ABR (active BCR-related) is a brain-enriched, 859-residue cytosolic regulator of Rho-family small GTPases. A single polypeptide carries two catalytic modules of opposite sign: a central Dbl-homology (DH) guanine-nucleotide exchange factor domain with an adjacent PH domain, and a C-terminal Rho-GAP domain whose arginine at R683 is required for full GAP activity, with a phospholipid-binding C2 domain between them and a class I PDZ-binding motif at the extreme C-terminus. R683 is assigned as the catalytic arginine finger, and hence a transition-state stabiliser, by sequence rule; what is measured is that R683A reduces and R683A/N795A abolishes GAP activity. Assayed as purified domains, the DH module loads GTP onto CDC42, RHOA, RAC1 and RAC2, while the GAP module accelerates hydrolysis on RAC1, RAC2 and CDC42 but not RhoA; the two modules bind substrate noncompetitively, so ABR can engage two GTPases at once rather than acting as a bidirectional switch on one. In cells the net output is predominantly restraint of Rac: reducing ABR raises active Rac1 in human epithelial cells, and loss of Abr in mice elevates Rac activity in macrophages, neutrophils, CD4+ T cells, astroglia and neurons, with consequences for macrophage shape, motility and phagocytosis, neutrophil granule and oxidant release, allergic airway inflammation, and postnatal cerebellar development. Rather than a simple switch, ABR behaves as a spatial segregation device: recruited to a zone of active Rho by binding the GTP-bound form, it amplifies Rho locally through its GEF domain while its GAP domain excludes Cdc42 from the same zone. At excitatory synapses ABR is tethered through its C-terminal valine to the PDZ domains of PSD-95/DLG4 and restrains basal Rac1, which is required for the maintenance rather than the induction of long-term potentiation and for spatial and object recognition memory. ABR is the paralog of BCR and shares its GEF and GAP architecture but lacks BCR's N-terminal oligomerisation and serine/threonine kinase region. It is also the host protein targeted by the enteropathogenic Escherichia coli effector EspH, which binds the ABR GAP domain to suppress Rac1 and Cdc42.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0005096 GTPase activator activity | IBA GO_REF:0000033 | ACCEPT | Summary: Core molecular function, and the best-supported annotation in the set. ABR's C-terminal Rho-GAP domain (UniProt FT 647-845) accelerates GTP hydrolysis on RAC1, RAC2 and CDC42. Reason: This is not a domain-name-derived guess in either direction. The activity is directly demonstrated for ABR itself: purified recombinant ABR GAP domain accelerates hydrolysis on Rac1, Rac2 and CDC42Hs [PMID:7479768], reproduced independently in two 1993 studies [PMID:8262969, PMID:8349582]. The catalytic machinery is also experimentally confirmed rather than merely inferred: UniProt annotates R683 as the arginine finger and records that R683A reduces GAP activity and R683A+N795A abolishes it (ECO:0000269|PubMed:17116687). Both residues lie inside the annotated Rho-GAP domain, and the human sequence carries R at 683 and N at 795. The propagation is also sound: all five protein donors carry their own experimental annotation at GO:0005096 itself, including the true ortholog (mouse Abr, IDA/IGI) and a self-referential entry for human ABR, which records a PAINT curator judging the function core rather than any circularity. GO no longer has substrate-specific GAP terms: GO:0005099 (Ras), GO:0005100 (Rho) and GO:0030675 (Rac) GTPase activator activity are all secondary ids of GO:0005096, which has no substrate-specific children (checked via QuickGO /ontology/go/terms/GO:0005096/complete). GO:0005096 is therefore already the maximally specific molecular-function term available. The measured substrate set is therefore recorded in the extensions block and in core_functions.substrates instead of by choosing a narrower term. Propagation Review Root cause: NO FAILURE CORE Sources checked: CGD:CAL0000181133 · Candida albicans BEM2 SUPPORTS TRANSFER Carries its own IGI/IPI at GO:0005096. Unreviewed (TrEMBL) entry, so its protein NAME is an automatic label; its GO annotations are nonetheless curated experimental (IGI/IPI). Counted as a donor with own evidence, not as an independently characterised protein name. MGI:MGI:107771 · mouse Abr (ortholog) SUPPORTS TRANSFER Carries its own IDA/IGI at GO:0005096 MGI:MGI:88141 · mouse Bcr (paralog) SUPPORTS TRANSFER Carries its own IDA/IGI at GO:0005096 PANTHER:PTN002754245 · PANTHER node: GEF/GAP + Rho-signalling clade SUPPORTS TRANSFER UniProtKB:P11274 · human BCR (paralog, same species) SUPPORTS TRANSFER Carries its own IDA at GO:0005096. Same-species paralog. Legitimate for IBA, but note it supplies no ortholog-strength inference. UniProtKB:Q12979 · human ABR (self) SUPPORTS TRANSFER Carries its own IDA at GO:0005096. Self-referential donor: a PAINT curator judging this function core for ABR, not a circular transfer. Supporting Evidence: PMID:7479768 The Dbl-homology domains of Bcr and Abr were active in stimulating GTP binding to CDC42Hs, RhoA, Rac1, and Rac2 (rank order, CDC42Hs > RhoA > Rac1 = Rac2) but were inactive toward Rap1A and Ha-Ras. Both Bcr and Abr acted as GAPs for Rac1, Rac2, and CDC42Hs but were inactive toward RhoA, Rap1A, and Ha-Ras. file:human/ABR/ABR-uniprot.txt stimulates GTP hydrolysis by RAC1, RAC2 and CDC42. file:human/ABR/ABR-uniprot.txt Arginine finger file:human/ABR/ABR-uniprot.txt R->A: Reduces GAP activity. file:human/ABR/ABR-bioinformatics/RESULTS.md 5 of 5 protein donors carry their own experimental evidence at GO:0005096 itself. |
| GO:0005886 plasma membrane | IBA GO_REF:0000033 | ACCEPT | Summary: The plasma membrane is where the GAP activity is exercised: ABR is recruited there on stimulation to act on membrane-anchored Rac. Reason: is_active_in is the right qualifier — ABR is a soluble protein that translocates to the membrane rather than residing in it. In primary macrophages Abr and Bcr transiently move to the plasma membrane on CSF-1 stimulation, and GAP-dead mutants accumulate around phagosomes [PMID:17116687]. Rac1/Rac2/CDC42 are prenylated and membrane-associated in their active state, so this is the compartment in which the annotated GAP activity acts. Both donors (mouse Abr, mouse Bcr) carry their own IDA at GO:0005886. Propagation Review Root cause: NO FAILURE CORE Sources checked: MGI:MGI:107771 · mouse Abr (ortholog) SUPPORTS TRANSFER Carries its own IDA at GO:0005886 MGI:MGI:88141 · mouse Bcr (paralog) SUPPORTS TRANSFER Carries its own IDA at GO:0005886 PANTHER:PTN002754245 · PANTHER node: GEF/GAP + Rho-signalling clade SUPPORTS TRANSFER Supporting Evidence: PMID:17116687 in response to CSF-1 stimulation, Abr and Bcr transiently translocated to the plasma membrane file:human/ABR/ABR-bioinformatics/RESULTS.md 2 of 2 protein donors carry their own experimental evidence at GO:0005886 itself. |
