ARFGEF1

UniProt ID: Q9Y6D6
Organism: Homo sapiens
Review Status: COMPLETE
πŸ“ Provide Detailed Feedback

Gene Description

ARFGEF1 encodes BIG1 (brefeldin A-inhibited guanine nucleotide-exchange protein 1), a 1849-residue peripheral membrane protein of the large, brefeldin A-sensitive Arf-GEF family. Its central SEC7 domain (residues 709-840) catalyses GDP-to-GTP exchange on the class I ADP-ribosylation factors ARF1 and ARF3, converting them to the membrane-bound active form that recruits vesicle coats, chiefly the AP-1 clathrin adaptor, at the trans-Golgi network. Brefeldin A acts as an uncompetitive inhibitor that traps an abortive BIG1-ARF complex, which is why the drug collapses Golgi morphology. BIG1 is recruited to the trans-Golgi by the Arf-like GTPase ARL1, which binds an N-terminal DCB domain that also mediates homodimerisation and packs against the adjacent HUS domain. Through this activity BIG1 supports traffic between the TGN and endosomes, including retrograde delivery of furin from late endosomes, and the surface delivery and correct Golgi glycosylation of cargo such as integrin beta-1. It is separately required, apparently independently of its exchange activity, to maintain normal Golgi ribbon morphology - depleting BIG1 breaks the Golgi into mini-stacks that remain polarised and competent for export. BIG1 also works as a scaffold, binding protein kinase A regulatory subunits in the manner of an A-kinase anchoring protein and holding together a myosin phosphatase complex containing non-muscle myosin IIA, PP1-delta and MYPT1, thereby restraining myosin regulatory light chain phosphorylation, F-actin accumulation and directed cell migration; it binds the tail of myosin IXb and blocks that protein's Rho-GAP activity by competing with RhoA. Its own catalysis is under cyclic-AMP control - PKA phosphorylation lowers exchange activity and PP1-gamma restores it - while a separate PKA event, phosphorylation of Ser-883 together with a 711-715 nuclear localisation signal, drives accumulation of BIG1 in the nucleus and nucleolus, where it associates with nucleolin, fibrillarin and U3 snoRNA in a nucleic-acid-dependent complex of undetermined function. Heterozygous loss of ARFGEF1 causes an autosomal dominant neurodevelopmental disorder with impaired intellectual development, language and motor delay, behavioural abnormalities and, in about half of affected individuals, seizures.

