported these 18. DLG2 carries 3 PDZ domains and is a postsynaptic/neuronal scaffold. Worth flagging: the postsynaptic compartment is the characterised territory of the paralogue ADGRA1, which is the family member PAINT gives GO:0014069 postsynaptic density and GO:0098978 glutamatergic synapse at the ADGRA1-only node PTN002914505. There is no evidence placing ADGRA3 at a synapse. The most economical reading is that a class I PBM is recognised by class I PDZ domains regardless of where those scaffolds actually reside, which is what a domain-array affinity screen is designed to detect. Real biochemistry, unestablished cell biology. GO:0030165 PDZ domain binding states what was measured and is true of every row in this block; 'protein binding' states nothing. The partner is preserved machine-readably in the has_input extension rather than being discarded by the term change. Reason: Bare protein binding replaced by GO:0030165: the assay measured the affinity of the ADGRA3 class I PBM for the PDZ domains of DLG2, which is the specific activity supported. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:36115835 Here, we measure the affinities of 65,000 interactions involving PDZ domains and their target PDZ-binding motifs (PBM) within a human interactome PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) file:human/ADGRA3/ADGRA3-uniprot.txt FT MOTIF 1319..1321 |
| GO:0005515 protein binding | IPI PMID:36115835 Quantitative fragmentomics allow affinity mapping of interac... | MODIFY | Summary: One of eighteen partners from a single quantitative PDZ-domain affinity screen; modified from bare protein binding to the activity the assay actually measures. PMID:36115835 is a holdup assay measuring 65,000 PDZ-domain/PBM affinities. Its ADGRA3 bait is the C-terminal peptide, whose last four residues are Glu-Thr-Thr-Val - the class I PDZ-binding motif UniProt annotates as MOTIF 1319..1321. Every one of the eighteen partners GOA imported from this study is a reviewed Swiss-Prot canonical entry carrying between 2 and 13 PDZ domains, with no TrEMBL clone or partial ORFeome construct among them, so there is no partner-identity defect of the kind found on ACRV1. What the screen establishes is a property of the motif, not eighteen independent discoveries: IntAct holds 117 holdup evidences for ADGRA3 spanning 80 distinct PDZ proteins, of which GOA imported these 18. PDZK1 carries 4 PDZ domains and is a basolateral/junctional polarity or scaffolding protein, so it is topologically compatible with GPR125, which is itself confined to the basolateral membrane domain in polarised epithelium and is retained there through this very motif. The affinity is therefore plausible in cells as well as in vitro. It is nonetheless a single measurement in a single study with no cellular assay for this pair, so the interaction is recorded at the level the data support - PDZ-domain binding by the ETTV motif - and is not claimed as a core function. GO:0030165 PDZ domain binding states what was measured and is true of every row in this block; 'protein binding' states nothing. The partner is preserved machine-readably in the has_input extension rather than being discarded by the term change. Reason: Bare protein binding replaced by GO:0030165: the assay measured the affinity of the ADGRA3 class I PBM for the PDZ domains of PDZK1, which is the specific activity supported. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:36115835 Here, we measure the affinities of 65,000 interactions involving PDZ domains and their target PDZ-binding motifs (PBM) within a human interactome PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) file:human/ADGRA3/ADGRA3-uniprot.txt FT MOTIF 1319..1321 PMID:36089063 Consistent with the association, in polarized MDCK cells, FLAG–GPR125 colocalized with endogenous Dlg1 at the basolateral membrane |
| GO:0005515 protein binding | IPI PMID:36115835 Quantitative fragmentomics allow affinity mapping of interac... | MODIFY | Summary: One of eighteen partners from a single quantitative PDZ-domain affinity screen; modified from bare protein binding to the activity the assay actually measures. PMID:36115835 is a holdup assay measuring 65,000 PDZ-domain/PBM affinities. Its ADGRA3 bait is the C-terminal peptide, whose last four residues are Glu-Thr-Thr-Val - the class I PDZ-binding motif UniProt annotates as MOTIF 1319..1321. Every one of the eighteen partners GOA imported from this study is a reviewed Swiss-Prot canonical entry carrying between 2 and 13 PDZ domains, with no TrEMBL clone or partial ORFeome construct among them, so there is no partner-identity defect of the kind found on ACRV1. What the screen establishes is a property of the motif, not eighteen independent discoveries: IntAct holds 117 holdup evidences for ADGRA3 spanning 80 distinct PDZ proteins, of which GOA imported these 18. FRMPD2 carries 3 PDZ domains and is a basolateral/junctional polarity or scaffolding protein, so it is topologically compatible with GPR125, which is itself confined to the basolateral membrane domain in polarised epithelium and is retained there through this very motif. The affinity is therefore plausible in cells as well as in vitro. It is nonetheless a single measurement in a single study with no cellular assay for this pair, so the interaction is recorded at the level the data support - PDZ-domain binding by the ETTV motif - and is not claimed as a core function. GO:0030165 PDZ domain binding states what was measured and is true of every row in this block; 'protein binding' states nothing. The partner is preserved machine-readably in the has_input extension rather than being discarded by the term change. Reason: Bare protein binding replaced by GO:0030165: the assay measured the affinity of the ADGRA3 class I PBM for the PDZ domains of FRMPD2, which is the specific activity supported. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:36115835 Here, we measure the affinities of 65,000 interactions involving PDZ domains and their target PDZ-binding motifs (PBM) within a human interactome PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) file:human/ADGRA3/ADGRA3-uniprot.txt FT MOTIF 1319..1321 PMID:36089063 Consistent with the association, in polarized MDCK cells, FLAG–GPR125 colocalized with endogenous Dlg1 at the basolateral membrane |
| GO:0005515 protein binding | IPI PMID:36115835 Quantitative fragmentomics allow affinity mapping of interac... | MODIFY | Summary: One of eighteen partners from a single quantitative PDZ-domain affinity screen; modified from bare protein binding to the activity the assay actually measures. PMID:36115835 is a holdup assay measuring 65,000 PDZ-domain/PBM affinities. Its ADGRA3 bait is the C-terminal peptide, whose last four residues are Glu-Thr-Thr-Val - the class I PDZ-binding motif UniProt annotates as MOTIF 1319..1321. Every one of the eighteen partners GOA imported from this study is a reviewed Swiss-Prot canonical entry carrying between 2 and 13 PDZ domains, with no TrEMBL clone or partial ORFeome construct among them, so there is no partner-identity defect of the kind found on ACRV1. What the screen establishes is a property of the motif, not eighteen independent discoveries: IntAct holds 117 holdup evidences for ADGRA3 spanning 80 distinct PDZ proteins, of which GOA imported these 18. MAGI2 carries 6 PDZ domains and is a basolateral/junctional polarity or scaffolding protein, so it is topologically compatible with GPR125, which is itself confined to the basolateral membrane domain in polarised epithelium and is retained there through this very motif. The affinity is therefore plausible in cells as well as in vitro. It is nonetheless a single measurement in a single study with no cellular assay for this pair, so the interaction is recorded at the level the data support - PDZ-domain binding by the ETTV motif - and is not claimed as a core function. GO:0030165 PDZ domain binding states what was measured and is true of every row in this block; 'protein binding' states nothing. The partner is preserved machine-readably in the has_input extension rather than being discarded by the term change. Reason: Bare protein binding replaced by GO:0030165: the assay measured the affinity of the ADGRA3 class I PBM for the PDZ domains of MAGI2, which is the specific activity supported. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:36115835 Here, we measure the affinities of 65,000 interactions involving PDZ domains and their target PDZ-binding motifs (PBM) within a human interactome PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) file:human/ADGRA3/ADGRA3-uniprot.txt FT MOTIF 1319..1321 PMID:36089063 Consistent with the association, in polarized MDCK cells, FLAG–GPR125 colocalized with endogenous Dlg1 at the basolateral membrane |