| GO:0014069 postsynaptic density | IBA GO_REF:0000033 | ACCEPT | Summary: ABR is a postsynaptic-density component, anchored there by binding the PDZ domains of PSD-95/DLG4. Reason: Well supported: ABR localises to excitatory synapses and interacts directly with PSD-95 [PMID:20962234], and UniProt records the interaction with ECO:0000269 (ECO:0000269|PubMed:20962234). The donor set was checked for the pattern where a propagation lands above its own donors, and it is present but does not warrant a downward move: mouse Bcr carries IDA at GO:0014069 itself, whereas rat Abr and rat Bcr hold GO:0014069 only by IBA/ISO/ISS and carry their experimental IDA one level down at GO:0099092 postsynaptic density, intracellular component. GO:0014069 is thus the least common ancestor of what the donors actually demonstrate, so it is the correct call rather than an over-broad one. Whether the PAINT node should instead propagate GO:0099092 is raised in suggested_questions; the human evidence does not itself distinguish the intracellular sub-compartment. Propagation Review Root cause: NO FAILURE CORE Sources checked: MGI:MGI:88141 · mouse Bcr (paralog) SUPPORTS TRANSFER Carries its own IDA at GO:0014069 PANTHER:PTN001142600 · PANTHER node: vertebrate synaptic clade SUPPORTS TRANSFER RGD:1306279 · rat Abr (ortholog) SUPPORTS TRANSFER Holds GO:0014069 only by IBA; IDA at the more specific GO:0099092 postsynaptic density, intracellular component. So this donor adds no independent support at the donated term; its grounding sits one level below it. RGD:1307993 · rat Bcr (paralog) SUPPORTS TRANSFER Holds GO:0014069 only by IBA/IEA/ISO/ISS; IDA/IEP at the more specific GO:0099092 postsynaptic density, intracellular component. Same pattern as rat Abr. Supporting Evidence: PMID:20962234 BCR and its close relative active BCR-related (ABR) localize at excitatory synapses and directly interact with PSD-95, an abundant postsynaptic scaffolding protein file:human/ABR/ABR-uniprot.txt Interacts with DLG4. file:human/ABR/ABR-bioinformatics/RESULTS.md 1 of 3 protein donors carry their own experimental evidence at GO:0014069 itself. |
| GO:0035023 regulation of Rho protein signal transduction | IBA GO_REF:0000033 | ACCEPT | Summary: ABR regulates Rho-branch signalling through its DH/GEF module, which loads GTP onto RHOA. Reason: The conclusion is correct but the propagation path is the weakest in the set and is worth recording: the only protein donor is UniProtKB:P11274, the same-species paralog BCR, with no ABR ortholog among the donors. Read mechanically this is a paralog-only transfer. It survives because ABR's own IDA independently establishes the activity: the purified ABR DH domain stimulates GTP binding to RhoA [PMID:7479768], and the Xenopus wound-repair work shows Abr amplifying Rho locally through its GEF domain in vivo [PMID:21295482]. So no failure mode is asserted, but note that GO's regulation branch keeps Rho and Rac disjoint — GO:0035021 is NOT a descendant of GO:0035023 — so this term covers only the Rho half of ABR's regulatory range and does not subsume its dominant Rac-directed activity. Propagation Review Root cause: NO FAILURE CORE Sources checked: PANTHER:PTN002754245 · PANTHER node: GEF/GAP + Rho-signalling clade SUPPORTS TRANSFER UniProtKB:P11274 · human BCR (paralog, same species) SUPPORTS TRANSFER Carries its own IMP at GO:0035023. Sole protein donor and a same-species paralog: this row's transfer rests entirely on BCR. ABR's own IDA (PMID:7479768) is what makes the term correct. Supporting Evidence: PMID:7479768 The Dbl-homology domains of Bcr and Abr were active in stimulating GTP binding to CDC42Hs, RhoA, Rac1, and Rac2 (rank order, CDC42Hs > RhoA > Rac1 = Rac2) but were inactive toward Rap1A and Ha-Ras. Both Bcr and Abr acted as GAPs for Rac1, Rac2, and CDC42Hs but were inactive toward RhoA, Rap1A, and Ha-Ras. PMID:21295482 Within the Rho zone, Abr promotes local Rho activation via its GEF domain and controls local crosstalk via its GAP domain, which limits Cdc42 activity within the Rho zone. |
| GO:0050804 modulation of chemical synaptic transmission | IBA GO_REF:0000033 | ACCEPT | Summary: ABR restrains synaptic Rac1 and is required for the maintenance phase of long-term potentiation, so it modulates synaptic transmission. Reason: Grounded in single-gene loss of function, not only the redundant double mutant: ABR-null mice have enhanced basal Rac1 activity and a marked, selective loss of LTP maintenance with impaired spatial and object recognition memory [PMID:20962234]. Both donors carry their own IDA and IMP at this term. ABR is brain-enriched [PMID:8262969], so this is a core physiological role rather than a peripheral one. Propagation Review Root cause: NO FAILURE CORE Sources checked: MGI:MGI:107771 · mouse Abr (ortholog) SUPPORTS TRANSFER Carries its own IDA/IMP at GO:0050804 MGI:MGI:88141 · mouse Bcr (paralog) SUPPORTS TRANSFER Carries its own IDA/IMP at GO:0050804 PANTHER:PTN001142600 · PANTHER node: vertebrate synaptic clade SUPPORTS TRANSFER Supporting Evidence: PMID:20962234 mice lacking BCR or ABR exhibit a marked decrease in the maintenance, but not induction, of long-term potentiation, and show impaired spatial and object recognition memory PMID:20962234 Mice deficient for BCR or ABR show enhanced basal Rac1 activity but only a small increase in spine density. file:human/ABR/ABR-uniprot.txt Highly enriched in the brain. |
| GO:0098978 glutamatergic synapse | IBA GO_REF:0000033 | ACCEPT | Summary: ABR acts at glutamatergic (excitatory) synapses. Reason: All four protein donors — mouse Abr, mouse Bcr, rat Abr, rat Bcr — carry their own IDA (and, for the Abr entries, IMP) at GO:0098978, and the human/rodent synaptic work localises ABR to excitatory synapses with direct PSD-95 binding [PMID:20962234]. Consistent with the Schaffer-collateral CA1 ISS row and with the postsynaptic density annotation. Propagation Review Root cause: NO FAILURE CORE Sources checked: MGI:MGI:107771 · mouse Abr (ortholog) SUPPORTS TRANSFER Carries its own IDA/IMP at GO:0098978 MGI:MGI:88141 · mouse Bcr (paralog) SUPPORTS TRANSFER Carries its own IDA/IMP at GO:0098978 PANTHER:PTN001142600 · PANTHER node: vertebrate synaptic clade SUPPORTS TRANSFER RGD:1306279 · rat Abr (ortholog) SUPPORTS TRANSFER Carries its own IDA at GO:0098978 RGD:1307993 · rat Bcr (paralog) SUPPORTS TRANSFER Carries its own IDA/IEP at GO:0098978 Supporting Evidence: PMID:20962234 BCR and its close relative active BCR-related (ABR) localize at excitatory synapses and directly interact with PSD-95, an abundant postsynaptic scaffolding protein file:human/ABR/ABR-bioinformatics/RESULTS.md 4 of 4 protein donors carry their own experimental evidence at GO:0098978 itself. |