Existing Annotations Review

GO Term Evidence Action Reason
GO:0000139 Golgi membrane
IDA
PMID:10716990
Identification and localization of two brefeldin A-inhibited...
ACCEPT
Summary: BIG1 co-fractionates with the Golgi marker beta-COP on density gradients of a microsomal fraction and partially colocalises with Golgi 58K protein and gamma-adaptin, placing the membrane-associated pool of the protein on Golgi membranes. This is where the SEC7 domain does its work, so the location is core.
Reason: Directly measured, consistent with every other localisation study of BIG1, and mechanistically necessary - a peripheral Arf-GEF has to be on the donor membrane to load ARF1/ARF3 with GTP there.
Supporting Evidence:
PMID:10716990
After density gradient centrifugation of a microsomal fraction, BIG1 and BIG2 were recovered in the same fraction as beta-COP, a marker for Golgi membranes.
GO:0005085 guanyl-nucleotide exchange factor activity
IBA
GO_REF:0000033
ACCEPT
Summary: Guanine-nucleotide exchange is the defining activity of the whole Sec7-domain superfamily, and the PAINT node behind this row spans it - yeast Sec7, Gea1, Gea2, Syt1 and Mon2, Arabidopsis GNOM, Drosophila Sec71 and garz, the human and mouse cytohesins, PSD, PSD3, IQSEC2, and ARFGEF2. ARFGEF1 itself appears in the WITH/FROM because it carries its own IDA for the term.
Reason: Sound at every level. All eighteen protein donors carry their own experimental evidence for this exact term (sixteen are Swiss-Prot; the two TrEMBL entries are Drosophila Sec71 and garz), and ARFGEF1 has the SEC7 domain (UniProt DOMAIN 709..840) plus a direct measurement of the activity on the human protein. Note that no more specific term is available: GO:0005085 carries 14 merged ids among its secondaryIds, including GO:0005086, and has no substrate-specific is_a children at all. Those merged ids now resolve to the parent, so their former labels are gone from the ontology; what identifies the absorbed activities is the narrow synonym list GO:0005085 still carries, naming ARF, Rab, Rac, Ral, Ran, Rap, Ras, Rho and Sar guanyl-nucleotide exchange factor activity. Consistently, not one of the eighteen donors holds any descendant of GO:0005085 - they all hold GO:0005085 itself. So this is already the maximal term and no child should be proposed.
Supporting Evidence:
PMID:10393931
A 39-kDa fragment spanning the Sec7 domain catalyzed loading of guanosine 5'-[gamma-thio]triphosphate onto class I ARFs and displayed clear sensitivity to BFA.
file:human/ARFGEF1/ARFGEF1-bioinformatics/RESULTS.md
All 18 protein donors of the GO:0005085 IBA carry their own experimental evidence for that term, 16 of them as Swiss-Prot entries, and none holds a descendant of it.
GO:0005085 guanyl-nucleotide exchange factor activity
IDA
PMID:15644318
BIG1 is a binding partner of myosin IXb and regulates its Rh...
ACCEPT
Summary: The direct-assay record of the exchange activity. BIG1 loads GTP onto the class I ARFs ARF1 and ARF3; the reaction is catalysed by the SEC7 domain and inhibited by brefeldin A.
Reason: Correct and core. The substrate identity, which is the single most characteristic fact about this protein, cannot be carried by the term itself because the substrate-specific GEF terms were merged into GO:0005085, so it is recorded here as has_input extensions on ARF1 and ARF3 instead. Two independent grounds for the substrate call: the cited paper itself describes BIG1 as a GEF for Arf1, and UniProt FUNCTION states it for both class I ARFs. A separate curation observation, not a defect in this row: the experiment that actually measured the exchange reaction and the brefeldin A inhibition constant on the human protein is PMID:10393931, which carries no GO annotation anywhere in GOA.
Supporting Evidence:
PMID:15644318
we found BIG1, a guanine nucleotide exchange factor for ADP-ribosylation factor (Arf1), as a potential binding partner for myosin IXb
file:human/ARFGEF1/ARFGEF1-uniprot.txt
Promotes guanine-nucleotide exchange on ARF1 and ARF3.
PMID:10393931
BFA did not compete with ARF for interaction with p200 but, rather, acted as an uncompetitive inhibitor that only targeted the p200-ARF complex with an inhibition constant of 7 microM.
GO:0005085 guanyl-nucleotide exchange factor activity
IEA
GO_REF:0000120
ACCEPT
Summary: Automatic assignment from the Sec7 domain signature (IPR000904 and its structural superfamily IPR035999), duplicating the IBA and the IDA.
Reason: The signature-derived call is right for once - this is not a name-implied catalytic term on a protein that lost its active site. UniProt annotates a genuine SEC7 domain at 709..840, and the activity has been measured directly on a fragment spanning that domain. Redundant with the IDA but not wrong.
Supporting Evidence:
PMID:10393931
A 39-kDa fragment spanning the Sec7 domain catalyzed loading of guanosine 5'-[gamma-thio]triphosphate onto class I ARFs and displayed clear sensitivity to BFA.
GO:0005096 GTPase activator activity
IDA NOT
PMID:15644318
BIG1 is a binding partner of myosin IXb and regulates its Rh...
ACCEPT
Summary: A NOT annotation, asserting that BIG1 does not stimulate GTP hydrolysis. It is the correct reading of an experiment in which BIG1 did the opposite - it inhibited the Rho-GAP activity of myosin IXb, with an IC50 of 0.06 microM, by competing with RhoA for the myosin IXb zinc finger/GAP domain.
Reason: The negation is well founded and worth keeping explicitly, because a Sec7-domain protein sitting in a paper about a GAP is exactly the situation in which a positive GAP annotation could be created by mistake. BIG1 is a GEF, not a GAP, and its effect on the Rho pathway runs through inhibiting someone else's GAP. This row and the GO:0034260 IDA from the same paper are two halves of one coherent result, not a contradiction.
Supporting Evidence:
PMID:15644318
These results suggest that BIG1 and RhoA compete with each other for the binding to myosin IXb, thus resulting in the inhibition of the GAP activity by BIG1.
GO:0005515 protein binding
IPI
PMID:10716990
Identification and localization of two brefeldin A-inhibited...
REMOVE
Summary: BIG1 and its paralogue BIG2 (ARFGEF2) co-purify in >670 kDa complexes from brain cytosol and co-immunoprecipitate reciprocally and near-quantitatively. The interaction is real and well replicated, but the term states nothing about function.
Reason: Removal is on the ground that GO:0005515 is uninformative, not that the interaction is doubtful - IntAct holds six distinct experiments across four publications for this pair, with a maximum MI score of 0.84. No more informative molecular function can be substituted, because BIG1 and BIG2 are paralogues rather than identical, and the shared assembly is described by UniProt only as the same or very similar macromolecular complexes, which does not support a named complex term.
Supporting Evidence:
PMID:10716990
All observations were consistent with the conclusion that significant fractions of BIG1 and BIG2 exist as components of the same macromolecular complexes in bovine brain cytosol and are similarly localized in cultured cells.
GO:0005515 protein binding
IPI
PMID:15644318
BIG1 is a binding partner of myosin IXb and regulates its Rh...
MODIFY
Summary: The BIG1-myosin IXb interaction, which is the central finding of the cited paper - found by yeast two-hybrid with the myosin IXb tail as bait, confirmed by co-immunoprecipitation of endogenous proteins, and shown to be direct with isolated proteins.
Reason: An informative molecular function term exists and is already used for this same interaction elsewhere in the same GOA record, so the bare protein binding row should be replaced by it rather than removed. MYO9B is a myosin and GO:0017022 myosin binding is supported by direct binding of the purified proteins.
Proposed replacements: myosin binding
Supporting Evidence:
PMID:15644318
Using the isolated proteins, it was demonstrated that myosin IXb and BIG1 directly bind to each other.
GO:0005515 protein binding
IPI
PMID:16417406
Hem-1 complexes are essential for Rac activation, actin poly...
REMOVE
Summary: ARFGEF1 was identified by mass spectrometry among the proteins co-purifying with Hem-1 (NCKAP1L) leading-edge complexes from neutrophil-like cells - one of the diverse set of potential polarity-regulating proteins the paper describes, with no BIG1-specific follow-up.
Reason: Uninformative term, and the evidence is a single affinity-purification co-complex identification rather than a measured binary interaction - IntAct holds no record for this pair at all. Nothing more specific can be substituted without inventing a function from a co-purification. The removal does not dispute that BIG1 was in the preparation, and a role for BIG1 in cell polarity is independently supported by PMID:22084092.
Supporting Evidence:
PMID:16417406
A subset of these leading edge complexes are biochemically separable from the WAVE2 complex and contain a diverse set of potential polarity-regulating proteins.
GO:0005515 protein binding
IPI
PMID:16866877
AMY-1 (associate of Myc-1) localization to the trans-Golgi n...
REMOVE
Summary: MYCBP (AMY-1) interacts with BIG1 in vitro, but the paper's own conclusion is that AMY-1 reaches the trans-Golgi network through BIG2 and not through BIG1.
Reason: Uninformative term with no substitutable molecular function. The interaction itself is well replicated across this directed study and three proteome-scale maps (six distinct IntAct experiments, four publications, MI 0.79), so this is a removal for lack of functional content, not for doubt about the binding. Worth recording that the cited paper explicitly assigns the functional consequence to the paralogue.
Supporting Evidence:
PMID:16866877
Furthermore, we have demonstrated that AMY-1 is associated with the TGN through interacting with BIG2 but not with BIG1 using an RNA interference approach, although AMY-1 can interact with both BIG1 and BIG2 in vitro.
GO:0005515 protein binding
IPI
PMID:18292223
Association of guanine nucleotide-exchange protein BIG1 in H...
REMOVE
Summary: Nucleolin co-immunoprecipitates with BIG1 from purified nuclei, but the association is abolished by RNase A or DNase, so it depends on nucleic acid and is not demonstrated to be a direct protein-protein contact.
Reason: Uninformative term, and the underlying evidence is the weakest of any partner here - one distinct IntAct experiment, three of the five records spoke-expanded, against a partner that has 303 IntAct partners of its own. The nucleic-acid dependence is stated by the authors themselves. The nucleolar co-assembly is retained through the GO:0030532 row (modified to the correct RNA class) rather than through a bare binding term.
Supporting Evidence:
PMID:18292223
Antibodies against BIG1 or nucleolin coprecipitated both proteins from nuclei, which was abolished by the incubation of nuclei with RNase A or DNase, indicating that the interaction depended on nucleic acids.
GO:0005515 protein binding
IPI
PMID:19020088
Interaction of brefeldin A-inhibited guanine nucleotide-exch...
MODIFY
Summary: The BIG1-KIF21A interaction, found by mass spectrometry of BIG1 immunoprecipitates and confirmed by reciprocal immunoprecipitation of the endogenous proteins, with the KIF21A C-terminal WD-40 tail mapped against the BIG1 C-terminal region.
Reason: KIF21A is a kinesin and GO:0019894 kinesin binding is an informative term that the evidence supports directly, so the bare binding row should be replaced rather than removed. The interaction is among the best supported in this record - six distinct IntAct experiments over two publications, maximum MI score 0.71, with domain-level mapping.
Proposed replacements: kinesin binding
Supporting Evidence:
PMID:19020088
Reciprocal immunoprecipitation (IP) of endogenous proteins and microscopically apparent overlap of immunoreactive BIG1 with overexpressed GFP-KIF21A in the perinuclear region were consistent with an interaction of KIF21A-BIG1.
GO:0005515 protein binding
IPI
PMID:19020088
Interaction of brefeldin A-inhibited guanine nucleotide-exch...
MODIFY
Summary: The same KIF21A interaction recorded against the second annotated isoform of KIF21A. The accession resolves to the canonical 1674-residue Swiss-Prot entry, so this is an isoform-level record of one interaction, not a second partner.
Reason: Same reasoning as the companion row for the canonical accession - replace the uninformative binding term with GO:0019894 kinesin binding, which the reciprocal endogenous immunoprecipitation supports. The isoform token is retained because it records which construct was tested.
Proposed replacements: kinesin binding
Supporting Evidence:
PMID:19020088
Overexpression of full-length KIF21A and BIG1 and their fragments in HEK293 cells followed by reciprocal IP revealed that the C-terminal tail of KIF21A, with seven WD-40 repeats, may interact with structure in the C-terminal region of BIG1.
GO:0005515 protein binding
IPI
PMID:19332778
Interaction of phosphodiesterase 3A with brefeldin A-inhibit...
REMOVE
Summary: PDE3A associates with BIG1 and BIG2 in AKAP complexes; depleting or inhibiting PDE3A disperses BIG1 from the perinuclear Golgi and lowers membrane ARF1-GTP, the interpretation being that local cAMP hydrolysis keeps PKA from inhibiting BIG1 exchange activity.
Reason: Uninformative term. The interaction is supported (two distinct IntAct experiments, MI 0.58) and the regulatory story it belongs to is real, but no more specific molecular function term is available for binding a phosphodiesterase, and inventing one from interaction evidence would not be justified. The cAMP arm of BIG1 regulation remains represented by the GO:0034237 rows.
Supporting Evidence:
PMID:19332778
Specific depletion of HeLa cell PDE3A with small interfering RNA significantly decreased membrane-associated BIG1 and BIG2, which by confocal immunofluorescence microscopy were widely dispersed from an initial perinuclear Golgi concentration.
GO:0005515 protein binding
IPI
PMID:19332778
Interaction of phosphodiesterase 3A with brefeldin A-inhibit...
REMOVE
Summary: The BIG1-BIG2 association recorded again, from the PDE3A study.
Reason: Same as the other ARFGEF2 rows - the paralogue association is genuine and repeatedly observed, but GO:0005515 conveys no function and there is no informative replacement. Removal does not dispute the interaction.
Supporting Evidence:
PMID:19332778
Specific depletion of HeLa cell PDE3A with small interfering RNA significantly decreased membrane-associated BIG1 and BIG2, which by confocal immunofluorescence microscopy were widely dispersed from an initial perinuclear Golgi concentration.
GO:0005515 protein binding
IPI
PMID:19651892
A role of histone H3 lysine 4 methyltransferase components i...
REMOVE
Summary: BIG1 targets mDpy-30, a subunit of H3K4 methyltransferase complexes, to the trans-Golgi network. The partner accession named in the WITH/FROM, UniProtKB:Q8K3E7, is the 99-residue rat Dpy30 entry.
Reason: Uninformative term with no informative replacement - BIG1 recruiting a cargo to the TGN is a trafficking event, already covered by the vesicle-transport and TGN rows, not a nameable binding function. Two separate weaknesses are worth recording rather than hiding - IntAct holds no record for this pair, and the named partner is a rodent accession on an annotation made against the human gene, so the row as written asserts a human-rat interaction that the paper did not set out to measure.
Supporting Evidence:
PMID:19651892
The TGN targeting of mDpy-30 is mediated by BIG1, a TGN-localized guanine nucleotide exchange factor for adenosine diphosphate ribosylation factor GTPases.
GO:0005515 protein binding
IPI
PMID:22084092
Effects of brefeldin A-inhibited guanine nucleotide-exchange...
REMOVE
Summary: KANK1 and BIG1 reciprocally immunoprecipitate, and depleting either produces the same defect in directed migration and Golgi/MTOC orientation, but the authors are explicit that the physical association involves only a small fraction of each protein and that colocalisation was not clear.
Reason: Uninformative term, and the physical evidence is deliberately hedged by the authors. The functional convergence that the paper actually establishes is captured by the GO:2000114 and migration rows from the same reference, which is the right place for it. No informative molecular function can be substituted.
Supporting Evidence:
PMID:22084092
Although colocalization of overexpressed KANK1 and endogenous BIG1 in HeLa cells was not clear microscopically, their reciprocal immunoprecipitation (IP) is compatible with the presence of small percentages of each protein in the same complexes.
GO:0005515 protein binding
IPI
PMID:22084092
Effects of brefeldin A-inhibited guanine nucleotide-exchange...
MODIFY
Summary: The BIG1-KIF21A interaction recorded again from the KANK1 study, which treats KIF21A as an established BIG1 partner.
Reason: Consistent with the other two KIF21A rows, replace the uninformative binding term with GO:0019894 kinesin binding. Judged per partner rather than per gene - this row gets an informative replacement because the partner is a kinesin and the interaction has domain-mapped, reciprocal endogenous support, whereas the KANK1 row from the same reference does not.
Proposed replacements: kinesin binding
Supporting Evidence:
PMID:22084092
Among proteins that interact with BIG1, kinesin family member 21A (KIF21A), a plus-end-directed motor protein, moves cargo away from the microtubule-organizing center (MTOC) on microtubules.
GO:0005515 protein binding
IPI
PMID:22084092
Effects of brefeldin A-inhibited guanine nucleotide-exchange...
REMOVE
Summary: The BIG1-BIG2 association recorded a third time, from the KANK1 study.
Reason: Same as the other ARFGEF2 rows - uninformative term, genuine interaction, no informative replacement available.
Supporting Evidence:
PMID:10716990
Western blotting using anti-peptide antibodies specific for BIG1 or BIG2 demonstrated that approximately 70% of BIG2 was immunoprecipitated along with 100% of BIG1 by the anti-BIG1 IgG
GO:0005515 protein binding
IPI
PMID:23572552
ACBD3 interaction with TBC1 domain 22 protein is differentia...
REMOVE
Summary: ARFGEF1 co-purified with TBC1D22A in affinity purification-mass spectrometry experiments from a study whose subject is ACBD3. The authors make a strong specificity argument - ARFGEF1 peptides appeared in only the six TBC1D22A/B purifications out of more than 2,100 runs in that laboratory.
Reason: Uninformative term. The specificity argument makes the co-purification credible, so this is not a removal for doubt, but binding a Rab33 GTPase-activating protein has no informative molecular function term, and the paper reports no functional consequence for ARFGEF1. IntAct holds no record for this pair.
Supporting Evidence:
PMID:23572552
This interaction is highly specific, as peptides to ARFGEF1 were detected in only those six deletion and full-length TBC1D22A/B AP-MS experiments out of a total of >2,100 AP-MS runs in our lab
GO:0005515 protein binding
IPI
PMID:23572552
ACBD3 interaction with TBC1 domain 22 protein is differentia...
REMOVE
Summary: The companion TBC1D22B co-purification from the same affinity purification-mass spectrometry study.
Reason: Uninformative term, same reasoning as the TBC1D22A row. This one is additionally the thinnest interaction record in the set - IntAct holds a single spoke-expanded pull-down with an MI score of 0.35, and TBC1D22B has 175 IntAct partners of its own.
Supporting Evidence:
PMID:23572552
Using affinity purification-mass spectrometry, we identified the putative Rab33 GTPase-activating proteins TBC1D22A and TBC1D22B as ACBD3-interacting factors.
GO:0005515 protein binding
IPI
PMID:33961781
Dual proteome-scale networks reveal cell-specific remodeling...
REMOVE
Summary: The BIG1-MYCBP interaction recovered by BioPlex, a proteome-scale affinity-purification interactome.
Reason: Uninformative term. The reference supports 9,514 annotations across GOA, so it is a proteome-scale screen rather than a study of this gene, and every row it contributes to ARFGEF1 is GO:0005515. What it contributes to the rest of GOA was not enumerated - the QuickGO result is paginated - and is not claimed here. It does add independent replication for the MYCBP interaction, which is why the removal rests on the term's lack of content and not on evidential weakness.
GO:0005515 protein binding
IPI
PMID:35271311
OpenCell: Endogenous tagging for the cartography of human ce...
REMOVE
Summary: The BIG1-MYCBP interaction recovered again by OpenCell endogenous tagging.
Reason: Uninformative term, proteome-scale source (2,876 annotations in GOA from this reference; the rows it contributes to ARFGEF1 are all GO:0005515, and the full set is paginated so it was not enumerated). Replication of the MYCBP interaction is noted; the removal is about the term.
GO:0005515 protein binding
IPI
PMID:35271311
OpenCell: Endogenous tagging for the cartography of human ce...
REMOVE
Summary: The BIG1-BIG2 association recovered by OpenCell endogenous tagging.
Reason: Uninformative term. Consistent with the other ARFGEF2 rows, and from a proteome-scale source that contributes only GO:0005515.
GO:0005515 protein binding
IPI
PMID:40205054
Multimodal cell maps as a foundation for structural and func...
REMOVE
Summary: The BIG1-MYCBP interaction recovered a fourth time, by the multimodal cell maps study.
Reason: Uninformative term, proteome-scale source (3,026 annotations in GOA from this reference). Four independent observations make MYCBP one of the better-replicated BIG1 partners, but replication of a contentless term does not create content.
GO:0005634 nucleus
EXP
PMID:14973189
Nuclear localization and molecular partners of BIG1, a brefe...
KEEP AS NON CORE
Summary: In serum-starved HepG2 cells a large fraction of endogenous BIG1 is nuclear, shown by confocal immunofluorescence and by Western blotting of purified subnuclear fractions.
Reason: The localisation is well demonstrated and conditional on serum starvation, but no nuclear activity has been identified in the two decades since - UniProt still says the function in the nucleus remains to be determined, and ARF was never detected in nuclear BIG1 immunoprecipitates, so the nuclear pool is not doing the protein's known job. Non-core rather than core.
Supporting Evidence:
PMID:14973189
After incubation overnight without serum, a large fraction of endogenous BIG1 was in the nuclei.
PMID:14973189
Also of note, ARF was never detected among proteins precipitated from purified nuclei by anti-BIG1 antibodies
GO:0005634 nucleus
IEA
GO_REF:0000044
KEEP AS NON CORE
Summary: The UniProt subcellular-location keyword mapping for Nucleus, tracking the experimental EXP row from the same study.
Reason: Correct and consistent with the experimental row; marked non-core for the same reason - the nuclear pool has no assigned function.
GO:0005654 nucleoplasm
IDA
GO_REF:0000052
KEEP AS NON CORE
Summary: Human Protein Atlas immunofluorescence places BIG1 in the nucleoplasm, alongside its Golgi and cytosolic calls.
Reason: Consistent with the independently demonstrated nuclear pool from PMID:14973189, so the antibody call is corroborated rather than standing alone. Non-core because no nuclear function is known.
GO:0005730 nucleolus
IDA
PMID:14973189
Nuclear localization and molecular partners of BIG1, a brefe...
KEEP AS NON CORE