| GO:0005515 protein binding | IPI PMID:36115835 Quantitative fragmentomics allow affinity mapping of interac... | MODIFY | Summary: One of eighteen partners from a single quantitative PDZ-domain affinity screen; modified from bare protein binding to the activity the assay actually measures. PMID:36115835 is a holdup assay measuring 65,000 PDZ-domain/PBM affinities. Its ADGRA3 bait is the C-terminal peptide, whose last four residues are Glu-Thr-Thr-Val - the class I PDZ-binding motif UniProt annotates as MOTIF 1319..1321. Every one of the eighteen partners GOA imported from this study is a reviewed Swiss-Prot canonical entry carrying between 2 and 13 PDZ domains, with no TrEMBL clone or partial ORFeome construct among them, so there is no partner-identity defect of the kind found on ACRV1. What the screen establishes is a property of the motif, not eighteen independent discoveries: IntAct holds 117 holdup evidences for ADGRA3 spanning 80 distinct PDZ proteins, of which GOA imported these 18. LNX2 carries 4 PDZ domains but is neither a junctional nor a postsynaptic polarity scaffold (Ligand of Numb protein X 2), so no cellular context connects it to ADGRA3. Recorded as measured PDZ-domain affinity with no claim beyond that. GO:0030165 PDZ domain binding states what was measured and is true of every row in this block; 'protein binding' states nothing. The partner is preserved machine-readably in the has_input extension rather than being discarded by the term change. Reason: Bare protein binding replaced by GO:0030165: the assay measured the affinity of the ADGRA3 class I PBM for the PDZ domains of LNX2, which is the specific activity supported. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:36115835 Here, we measure the affinities of 65,000 interactions involving PDZ domains and their target PDZ-binding motifs (PBM) within a human interactome PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) file:human/ADGRA3/ADGRA3-uniprot.txt FT MOTIF 1319..1321 |
| GO:0005515 protein binding | IPI PMID:36115835 Quantitative fragmentomics allow affinity mapping of interac... | MODIFY | Summary: One of eighteen partners from a single quantitative PDZ-domain affinity screen; modified from bare protein binding to the activity the assay actually measures. PMID:36115835 is a holdup assay measuring 65,000 PDZ-domain/PBM affinities. Its ADGRA3 bait is the C-terminal peptide, whose last four residues are Glu-Thr-Thr-Val - the class I PDZ-binding motif UniProt annotates as MOTIF 1319..1321. Every one of the eighteen partners GOA imported from this study is a reviewed Swiss-Prot canonical entry carrying between 2 and 13 PDZ domains, with no TrEMBL clone or partial ORFeome construct among them, so there is no partner-identity defect of the kind found on ACRV1. What the screen establishes is a property of the motif, not eighteen independent discoveries: IntAct holds 117 holdup evidences for ADGRA3 spanning 80 distinct PDZ proteins, of which GOA imported these 18. PATJ carries 10 PDZ domains and is a basolateral/junctional polarity or scaffolding protein, so it is topologically compatible with GPR125, which is itself confined to the basolateral membrane domain in polarised epithelium and is retained there through this very motif. The affinity is therefore plausible in cells as well as in vitro. It is nonetheless a single measurement in a single study with no cellular assay for this pair, so the interaction is recorded at the level the data support - PDZ-domain binding by the ETTV motif - and is not claimed as a core function. GO:0030165 PDZ domain binding states what was measured and is true of every row in this block; 'protein binding' states nothing. The partner is preserved machine-readably in the has_input extension rather than being discarded by the term change. Reason: Bare protein binding replaced by GO:0030165: the assay measured the affinity of the ADGRA3 class I PBM for the PDZ domains of PATJ, which is the specific activity supported. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:36115835 Here, we measure the affinities of 65,000 interactions involving PDZ domains and their target PDZ-binding motifs (PBM) within a human interactome PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) file:human/ADGRA3/ADGRA3-uniprot.txt FT MOTIF 1319..1321 PMID:36089063 Consistent with the association, in polarized MDCK cells, FLAG–GPR125 colocalized with endogenous Dlg1 at the basolateral membrane |
| GO:0005515 protein binding | IPI PMID:36115835 Quantitative fragmentomics allow affinity mapping of interac... | MODIFY | Summary: One of eighteen partners from a single quantitative PDZ-domain affinity screen; modified from bare protein binding to the activity the assay actually measures. PMID:36115835 is a holdup assay measuring 65,000 PDZ-domain/PBM affinities. Its ADGRA3 bait is the C-terminal peptide, whose last four residues are Glu-Thr-Thr-Val - the class I PDZ-binding motif UniProt annotates as MOTIF 1319..1321. Every one of the eighteen partners GOA imported from this study is a reviewed Swiss-Prot canonical entry carrying between 2 and 13 PDZ domains, with no TrEMBL clone or partial ORFeome construct among them, so there is no partner-identity defect of the kind found on ACRV1. What the screen establishes is a property of the motif, not eighteen independent discoveries: IntAct holds 117 holdup evidences for ADGRA3 spanning 80 distinct PDZ proteins, of which GOA imported these 18. DLG3 carries 3 PDZ domains and is a postsynaptic/neuronal scaffold. Worth flagging: the postsynaptic compartment is the characterised territory of the paralogue ADGRA1, which is the family member PAINT gives GO:0014069 postsynaptic density and GO:0098978 glutamatergic synapse at the ADGRA1-only node PTN002914505. There is no evidence placing ADGRA3 at a synapse. The most economical reading is that a class I PBM is recognised by class I PDZ domains regardless of where those scaffolds actually reside, which is what a domain-array affinity screen is designed to detect. Real biochemistry, unestablished cell biology. GO:0030165 PDZ domain binding states what was measured and is true of every row in this block; 'protein binding' states nothing. The partner is preserved machine-readably in the has_input extension rather than being discarded by the term change. Reason: Bare protein binding replaced by GO:0030165: the assay measured the affinity of the ADGRA3 class I PBM for the PDZ domains of DLG3, which is the specific activity supported. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:36115835 Here, we measure the affinities of 65,000 interactions involving PDZ domains and their target PDZ-binding motifs (PBM) within a human interactome PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) file:human/ADGRA3/ADGRA3-uniprot.txt FT MOTIF 1319..1321 |