| GO:0005085 guanyl-nucleotide exchange factor activity | IEA GO_REF:0000120 | ACCEPT | Summary: Domain-derived GEF call that is independently confirmed by direct assay of ABR's own purified DH domain. Reason: The supporting signatures are the Dbl-homology domain (InterPro:IPR000219) and the CDC24 guanine-nucleotide dissociation stimulator conserved site (InterPro:IPR001331), which map to UniProt FT DOMAIN 91-284 'DH'. Ordinarily a domain name is not an activity, but here the activity was measured for ABR directly: the purified DH domain stimulates GTP binding to CDC42Hs, RhoA, Rac1 and Rac2 [PMID:7479768]. As with the GAP side, the substrate-specific GEF terms have been merged away — GO:0005089 (Rho) and GO:0030676 (Rac) guanyl-nucleotide exchange factor activity are secondary ids of GO:0005085 — so this term is already maximal and the substrates are recorded in core_functions instead. Propagation Review Root cause: NO FAILURE CORE Sources checked: ARBA:ARBA00028188 · ARBA machine-learning rule SUPPORTS TRANSFER InterPro:IPR000219 · InterPro domain: Dbl homology domain SUPPORTS TRANSFER InterPro:IPR001331 · InterPro conserved site: guanine-nucleotide dissociation stimulator, CDC24 SUPPORTS TRANSFER Supporting Evidence: PMID:7479768 The Dbl-homology domains of Bcr and Abr were active in stimulating GTP binding to CDC42Hs, RhoA, Rac1, and Rac2 (rank order, CDC42Hs > RhoA > Rac1 = Rac2) but were inactive toward Rap1A and Ha-Ras. Both Bcr and Abr acted as GAPs for Rac1, Rac2, and CDC42Hs but were inactive toward RhoA, Rap1A, and Ha-Ras. file:human/ABR/ABR-uniprot.txt DH |
| GO:0005096 GTPase activator activity | IEA GO_REF:0000002 | ACCEPT | Summary: InterPro2GO mapping from a gene-family-specific signature, agreeing with ABR's own direct assays. Reason: Worth distinguishing from a bare-fold inference: the supporting signature InterPro:IPR037769 is a *family* entry named 'Abr/Bcr', not a generic RhoGAP fold, so the mapping is made at the level of this specific gene family and is much better founded than a domain-level guess. It agrees with the direct biochemistry [PMID:7479768, PMID:8262969] and with the experimentally confirmed arginine finger at R683. Propagation Review Root cause: NO FAILURE CORE Sources checked: InterPro:IPR037769 · InterPro family signature Abr/Bcr SUPPORTS TRANSFER Family-level signature 'Abr/Bcr' (InterPro type: family), not a bare RhoGAP fold signature. Supporting Evidence: PMID:8262969 The ABR GAP domain expressed as an Escherichia coli fusion protein was active against Rac1 and Cdc42 of the rho subfamily. file:human/ABR/ABR-uniprot.txt R->A: Reduces GAP activity. |
| GO:0007165 signal transduction | IEA GO_REF:0000002 | MODIFY | Summary: 'signal transduction' is true but uninformative, and it types ABR as a transducer rather than as the regulator it is. Reason: The mapping comes from InterPro:IPR000198, the Rho GTPase-activating protein domain. What a GAP domain licenses is not participation in signal transduction generally but modulation of a small-GTPase switch, so the informative term is GO:0051056 regulation of small GTPase mediated signal transduction. GO:0051056 is chosen over the narrower GO:0035023 deliberately: GO keeps the Rho and Rac regulation branches disjoint (GO:0035021 is not a descendant of GO:0035023), and ABR acts on RHOA, RAC1, RAC2 and CDC42, so GO:0051056 is the true least common ancestor of its regulatory range. The replacement term is independently present in GOA from Reactome, so this is a consolidation onto a better-supported term rather than the introduction of a new claim. Propagation Review Root cause: TERM SCOPING PROBLEM Failure modes: GRANULARITY MISMATCH ROLE CONFLATION Sources checked: InterPro:IPR000198 · InterPro domain: Rho GTPase-activating protein domain SUPPORTS TRANSFER RhoGAP domain signature; supports a small-GTPase regulatory role, not signal transduction in general. Proposed replacements: regulation of small GTPase mediated signal transduction Supporting Evidence: PMID:7479768 The Dbl-homology domains of Bcr and Abr were active in stimulating GTP binding to CDC42Hs, RhoA, Rac1, and Rac2 (rank order, CDC42Hs > RhoA > Rac1 = Rac2) but were inactive toward Rap1A and Ha-Ras. Both Bcr and Abr acted as GAPs for Rac1, Rac2, and CDC42Hs but were inactive toward RhoA, Rap1A, and Ha-Ras. |
| GO:0007264 small GTPase-mediated signal transduction | IEA GO_REF:0000117 | MODIFY | Summary: Role conflation: GO:0007264 is defined as the cassette in which the GTPase itself relays the signal. ABR is the regulator of that cassette. Reason: GO:0007264's definition is 'An intracellular signaling cassette in which a small monomeric GTPase relays a signal'. ABR is not the relaying GTPase; it is the GAP and GEF that switches one off and on. The regulation-branch counterpart GO:0051056 states exactly this and is already in GOA from Reactome. Supported only by an ARBA machine-learning rule (ARBA:ARBA00044393) with no gene-specific evidence, so there is no experimental call being overruled here. Propagation Review Root cause: TERM SCOPING PROBLEM Failure modes: ROLE CONFLATION Sources checked: ARBA:ARBA00044393 · ARBA machine-learning rule SUPPORTS TRANSFER Machine-learning rule; carries no ABR-specific evidence. Proposed replacements: regulation of small GTPase mediated signal transduction Supporting Evidence: PMID:7479768 The Dbl-homology domains of Bcr and Abr were active in stimulating GTP binding to CDC42Hs, RhoA, Rac1, and Rac2 (rank order, CDC42Hs > RhoA > Rac1 = Rac2) but were inactive toward Rap1A and Ha-Ras. Both Bcr and Abr acted as GAPs for Rac1, Rac2, and CDC42Hs but were inactive toward RhoA, Rap1A, and Ha-Ras. PMID:27481945 we propose that other domains of RHOGAPs confer substrate specificity and fine-tune their catalytic efficiency in cells |
| GO:0030424 axon | IEA GO_REF:0000044 | KEEP AS NON CORE | Summary: Axonal localisation, transferred from the rodent orthologs via a UniProt subcellular-location keyword. Plausible but not where the characterised activity acts. Reason: Traces to UniProt SUBCELLULAR LOCATION 'Cell projection, axon', whose evidence tag is ECO:0000250|UniProtKB:Q5SSL4 — by similarity from mouse Abr, the true ortholog, so the transfer is ortholog-based rather than paralog-based. For a brain-enriched protein this is entirely plausible, but the functionally informative locations are the postsynaptic density and the plasma membrane, which are separately annotated with stronger evidence. Retained as non-core rather than removed: there is no evidence against it. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: UniProtKB-SubCell:SL-0279 · UniProt subcellular-location keyword: Axon SUPPORTS TRANSFER Keyword mapping; underlying UniProt evidence is ECO:0000250 from mouse Abr (Q5SSL4). Supporting Evidence: file:human/ABR/ABR-uniprot.txt Cell projection, axon |