Summary: BIG1 concentrates in nucleoli of serum-starved HepG2 cells, visible against weaker nuclear matrix staining, and is recovered in purified nucleolar fractions.
Reason: Directly observed by two independent methods in the same study and followed up in PMID:18292223, which identified nucleolin, fibrillarin and U3 snoRNA as the associated molecules. Still non-core - the nucleolar role remains undetermined.
Supporting Evidence:
PMID:14973189
BIG1 was localized with nucleoporin p62 at the nuclear envelope (probably during nucleocytoplasmic transport) and also in nucleoli, clearly visible against the less concentrated overall matrix staining.
GO:0005730 nucleolus
IEA
GO_REF:0000044
KEEP AS NON CORE
Summary: The UniProt subcellular-location keyword mapping for Nucleolus, tracking the IDA from PMID:14973189.
Reason: Correct, redundant with the IDA, and non-core for the same reason.
GO:0005737 cytoplasm
IEA
GO_REF:0000044
ACCEPT
Summary: The UniProt subcellular-location keyword mapping for Cytoplasm. BIG1 is substantially cytosolic, being a peripheral protein that cycles on and off membranes.
Reason: Broader than the cytosol IDA rows but not wrong, and the cytosolic pool is quantitatively the major one - both BIG1 and BIG2 were clearly more abundant in cytosol than in microsomes. A general term that is correct does not need modifying.
Supporting Evidence:
PMID:10716990
On Western blot analysis, both BIG1 and BIG2 were clearly more abundant in the cytosol than in the microsomal fractions.
GO:0005794 Golgi apparatus
EXP
PMID:12571360
Protein kinase A-anchoring (AKAP) domains in brefeldin A-inh...
ACCEPT
Summary: Subcellular fractionation showed BIG1, along with BIG2, translocating from cytosol to Golgi and other membranes after treatment with 8-Br-cAMP or forskolin.
Reason: Directly measured and core - this is where the exchange activity acts. A caveat for the record rather than an objection: this paper is titled and framed for the paralogue BIG2, and BIG1 appears in the abstract only in the translocation result. That result is nonetheless explicitly about BIG1, and the localisation is independently established by several other studies.
Supporting Evidence:
PMID:12571360
Western blot analysis of subcellular fractions demonstrated translocation of BIG2 (and BIG1) from cytosol to the Golgi and other membrane structures after incubation of cells with 8-Br-cAMP or forskolin.
GO:0005794 Golgi apparatus
IDA
GO_REF:0000052
ACCEPT
Summary: Human Protein Atlas immunofluorescence Golgi call.
Reason: Corroborates the experimental Golgi localisation from several independent studies. Core.
GO:0005794 Golgi apparatus
IEA
GO_REF:0000120
ACCEPT
Summary: Automatic Golgi assignment from the UniProt subcellular-location keyword and an ARBA rule.
Reason: Correct and redundant with two experimental rows for the same term.
GO:0005802 trans-Golgi network
IBA
GO_REF:0000033
ACCEPT
Summary: Phylogenetic propagation placing BIG1 as active in the trans-Golgi network, from a narrow node whose donors are Drosophila Sec71 and rat Arfgef1 and Arfgef2.
Reason: The best-specified localisation row in the record, and the is_active_in qualifier is the right one - this is where the exchange reaction happens. Every protein donor was checked: all three carry their own IDA for GO:0005802 itself, not for a parent, so this propagation lands at the same specificity as its donors rather than above them. Independent human evidence agrees - ARL1 recruitment targets BIG1 specifically to the trans-Golgi.
Supporting Evidence:
PMID:22291037
We show that in mammalian cells, Arl1 is necessary for Golgi recruitment of BIG1 and BIG2 but not GBF1. Thus, Arl1 acts to direct a trans-Golgi-specific Arf1 GEF, and hence active Arf1, to the trans side of the Golgi.
file:human/ARFGEF1/ARFGEF1-bioinformatics/RESULTS.md
All 3 protein donors of the GO:0005802 IBA carry their own experimental evidence for GO:0005802 itself rather than for a parent of it.
GO:0005802 trans-Golgi network
IEA
GO_REF:0000107
ACCEPT
Summary: Ensembl Compara orthology transfer of the trans-Golgi network localisation from rat Arfgef1.
Reason: A one-to-one ortholog transfer between the same gene in rat and human, where the donor carries its own IDA for exactly this term. Redundant with the IBA and the ISS but sound.
GO:0005802 trans-Golgi network
ISS
GO_REF:0000024
ACCEPT
Summary: Curator-judged sequence-similarity transfer from rat Arfgef1, the direct ortholog.
Reason: Same donor and same term as the Ensembl row, with a curator in the loop. Sound; redundant with the IBA.
GO:0005829 cytosol
IDA
GO_REF:0000052
ACCEPT
Summary: Human Protein Atlas immunofluorescence cytosol call.
Reason: Consistent with the biochemical demonstration that the cytosolic pool is the larger one. A peripheral Arf-GEF is expected in both compartments.
GO:0005829 cytosol
IDA
PMID:10716990
Identification and localization of two brefeldin A-inhibited...
ACCEPT
Summary: Endogenous BIG1 shows a punctate cytosolic distribution by immunofluorescence and is more abundant in cytosol than in microsomes by Western blot; gel filtration puts most of it in >670 kDa cytosolic complexes.
Reason: Directly measured by three methods in one study. The cytosolic pool is the reservoir from which BIG1 is recruited to membranes, so the location is real and functionally meaningful rather than a fractionation artefact.
Supporting Evidence:
PMID:10716990
When observed by immunofluorescence in HeLa S3 and HepG2 cells, endogenous BIG1 and coexpressed BIG2 were distributed in a punctate pattern throughout the cytosol, and also concentrated in the perinuclear region
GO:0006887 exocytosis
IEA
GO_REF:0000117
MARK AS OVER ANNOTATED
Summary: An ARBA machine-learning rule assigns exocytosis to BIG1 at family level. No experiment on human BIG1 has measured a plasma-membrane fusion or secretion step.
Reason: GO:0006887 is defined as a secretion process that ends when molecules are secreted from the cell. BIG1 demonstrated territory is upstream of that - the trans-Golgi network and endosomes. Where BIG1 has been implicated in surface delivery the cargo is an integral membrane protein, ABCA1 or GABA-A receptors, which is Golgi-to-plasma-membrane transport, a sibling process that by its own definition precedes exocytosis. The supported level is GO:0016192 vesicle-mediated transport, which this record already carries by IBA, so no replacement term is proposed.
Propagation Review
Root cause: TERM SCOPING PROBLEM
Failure modes: GRANULARITY MISMATCH
Sources checked:
ARBA:ARBA00027178 Β· ARBA machine-learning rule SOURCE WEAK OR INFERRED
Family-level statistical rule with no BIG1-specific evidence. It lands on a descendant of GO:0016192 that requires the terminal secretion step, which no experiment on BIG1 has measured.
GO:0006887 exocytosis
TAS
PMID:10212200
Purification and cloning of a brefeldin A-inhibited guanine ...
MARK AS OVER ANNOTATED
Summary: A 2003 ProtInc traceable-author statement drawn from the BIG1/BIG2 cloning paper. That paper's statements about secretion are background about the ARF family and about brefeldin A, and its only biochemical assay is on BIG2, not BIG1.
Reason: The cited reference contains no experiment on BIG1 at all - the recombinant nucleotide-exchange assay in it was performed on BIG2. The exocytosis claim traces to an introductory statement that brefeldin A blocks protein secretion, which is a property of the drug and the pathway, not a measured BIG1 function. Keeping the row would let a background sentence stand as a process annotation. GO:0016192, already present by IBA, is the level the evidence supports.
Supporting Evidence:
PMID:10212200
BIG2, synthesized as a His6 fusion protein in Sf9 cells, accelerated guanosine 5'-3-O-(thio)triphosphate binding by recombinant ARF1, ARF5, and ARF6.
GO:0007030 Golgi organization
IBA
GO_REF:0000033
ACCEPT
Summary: Phylogenetic propagation of Golgi organization from a node whose only protein donor is ARFGEF1 itself, reflecting a PAINT curator's judgement that ARFGEF1's own experimental evidence establishes the function at that node.
Reason: The self-referential WITH/FROM is correct and expected, not circular - it marks that the experimental grounding sits on the target gene, and ARFGEF1 carries two independent IMPs for this term. The function is also the clearest non-catalytic role BIG1 has.
Propagation Review
Root cause: NO FAILURE CORE
Sources checked:
PANTHER:PTN002580171 Β· PAINT ancestral node SUPPORTS TRANSFER
Ancestral node at which the function was placed; not a protein.
UniProtKB:Q9Y6D6 Β· ARFGEF1 SUPPORTS TRANSFER
The target's own gene product, carrying two IMPs for GO:0007030. A descendant evidence used to place the IBD, not duplicated support.
GO:0007030 Golgi organization
IMP
PMID:17227842
BIG1, a brefeldin A-inhibited guanine nucleotide-exchange pr...
ACCEPT
Summary: Electron microscopy of BIG1-depleted HepG2 cells showed Golgi membranes that were less sharply defined, with more vesicle-like structures at the trans face; BIG2 siRNA did not do this.
Reason: A direct, paralogue-controlled loss-of-function result in human cells. Together with PMID:20360857 this makes Golgi structural maintenance one of BIG1's two best-established roles.
Supporting Evidence:
PMID:17227842
By electron microscopy, Golgi membranes in BIG1-depleted cells were less sharply defined than those in mock or BIG2 siRNA-treated cells, with more vesicle-like structures at the transface.
GO:0007030 Golgi organization
IMP
PMID:20360857
Specific functions of BIG1 and BIG2 in endomembrane organiza...
ACCEPT
Summary: siRNA in human cells, with live and fixed imaging, showed that suppressing BIG1 fragments the Golgi into mini-stacks that remain polarised and functional for cargo export, and that BIG2 cannot substitute.
Reason: The strongest evidence for a BIG1 role that is structural rather than simply catalytic, and the reason UniProt records the Golgi-maintenance function as possibly independent of GEF activity. The mini-stacks remaining export-competent is the key observation - it separates Golgi morphology from Golgi throughput.
Supporting Evidence:
PMID:20360857
In contrast, suppression of BIG1 induces the formation of Golgi mini-stacks still polarized and functional in terms of cargo export.
PMID:20360857
BIG1 is required to maintain the normal morphology of the Golgi; BIG2 is important for endosomal compartment integrity and cannot replace the function of BIG1 in Golgi organization.
GO:0009101 glycoprotein biosynthetic process
IMP
PMID:17227842
BIG1, a brefeldin A-inhibited guanine nucleotide-exchange pr...
ACCEPT
Summary: BIG1 siRNA, but not BIG2 siRNA, produced integrin beta-1 with aberrant electrophoretic mobility that was still N-glycosylated but incorrectly processed, and the defect was reversible on BIG1 recovery.
Reason: The acts_upstream_of_positive_effect qualifier is exactly right and is what makes this row defensible. BIG1 does not perform any glycosyl transfer; it maintains the Golgi compartment in which the resident enzymes act, so it acts upstream of the process rather than participating in it. Substituting involved_in would overstate the claim.
Supporting Evidence:
PMID:17227842
In HepG2 cells treated for 48 or 72 h with BIG1, but not BIG2, siRNA, both the amount and electrophoretic mobility of the initially 130-kDa beta1 were increased.
PMID:17227842
These results indicate a previously unrecognized role for BIG1 in the glycosylation of beta1 by Golgi enzymes, which is critical for its function in developmental and other vital cell processes.
GO:0010256 endomembrane system organization
IEA
GO_REF:0000117
ACCEPT
Summary: An ARBA rule assigning endomembrane system organization, duplicating the IMP from PMID:20360857.
