AGFG2 (HRBL) is a human vesicle-trafficking protein with an N-terminal ArfGAP-type zinc-finger domain and a long C-terminal region containing FG repeats and four NPF motifs. In endothelial cells it promotes stimulus-evoked release of von Willebrand factor from Weibel-Palade bodies. Depletion reduces stimulated secretion, whereas overexpression promotes granule release; the precise molecular effector step remains unresolved. Inferred additional roles include acrosome assembly, spermatid nuclear differentiation and intermediate filament organization. AGFG2 acts as a host cofactor for HIV-1 Nef- and Vpu-mediated CD4 downregulation and enhances a Rev-dependent reporter in cooperation with Eps15/Eps15R. It retains the ArfGAP arginine finger at Arg75 but has Thr89 at a site occupied by an Arf-contacting aspartate in ASAP3. Whether this domain has physiologically relevant GTPase-activating activity remains unresolved.
Definition: The regulated exocytosis of a Weibel-Palade body, the rod-shaped secretory granule characteristic of vascular endothelial cells, in which the granule fuses with the plasma membrane and releases its contents, principally von Willebrand factor and P-selectin, to the cell surface and the vessel lumen following secretagogue stimulation.
Justification: A dedicated term could distinguish endothelial Weibel-Palade body release from other forms of regulated exocytosis. The current broad process plus cargo extension already represents the demonstrated AGFG2 role; a new term is an ontology question, not a prerequisite for this review.
Parent term: regulated exocytosis
Supporting Evidence:
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0005737 cytoplasm | IBA GO_REF:0000033 | ACCEPT | Summary: AGFG2 has an inherited cytoplasmic localization. Reason: The target lies below the cytoplasm IBD at PTN002353603 in the actual family tree. A broad compartment remains a correct central annotation when a vesicle-associated pool is also supported; parent-term overlap is not a biological reason to demote an existing source row. Propagation Review Root cause: NO FAILURE CORE Sources checked: FB:FBgn0020304 SUPPORTS TRANSFER Experimental descendant evidence contributes to the ancestral IBD. Transfer is evaluated through target descent and evidence of loss, not pairwise identity or donor number. MGI:MGI:1333754 SUPPORTS TRANSFER Experimental descendant evidence contributes to the ancestral IBD. Transfer is evaluated through target descent and evidence of loss, not pairwise identity or donor number. PANTHER:PTN002353603 SUPPORTS TRANSFER The actual target tree path descends from this positive IBD node; the node records the ancestral functional judgment, not an irrelevant nonprotein token. UniProtKB:P52594 SUPPORTS TRANSFER Experimental descendant evidence contributes to the ancestral IBD. Transfer is evaluated through target descent and evidence of loss, not pairwise identity or donor number. Supporting Evidence: file:human/AGFG2/AGFG2-primary-and-phylogeny-scope.md The actual PTHR46134 tree places O95081 leaf PTN002509056 below PTN002919572 and root PTN002353603. |
| GO:0031410 cytoplasmic vesicle | IBA GO_REF:0000033 | ACCEPT | Summary: AGFG2 inherits association with cytoplasmic vesicles. Reason: The actual target lineage passes through the positive PTN002919572 IBD. Donor vesicle-localization evidence supports that ancestral judgment. Direct AGFG2 secretion experiments provide compatible functional context but did not themselves image target localization, so the location remains an inherited assertion. Propagation Review Root cause: NO FAILURE CORE Sources checked: MGI:MGI:1333754 SUPPORTS TRANSFER Experimental descendant evidence contributes to the ancestral IBD. Transfer is evaluated through target descent and evidence of loss, not pairwise identity or donor number. PANTHER:PTN002919572 SUPPORTS TRANSFER The actual target tree path descends from this positive IBD node; the node records the ancestral functional judgment, not an irrelevant nonprotein token. UniProtKB:P52594 SUPPORTS TRANSFER Experimental descendant evidence contributes to the ancestral IBD. Transfer is evaluated through target descent and evidence of loss, not pairwise identity or donor number. Supporting Evidence: file:human/AGFG2/AGFG2-primary-and-phylogeny-scope.md The actual PTHR46134 tree places O95081 leaf PTN002509056 below PTN002919572 and root PTN002353603. PMID:11711676 Although proacrosomic vesicles form in spermatids that lack Hrb, the vesicles are unable to fuse, blocking acrosome development at step 2. |