| GO:0035556 intracellular signal transduction | IEA GO_REF:0000002 | MODIFY | Summary: Uninformative parent; the specific regulatory role is what the DH domain supports. Reason: Mapped from InterPro:IPR001331, the CDC24 guanine-nucleotide dissociation stimulator conserved site, i.e. the GEF module. What that signature supports is regulation of a small-GTPase switch, not intracellular signal transduction at large. Replaced by GO:0051056 for the same reason as the GO:0007165 and GO:0007264 rows, giving these four vague domain-derived process terms a single accurate target. Propagation Review Root cause: TERM SCOPING PROBLEM Failure modes: GRANULARITY MISMATCH Sources checked: InterPro:IPR001331 · InterPro conserved site: guanine-nucleotide dissociation stimulator, CDC24 SUPPORTS TRANSFER DH/CDC24 conserved-site signature; supports GEF-mediated regulation of small GTPases. Proposed replacements: regulation of small GTPase mediated signal transduction Supporting Evidence: PMID:7479768 The Dbl-homology domains of Bcr and Abr were active in stimulating GTP binding to CDC42Hs, RhoA, Rac1, and Rac2 (rank order, CDC42Hs > RhoA > Rac1 = Rac2) but were inactive toward Rap1A and Ha-Ras. Both Bcr and Abr acted as GAPs for Rac1, Rac2, and CDC42Hs but were inactive toward RhoA, Rap1A, and Ha-Ras. |
| GO:0043197 dendritic spine | IEA GO_REF:0000044 | ACCEPT | Summary: Dendritic spine, the compartment in which ABR restrains Rac1 to permit LTP maintenance. Reason: From UniProt 'Cell projection, dendritic spine' (ECO:0000250|UniProtKB:Q5SSL4, i.e. from the mouse ortholog). Unlike the generic axon and synapse rows this one lines up directly with the characterised function: ABR-null mice show enhanced basal Rac1 activity together with a small increase in spine density [PMID:20962234], and Rac1 is the principal regulator of spine morphogenesis. Accepted rather than kept as non-core because it names the site of the core synaptic activity. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniProtKB-SubCell:SL-0284 · UniProt subcellular-location keyword: Dendritic spine SUPPORTS TRANSFER Keyword mapping; underlying UniProt evidence is ECO:0000250 from mouse Abr (Q5SSL4). Supporting Evidence: file:human/ABR/ABR-uniprot.txt Cell projection, dendritic spine PMID:20962234 Mice deficient for BCR or ABR show enhanced basal Rac1 activity but only a small increase in spine density. |
| GO:0045202 synapse | IEA GO_REF:0000044 | KEEP AS NON CORE | Summary: True but redundant: the specific synapse terms GO:0098978 and GO:0014069 are separately annotated with experimental support. Reason: From the UniProt 'Synapse' keyword (ECO:0000250|UniProtKB:A0A0G2JTR4, rat Abr). Correct, but GO:0098978 glutamatergic synapse, GO:0098685 Schaffer collateral - CA1 synapse and GO:0014069 postsynaptic density all sit below it and are independently annotated, so this row adds no information. Kept rather than removed because it is not wrong and general CC parents are harmless; flagged as non-core to keep the informative children foremost. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: UniProtKB-SubCell:SL-0258 · UniProt subcellular-location keyword: Synapse SUPPORTS TRANSFER Keyword mapping; underlying UniProt evidence is ECO:0000250 from rat Abr (A0A0G2JTR4). Supporting Evidence: PMID:20962234 BCR and its close relative active BCR-related (ABR) localize at excitatory synapses and directly interact with PSD-95, an abundant postsynaptic scaffolding protein |
| GO:0051056 regulation of small GTPase mediated signal transduction | TAS Reactome:R-HSA-9012999 | ACCEPT | Summary: The correct least common ancestor of ABR's regulatory range, and the target chosen for the four vague domain-derived process rows. Reason: ABR both activates and inactivates members of the Rho family: GEF activity toward CDC42, RHOA, RAC1 and RAC2, GAP activity toward RAC1, RAC2 and CDC42 [PMID:7479768]. Because GO's regulation branch keeps Rho, Rac and Cdc42 signalling separate — GO:0035021 negative regulation of Rac protein signal transduction is not a descendant of GO:0035023 regulation of Rho protein signal transduction — no single child covers ABR's range, and GO:0051056 is genuinely the LCA rather than a lazy generalisation. Reactome's placement in the RHO GTPase cycle is traceable and appropriate. Supporting Evidence: PMID:7479768 The Dbl-homology domains of Bcr and Abr were active in stimulating GTP binding to CDC42Hs, RhoA, Rac1, and Rac2 (rank order, CDC42Hs > RhoA > Rac1 = Rac2) but were inactive toward Rap1A and Ha-Ras. Both Bcr and Abr acted as GAPs for Rac1, Rac2, and CDC42Hs but were inactive toward RhoA, Rap1A, and Ha-Ras. PMID:7479768 Each individual domain bound in a noncompetitive manner to GTP-binding protein substrates. |
| GO:0045202 synapse | ISS GO_REF:0000024 | KEEP AS NON CORE | Summary: Ortholog-based synapse transfer; correct but subsumed by the specific synapse terms. Reason: Curator transfer from rat Abr (UniProtKB:A0A0G2JTR4), which carries experimental evidence at GO:0045202 and IDA at the more specific GO:0098978 and GO:0099092. A genuine ortholog transfer, so the inference is sound; it is non-core only because the informative children are separately annotated. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: UniProtKB:A0A0G2JTR4 · rat Abr (ortholog) SUPPORTS TRANSFER Carries its own EXP at GO:0045202; IDA at the more specific GO:0098978 glutamatergic synapse; IDA at the more specific GO:0099092 postsynaptic density, intracellular component Supporting Evidence: PMID:20962234 BCR and its close relative active BCR-related (ABR) localize at excitatory synapses and directly interact with PSD-95, an abundant postsynaptic scaffolding protein file:human/ABR/ABR-bioinformatics/RESULTS.md 1 of 1 protein donors carry their own experimental evidence at GO:0045202 itself. |
| GO:0005085 guanyl-nucleotide exchange factor activity | TAS Reactome:R-HSA-9014296 | ACCEPT | Summary: Traceable Reactome statement crediting the primary GEF biochemistry. Reason: Reactome's 'RAC2 GEFs activate RAC2' attributes ABR to Chuang et al. 1995 = PMID:7479768, which did measure RAC2 loading by the purified ABR DH domain. Rac2 sits at the weak end of the measured rank order (CDC42Hs > RhoA > Rac1 = Rac2) but was assayed, so the citation is traceable and the term correct. Supporting Evidence: PMID:7479768 The Dbl-homology domains of Bcr and Abr were active in stimulating GTP binding to CDC42Hs, RhoA, Rac1, and Rac2 (rank order, CDC42Hs > RhoA > Rac1 = Rac2) but were inactive toward Rap1A and Ha-Ras. Both Bcr and Abr acted as GAPs for Rac1, Rac2, and CDC42Hs but were inactive toward RhoA, Rap1A, and Ha-Ras. |