Reason: Unlike the exocytosis ARBA rule, this one lands on a term that a direct human loss-of-function experiment supports - BIG1 depletion reorganises the Golgi, which is an endomembrane compartment. Correct, if redundant.
GO:0010256 endomembrane system organization
IMP
PMID:20360857
Specific functions of BIG1 and BIG2 in endomembrane organiza...
ACCEPT
Summary: The parent-level process record of the same siRNA experiment that gives the Golgi organization IMP, from a study whose explicit scope is endomembrane organisation.
Reason: Correct and appropriately general - the paper compares BIG1 and BIG2 across the Golgi and the recycling endosome, and shows they have distinct, non-redundant roles in the endomembrane system as a whole. Kept alongside the more specific GO:0007030 rows rather than replacing them.
Supporting Evidence:
PMID:20360857
Our data indicate that the human BFA-sensitive large Arf-GEFs have non-redundant functions in cell organization and membrane trafficking.
GO:0016192 vesicle-mediated transport
IBA
GO_REF:0000033
ACCEPT
Summary: Phylogenetic propagation of vesicle-mediated transport from a deep node spanning the Sec7-domain superfamily - Arabidopsis GNOM, GNL1, BIG2, BIG3 and BIG5, yeast Sec7, Gea1, Gea2, Syt1 and Mon2, fission yeast, Drosophila Sec71, C. elegans gbf-1, Dictyostelium, human GBF1, CYTH3 and ARFGEF2, and rat Arfgef2.
Reason: The general term is the correct least common ancestor here, not lazy curation. Seventeen of the eighteen protein donors carry their own experimental evidence (the exception is the Dictyostelium entry, which holds GO:0016192 by IBA only), but they carry it for thirteen different descendant processes - among them GO:0006888, GO:0006890, GO:0006891, GO:0006893, GO:0006895, GO:0042147, GO:0016197 and GO:0048193 - so the donor set genuinely disagrees about which transport step is involved and refining the node term would mean arbitrarily picking one donor. Human ARFGEF1 evidence supports a specific step of its own, endosome-to-Golgi retrograde transport, which is proposed separately as a NEW row rather than by narrowing this one.
Supporting Evidence:
file:human/ARFGEF1/ARFGEF1-bioinformatics/RESULTS.md
the donors' own terms are spread across 13 distinct descendant processes
GO:0016363 nuclear matrix
EXP
PMID:14973189
Nuclear localization and molecular partners of BIG1, a brefe...
KEEP AS NON CORE
Summary: BIG1 was identified by Western blot in purified subnuclear fractions including the nuclear matrix.
Reason: Fractionation evidence from a study that also showed nucleolar and nuclear-envelope pools. Real but non-core - it is part of the undetermined nuclear localisation, not of the Golgi exchange function.
Supporting Evidence:
PMID:14973189
BIG1 was also identified by Western blot analyses in purified subnuclear fractions (e.g., nucleoli and nuclear matrix).
GO:0016363 nuclear matrix
IEA
GO_REF:0000044
KEEP AS NON CORE
Summary: The UniProt subcellular-location keyword mapping for Nucleus matrix, tracking the EXP row.
Reason: Correct, redundant with the EXP row, non-core for the same reason.
GO:0017022 myosin binding
IPI
PMID:15644318
BIG1 is a binding partner of myosin IXb and regulates its Rh...
UNDECIDED
Summary: A myosin binding annotation whose named partner is MYO9A (unconventional myosin-IXa), from a paper whose title, abstract and described experiments all concern myosin IXb.
Reason: The term itself is correct for BIG1 and is independently established by the companion MYO9B row and by the endogenous myosin IIA co-immunoprecipitation in PMID:23918382, so nothing about BIG1 biology turns on this row. What cannot be settled is the partner. The cached record is abstract-only and the publisher full text returns HTTP 403, so I cannot see what the curator saw, and this repository standing rule is that an experimental annotation is not overruled from an abstract. The observation worth passing to the assigning curator is structural rather than textual: querying GOA by this reference returns eighteen annotations in which MYO9A has acquired an exact copy of the MYO9B partner set - RHOA and ARFGEF1 - but none of the MYO9B functional rows (GO:0005096, and the NOT GO:0032011). Four annotations across three genes are involved. I also excluded the obvious alternative explanation: B2RTY4 twelve secondary accessions do not include Q13459 or any of its secondaries, so this is not an accession-identity migration. UNDECIDED, pending someone with full-text access.
Supporting Evidence:
PMID:15644318
Myosin IXb, a member of the myosin superfamily, is a molecular motor that possesses a GTPase activating protein (GAP) for Rho.
file:human/ARFGEF1/ARFGEF1-bioinformatics/RESULTS.md
MYO9A has acquired an exact copy of the MYO9B partner set (ARFGEF1, RHOA) but none of the MYO9B functional rows
GO:0017022 myosin binding
IPI
PMID:15644318
BIG1 is a binding partner of myosin IXb and regulates its Rh...
ACCEPT
Summary: BIG1 binds the myosin IXb tail directly, mapped to the zinc finger/GAP domain, and the interaction was confirmed between the endogenous proteins in normal rat kidney cells as well as with isolated proteins.
Reason: Direct binding of purified proteins with a mapped binding site is about as good as an interaction annotation gets, and myosin binding is an informative molecular function rather than a bare binding term. It is also the basis of a real regulatory effect - occupying the myosin IXb GAP domain excludes RhoA. The function is non-catalytic and secondary to the exchange activity, but it is well enough characterised to keep as a molecular function of BIG1.
Supporting Evidence:
PMID:15644318
Various truncation mutants of the myosin IXb tail domain were produced, and it was revealed that the binding region of myosin IXb to BIG1 is the zinc finger/GAP domain.
GO:0030532 small nuclear ribonucleoprotein complex
IDA
PMID:18292223
Association of guanine nucleotide-exchange protein BIG1 in H...
MODIFY
Summary: RNA immunoprecipitated with BIG1 from HepG2 nuclei ran at about 210 bases and hybridised with oligonucleotides specific for U3, and fibrillarin was found in the same immunoprecipitates. U3 is a small nucleolar RNA, not a small nuclear RNA.
Reason: Wrong RNA class, and the two term definitions make it unambiguous rather than a matter of granularity. GO:0030532 requires at least one RNA of the small nuclear RNA (snRNA) class; GO:0005732 is a ribonucleoprotein complex that contains an RNA molecule of the snoRNA family and associated proteins. They are sibling is_a children of GO:0030529, so GO:0030532 is not a parent of the right answer, it is a different branch. The likeliest cause is a sno-to-sn transcription slip. I stopped at GO:0005732 rather than going on to GO:0031428 box C/D methylation guide snoRNP complex, because U3 is a box C/D snoRNA but guides pre-rRNA cleavage rather than 2-O-methylation, and GO:0031428 requires the complex be capable of ribose-2-O-methylation. The part_of qualifier is kept because it is the curator call and only the RNA class is being corrected - though the authors themselves hedge, and the association is nucleic-acid-dependent.
Supporting Evidence:
PMID:18292223
(32)P labeling of RNAs immunoprecipitated with BIG1 or nucleolin from nuclei revealed bands of approximately 210 bases that also hybridized with U3 small nucleolar (sno)RNA-specific oligonucleotides.
PMID:18292223
Our data demonstrate that BIG1, nucleolin, U3, the U3-binding protein fibrillarin, and the RNA-binding protein La may exist together in nuclear complexes, consistent with a potential role for BIG1 in nucleolar processes.
GO:0030837 negative regulation of actin filament polymerization
IMP
PMID:22084092
Effects of brefeldin A-inhibited guanine nucleotide-exchange...
KEEP AS NON CORE
Summary: An IMP from the KANK1/wound-healing study. The abstract of that paper describes migration and Golgi/MTOC orientation rather than actin measurements, but an independent study from the same laboratory later showed that selective depletion of BIG1 raises F-actin content.
Reason: Retained rather than questioned, because a second paper independently supports the direction of the effect: depleting BIG1 increases F-actin, which is what negative regulation of actin polymerisation predicts. The mechanism is indirect - BIG1 scaffolds a myosin phosphatase complex and restrains myosin regulatory light chain phosphorylation - so this is a downstream consequence rather than a core activity of an Arf exchange factor. Non-core.
Supporting Evidence:
PMID:23918382
Selective depletion of BIG1 or BIG2 enhanced specific phosphorylation of myosin regulatory light chain (T18/S19) and F-actin content, which impaired cell migration in Transwell assays.
GO:0031175 neuron projection development
IEA
GO_REF:0000107
KEEP AS NON CORE
Summary: Ensembl Compara transfer from rat Arfgef1, which carries this term by IMP from a study showing that BIG1 regulates neurite development and that BIG1 expression rises during brain development.
Reason: A one-to-one ortholog transfer whose donor holds its own IMP for exactly this term, so the propagation is sound. Marked non-core because it is a tissue-specific consequence of the general trafficking function rather than a distinct activity - although it is the arm of BIG1 biology that matters clinically, since heterozygous loss of ARFGEF1 causes a neurodevelopmental disorder.
Supporting Evidence:
PMID:34113008
Haploinsufficiency of ARFGEF1 is associated with developmental delay, intellectual disability, and epilepsy with variable expressivity
GO:0032012 regulation of ARF protein signal transduction
IEA
GO_REF:0000002
MODIFY
Summary: The interpro2go mapping from the Sec7 domain signature. It assigns direction-neutral regulation, whereas what a Sec7 domain does is specifically activate - it loads GTP and converts ARF to the active conformation.
Reason: The direction is measured, not assumed: a fragment spanning the BIG1 Sec7 domain catalyses GTP loading onto class I ARFs, and GO:0032014 positive regulation of ARF protein signal transduction exists and is current. Replacing the neutral parent with the signed child loses nothing and states what is known. Whether the same sharpening should be applied to the interpro2go mapping for IPR000904 family-wide is a question for InterPro rather than something to assert from one gene, and it is raised as such in suggested_questions.
Propagation Review
Root cause: TERM SCOPING PROBLEM
Failure modes: GRANULARITY MISMATCH
Sources checked:
InterPro:IPR000904 Β· Sec7 domain SUPPORTS TRANSFER
The domain is genuinely present in ARFGEF1 (UniProt DOMAIN 709..840) and is the catalytic module; only the mapped term direction is unnecessarily neutral.
InterPro:IPR035999 Β· Sec7 domain superfamily SUPPORTS TRANSFER
Structural superfamily signature for the same domain; same assessment.
Supporting Evidence:
PMID:10393931
A 39-kDa fragment spanning the Sec7 domain catalyzed loading of guanosine 5'-[gamma-thio]triphosphate onto class I ARFs and displayed clear sensitivity to BFA.
GO:0034237 protein kinase A regulatory subunit binding
IDA
PMID:12571360
Protein kinase A-anchoring (AKAP) domains in brefeldin A-inh...
ACCEPT
Summary: BIG1 binds PKA regulatory subunits, the basis of its proposed A-kinase anchoring role. UniProt records the specific partners as PRKAR1A and PRKAR2A on the strength of this reference.