| GO:0005515 protein binding | IPI PMID:36115835 Quantitative fragmentomics allow affinity mapping of interac... | MODIFY | Summary: One of eighteen partners from a single quantitative PDZ-domain affinity screen; modified from bare protein binding to the activity the assay actually measures. PMID:36115835 is a holdup assay measuring 65,000 PDZ-domain/PBM affinities. Its ADGRA3 bait is the C-terminal peptide, whose last four residues are Glu-Thr-Thr-Val - the class I PDZ-binding motif UniProt annotates as MOTIF 1319..1321. Every one of the eighteen partners GOA imported from this study is a reviewed Swiss-Prot canonical entry carrying between 2 and 13 PDZ domains, with no TrEMBL clone or partial ORFeome construct among them, so there is no partner-identity defect of the kind found on ACRV1. What the screen establishes is a property of the motif, not eighteen independent discoveries: IntAct holds 117 holdup evidences for ADGRA3 spanning 80 distinct PDZ proteins, of which GOA imported these 18. MAGI1 carries 6 PDZ domains and is a basolateral/junctional polarity or scaffolding protein, so it is topologically compatible with GPR125, which is itself confined to the basolateral membrane domain in polarised epithelium and is retained there through this very motif. The affinity is therefore plausible in cells as well as in vitro. It is nonetheless a single measurement in a single study with no cellular assay for this pair, so the interaction is recorded at the level the data support - PDZ-domain binding by the ETTV motif - and is not claimed as a core function. GO:0030165 PDZ domain binding states what was measured and is true of every row in this block; 'protein binding' states nothing. The partner is preserved machine-readably in the has_input extension rather than being discarded by the term change. Reason: Bare protein binding replaced by GO:0030165: the assay measured the affinity of the ADGRA3 class I PBM for the PDZ domains of MAGI1, which is the specific activity supported. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:36115835 Here, we measure the affinities of 65,000 interactions involving PDZ domains and their target PDZ-binding motifs (PBM) within a human interactome PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) file:human/ADGRA3/ADGRA3-uniprot.txt FT MOTIF 1319..1321 PMID:36089063 Consistent with the association, in polarized MDCK cells, FLAG–GPR125 colocalized with endogenous Dlg1 at the basolateral membrane |
| GO:0005515 protein binding | IPI PMID:36115835 Quantitative fragmentomics allow affinity mapping of interac... | MODIFY | Summary: One of eighteen partners from a single quantitative PDZ-domain affinity screen; modified from bare protein binding to the activity the assay actually measures. PMID:36115835 is a holdup assay measuring 65,000 PDZ-domain/PBM affinities. Its ADGRA3 bait is the C-terminal peptide, whose last four residues are Glu-Thr-Thr-Val - the class I PDZ-binding motif UniProt annotates as MOTIF 1319..1321. Every one of the eighteen partners GOA imported from this study is a reviewed Swiss-Prot canonical entry carrying between 2 and 13 PDZ domains, with no TrEMBL clone or partial ORFeome construct among them, so there is no partner-identity defect of the kind found on ACRV1. What the screen establishes is a property of the motif, not eighteen independent discoveries: IntAct holds 117 holdup evidences for ADGRA3 spanning 80 distinct PDZ proteins, of which GOA imported these 18. APBA2 carries 2 PDZ domains and is a postsynaptic/neuronal scaffold. Worth flagging: the postsynaptic compartment is the characterised territory of the paralogue ADGRA1, which is the family member PAINT gives GO:0014069 postsynaptic density and GO:0098978 glutamatergic synapse at the ADGRA1-only node PTN002914505. There is no evidence placing ADGRA3 at a synapse. The most economical reading is that a class I PBM is recognised by class I PDZ domains regardless of where those scaffolds actually reside, which is what a domain-array affinity screen is designed to detect. Real biochemistry, unestablished cell biology. GO:0030165 PDZ domain binding states what was measured and is true of every row in this block; 'protein binding' states nothing. The partner is preserved machine-readably in the has_input extension rather than being discarded by the term change. Reason: Bare protein binding replaced by GO:0030165: the assay measured the affinity of the ADGRA3 class I PBM for the PDZ domains of APBA2, which is the specific activity supported. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:36115835 Here, we measure the affinities of 65,000 interactions involving PDZ domains and their target PDZ-binding motifs (PBM) within a human interactome PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) file:human/ADGRA3/ADGRA3-uniprot.txt FT MOTIF 1319..1321 |
| GO:0005515 protein binding | IPI PMID:36115835 Quantitative fragmentomics allow affinity mapping of interac... | MODIFY | Summary: One of eighteen partners from a single quantitative PDZ-domain affinity screen; modified from bare protein binding to the activity the assay actually measures. PMID:36115835 is a holdup assay measuring 65,000 PDZ-domain/PBM affinities. Its ADGRA3 bait is the C-terminal peptide, whose last four residues are Glu-Thr-Thr-Val - the class I PDZ-binding motif UniProt annotates as MOTIF 1319..1321. Every one of the eighteen partners GOA imported from this study is a reviewed Swiss-Prot canonical entry carrying between 2 and 13 PDZ domains, with no TrEMBL clone or partial ORFeome construct among them, so there is no partner-identity defect of the kind found on ACRV1. What the screen establishes is a property of the motif, not eighteen independent discoveries: IntAct holds 117 holdup evidences for ADGRA3 spanning 80 distinct PDZ proteins, of which GOA imported these 18. GRIP2 carries 7 PDZ domains and is a postsynaptic/neuronal scaffold. Worth flagging: the postsynaptic compartment is the characterised territory of the paralogue ADGRA1, which is the family member PAINT gives GO:0014069 postsynaptic density and GO:0098978 glutamatergic synapse at the ADGRA1-only node PTN002914505. There is no evidence placing ADGRA3 at a synapse. The most economical reading is that a class I PBM is recognised by class I PDZ domains regardless of where those scaffolds actually reside, which is what a domain-array affinity screen is designed to detect. Real biochemistry, unestablished cell biology. GO:0030165 PDZ domain binding states what was measured and is true of every row in this block; 'protein binding' states nothing. The partner is preserved machine-readably in the has_input extension rather than being discarded by the term change. Reason: Bare protein binding replaced by GO:0030165: the assay measured the affinity of the ADGRA3 class I PBM for the PDZ domains of GRIP2, which is the specific activity supported. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:36115835 Here, we measure the affinities of 65,000 interactions involving PDZ domains and their target PDZ-binding motifs (PBM) within a human interactome PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) file:human/ADGRA3/ADGRA3-uniprot.txt FT MOTIF 1319..1321 |