| GO:0005096 GTPase activator activity | TAS Reactome:R-HSA-9013022 | ACCEPT | Summary: Correct term. The RHOB substrate assignment behind it is in apparent tension with the 1995 biochemistry, and the resolution is that both measured an isolated, intrinsically nonselective domain. Reason: Reactome lists ABR as a confirmed RHOB GAP (Amin et al. 2016, supported by Bagci et al. 2020) while listing its paralog BCR only as a candidate. That looks to contradict PMID:7479768, which found the ABR GAP domain inactive toward RhoA, and PMID:36219160, which restates that the ABR GAP domain acts on Rac1/Rac2 and Cdc42 but not RhoA. RhoB is not RhoA, so there is no formal contradiction, and the deeper resolution is in Amin et al. itself: having assayed 14 RHOGAPs against 12 RHO proteins, they conclude the RHOGAP domain is nonselective in cell-free conditions and that specificity comes from other domains in cells [PMID:27481945]. Both the 1995 and 2016 experiments used isolated recombinant domains [PMID:7479768], so they are measurements of an intrinsically promiscuous module rather than conflicting claims about ABR's cellular substrate. GO:0005096 is true on any reading, and since the substrate-specific GAP terms are merged the disagreement is invisible to GO in any case. Neither cached abstract (PMID:27481945, PMID:31871319) names ABR, so the ABR-specific RHOB assignment rests on data not verifiable from the abstracts; recorded rather than relied upon. Supporting Evidence: PMID:27481945 We have found that the RHOGAP domain itself is nonselective and in some cases rather inefficient under cell-free conditions. PMID:27481945 we propose that other domains of RHOGAPs confer substrate specificity and fine-tune their catalytic efficiency in cells PMID:7479768 We purified as recombinant fusion proteins the GAP- and Dbl-homology domains of both Abr and Bcr. PMID:36219160 The GAP domain of ABR (and Bcr) acts on Rac1/Rac2 and Cdc42, but not RhoA. |
| GO:0050804 modulation of chemical synaptic transmission | ISS GO_REF:0000024 | ACCEPT | Summary: Ortholog transfer from mouse Abr, which carries IDA and IMP at this exact term. Reason: Curator transfer from UniProtKB:Q5SSL4 (mouse Abr), a true ortholog of identical length (859 aa) that carries its own IDA and IMP at GO:0050804. The underlying phenotype — loss of LTP maintenance and impaired memory in ABR-null mice — is single-gene, not dependent on the redundant Abr;Bcr double mutant [PMID:20962234]. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniProtKB:Q5SSL4 · mouse Abr (ortholog) SUPPORTS TRANSFER Carries its own IDA/IMP at GO:0050804 Supporting Evidence: PMID:20962234 mice lacking BCR or ABR exhibit a marked decrease in the maintenance, but not induction, of long-term potentiation, and show impaired spatial and object recognition memory |
| GO:0098685 Schaffer collateral - CA1 synapse | ISS GO_REF:0000024 | ACCEPT | Summary: Schaffer collateral - CA1 synapse: the specific synapse at which the LTP phenotype was measured. Reason: Unusually specific for an ISS row, but the donor mouse Abr (UniProtKB:Q5SSL4) carries both IDA and IMP at GO:0098685, and the specificity is earned — the LTP maintenance deficit in ABR-null mice was recorded in hippocampal CA1, the postsynaptic target of the Schaffer collaterals [PMID:20962234]. This is the reverse of the usual over-propagation pattern: the donor holds the specific term experimentally and the transfer preserves it. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniProtKB:Q5SSL4 · mouse Abr (ortholog) SUPPORTS TRANSFER Carries its own IDA/IMP at GO:0098685 Supporting Evidence: PMID:20962234 mice lacking BCR or ABR exhibit a marked decrease in the maintenance, but not induction, of long-term potentiation, and show impaired spatial and object recognition memory |
| GO:0098978 glutamatergic synapse | ISS GO_REF:0000024 | ACCEPT | Summary: Ortholog transfer of glutamatergic-synapse localisation; donor carries IDA and IMP. Reason: From mouse Abr (UniProtKB:Q5SSL4), which holds IDA and IMP at GO:0098978. Redundant with the IBA row at the same term but independently sourced by a curator, and the human protein's direct PSD-95 interaction supports excitatory-synapse localisation [PMID:20962234]. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniProtKB:Q5SSL4 · mouse Abr (ortholog) SUPPORTS TRANSFER Carries its own IDA/IMP at GO:0098978 Supporting Evidence: PMID:20962234 BCR and its close relative active BCR-related (ABR) localize at excitatory synapses and directly interact with PSD-95, an abundant postsynaptic scaffolding protein |
| GO:0005515 protein binding | IPI PMID:20962234 Regulation of synaptic Rac1 activity, long-term potentiation... | MODIFY | Summary: A real, well-mapped interaction recorded under an uninformative term. The partner is PSD-95/DLG4 and the binding is PDZ-motif mediated. Reason: This is not screen noise, so the right response is to make the term informative rather than to down-weight the row. The WITH/FROM accession UniProtKB:P78352 resolves to reviewed canonical DLG4_HUMAN at the expected 724 aa — no unreviewed ORFeome fragment substituted for the canonical partner. The interaction is direct [PMID:20962234] and UniProt records it with ECO:0000269. The mechanism is mapped to a specific motif: ABR ends in ...S-T-D-V, a canonical class I PDZ-binding motif (Thr at -2, Val at position 0), V859A abolishes DLG4 binding while leaving synaptic targeting intact, and an independent study describes ABR as possessing 'a PDZ binding motif located at the C-terminus' [PMID:36219160]. Since PSD-95 engages it through its PDZ domains, GO:0030165 PDZ domain binding states what ABR's molecular function actually is here, in line with the project's guidance against bare protein binding. Proposed replacements: PDZ domain binding Supporting Evidence: PMID:20962234 BCR and its close relative active BCR-related (ABR) localize at excitatory synapses and directly interact with PSD-95, an abundant postsynaptic scaffolding protein file:human/ABR/ABR-uniprot.txt V->A: Abolishes interaction with DLG4. file:human/ABR/ABR-uniprot.txt Interacts with DLG4. PMID:36219160 ABR possesses a phospholipid-binding C2 domain positioned between the two domains and a PDZ binding motif located at the C-terminus of the protein. |
| GO:0005085 guanyl-nucleotide exchange factor activity | IDA PMID:7479768 Abr and Bcr are multifunctional regulators of the Rho GTP-bi... | ACCEPT | Summary: Direct assay of the purified ABR DH domain: GTP loading of CDC42, RHOA, RAC1 and RAC2. Core molecular function. Reason: The primary evidence for ABR's GEF half. Purified recombinant DH domain stimulated GTP binding to CDC42Hs, RhoA, Rac1 and Rac2 in rank order CDC42Hs > RhoA > Rac1 = Rac2, and was inactive toward Rap1A and Ha-Ras — a specificity control that makes the positive result interpretable. The GEF and GAP domains bind substrate noncompetitively, so ABR can engage two GTPases at once rather than acting as a bidirectional switch on one. Substrate-specific GEF terms are merged into GO:0005085, so the substrates are recorded in the extensions block. Supporting Evidence: PMID:7479768 The Dbl-homology domains of Bcr and Abr were active in stimulating GTP binding to CDC42Hs, RhoA, Rac1, and Rac2 (rank order, CDC42Hs > RhoA > Rac1 = Rac2) but were inactive toward Rap1A and Ha-Ras. Both Bcr and Abr acted as GAPs for Rac1, Rac2, and CDC42Hs but were inactive toward RhoA, Rap1A, and Ha-Ras. PMID:7479768 Each individual domain bound in a noncompetitive manner to GTP-binding protein substrates. PMID:7479768 We purified as recombinant fusion proteins the GAP- and Dbl-homology domains of both Abr and Bcr. |