Reason: A genuinely informative molecular function, and one that is mechanistically coupled to the core activity rather than incidental - PKA phosphorylation lowers the exchange activity of BIG1 and PP1-gamma restores it, and a PDE3A-containing AKAP complex keeps local cAMP low to prevent that. Anchoring PKA next to the catalytic subunit is exactly the arrangement such a switch needs. Two PKA consequences are easily conflated here and should be kept apart: phosphorylation of Ser-883 is what drives nuclear accumulation, while the site whose phosphorylation lowers exchange activity is unidentified - BIG2 carries alanine at the position corresponding to Ser-883, and its activity falls in parallel with that of BIG1. A caveat recorded rather than acted on: this paper is titled and framed for BIG2 and the AKAP domain mapping in its abstract is that of BIG2, so the BIG1 result sits in the full text, which is not in our cache. UniProt curators, who read it, assert the PRKAR1A and PRKAR2A interactions for the human BIG1 entry with experimental evidence, so I defer to that rather than second-guessing from an abstract.
Supporting Evidence:
PMID:12571360
All findings are consistent with a role for BIG2 as an A kinase-anchoring protein (or AKAP) that could coordinate cAMP and ARF regulatory pathways.
GO:0034237 protein kinase A regulatory subunit binding
IEA
GO_REF:0000117
ACCEPT
Summary: An ARBA rule assigning the same PKA regulatory subunit binding term, duplicating the IDA.
Reason: The automatic call agrees with a curated experimental annotation on the same gene for the same term. Redundant but correct.
GO:0034260 negative regulation of GTPase activity
IDA
PMID:15644318
BIG1 is a binding partner of myosin IXb and regulates its Rh...
ACCEPT
Summary: BIG1 inhibits the Rho-GAP activity of myosin IXb with an IC50 of about 0.06 microM by competing with RhoA for the myosin IXb zinc finger/GAP domain, so the net effect is to slow GTP hydrolysis on RhoA.
Reason: Correct at the level the ontology can express. GO:0034260 is defined as any process that stops or reduces the rate of GTP hydrolysis by a GTPase, which is exactly the outcome measured. A term for negatively regulating a GTPase activator would describe the mechanism more precisely and does not exist - GO:1905098 provides the analogous term on the exchange-factor side but there is no GAP counterpart. The row is consistent with, not contradictory to, the NOT GO:0005096 annotation from the same paper: BIG1 is not a GAP, it disables one.
Supporting Evidence:
PMID:15644318
Interestingly, the GAP activity of myosin IXb was significantly inhibited by the addition of BIG1 with IC(50) of 0.06 microm.
GO:0048471 perinuclear region of cytoplasm
IDA
PMID:22084092
Effects of brefeldin A-inhibited guanine nucleotide-exchange...
ACCEPT
Summary: Endogenous BIG1 concentrates in the perinuclear region, where the Golgi and the microtubule-organizing centre sit - the structure whose orientation toward the leading edge is the measured phenotype of this paper.
Reason: Consistent with the perinuclear concentration seen in every imaging study of BIG1, and functionally meaningful in this paper because Golgi/MTOC positioning relative to the leading edge is what was scored.
Supporting Evidence:
PMID:22084092
Treatment of cells with BIG1- or KANK1-specific siRNA interfered significantly with directed cell migration and initial orientation of Golgi/MTOC toward the leading edge, which was not mimicked by KIF21A depletion.
PMID:10716990
When observed by immunofluorescence in HeLa S3 and HepG2 cells, endogenous BIG1 and coexpressed BIG2 were distributed in a punctate pattern throughout the cytosol, and also concentrated in the perinuclear region
GO:0048471 perinuclear region of cytoplasm
IEA
GO_REF:0000044
ACCEPT
Summary: The UniProt subcellular-location keyword mapping for Cytoplasm, perinuclear region.
Reason: Correct and redundant with the IDA.
GO:0051897 positive regulation of phosphatidylinositol 3-kinase/protein kinase B signal transduction
IEA
GO_REF:0000107
KEEP AS NON CORE
Summary: Ensembl Compara transfer from rat Arfgef1, whose IMP comes from a study reporting that BIG1 regulates neurite development through PI3K-AKT signalling in a GEF-activity-dependent manner.
Reason: A one-to-one ortholog transfer with a donor holding its own IMP for exactly this term, so the propagation mechanism is sound. Non-core because it is a downstream signalling consequence in one cell type rather than an activity of the protein, and because no human experiment has reproduced it.
GO:0090303 positive regulation of wound healing
IMP
PMID:22084092
Effects of brefeldin A-inhibited guanine nucleotide-exchange...
MODIFY
Summary: BIG1 siRNA impaired directed migration and the initial orientation of the Golgi/MTOC toward the leading edge in a scratch-wound assay on a cell monolayer.
Reason: The term overshoots the assay. GO:0090303 is defined at tissue level, as the series of events that restore integrity to a damaged tissue following an injury, whereas what was measured is single-cell-layer directed migration and organelle reorientation in HeLa cells. GO:0030335 positive regulation of cell migration states the measured phenotype without asserting tissue repair. The companion annotation from the same reference, GO:2000114, is accepted unchanged, because Golgi/MTOC reorientation is exactly what that term covers.
Supporting Evidence:
PMID:22084092
Treatment of cells with BIG1- or KANK1-specific siRNA interfered significantly with directed cell migration and initial orientation of Golgi/MTOC toward the leading edge, which was not mimicked by KIF21A depletion.
GO:2000114 regulation of establishment of cell polarity
IMP
PMID:22084092
Effects of brefeldin A-inhibited guanine nucleotide-exchange...
ACCEPT
Summary: Depleting BIG1 disrupts the orientation of the Golgi/MTOC toward the leading edge, the canonical readout of polarity establishment in a migrating cell.
Reason: The term matches the measurement precisely, and it is the half of this paper's phenotype that was assayed directly rather than inferred. It also connects to the structural role of BIG1 at the Golgi, since Golgi positioning is what is being scored.
Supporting Evidence:
PMID:22084092
Our data identify actions of both BIG1 and KANK1 in regulating cell polarity during directed migration
GO:0031267 small GTPase binding
IPI
PMID:27373159
Structural Insights into Arl1-Mediated Targeting of the Arf-...
NEW
Summary: ARL1, an Arf-like small GTPase, binds the N-terminal DCB domain of BIG1. There is a crystal structure of human ARL1 bound to that domain, and the DCB residues that contact ARL1 are required for BIG1 to reach the Golgi in cells.
Reason: GOA carries no ARL1-binding annotation on ARFGEF1 at all, despite two independent 2016 crystal structures of the complex and four UniProt MUTAGEN entries - K105D, Y109K, L156D and Q200E - each recorded as abolishing the ARL1 interaction. This is the recruitment step that puts a soluble exchange factor on the right membrane, so its absence is a substantive gap rather than a missing detail. GO:0031267 is the maximal available term: eleven ids are merged into it, and its narrow synonyms include "ADP-ribosylation factor binding" and "ARF binding" - so an Arf-specific GTPase-binding term did once exist and is now only a synonym. There is no Arf-family child left to propose.
Supporting Evidence:
PMID:27373159
We find that Arl1 binds to the dimerization and cyclophilin binding (DCB) domain in BIG1 and report a crystal structure of human Arl1 bound to this domain.
PMID:27373159
Residues in the DCB domain that bind Arl1 are required for BIG1 to locate to the Golgi in vivo.
PMID:27436755
Structural basis for targeting BIG1 to Golgi apparatus through interaction of its DCB domain with Arl1.
GO:0042802 identical protein binding
IPI
PMID:17640864
Interactions between conserved domains within homodimers in ...
NEW
Summary: BIG1 is a homodimer, held together by an interaction between the DCB domains of the two protomers, with an additional intramolecular DCB-HUS contact mediated by the HUS box.
Reason: UniProt states the homodimer as an experimental fact for the human entry and no GO annotation records it. The oligomeric state is not decorative - the same DCB domain that dimerises is the one ARL1 binds for Golgi recruitment, so dimerisation and targeting are structurally coupled. GO:0042802 is the term GO provides for self-association.
Supporting Evidence:
PMID:17640864
Our data demonstrate a strong interaction between DCB domains within GBF1, BIG1, and BIG2 to maintain homodimers and an interaction between DCB and HUS domains within each homodimer.
file:human/ARFGEF1/ARFGEF1-uniprot.txt
Homodimer (PubMed:17640864).
GO:0030674 protein-macromolecule adaptor activity
IMP
PMID:23918382
Arf guanine nucleotide-exchange factors BIG1 and BIG2 regula...
NEW
Summary: BIG1 holds together a myosin phosphatase complex containing non-muscle myosin IIA, protein phosphatase 1-delta and MYPT1. The association with myosin IIA is independent of exchange activity, and depletion phenotypes are rescued by the BIG1 C-terminal sequence that lies downstream of the catalytic site.
Reason: The scaffolding function of BIG1 is documented twice over - the myosin phosphatase complex here, and PKA regulatory subunit anchoring recorded by GO:0034237 - yet no molecular function term captures the bridging activity itself. GO:0030674 is the term for a molecule that brings other molecules together so they can act in a coordinated way, which is what the authors conclude. Importantly this is not a restatement of the exchange activity: the control in the paper shows the myosin IIA association survives without it, so a GEF-independent function is being annotated. Experiments were in HeLa cells, so an evidence code asserting a human experiment is correct.
Supporting Evidence:
PMID:23918382
Reciprocal coimmunoprecipitation of endogenous HeLa cell BIG1 and BIG2 with myosin IIA was demonstrably independent of Arf guanine nucleotide-exchange factor activity, because effects of BIG1 and BIG2 depletion were reversed by overexpression of the cognate BIG molecule C-terminal sequence that follows the Arf activation site.
PMID:23918382
Thus, by anchoring or scaffolding the assembly, organization, and efficient operation of multimolecular myosin phosphatase complexes that include myosin IIA, protein phosphatase 1Ξ΄, and myosin phosphatase-targeting subunit 1, BIG1 and BIG2 serve to integrate diverse biophysical and biochemical events in cells.
GO:0042147 retrograde transport, endosome to Golgi
IMP
PMID:18417613
Redundant roles of BIG2 and BIG1, guanine-nucleotide exchang...
NEW
Summary: RNAi against both BIG1 and BIG2 in HeLa cells mislocalises TGN and recycling-endosome proteins and blocks retrograde transport of furin from late endosomes to the trans-Golgi network, phenocopying depletion of the AP-1 clathrin adaptor.
Reason: The only named, directional transport step demonstrated for human BIG1, and GOA records nothing from this reference on any gene. It passes the participation test: BIG1 is not cargo, it catalyses the nucleotide exchange that puts ARF-GTP on the membrane and thereby recruits AP-1, so it performs a step the process depends on. Comparator check - GBF1, the other large human Arf-GEF, carries GO:0042147 by IMP, so the term is used for a protein in exactly this role. The knockdown was double, so the row states a requirement that BIG1 shares redundantly with BIG2 rather than a BIG1-specific one; the title of the paper says as much. Cell line confirmed as HeLa from the Materials and Methods of the full text, so IMP on the human gene is species-correct.
Supporting Evidence:
PMID:18417613
We here show that knockdown of both BIG2 and BIG1 by RNAi causes mislocalization of a subset of proteins associated with the TGN and recycling endosomes and blocks retrograde transport of furin from late endosomes to the TGN.
PMID:18417613
Taken together with previous reports, these observations indicate that BIG2 and BIG1 play redundant roles in trafficking between the TGN and endosomes that involves the AP-1 complex.