| GO:0005515 protein binding | IPI PMID:36115835 Quantitative fragmentomics allow affinity mapping of interac... | MODIFY | Summary: One of eighteen partners from a single quantitative PDZ-domain affinity screen; modified from bare protein binding to the activity the assay actually measures. PMID:36115835 is a holdup assay measuring 65,000 PDZ-domain/PBM affinities. Its ADGRA3 bait is the C-terminal peptide, whose last four residues are Glu-Thr-Thr-Val - the class I PDZ-binding motif UniProt annotates as MOTIF 1319..1321. Every one of the eighteen partners GOA imported from this study is a reviewed Swiss-Prot canonical entry carrying between 2 and 13 PDZ domains, with no TrEMBL clone or partial ORFeome construct among them, so there is no partner-identity defect of the kind found on ACRV1. What the screen establishes is a property of the motif, not eighteen independent discoveries: IntAct holds 117 holdup evidences for ADGRA3 spanning 80 distinct PDZ proteins, of which GOA imported these 18. GRIP1 carries 7 PDZ domains and is a postsynaptic/neuronal scaffold. Worth flagging: the postsynaptic compartment is the characterised territory of the paralogue ADGRA1, which is the family member PAINT gives GO:0014069 postsynaptic density and GO:0098978 glutamatergic synapse at the ADGRA1-only node PTN002914505. There is no evidence placing ADGRA3 at a synapse. The most economical reading is that a class I PBM is recognised by class I PDZ domains regardless of where those scaffolds actually reside, which is what a domain-array affinity screen is designed to detect. Real biochemistry, unestablished cell biology. GO:0030165 PDZ domain binding states what was measured and is true of every row in this block; 'protein binding' states nothing. The partner is preserved machine-readably in the has_input extension rather than being discarded by the term change. Reason: Bare protein binding replaced by GO:0030165: the assay measured the affinity of the ADGRA3 class I PBM for the PDZ domains of GRIP1, which is the specific activity supported. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:36115835 Here, we measure the affinities of 65,000 interactions involving PDZ domains and their target PDZ-binding motifs (PBM) within a human interactome PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) file:human/ADGRA3/ADGRA3-uniprot.txt FT MOTIF 1319..1321 |
| GO:0016020 membrane | ISS GO_REF:0000024 | MODIFY | Summary: An inference that landed one step above the very donor it was drawn from, when the donor's own term was available for free. The single WITH/FROM entity is UniProtKB:S4X0Q8, zebrafish adgra3 - the reviewed 1:1 orthologue, so the entity type is right for an ISS row (sequence-similar entity, not an interactor). But QuickGO shows that donor carries GO:0005886 plasma membrane by IDA from PMID:23821037, not merely 'membrane'. GO:0005886 is a direct is_a child of GO:0016020 - one edge, confirmed on the QuickGO graph endpoint - so the transfer gave up the single step of precision its own source supplies. The step is small but it is free: this is the same defect shape recorded on ACRV1, where the loss was larger. There is no heterogeneous-donor excuse available: the WITH/FROM has exactly one entity, it agrees with the target completely, and human ADGRA3 has independently been shown at the plasma membrane. Modified downward to the term the donor actually holds. Reason: The sole ISS donor, zebrafish adgra3 (S4X0Q8), holds GO:0005886 by IDA; the propagated term is its direct is_a parent, one edge above, with no heterogeneous clade forcing even that generalisation. Propagation Review Root cause: TERM SCOPING PROBLEM Failure modes: GRANULARITY MISMATCH Sources checked: UniProtKB:S4X0Q8 · zebrafish adgra3 (reviewed/Swiss-Prot, 1346 aa) - the 1:1 orthologue and the only entity on this row; holds GO:0005886 plasma membrane by IDA from PMID:23821037, i.e. a term more specific than the one propagated SUPPORTS TRANSFER Proposed replacements: plasma membrane Supporting Evidence: file:human/ADGRA3/ADGRA3-uniprot.txt CC -!- SUBCELLULAR LOCATION: Membrane {ECO:0000250|UniProtKB:S4X0Q8}; Multi- PMID:31659746 In summary, GPR125 is mainly expressed on the cell surface and undergoes constitutive internalization under physiological temperatures |
| GO:0004930 G protein-coupled receptor activity | TAS PMID:15203201 The human and mouse repertoire of the adhesion family of G-p... | ACCEPT | Summary: Right conclusion, weak provenance - kept because the conclusion has since been earned. This is a 2004 GDB annotation citing a bioinformatic repertoire paper. Its basis for calling GPR125 a G protein-coupled receptor is sequence classification into the 33-member human adhesion GPCR family; no coupling assay appears anywhere in it, so as a Traceable Author Statement it traces to an inference rather than to a measurement. That is precisely the pattern this campaign is calibrated to distrust. The distrust is not warranted here, because the inference has since been tested and confirmed: ADGRA3 activates Gi and Gs in human cell lines, and abolishing the stachel tethered agonist abolishes it. Accepted, with the recommendation that GOA re-evidence the term as IDA against PMID:40127866 rather than leaving a demonstrated molecular function resting on a TAS to a catalogue paper. Reason: Term is correct and now experimentally demonstrated; only the evidence provenance (TAS to a sequence-catalogue paper) is inadequate, and that is a re-evidencing recommendation, not a term error. Supporting Evidence: PMID:15203201 The adhesion G-protein-coupled receptors (GPCRs) (also termed LN-7TM or EGF-7TM receptors) are membrane-bound proteins with long N-termini containing multiple domains. PMID:40127866 We found low-level activation of Gi and Gs by ADGRA3 and slightly more by its CTF. PMID:40127866 Collectively, this establishes classical G protein-mediated signaling for ADGRA3. |
| GO:0007186 G protein-coupled receptor signaling pathway | TAS PMID:15203201 The human and mouse repertoire of the adhesion family of G-p... | ACCEPT | Summary: Same provenance issue and same resolution as the GO:0004930 TAS row. The cited repertoire paper contains no signalling experiment on GPR125; the annotation is a family-membership inference recorded as a Traceable Author Statement. The process it asserts has since been demonstrated directly for human ADGRA3 - Gi and Gs activation, stachel-dependent, and sustained in DVL1-3-null cells, so it is a G protein pathway in its own right rather than a Wnt readout. Accepted, with the same recommendation to re-evidence against PMID:40127866. Reason: Correct process, now directly demonstrated; the TAS provenance should be replaced with the experimental reference. Supporting Evidence: PMID:15203201 The adhesion G-protein-coupled receptors (GPCRs) (also termed LN-7TM or EGF-7TM receptors) are membrane-bound proteins with long N-termini containing multiple domains. PMID:40127866 Collectively, this suggests that ADGRA3 is capable of activating Gai signaling axes independently of DVL presence. PMID:40127866 Collectively, this establishes classical G protein-mediated signaling for ADGRA3. |
| GO:0016020 membrane | TAS PMID:15203201 The human and mouse repertoire of the adhesion family of G-p... | MODIFY | Summary: The weakest of the three membrane rows, and the least precise. PMID:15203201 is a genome-wide catalogue of the adhesion GPCR repertoire: it identifies GPR133 and GPR144, assigns orthologue pairs, and charts EST expression. Its only statement bearing on this term is the definitional one that adhesion GPCRs are membrane-bound proteins with long N-termini - a family-level generality, not an observation about GPR125. Two decades later the localisation of the human protein is established at plasma-membrane and basolateral resolution. Modified rather than removed for the same reason as the other two membrane rows: the claim is true, it is simply superseded. Reason: The cited paper is a sequence-based repertoire survey whose membrane statement is definitional for the family; the specific compartment is now experimentally established for the human protein. Proposed replacements: plasma membrane Supporting Evidence: PMID:15203201 The adhesion G-protein-coupled receptors (GPCRs) (also termed LN-7TM or EGF-7TM receptors) are membrane-bound proteins with long N-termini containing multiple domains. PMID:31659746 In summary, GPR125 is mainly expressed on the cell surface and undergoes constitutive internalization under physiological temperatures PMID:36089063 Consistent with the association, in polarized MDCK cells, FLAG–GPR125 colocalized with endogenous Dlg1 at the basolateral membrane |