| GO:0005096 GTPase activator activity | IDA PMID:7479768 Abr and Bcr are multifunctional regulators of the Rho GTP-bi... | ACCEPT | Summary: Direct assay of the purified ABR GAP domain on Rac1, Rac2 and CDC42Hs, with RhoA, Rap1A and Ha-Ras as negative controls. Core molecular function. Reason: The primary evidence for ABR's GAP half, reproduced independently in 1993 by two groups [PMID:8262969, PMID:8349582]. The negative results on RhoA, Rap1A and Ha-Ras are what make it a specificity claim rather than a generic activity claim. Note the caveat that the assay used the isolated domain, which Amin et al. later showed to be intrinsically nonselective [PMID:27481945]; the in-cell specificity is established separately by the loss-of-function work. GO no longer has substrate-specific GAP terms: GO:0005099 (Ras), GO:0005100 (Rho) and GO:0030675 (Rac) GTPase activator activity are all secondary ids of GO:0005096, which has no substrate-specific children (checked via QuickGO /ontology/go/terms/GO:0005096/complete). GO:0005096 is therefore already the maximally specific molecular-function term available. Supporting Evidence: PMID:7479768 The Dbl-homology domains of Bcr and Abr were active in stimulating GTP binding to CDC42Hs, RhoA, Rac1, and Rac2 (rank order, CDC42Hs > RhoA > Rac1 = Rac2) but were inactive toward Rap1A and Ha-Ras. Both Bcr and Abr acted as GAPs for Rac1, Rac2, and CDC42Hs but were inactive toward RhoA, Rap1A, and Ha-Ras. PMID:8262969 The ABR GAP domain expressed as an Escherichia coli fusion protein was active against Rac1 and Cdc42 of the rho subfamily. PMID:8349582 A domain of ABR with similarity to GAPrho was expressed as a fusion protein in Escherichia coli and was shown to have GAP activity toward rac. |
| GO:0090630 activation of GTPase activity | IDA PMID:7479768 Abr and Bcr are multifunctional regulators of the Rho GTP-bi... | ACCEPT | Summary: The biological-process counterpart of the GAP activity: ABR accelerates the GTPase cycle of its Rho-family targets. Reason: Correctly paired with the GO:0005096 IDA from the same experiment. Worth recording that GO:0090630 has undergone the same consolidation as the molecular-function branch — GO:0032856 through GO:0032864, the substrate-specific 'activation of X GTPase activity' terms, are all secondary ids of GO:0090630, which now has no children — so as with GO:0005096 there is no more specific term to move to and no case for proposing one. Supporting Evidence: PMID:7479768 The Dbl-homology domains of Bcr and Abr were active in stimulating GTP binding to CDC42Hs, RhoA, Rac1, and Rac2 (rank order, CDC42Hs > RhoA > Rac1 = Rac2) but were inactive toward Rap1A and Ha-Ras. Both Bcr and Abr acted as GAPs for Rac1, Rac2, and CDC42Hs but were inactive toward RhoA, Rap1A, and Ha-Ras. file:human/ABR/ABR-uniprot.txt R->A: Reduces GAP activity. |
| GO:0016020 membrane | HDA PMID:19946888 Defining the membrane proteome of NK cells. | MARK AS OVER ANNOTATED | Summary: Generic membrane call from a crude NK-cell membrane-fraction sweep, for a protein with no transmembrane segment. Reason: ABR has no transmembrane region in its UniProt feature table; its domain complement is DH, PH, C2 and Rho-GAP, all soluble. The source study's own abstract limits what the dataset can support: of 1843 proteins identified, only 'approximately 40%' were predicted plausible membrane proteins, and the remainder were expected to be 'transiently associated with membranes'. ABR belongs to that second class — it is a cytosolic protein that translocates to the plasma membrane on stimulation [PMID:17116687], which is exactly what the separately annotated GO:0005886 with is_active_in captures. Marked as over-annotated rather than removed because the peptides were genuinely detected in a membrane preparation and HDA is an experimental code; the objection is to the generic term's implication of membrane residence, not to the observation. Supporting Evidence: PMID:19946888 approximately 40% of the identified proteins were predicted as plausible membrane proteins PMID:19946888 The remaining species were largely involved in cellular processes and molecular functions that could be predicted to be transiently associated with membranes. PMID:17116687 in response to CSF-1 stimulation, Abr and Bcr transiently translocated to the plasma membrane |
| GO:0005829 cytosol | TAS Reactome:R-HSA-205039 | KEEP AS NON CORE | Summary: Correct resting-state compartment; not where the activity is exercised. Reason: ABR is a soluble cytosolic protein, so cytosol is correct, and it is the resting-state pool from which ABR is recruited on stimulation. It is marked non-core because the compartments where the activity is actually exercised — the plasma membrane on CSF-1 stimulation [PMID:17116687] and the postsynaptic density at excitatory synapses [PMID:20962234] — are separately annotated with is_active_in and part_of. Four identical cytosol rows differing only in the citing Reactome reaction are treated identically. This particular reaction ('p75NTR indirectly activates RAC and Cdc42 via a guanyl-nucleotide exchange factor') names no GEF in its summary and places ABR in an unnamed candidate set, so it carries no ABR-specific information beyond the compartment. Supporting Evidence: PMID:17116687 in response to CSF-1 stimulation, Abr and Bcr transiently translocated to the plasma membrane |
| GO:0005829 cytosol | TAS Reactome:R-HSA-419166 | KEEP AS NON CORE | Summary: Correct resting-state compartment; not where the activity is exercised. Reason: ABR is a soluble cytosolic protein, so cytosol is correct, and it is the resting-state pool from which ABR is recruited on stimulation. It is marked non-core because the compartments where the activity is actually exercised — the plasma membrane on CSF-1 stimulation [PMID:17116687] and the postsynaptic density at excitatory synapses [PMID:20962234] — are separately annotated with is_active_in and part_of. Four identical cytosol rows differing only in the citing Reactome reaction are treated identically. Supporting Evidence: PMID:17116687 in response to CSF-1 stimulation, Abr and Bcr transiently translocated to the plasma membrane |