Core Functions

Catalyses GDP-to-GTP exchange on the class I ADP-ribosylation factors ARF1 and ARF3 at the trans-Golgi network, through a SEC7 domain that brefeldin A inhibits uncompetitively by trapping an abortive enzyme-substrate complex. This is the activity that recruits vesicle coats, chiefly AP-1, and drives traffic between the TGN and endosomes, including retrograde delivery of furin from late endosomes.

Supporting Evidence:
  • PMID:10393931
    A 39-kDa fragment spanning the Sec7 domain catalyzed loading of guanosine 5'-[gamma-thio]triphosphate onto class I ARFs and displayed clear sensitivity to BFA.
  • file:human/ARFGEF1/ARFGEF1-uniprot.txt
    Promotes guanine-nucleotide exchange on ARF1 and ARF3.
  • PMID:18417613
    Taken together with previous reports, these observations indicate that BIG2 and BIG1 play redundant roles in trafficking between the TGN and endosomes that involves the AP-1 complex.

Maintains the morphology of the Golgi ribbon, apparently independently of exchange activity. Depleting BIG1, but not BIG2, breaks the Golgi into mini-stacks that stay polarised and competent for cargo export, and leaves resident Golgi enzymes unable to process integrin beta-1 correctly.

Supporting Evidence:
  • PMID:20360857
    BIG1 is required to maintain the normal morphology of the Golgi; BIG2 is important for endosomal compartment integrity and cannot replace the function of BIG1 in Golgi organization.
  • PMID:17227842
    By electron microscopy, Golgi membranes in BIG1-depleted cells were less sharply defined than those in mock or BIG2 siRNA-treated cells, with more vesicle-like structures at the transface.

Acts as a scaffold that positions regulatory enzymes on their targets. It anchors protein kinase A through its regulatory subunits, in the manner of an A-kinase anchoring protein, which is also how its own catalysis is controlled - PKA phosphorylation lowers exchange activity and PP1-gamma restores it, at a site that has not been identified. Separately it holds together a myosin phosphatase complex of non-muscle myosin IIA, PP1-delta and MYPT1, restraining myosin regulatory light chain phosphorylation and F-actin accumulation, and so tuning directed cell migration and the orientation of the Golgi and MTOC toward a leading edge.