| GO:0001675 acrosome assembly | IBA GO_REF:0000033 | ACCEPT | Summary: Retain inherited participation in acrosome assembly. Reason: O95081 descends from the experimentally grounded PTN002919572 assertion. Mouse Hrb promotes proacrosomal-vesicle docking/fusion, an active assembly contribution. Neither the number of descendants supplying experiments, the existence of a closer unassayed ortholog, nor salivary enrichment refutes the ancestral capacity. No target-specific loss was established; this is phylogenetic evidence rather than a direct human AGFG2 germ-cell assay. Propagation Review Root cause: NO FAILURE CORE Sources checked: MGI:MGI:1333754 SUPPORTS TRANSFER Experimental descendant evidence contributes to the ancestral IBD. Transfer is evaluated through target descent and evidence of loss, not pairwise identity or donor number. PANTHER:PTN002919572 SUPPORTS TRANSFER The actual target tree path descends from this positive IBD node; the node records the ancestral functional judgment, not an irrelevant nonprotein token. Supporting Evidence: file:human/AGFG2/AGFG2-primary-and-phylogeny-scope.md The actual PTHR46134 tree places O95081 leaf PTN002509056 below PTN002919572 and root PTN002353603. PMID:11711676 Although proacrosomic vesicles form in spermatids that lack Hrb, the vesicles are unable to fuse, blocking acrosome development at step 2. |
| GO:0007289 spermatid nucleus differentiation | IBA GO_REF:0000033 | ACCEPT | Summary: Retain inherited participation in spermatid nucleus differentiation. Reason: The positive IBD at PTN002919572 is ancestral to AGFG2. Mouse Hrb-deficient spermatids show altered nuclear shaping; PMID:16765935 explicitly includes hrb-null cells despite its H1T2 title. Its preserved early H1T2 polarity marker does not refute every role in nuclear differentiation. Abstract-only access limits mechanistic detail and does not invalidate the curator-supported ancestral assertion. Propagation Review Root cause: NO FAILURE CORE Sources checked: MGI:MGI:1333754 SUPPORTS TRANSFER Experimental descendant evidence contributes to the ancestral IBD. Transfer is evaluated through target descent and evidence of loss, not pairwise identity or donor number. PANTHER:PTN002919572 SUPPORTS TRANSFER The actual target tree path descends from this positive IBD node; the node records the ancestral functional judgment, not an irrelevant nonprotein token. Supporting Evidence: file:human/AGFG2/AGFG2-primary-and-phylogeny-scope.md The actual PTHR46134 tree places O95081 leaf PTN002509056 below PTN002919572 and root PTN002353603. PMID:14724135 As spermiogenesis advances, round-shaped spermatid nuclei of the mutant display several nuclear protrusions, designated nucleopodes. |
| GO:0045109 intermediate filament organization | IBA GO_REF:0000033 | ACCEPT | Summary: Retain inherited intermediate filament organization. Reason: Actual descent from PTN002919572 supports the capacity. The mouse source experiment connects Hrb loss to deficient keratin-5 organization in the acroplaxome. A vesicle-trafficking protein can contribute to filament organization without being a filament polymerization enzyme. Neither the donor count nor incomplete target experimentation establishes loss; the detailed molecular route remains unresolved. Propagation Review Root cause: NO FAILURE CORE Sources checked: MGI:MGI:1333754 SUPPORTS TRANSFER Experimental descendant evidence contributes to the ancestral IBD. Transfer is evaluated through target descent and evidence of loss, not pairwise identity or donor number. PANTHER:PTN002919572 SUPPORTS TRANSFER The actual target tree path descends from this positive IBD node; the node records the ancestral functional judgment, not an irrelevant nonprotein token. Supporting Evidence: file:human/AGFG2/AGFG2-primary-and-phylogeny-scope.md The actual PTHR46134 tree places O95081 leaf PTN002509056 below PTN002919572 and root PTN002353603. PMID:14724135 Yet the acroplaxome in the mutant contains F-actin but is deficient in keratin 5. |
| GO:0005096 GTPase activator activity | IEA GO_REF:0000002 | UNDECIDED | Summary: The ArfGAP-domain activity inference is unresolved. Reason: The target retains Arg75 but carries Thr89 at the aligned ASAP3 Asp484 site. PMID:23433073 predicts reduced AGFG activity from conserved-site divergence; these sequence observations warrant scrutiny of InterPro:IPR001164 transfer. They do not prove inactivity: the same paper reports loss of the Asp-homologous site across GIT proteins, while PMID:10788515 directly measures GIT1/GIT2 GAP activity. R75Q secretion rescue/overexpression in PMID:34369554 is not a GTP-hydrolysis assay and is reported as data not shown. A focused neutral adjudication will distinguish target biochemical evidence, altered catalytic mechanisms and absent assays. Propagation Review Root cause: UNRESOLVED Sources checked: InterPro:IPR001164 UNRESOLVED ArfGAP-domain recognition supports the inference, but AGFG site divergence and the catalytically active GIT counterexample leave its target-specific activity unresolved. Supporting Evidence: PMID:23433073 Only two of the 40 AGFG sequences contain an aspartate at the position homologous to D47 in the other subfamilies (D484 in ASAP3 structure) PMID:23433073 In fact, it is absent from all 18 GIT sequences used in our analyses. PMID:10788515 Unlike ARF-GAP1, GIT1 and GIT2 stimulate hydrolysis of GTP bound to ARF6. PMID:34369554 AGFG2[R75Q] expression also induced secretion of vWF and we have not detected the specific effects of AGFG2[R75Q] expression compared with that of wild-type AGFG2 so far (data not shown). file:human/AGFG2/AGFG2-primary-and-phylogeny-scope.md The stored MAFFT domain alignment maps ASAP3 Arg469 to AGFG2 Arg75 and ASAP3 Asp484 to AGFG2 Thr89. |