| GO:0030165 PDZ domain binding | IPI PMID:36089063 GPR125 (ADGRA3) is an autocleavable adhesion GPCR that traff... | NEW | Summary: Proposed so that the gene's best-evidenced molecular function rests on a cellular experiment rather than only on peptide-array affinity. The nineteen GO:0005515 rows are all modified to GO:0030165 above, but every one of them cites an in vitro domain/peptide assay. This row records the interaction as it was demonstrated in cells: FLAG-Dlg1 co-immunoprecipitates full-length Myc-GPR125 from HEK293 cells, endogenous Dlg1 co-precipitates with FLAG-GPR125 from MDCK cells, and both are abolished by deleting the four C-terminal residues ETTV. The deletion control is what makes this a PDZ-domain binding assignment rather than a generic association. IPI rather than IDA because the assay is a co-immunoprecipitation identifying the physical partner; the partner is carried in the has_input extension. Reason: The C-terminal ETTV class I PDZ-binding motif mediates a co-immunoprecipitable interaction with DLG1 in both HEK293 and MDCK cells, lost on deletion of the motif; GOA currently records this only as bare protein binding from in vitro peptide screens. Supporting Evidence: PMID:36089063 The CTT of GPR125 contains the extreme C-terminal tetrapeptide Glu-Thr-Thr-Val (amino acid residues 1318–1321), which corresponds to the type I PDZ domain–binding motif (PBM) PMID:36089063 These findings indicate that GPR125 associates with Dlg1 in a PBM-dependent manner. PMID:36089063 Similarly, endogenous Dlg1 in MDCK cells was coprecipitated with FLAG–GRP125-FL but not with FLAG–GRP125-ΔETTV PMID:15021905 hDlg furthermore bound a novel seven-pass transmembrane protein, which was homologous to TEM5, and was named here a TEM5-like protein (TEM5-like). file:human/ADGRA3/ADGRA3-uniprot.txt CC -!- SUBUNIT: Interacts (via PDZ-binding motif) with DLG1. |
| GO:0016323 basolateral plasma membrane | IDA PMID:36089063 GPR125 (ADGRA3) is an autocleavable adhesion GPCR that traff... | NEW | Summary: The one compartment statement in the ADGRA3 record that is both specific and directly measured on the human protein, and it is absent from GOA. In polarised MDCK monolayers, human GPR125 is confined to the basolateral domain of the plasma membrane and colocalises there with endogenous Dlg1. Targeting requires the C-terminal PBM: a construct lacking ETTV is not retained basolaterally, and neither is a GPS-tripeptide mutant that cannot self-cleave. The protein assayed is human, so this is an annotation of the human gene product even though the host cell line is canine. This term also supersedes the three generic GO:0016020 rows in informativeness, and sits below the GO:0005886 that the IBA already supplies. Reason: Human GPR125 expressed in polarised MDCK cells is exclusively basolateral and colocalises with endogenous Dlg1 there; GOA has no compartment term more specific than plasma membrane. Supporting Evidence: PMID:36089063 in polarized MDCK cells, GPR125 is exclusively recruited to the basolateral domain of the plasma membrane PMID:36089063 Consistent with the association, in polarized MDCK cells, FLAG–GPR125 colocalized with endogenous Dlg1 at the basolateral membrane |
| GO:0016540 protein autoprocessing | IDA PMID:36089063 GPR125 (ADGRA3) is an autocleavable adhesion GPCR that traff... | NEW | Summary: A demonstrated property of the human protein that contradicts the standing prediction, and a term no human adhesion GPCR currently carries. Human GPR125 expressed in HEK293 and MDCK cells is cleaved into an N-terminal and a C-terminal fragment that remain associated, cleavage is lost in GAIN-domain deletions, and substituting the GPS tripeptide Ser736-Leu737-Ser738 with alanines abolishes it. Cleavage occurs early in biosynthesis. This is GAIN-mediated cis-autoproteolysis, which GO:0016540 defines as processing a protein carries out on itself to generate its mature form. It matters twice over. First, it overturns the prediction that ADGRA3 is non-cleavable because its GPS tripeptide is Ser-Leu-Ser rather than the canonical His-Leu/Ile-Ser/Thr - a prediction UniProt still records. Second, autoproteolysis is what liberates the stachel tethered agonist whose removal abolishes ADGRA3's G protein signalling, so it is mechanistically upstream of the receptor's only demonstrated molecular function. A QuickGO census of GO:0016540 and its descendants in human returns 68 annotations across 36 gene products, and not one is an adhesion GPCR - so the defining biochemical feature of the aGPCR class is unannotated in GO for all 33 human members. Raised for the family in suggested_questions; proposed here for the gene where the experiment exists. Reason: Direct demonstration of cis-autoproteolysis of human GPR125 at the atypical GPS motif Ser736-Leu737-Ser738, abolished by alanine substitution of the tripeptide; no human adhesion GPCR currently carries this term. Supporting Evidence: PMID:36089063 In the present study, we show that human GPR125, expressed in kidney epithelial Madin-Darby canine kidney (MDCK) cells and embryonic kidney HEK293 cells, undergoes cis-autoproteolysis at the atypical GPS during an early stage of receptor biosynthesis. PMID:36089063 GPR125 has been predicted to be noncleavable in the GAIN domain because of its atypical GPS motif (Ser736-Leu737-Ser738) but not the canonical one (His-Leu/Ile-Ser/Thr) file:human/ADGRA3/ADGRA3-uniprot.txt ADGRA3 is predicted non-cleavable PMID:40127866 The GPS in most ADGRA3 vertebrate orthologs is SL|S/G (where | denotes the potential cleavage point). |
| GO:0007189 adenylate cyclase-activating G protein-coupled receptor signaling pathway | IDA PMID:40127866 Complex G-protein signaling of the adhesion GPCR, ADGRA3. | NEW | Summary: The Gs arm of the coupling that finally removed ADGRA3 from orphan status, recorded at the specific level the assays reach. In human cell lines ADGRA3 and its GPS-truncated C-terminal fragment give a CRE-reporter response that is abolished in Gs-knockout HEK293 cells, and removing the first three residues of the CTF - the stachel tethered agonist - abolishes G protein signalling altogether. Independently, ADGRA3 is reported to act as a constitutively active receptor driving a Gs-PKA-CREB thermogenic programme in adipocytes without added ligand. GOA currently holds only GO:0007186, the generic parent, from InterPro and a 2004 TAS. Proposed as an addition rather than a replacement, since the parent remains correct and also covers the Gi arm. Reason: Gs-dependent CRE responses to ADGRA3 and its CTF in human cell lines, lost in Gs-null cells and on removal of the stachel peptide, with corroborating constitutive Gs-PKA-CREB signalling in adipocytes. Supporting Evidence: PMID:40127866 We found low-level activation of Gi and Gs by ADGRA3 and slightly more by its CTF. PMID:40127866 This resulted in abrogated G protein-mediated signaling, as observed for other aGPCRs. PMID:40127866 Collectively, this establishes classical G protein-mediated signaling for ADGRA3. PMID:39718208 Conversely, Adgra3 overexpression activated the adipose thermogenic program and improved metabolic homeostasis in mice without exogenous ligand. |