| GO:0005829 cytosol | TAS Reactome:R-HSA-9013022 | KEEP AS NON CORE | Summary: Correct resting-state compartment; not where the activity is exercised. Reason: ABR is a soluble cytosolic protein, so cytosol is correct, and it is the resting-state pool from which ABR is recruited on stimulation. It is marked non-core because the compartments where the activity is actually exercised — the plasma membrane on CSF-1 stimulation [PMID:17116687] and the postsynaptic density at excitatory synapses [PMID:20962234] — are separately annotated with is_active_in and part_of. Four identical cytosol rows differing only in the citing Reactome reaction are treated identically. Supporting Evidence: PMID:17116687 in response to CSF-1 stimulation, Abr and Bcr transiently translocated to the plasma membrane |
| GO:0005829 cytosol | TAS Reactome:R-HSA-9014296 | KEEP AS NON CORE | Summary: Correct resting-state compartment; not where the activity is exercised. Reason: ABR is a soluble cytosolic protein, so cytosol is correct, and it is the resting-state pool from which ABR is recruited on stimulation. It is marked non-core because the compartments where the activity is actually exercised — the plasma membrane on CSF-1 stimulation [PMID:17116687] and the postsynaptic density at excitatory synapses [PMID:20962234] — are separately annotated with is_active_in and part_of. Four identical cytosol rows differing only in the citing Reactome reaction are treated identically. Supporting Evidence: PMID:17116687 in response to CSF-1 stimulation, Abr and Bcr transiently translocated to the plasma membrane |
| GO:0005096 GTPase activator activity | TAS PMID:8262969 The human active breakpoint cluster region-related gene enco... | ACCEPT | Summary: Traceable author statement that is in fact backed by a direct assay in the cited paper itself. Reason: The evidence code understates the source. Although the paper is best known for identifying the brain ABR transcript by homology to GEF and GAP domains, it also expressed the ABR GAP domain as an E. coli fusion protein and showed activity against Rac1 and Cdc42 — a direct biochemical result that would support IDA. Accepted as correct; the evidence-code choice is the curator's. Supporting Evidence: PMID:8262969 The ABR GAP domain expressed as an Escherichia coli fusion protein was active against Rac1 and Cdc42 of the rho subfamily. PMID:8262969 The ABR mRNA is highly enriched in the brain. |
| GO:0007264 small GTPase-mediated signal transduction | TAS PMID:8262969 The human active breakpoint cluster region-related gene enco... | MODIFY | Summary: Same role conflation as the IEA row at this term: ABR regulates the small-GTPase cassette rather than relaying the signal through it. Reason: GO:0007264 is defined as the cassette in which a small monomeric GTPase relays a signal. The cited paper characterises ABR as a GAP acting on Rac1 and Cdc42 and as containing a dbl-related GEF domain, i.e. as a regulator of that cassette on both sides. GO:0051056 is the accurate term and is already independently present in GOA. Both GO:0007264 rows are modified identically so the two do not diverge. Proposed replacements: regulation of small GTPase mediated signal transduction Supporting Evidence: PMID:8262969 The ABR GAP domain expressed as an Escherichia coli fusion protein was active against Rac1 and Cdc42 of the rho subfamily. PMID:7479768 The Dbl-homology domains of Bcr and Abr were active in stimulating GTP binding to CDC42Hs, RhoA, Rac1, and Rac2 (rank order, CDC42Hs > RhoA > Rac1 = Rac2) but were inactive toward Rap1A and Ha-Ras. Both Bcr and Abr acted as GAPs for Rac1, Rac2, and CDC42Hs but were inactive toward RhoA, Rap1A, and Ha-Ras. |
| GO:0035021 negative regulation of Rac protein signal transduction | IMP PMID:36219160 EspH interacts with the host active Bcr related (ABR) protei... | NEW | Summary: The single largest gap in ABR's GO record: its dominant, most reproducible cellular role — restraining Rac signalling — is not represented anywhere in GOA. Reason: Every loss-of-function system points the same way, and the effect is Rac-selective. In human cells: shRNA knockdown of ABR in HeLa raised active Rac1 but not Cdc42 in uninfected cells [PMID:36219160], which is direct human evidence (IMP) for negative regulation of Rac specifically. Corroborated across species and cell types by single-gene knockouts — ABR-null mouse CD4+ T cells carry elevated GTP-bound Rac [PMID:24058174], ABR-null mice have enhanced basal synaptic Rac1 activity [PMID:20962234] — and by double mutants in macrophages [PMID:17116687], neutrophils [PMID:19703997] and glia [PMID:11684658]. Cho et al. state the conclusion at family level: Abr and Bcr are 'the only GAPs to date that specifically negatively regulate Rac function in vivo'. This term is required rather than merely desirable because GO's regulation branch keeps Rac and Rho disjoint: GO:0035021 descends through GO:0035020 and GO:0051058, NOT through the already-annotated GO:0035023, so nothing currently in GOA implies it. Note the useful asymmetry — GO merged away the substrate-specific GAP *activity* terms because the isolated domain is nonselective [PMID:27481945], but retained substrate-specific *process* terms, which is where ABR's real in-cell selectivity belongs. The affinage record is cited here only for the corpus-level direction of the literature, which it independently summarises the same way; every mechanistic and quantitative claim above is taken from the primary papers, and the provider record does not cite Amin et al. 2016 at all. Supporting Evidence: file:human/ABR/ABR-deep-research-affinage.md In vivo, ABR acts predominantly as a negative regulator of Rac PMID:36219160 Intriguingly, in uninfected cells, the Rac1, but not Cdc42 activity levels were increased in the ABR knockdown compared to the Control shRNA-treated cells PMID:24058174 CD4(+) T cells from CRA-immunized and challenged abr(-/-) mice contained elevated levels of activated GTP-bound Rac compared with wild-type controls. PMID:17116687 These results identify Abr and Bcr as the only GAPs to date that specifically negatively regulate Rac function in vivo in primary macrophages. PMID:20962234 Mice deficient for BCR or ABR show enhanced basal Rac1 activity but only a small increase in spine density. PMID:19703997 These data show that Abr and Bcr normally curb very specific functions of mature tissue innate immune cells |