Supporting Evidence:
  • PMID:23918382
    Thus, by anchoring or scaffolding the assembly, organization, and efficient operation of multimolecular myosin phosphatase complexes that include myosin IIA, protein phosphatase 1Ξ΄, and myosin phosphatase-targeting subunit 1, BIG1 and BIG2 serve to integrate diverse biophysical and biochemical events in cells.
  • PMID:22084092
    Our data identify actions of both BIG1 and KANK1 in regulating cell polarity during directed migration

Binds the tail of myosin IXb at its zinc finger/Rho-GAP domain and blocks the GAP activity of that protein by competing with RhoA for the same surface, so that RhoA stays GTP-loaded. BIG1 is itself not a GTPase activator; it inhibits one.

Molecular Function:
myosin binding
Supporting Evidence:
  • PMID:15644318
    These results suggest that BIG1 and RhoA compete with each other for the binding to myosin IXb, thus resulting in the inhibition of the GAP activity by BIG1.
  • PMID:15644318
    Various truncation mutants of the myosin IXb tail domain were produced, and it was revealed that the binding region of myosin IXb to BIG1 is the zinc finger/GAP domain.

Is recruited to the trans-Golgi by the Arf-like GTPase ARL1, which binds the N-terminal DCB domain; the same domain mediates homodimerisation through a DCB-DCB contact and packs against the HUS domain of its own protomer. This is how a soluble exchange factor is delivered to the membrane where its substrates are.

Molecular Function:
small GTPase binding
Cellular Locations:
Supporting Evidence:
  • PMID:27373159
    We find that Arl1 binds to the dimerization and cyclophilin binding (DCB) domain in BIG1 and report a crystal structure of human Arl1 bound to this domain.
  • PMID:17640864
    Our data demonstrate a strong interaction between DCB domains within GBF1, BIG1, and BIG2 to maintain homodimers and an interaction between DCB and HUS domains within each homodimer.

References

Loading supporting content…

Download this section (compressed HTML)

Suggested Questions for Experts

Q: Does the full text of PMID:15644318 contain any experiment on myosin IXa (MYO9A, UniProtKB:B2RTY4)? Querying GOA by that reference returns four annotations across three genes in which MYO9A has an exact copy of the MYO9B partner set (RHOA and ARFGEF1) but none of the MYO9B functional rows, while the abstract of the paper describes only myosin IXb. If the full text has no MYO9A experiment, the affected rows are ARFGEF1 GO:0017022 with B2RTY4, MYO9A GO:0005515 with Q9Y6D6, MYO9A GO:0005515 with P61586, and RHOA GO:0005515 with B2RTY4.

Suggested experts: Ikebe M

Q: Should the interpro2go mapping from IPR000904 (Sec7 domain) be sharpened from GO:0032012 regulation of ARF protein signal transduction to GO:0032014 positive regulation of ARF protein signal transduction? Every characterised Sec7-domain protein is an activator, but the node also covers proteins such as yeast Mon2 whose Sec7 region is degenerate, so the family-wide safety of the change is a question for InterPro rather than something to infer from one gene.

Q: GO:0005085 is now the maximal molecular-function term for any guanine-nucleotide exchange factor. Fourteen ids were merged into it and it has no substrate-specific is_a children; its narrow synonyms still name ARF, Rab, Rac, Ral, Ran, Rap, Ras, Rho and Sar guanyl-nucleotide exchange factor activity, which is the only retrievable record of what was absorbed. Substrate specificity is the single most characteristic property of a GEF and is now expressible only through annotation extensions. Was this consequence intended, and is has_input the recommended way to record it?

Q: PMID:10393931 measured guanine-nucleotide exchange by a fragment spanning the BIG1 Sec7 domain and determined the brefeldin A inhibition constant on the human protein, and it carries no GO annotation anywhere in GOA. The GO:0005085 IDA on ARFGEF1 instead cites PMID:15644318, a study of myosin IXb. Should the exchange annotation be re-referenced or supplemented?

Suggested experts: MelanΓ§on P

Q: ARFGEF1 haploinsufficiency causes a neurodevelopmental disorder in which about half of affected individuals have seizures, and the mouse model attributes the lowered seizure threshold to reduced surface GABA-A receptors. Has GABA-A receptor trafficking been tested in human neurons? PMID:24198228 does not state the species of the neurons used, which is why no human GABA-A trafficking annotation is proposed here.

Suggested experts: Shen X

Suggested Experiments

Experiment: Rescue BIG1-depleted cells with E793K, the substitution of the catalytic glutamate already used as a dominant negative in the literature, and score Golgi ribbon integrity by electron microscopy alongside cargo export and ARF1-GTP loading. If morphology is rescued while ARF1-GTP is not, the structural role is separable and deserves its own molecular function term.

Hypothesis: The requirement of BIG1 for normal Golgi morphology is genuinely independent of its exchange activity, rather than reflecting residual catalysis below the detection threshold of the rescue assays.

Type: mutagenesis and rescue

Experiment: Use auxin-inducible degradation to deplete BIG1 acutely and measure pre-rRNA processing intermediates by northern blot, with U3 snoRNP integrity assessed by fibrillarin and nucleolin co-immunoprecipitation. Compare an NLS-dead mutant (the 711-715 KPK motif) that cannot enter the nucleus with wild type, so the nuclear and Golgi pools can be separated. This is the experiment that would decide whether GO:0005732 for BIG1 should ever acquire a companion biological-process term.

Hypothesis: The nuclear and nucleolar pool of BIG1 has a function in ribosome biogenesis rather than being a storage or sequestration form.

Type: acute depletion and RNA processing

Experiment: Knock in the K105D, Y109K, L156D or Q200E substitutions at the endogenous ARFGEF1 locus and measure Golgi ribbon integrity, furin retrograde transport and integrin beta-1 glycosylation against a full knockout. A separation of phenotypes would show which BIG1 functions require membrane recruitment and which are satisfied by the cytosolic pool.

Hypothesis: ARL1 binding is the rate-limiting step for BIG1 recruitment, so the DCB point mutants that abolish it should phenocopy BIG1 loss.

Type: genome editing and trafficking assays

Experiment: Map the myosin IIA, PP1-delta and MYPT1 binding sites within the C-terminal region that rescues depletion, and test whether the same fragment competes with PKA regulatory subunit binding. If the sites are separable, GO:0030674 is correctly stated once for the protein; if they overlap, the adaptor activity is a single surface and should be described as such.

Hypothesis: The myosin phosphatase scaffold and the AKAP function are two instances of one adaptor surface in the BIG1 C-terminus.

Type: domain mapping and competition binding

Knowledge Gaps

What is not known β€” curated, literature-grounded statements of the open unknowns (the inverse of core functions).

Gap: What BIG1 does in the nucleus and nucleolus is unknown. It concentrates there on serum starvation, associates with nucleolin, fibrillarin and U3 snoRNA in a nucleic-acid-dependent complex, and notably does not bring ARF with it, so the nuclear pool is not performing the known catalytic role of the protein.

BIOLOGY

Provenance (the field's own admissions):

Gap: The PKA phosphorylation site that lowers BIG1 exchange activity has not been identified. Ser-883 is often named for it, but that residue's demonstrated consequence is cAMP-induced nuclear accumulation, and BIG2 - which has alanine at the corresponding position - loses exchange activity in parallel with BIG1 after PKA treatment, so Ser-883 cannot be the responsible site. The authors of the measurement say so themselves.

BIOLOGY

Provenance (the field's own admissions):

Gap: GO can no longer express the substrate specificity of a guanine-nucleotide exchange factor. Fourteen ids were merged into GO:0005085, which has no substrate-specific is_a children; the activities they named survive only as narrow synonyms on the parent, which still lists ARF, Rab, Rac, Ral, Ran, Rap, Ras, Rho and Sar guanyl-nucleotide exchange factor activity. So for ARFGEF1 - whose defining property is that it is an ARF1/ARF3 exchange factor - the most informative molecular function term available says only guanyl-nucleotide exchange factor activity, and the specificity survives only as an annotation extension.

ONTOLOGY

Provenance (the field's own admissions):

Gap: The great majority of what is known about human BIG1 has never reached GO. Of the twenty-six functional papers returned by the affinage deep-research provider, only six have produced any GO annotation on ARFGEF1 and seventeen have produced none anywhere in GOA - including the paper that measured the exchange reaction and the brefeldin A inhibition constant, the PKA/PP1-gamma regulatory switch, both ARL1 crystal structures, the homodimer, the retrograde transport step and the myosin phosphatase scaffold. This is a curation-coverage gap, not an over-annotation problem.

CURATION

Provenance (the field's own admissions):

Deep Research

Affinage

(ARFGEF1-deep-research-affinage.md)

Loading supporting content…

Download this section (compressed HTML)

πŸ“š Additional Documentation

Notes

(ARFGEF1-notes.md)

Loading supporting content…

Download this section (compressed HTML)

Bioinformatics Results

(RESULTS.md)

Loading supporting content…

Download this section (compressed HTML)

πŸ“„ View Raw YAML

Loading supporting content…

Download this section (compressed HTML)