| GO:0007193 adenylate cyclase-inhibiting G protein-coupled receptor signaling pathway | IDA PMID:40127866 Complex G-protein signaling of the adhesion GPCR, ADGRA3. | NEW | Summary: The Gi arm of the same study, proposed alongside the Gs term because the receptor demonstrably engages both. ADGRA3 activates Gi as well as Gs; Gi activity is read out both as a downshifted CRE response when Galpha-s is depleted and directly with a BRET G protein activity sensor, and it is sustained in DVL1-3-null cells. The DVL-null control matters: it shows this is genuine heterotrimeric G protein signalling and not a Dishevelled-dependent Wnt/PCP readout in disguise, which is the obvious alternative given the zebrafish literature. Proposing both GO:0007189 and GO:0007193 is deliberate. The paper's own framing is that the signalling is complex and balanced between the two axes, so recording only one would misstate the result. Reason: Direct BRET and reporter evidence of Gi activation by ADGRA3 and its CTF in human cells, sustained in the absence of DVL1-3. Supporting Evidence: PMID:40127866 We found low-level activation of Gi and Gs by ADGRA3 and slightly more by its CTF. PMID:40127866 Collectively, this suggests that ADGRA3 is capable of activating Gai signaling axes independently of DVL presence. PMID:40127866 Collectively, this establishes classical G protein-mediated signaling for ADGRA3. |
| GO:0031623 receptor internalization | IDA PMID:31659746 Arrestin-independent constitutive endocytosis of GPR125/ADGR... | NEW | Summary: Closes the one annotatable claim this review asserted in prose without proposing a term - a gap that would sit badly in a review whose central thesis is that absence is this gene's dominant curation defect. Human GPR125 carrying an N-terminal FLAG tag is quantified at the surface of non-permeabilised HEK293 cells by ELISA, then shown to disappear from the surface over 30 min at 37 degrees and to appear in intracellular puncta that moderately overlap GFP-TfR1. Internalisation is unaffected in beta-arrestin-1/2 double-knockout cells and proceeds to the same extent as agonist-driven internalisation of the control receptor GPR183. The authors' conclusion is that this constitutive retrieval sets the receptor's surface level, which matters here because surface level is what gates every other function ADGRA3 has - G protein coupling and PDZ-scaffold engagement both happen at the plasma membrane. Two term choices were tested against their definitions rather than their labels. GO:0072583 clathrin-dependent endocytosis was rejected: in human it is annotated to 19 gene products and every one is endocytic machinery - AP2 subunits, FCHO1/2, PICALM, SGIP1, SNAP91, HSPA8 - so placing a cargo receptor there would be role conflation. GO:0031623 is the term GO actually uses for internalised receptors, including cargo: TFRC itself holds it involved_in by IDA, and TFRC is the constitutively recycling receptor used as the marker in this very experiment. One caveat is recorded rather than smoothed over. GO:0031623's definition states that the process 'begins when cell surface receptors are monoubiquitinated following ligand-induced activation', and ADGRA3's internalisation is neither ligand-induced nor beta-arrestin-dependent. GO's curated usage is plainly broader than that definition - TFRC is the proof - but the mismatch is real and is raised as an ontology point in suggested_questions. Reason: Constitutive, beta-arrestin-independent, clathrin-dependent internalisation of human GPR125 from the surface of HEK293 cells, controlling receptor surface expression; asserted in this review's description and core functions but absent from GOA. GO:0072583 was considered and rejected because its human usage is confined to endocytic machinery rather than cargo. Supporting Evidence: PMID:31659746 In summary, GPR125 is mainly expressed on the cell surface and undergoes constitutive internalization under physiological temperatures PMID:31659746 Moreover, we show that the internalization of GPR125 happens in a β‐arrestin−independent, but TfR1 colocalizing/clathrin‐dependent manner. PMID:31659746 These data support that the constitutive internalization of GPR125 contributes to its biological functions by controlling receptor surface expression |
| GO:0045197 establishment or maintenance of epithelial cell apical/basal polarity | ISS PMID:36089063 GPR125 (ADGRA3) is an autocleavable adhesion GPCR that traff... | NEW | Summary: Proposed, but deliberately as ISS rather than IMP, because the loss-of-function was not performed on the human gene. Two CRISPR GPR125-knockout MDCK lines form Matrigel cysts with multiple lumens, and their mitotic spindles are randomised relative to the apical-basal axis instead of lying perpendicular to it; siRNA knockdown of GPR125 or of Dlg1 reproduces the multilumen phenotype. That establishes the process for the gene product that was removed - endogenous canine Gpr125 (the nearest UniProt entity is the unreviewed TrEMBL entry A0A8I3MZC6, 1313 aa; there is no reviewed dog entry). What is directly human in the same paper is the molecular arm: human GPR125 is basolaterally targeted through its ETTV motif and co-immunoprecipitates DLG1. Human and dog ADGRA3 are close orthologues and the polarity machinery is shared, so the inference is sound - but calling it IMP would overstate which organism's gene was perturbed, and this campaign has repeatedly paid for exactly that kind of slippage. Recorded honestly at ISS with the canine entity in supporting_entities, per the rule that ISS WITH/FROM takes the sequence-similar entity. Reason: Knockout and knockdown of endogenous Gpr125 in MDCK cells randomise mitotic spindle orientation and produce multilumen cysts; the human orthologue supplies the molecular arm (ETTV-dependent DLG1 binding and basolateral targeting) but was not itself the perturbed gene, so the transfer is by similarity. Supporting Evidence: PMID:36089063 GPR125 is indispensable for correct cystogenesis and mitotic spindle orientation in 3D-cultured MDCK cells, indicating its crucial role in epithelial apicobasal polarization. PMID:36089063 Consistent with the association, in polarized MDCK cells, FLAG–GPR125 colocalized with endogenous Dlg1 at the basolateral membrane PMID:36089063 These findings indicate that GPR125 associates with Dlg1 in a PBM-dependent manner. |
Loading supporting content…
Download this section (compressed HTML)Q: UniProt correction. The Q8IWK6 entry still states, under MISCELLANEOUS, that 'ADGRA3 is predicted non-cleavable because of the lack of a consensus catalytic triad sequence within GPS region' (ECO:0000305, a curator inference). PMID:36089063 tested that prediction and refuted it: human GPR125 expressed in HEK293 and MDCK cells is cleaved into associated N- and C-terminal fragments, cleavage requires the GAIN domain, and alanine substitution of the atypical GPS tripeptide Ser736-Leu737-Ser738 abolishes it. PMID:40127866 concurs that the ADGRA3 GPS is SL|S/G across vertebrates and builds its CTF constructs on that cleavage point. Should the MISCELLANEOUS line be replaced with a PTM/PROCESSING statement citing PMID:36089063, and a SITE or CHAIN feature added at the GPS?
Suggested experts: UniProt curators, adhesion GPCR structural biologists
Q: UniProt correction, second item. The SUBCELLULAR LOCATION line reads 'Membrane {ECO:0000250|UniProtKB:S4X0Q8}; Multi-pass membrane protein' - a by-similarity inference from the zebrafish orthologue made before the human protein was localised. Human GPR125 is now shown at the surface of non-permeabilised HEK293 cells (PMID:31659746) and confined to the basolateral domain in polarised MDCK cells (PMID:36089063). Should this become 'Cell membrane', with a Note recording basolateral restriction in polarised epithelia? The three GO:0016020 rows in GOA all trace back to this line, so correcting it upstream fixes the GO record too.