| GO:0035025 positive regulation of Rho protein signal transduction | IDA PMID:7479768 Abr and Bcr are multifunctional regulators of the Rho GTP-bi... | NEW | Summary: The GEF half of ABR has a known direction that GOA records only as the undirected GO:0035023: ABR activates Rho signalling. Reason: GOA carries GO:0035023 regulation of Rho protein signal transduction by IBA, which is correct but undirected, and its donor (human BCR) holds only the undirected term by IMP — so sharpening that row would over-reach its own donor. Proposed additively instead, on ABR's own evidence. The purified ABR DH domain stimulates GTP binding to RhoA [PMID:7479768], and in vivo Abr promotes local Rho activation through its GEF domain within the Rho activity zone of a healing wound [PMID:21295482]. Both give a sign, so the directional child is warranted. Deliberately not extended to the Rac or Cdc42 branches: the DH domain also loads GTP onto CDC42, RAC1 and RAC2, but that is isolated-domain biochemistry, and by the same argument this review applies to the GAP side, an isolated domain's substrate range does not establish in-cell specificity — in cells ABR's net effect on Rac is negative, not positive. The in vitro range is recorded in core_functions.substrates and the GO:0005085 extensions instead. Supporting Evidence: PMID:7479768 The Dbl-homology domains of Bcr and Abr were active in stimulating GTP binding to CDC42Hs, RhoA, Rac1, and Rac2 (rank order, CDC42Hs > RhoA > Rac1 = Rac2) but were inactive toward Rap1A and Ha-Ras. Both Bcr and Abr acted as GAPs for Rac1, Rac2, and CDC42Hs but were inactive toward RhoA, Rap1A, and Ha-Ras. PMID:21295482 Within the Rho zone, Abr promotes local Rho activation via its GEF domain and controls local crosstalk via its GAP domain, which limits Cdc42 activity within the Rho zone. |
| GO:1900273 positive regulation of long-term synaptic potentiation | ISS PMID:20962234 Regulation of synaptic Rac1 activity, long-term potentiation... | NEW | Summary: ABR is required for the maintenance phase of LTP; the existing GO:0050804 is three levels more general than what was measured. Reason: ABR-null mice show a marked decrease in the maintenance, but not the induction, of long-term potentiation, together with impaired spatial and object recognition memory [PMID:20962234]. GO:1900273 is a descendant of the already-annotated GO:0050804 modulation of chemical synaptic transmission and states both the specific process and its direction. Ancestry confirmed by QuickGO ancestor query: GO:1900273 -> GO:1900271 regulation of long-term synaptic potentiation -> GO:0048167 regulation of synaptic plasticity -> GO:0050804, i.e. three is_a steps. Proposed as ISS rather than IMP because the electrophysiology is in the mouse ortholog (UniProtKB:Q5SSL4, 859 aa, same length as human ABR), consistent with how GOA already handles the other synaptic transfers for this gene. Supporting Evidence: PMID:20962234 mice lacking BCR or ABR exhibit a marked decrease in the maintenance, but not induction, of long-term potentiation, and show impaired spatial and object recognition memory PMID:20962234 Mice deficient for BCR or ABR show enhanced basal Rac1 activity but only a small increase in spine density. |
Loading supporting content…
Download this section (compressed HTML)Q: PAINT node PTN001142600 propagates GO:0014069 postsynaptic density to ABR, but two of the three protein donors (rat Abr, rat Bcr) hold GO:0014069 only by IBA/ISO/ISS and carry their experimental IDA one level down at GO:0099092 postsynaptic density, intracellular component. Should the node propagate the more specific term, given that ABR is a soluble PSD-95-binding protein rather than a membrane-embedded one?
Suggested experts: Eunjoon Kim, Nora Heisterkamp
Q: The GO:0035023 IBA for ABR lists only PANTHER:PTN002754245 and UniProtKB:P11274 (human BCR) in WITH/FROM — the same-species paralog, with no ABR ortholog among the donors — even though ABR carries its own IDA for both GEF and GAP activity. Should ABR's direct experimental annotation be contributing at this node, and is a paralog-only donor set for a same-species pair intended?
Q: Reactome R-HSA-9013022 lists ABR as a confirmed RHOB GAP while listing BCR only as a candidate, on the strength of Amin et al. 2016 and Bagci et al. 2020 — yet Amin et al. conclude that the RHOGAP domain is nonselective under cell-free conditions, and three independent studies report the ABR GAP domain as inactive toward RhoA. Is ABR a physiological RHOB GAP in cells, or is the assignment an artefact of isolated-domain assays?
Suggested experts: Mohammad Reza Ahmadian, Jean-François Côté
Q: GO merged the substrate-specific GAP and GEF activity terms into GO:0005096 and GO:0005085 while retaining substrate-specific process terms such as GO:0035021. For a dual GEF-GAP protein like ABR, whose two domains act on overlapping but different GTPase sets, the molecular-function annotations are now indistinguishable from those of a GAP with the opposite specificity. Should substrate identity for GAPs and GEFs be captured by mandatory has_input extensions in GO-CAM, rather than being left optional?
Q: ABR's GEF domain loads GTP onto RHOA while its GAP domain does not act on RhoA, and the two domains bind substrate noncompetitively. Does ABR ever act as a net Rho activator and Rac inactivator simultaneously on the same membrane, as the Xenopus wound-repair zone model implies for the human protein?
Suggested experts: William M Bement
Experiment: Compare GAP activity and substrate preference of the isolated ABR Rho-GAP domain (647-845) against full-length ABR and PH- or C2-deleted variants, assayed on RAC1, RAC2, CDC42, RHOA and RHOB in parallel by real-time nucleotide-hydrolysis measurement, and then in cells by GTPase-specific biosensors after re-expression in ABR-null cells at native levels. This directly tests Amin et al.'s proposal and would settle whether the RHOB assignment in Reactome reflects cellular physiology.
Hypothesis: ABR's in-cell substrate selectivity is set by its non-catalytic domains (PH and C2) rather than by the Rho-GAP domain, which is intrinsically nonselective.
Type: comparative enzymology plus biosensor imaging
Experiment: Compare rescue of the LTP-maintenance deficit in ABR-null hippocampal slices by re-expression of wild-type ABR, the PDZ-motif mutant V859A, and the GAP-dead R683A/N795A double mutant. Separating targeting from catalysis would show whether the two proposed core functions are independent or strictly sequential.
Hypothesis: ABR is recruited to the postsynaptic density by its C-terminal PDZ motif, and that recruitment, not GAP activity per se, is what LTP maintenance requires.
Type: slice electrophysiology with structure-function rescue
Experiment: Measure phagocytic index after ABR depletion and after ABR overexpression in matched conditions in primary macrophages and in human trabecular meshwork cells, with Rac1 activity measured in the same samples. Mouse Abr;Bcr double nulls over-phagocytose whereas human trabecular-meshwork siRNA reduces phagocytosis by ~40%; resolving this would determine whether a phagocytosis term can be annotated for human ABR at all.
Hypothesis: The opposite-signed phagocytosis phenotypes in the literature reflect cell type rather than a real contradiction.
Type: parallel loss- and gain-of-function phagocytosis assay
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)