Suggested experts: UniProt curators
Q: PAINT node placement, and it is the reciprocal of the question this family invites. PANTHER node PTN002914494 has a reach of exactly the three ADGRA3 orthologues - human ADGRA3, mouse Adgra3, zebrafish adgra3 - and the single term it propagates to them is GO:0009897 external side of plasma membrane. Yet one member of that very node, zebrafish adgra3, carries GO:0060071 Wnt signaling pathway planar cell polarity pathway by IDA and GO:2000095 by IMP and IPI from PMID:23821037, and neither term reaches human or mouse. The terms are not IBA-ineligible: QuickGO returns 70 IBA annotations to GO:0060071 and 36 to GO:2000095 across many taxa, including human. Meanwhile the ADGRA2-only node PTN002914520 and the ADGRA1-only node PTN002914505 both do propagate their members' specific biology (sprouting angiogenesis, Wnt signalosome assembly and positive regulation of canonical Wnt signalling for ADGRA2; postsynaptic density and glutamatergic synapse for ADGRA1). Is the ADGRA3 node's silence a deliberate judgement that gastrulation PCP is fish-specific, or a coverage gap? If deliberate, an IRD or a recorded note would make it legible; if not, it is the one place in this family where PAINT under-transfers.
Suggested experts: GO Central PAINT curators
Q: Family-wide ontology and curation gap. GAIN-domain cis-autoproteolysis is the defining biochemical feature of the adhesion GPCR class, and it is what liberates the stachel tethered agonist that these receptors signal through. A QuickGO census of GO:0016540 protein autoprocessing and its descendants in human returns 68 annotations over 36 gene products - caspases, hedgehog ligands, MYRF, PCSK9, SPRTN and so on - and not a single adhesion GPCR, although cleavage has been demonstrated for many of the 33 human members. Is GO:0016540 the intended term for GAIN autoproteolysis, or is a dedicated child needed? Either way the class appears to be systematically unannotated for its own hallmark reaction.
Suggested experts: GO ontology editors, adhesion GPCR consortium (IUPHAR)
Q: Ontology definition versus curated usage on GO:0031623 receptor internalization. Its definition states that the process 'begins when cell surface receptors are monoubiquitinated following ligand-induced activation', yet the term is curated onto receptors whose internalisation is constitutive and ligand-independent - TFRC holds it involved_in by IDA, and TFRC is the textbook constitutive recycler. ADGRA3 is the same case: beta-arrestin-independent, clathrin-dependent, no ligand known. The alternative GO:0072583 clathrin-dependent endocytosis is not usable for cargo, since all 19 human gene products annotated to it are endocytic machinery. Should the GO:0031623 definition be broadened to match its usage, or is a constitutive-internalisation sibling needed?
Suggested experts: GO ontology editors, membrane trafficking curators
Q: Evidence-code upgrade in GOA. Three of ADGRA3's annotations - GO:0004930, GO:0007186 and GO:0016020 - are TAS rows made by GDB in 2005 citing PMID:15203201, a bioinformatic survey of the adhesion GPCR repertoire that contains no experiment on GPR125. Two of those three terms have since been demonstrated directly (PMID:40127866). Should these be re-evidenced as IDA against the experimental paper rather than left as twenty-year-old traceable author statements pointing at a sequence catalogue?
Suggested experts: GOA curators
Q: How should a class I PDZ-binding motif that binds many PDZ domains in vitro be annotated? IntAct holds 117 holdup-assay evidences for ADGRA3 spanning 80 distinct PDZ proteins from PMID:36115835 alone, of which GOA imported 18. All 18 are reviewed canonical PDZ proteins, so there is no partner-identity problem, but the set is a property of the ETTV motif rather than 18 independent biological findings - and it includes seven postsynaptic scaffolds (DLG2, DLG3, DLG4, GRIP1, GRIP2, APBA2, GRID2IP) in a compartment where ADGRA3 has never been observed and where its paralogue ADGRA1 is the characterised family member. Would a single GO:0030165 row with the measured partners in extensions serve users better than 18 GO:0005515 rows?
Suggested experts: GOA curators, IntAct curators
Experiment: Generate two knock-in mouse lines: Adgra3-DeltaETTV, which cannot bind PDZ scaffolds but retains autoproteolysis and G protein coupling, and Adgra3-GPS(AAA), which cannot self-cleave and therefore cannot expose the stachel agonist. Score both for the published null phenotypes - ejaculatory-duct obstruction in males, imperforate vagina in females - and for epithelial architecture in the affected ducts. The two alleles dissociate the receptor's two demonstrated molecular capabilities, which no existing experiment does.
Hypothesis: ADGRA3's urogenital knockout phenotypes are caused by loss of PDZ-scaffold-dependent epithelial polarity rather than by loss of G protein signalling.
Type: mouse genetics / separation-of-function alleles
Experiment: Knock out ADGRA3 in a human polarised epithelial model (Caco-2 or human intestinal or kidney organoids), score lumen number and mitotic spindle orientation in 3D culture, then rescue with wild-type ADGRA3 versus the Delta-ETTV and GPS(AAA) mutants. This would convert the proposed GO:0045197 annotation from ISS to IMP on the human gene and would establish whether polarity requires the motif, the cleavage, or both.
Hypothesis: The apicobasal-polarity function demonstrated in canine MDCK cells is conserved in human epithelium and depends on the ETTV motif.
Type: human organoid CRISPR knockout with structure-function rescue
Experiment: In ADGRA3-null human cells and in Adgra3-null mice, assay a mammalian PCP readout that does not involve beta-catenin: asymmetric membrane localisation of VANGL2 and PRICKLE, oriented cell division in a PCP-dependent epithelium, and cochlear hair-cell stereocilia orientation. Pair with a DVL-binding-deficient ADGRA3 intracellular-domain mutant. This tests the one function UniProt asserts for the human protein and that GOA does not record, and it distinguishes the PCP branch from the canonical Wnt branch that ADGRA3 was already shown not to engage.
Hypothesis: Human ADGRA3 retains the Dishevelled-dependent planar-cell-polarity output demonstrated for zebrafish adgra3, separately from its DVL-independent G protein coupling.
Type: planar cell polarity readouts in mammalian cells and mouse
Experiment: Screen the purified ADGRA3 ectodomain (LRR-Ig-HBD-GAIN, residues ~34-761) against an extracellular-protein interaction library by avidity-based extracellular interaction screening, and test hits for their effect on cAMP output of full-length ADGRA3 versus the constitutively active CTF. Comparing the two constructs separates a true agonist from a modulator of autoproteolysis or of tethered-agonist accessibility, and the ectodomain is the ~730-residue region for which no molecular function is currently assignable.
Hypothesis: ADGRA3 has an endogenous extracellular ligand or adhesion partner that modulates its constitutive activity.
Type: extracellular interactome screen with functional counter-assay
Experiment: Perform proximity labelling (TurboID or APEX2) from the ADGRA3 C-terminus in a polarised human epithelial line and in a cell type where ADGRA3 is natively expressed, and compare the recovered PDZ proteins with the 80 holdup partners and the 18 GOA-imported ones. Include the Delta-ETTV mutant as the specificity control. This would show which of the measured affinities are realised in a cell and would give GOA a principled basis for which partners to retain.
Hypothesis: The eighteen PDZ partners imported into GOA from the holdup screen are largely in vitro consequences of the ETTV motif rather than cellular interactions of ADGRA3.
Type: proximity-dependent biotinylation with motif-deletion control
What is not known — curated, literature-grounded statements of the open unknowns (the inverse of core functions).
Gap: ADGRA3 has no known ligand. Every signalling result to date is constitutive activity of the receptor or of its cleaved C-terminal fragment; nothing extracellular has been shown to bind the receptor and change its output. The large ectodomain - a leucine-rich-repeat block, two immunoglobulin domains, a hormone-binding domain and the GAIN domain, roughly 730 residues - is therefore functionally unassigned, and no GO molecular function describes what it does.
NARROWING BIOLOGY MF_DARK
What is known: What is established is the transduction machinery: cis-autoproteolysis at the atypical GPS Ser736-Leu737-Ser738, a stachel tethered agonist whose first three residues are required, and coupling to both Gs and Gi in human cells. What is missing is the input. A flavonoid, hesperetin, has been put forward as a candidate agonist in adipocytes, but that is a single pharmacological report and not an endogenous ligand.
Significance: The receptor is a proposed target for adipose thermogenesis and for osteoporosis, and target validation in both cases rests on constitutive activity rather than on a ligand-receptor pair. Without a ligand the ectodomain's adhesion-related domains cannot be assigned a binding function, which is the main reason the molecular-function half of this gene's GO record consists of one receptor-activity term and nineteen rows of bare protein binding.
What would resolve it: Deorphanisation: an unbiased extracellular-interactome screen against the purified ADGRA3 ectodomain, combined with a comparison of full-length versus stachel-exposed constructs to distinguish a true agonist from a modulator of autoproteolysis or of tethered-agonist accessibility.
Provenance (the field's own admissions):
Gap: Whether the Wnt/planar-cell-polarity role established for zebrafish adgra3 holds for the human protein is unresolved, and the available human evidence points in two directions at once. Zebrafish Gpr125 recruits Dishevelled to the membrane and modulates convergence-extension; UniProt transfers this to human ADGRA3 by similarity as its only FUNCTION statement. But human ADGRA3 produced no beta-catenin/TCF-LEF response, and its G protein signalling is unchanged in DVL1-3-null cells, and an independent overexpression study in HCT116 cells found GPR125 to suppress beta-catenin transcriptional activity outright - while the same laboratory that showed the DLG1-dependent polarity role also detected a PBM-independent interaction of human GPR125 with Dvl1.
OPEN BIOLOGYCURATION RESIDUAL_SUBGAP
What is known: The negatives are specific and do not settle the question: absence of beta-catenin signalling says nothing about the non-canonical PCP branch, and DVL-independence of G protein coupling does not exclude a separate DVL-dependent output. No mammalian PCP readout - Vangl/Prickle asymmetry, oriented cell division in a PCP-dependent tissue, convergent extension - has been assayed with ADGRA3 perturbed.
Significance: This is the only proposed function that would connect ADGRA3's molecular activity to the mouse knockout phenotypes, since obstructive azoospermia and imperforate vagina are both tubular-morphogenesis defects of the kind PCP signalling produces. It is also the reason UniProt and GOA currently disagree: UniProt asserts a PCP role by similarity while GOA carries no Wnt or PCP term for human ADGRA3 at all.
What would resolve it: Assay a mammalian PCP output directly in ADGRA3-null cells or mice - core PCP protein asymmetry, oriented cell division, or cochlear hair-cell orientation - rather than continuing to infer the pathway from the zebrafish gastrulation phenotype.
Provenance (the field's own admissions):
Gap: None of the seven papers that experimentally characterised ADGRA3 has produced a single GO annotation. QuickGO returns zero annotations, for any species and any gene, for each of PMID:36089063 (autoproteolysis, DLG1, basolateral targeting, apicobasal polarity), PMID:40127866 (Gs and Gi coupling), PMID:31659746 (surface expression and constitutive clathrin-dependent endocytosis), PMID:39718208 (adipose thermogenesis), PMID:35414778 (osteoclastogenesis), PMID:36688818 (obstructive azoospermia) and PMID:38589878 (imperforate vagina). The entire human ADGRA3 GO record consists of three IBA rows, five IEA rows, one ISS row, three TAS rows citing a 2004 sequence-catalogue paper, and nineteen bare protein-binding rows from two PDZ affinity screens.
OPEN CURATION RESIDUAL_SUBGAP
What is known: This is a curation backlog, not an absence of evidence, and it is why the gene reads as dark from GOA alone while UniProt records it as PE1 with a curated DLG1 SUBUNIT line. The bibliography is discoverable: three of the seven papers are cached with full text and two more are indexed under GPR125 rather than ADGRA3.
Significance: Six of this review's proposed annotations come from papers GO has never touched. A reader consulting GOA would conclude ADGRA3 is an uncharacterised orphan whose only molecular function is inferred from a family signature, when in fact its G protein coupling, its autoproteolysis, its polarity function and its surface trafficking have all been measured.
What would resolve it: Curate PMID:36089063 and PMID:40127866 first: between them they supply a molecular function, two specific signalling processes, a specific compartment, a partner and an autoprocessing term.
Provenance (the field's own admissions):
Gap: The mouse knockout phenotypes have no mechanistic link to any measured ADGRA3 activity. Adgra3-null males are infertile with a blockage between ejaculatory duct and urethra despite normal spermatogenesis, and Adgra3-null females develop imperforate vagina; both are attributed to altered sex-hormone receptor expression. Neither phenotype has been connected to Gs/Gi coupling, to the PDZ motif, to autoproteolysis or to epithelial polarity, and none of the cell-based mechanisms has been tested in the affected tissues. The same disconnect holds for the cellular roles reported outside the urogenital tract: knockdown cripples osteoclast differentiation and bone resorption, but through MAPK and AKT-NF-kB readouts that no one has joined to the receptor's Gs/Gi coupling or to its PDZ motif.
OPEN BIOLOGY BP_DARK
What is known: What is solid is the genetics - two independent, high-penetrance developmental phenotypes in the same tubular epithelia where the receptor is expressed - and, separately, the cell biology in kidney epithelial lines. What is absent is any experiment joining them: no knock-in of a signalling-dead or PBM-deleted allele, no conditional deletion in urogenital epithelium.
Significance: Whether ADGRA3's physiological role is receptor signalling or motif-dependent scaffolding is currently undecidable, and it determines which biological-process terms the gene should eventually carry. The apicobasal-polarity term proposed here is the only bridging hypothesis with any direct support, and it rests on a canine cell line.
What would resolve it: Separation-of-function alleles in mouse: a GPS mutant that cannot self-cleave and a Delta-ETTV mutant that cannot bind PDZ scaffolds, each scored for the urogenital phenotypes.
Provenance (the field's own admissions):
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)