| GO:0016020 membrane | HDA PMID:19946888 Defining the membrane proteome of NK cells. | ACCEPT | Summary: Retain experimentally curated membrane association. Reason: The NK-cell membrane-proteome HDA is consistent with a vesicle-trafficking protein. The abstract explicitly includes transiently associated proteins; no transmembrane segment or covalent lipid anchor is required for peripheral membrane binding. High throughput limits spatial/mechanistic detail but does not make the broad location non-core or erroneous. The source supports membrane association without a specific plasma-membrane claim. Supporting Evidence: PMID:19946888 The remaining species were largely involved in cellular processes and molecular functions that could be predicted to be transiently associated with membranes. |
| GO:0045055 regulated exocytosis | IMP PMID:34369554 Arf GTPase-activating proteins SMAP1 and AGFG2 regulate the ... | NEW | Summary: Retain the existing proposal for regulated exocytosis. Reason: Direct human endothelial perturbation and siRNA-resistant expression place AGFG2 in the release machinery for VWF-containing granules. AGFG2 is a trafficking cofactor, not the cargo consumed/released by the process; overexpression promotes release, supporting active regulatory participation beyond generic necessity. The precise effector interaction and fusion step remain unisolated. Cargo delivery and morphology are largely preserved but modest VWF-processing changes preclude an unqualified claim of normal maturation. Role comparators RAB11FIP1/2/5 carry regulated-exocytosis annotations; no existing developmental-process ancestor/descendant or GO-CAM target assertion duplicates this proposal. Supporting Evidence: PMID:34369554 These results suggest that AGFG2 plays an important role in the stimulation-dependent secretion of vWF. file:human/AGFG2/AGFG2-primary-and-phylogeny-scope.md AGFG2 is not the released VWF cargo: it performs a trafficking-cofactor/regulatory step. |
| GO:0044794 host-mediated activation of viral process | IMP PMID:25496667 Genome-wide shRNA screening identifies host factors involved... | NEW | Summary: Retain the existing broad host-cofactor annotation for a viral process. Reason: Human HRBL supports Nef- and Vpu-mediated CD4 removal, with independent Hrbl/Eps15 synergy in a Rev-dependent reporter. CD4 is the removed substrate, whereas AGFG2 assists the host trafficking machinery exploited by viral proteins. GO:0044794 covers promotion of a viral process and does not require a measurable replication advantage; PMID:25496667 found no significant replication reduction after HRBL depletion. Trafficking/cytoskeletal host factors CAV2 and CFL1 provide role-matched GO comparators. Retain this single broad proposal, without asserting direct RNA transport, an identified molecular adaptor mechanism, or redundant viral subprocess terms. The Affinage narrative supports dual Nef/Vpu activity, but its contradictory table and localization/paralog conflations are not used as evidence. Supporting Evidence: PMID:26701340 These results identify HRB as a previously unknown co-factor for HIV-1 Nef-mediated CD4 downregulation and highlight differences with the related protein HRBL, which affects the CD4 downregulation in a dual role as co-factor of both HIV-1 Nef and Vpu. PMID:10613896 A synergistic effect of coexpressing Eps15 or Eps15R with either Hrb or Hrbl, was readily observable (Fig. file:human/AGFG2/AGFG2-deep-research-affinage.md AGFG2 serves as a co-factor for both Nef- and Vpu-mediated CD4 downregulation file:human/AGFG2/AGFG2-primary-and-phylogeny-scope.md Host-mediated activation of viral process means promoting a viral process, not necessarily increasing viral replication; human CAV2 and CFL1 have this annotation from PMID:25556234. |
Loading supporting content…
Download this section (compressed HTML)Q: Does AGFG2 exhibit physiologically relevant GTPase-activating activity, and with which ARF substrate, membrane context or cofactors? Resolve R75 retention and T89 substitution against the catalytically active GIT counterexample; the R75Q secretion phenotype does not measure GAP activity.
Suggested experts: ArfGAP biochemists, InterPro curators
Q: Which germ-cell trafficking and cytoskeletal steps implement the inherited AGFG2 capacities? The positive ancestral placement is verified; target-specific experiments would resolve mechanism rather than justify discarding the inference.
Suggested experts: PAINT curators, germ-cell biologists
Q: Does a more precise host-cofactor term cover viral receptor downregulation, and would a dedicated Weibel-Palade-body exocytosis term add useful detail to the existing process and cargo representation?
Suggested experts: GO Central
Q: Can UniProt incorporate the direct endothelial secretion and HIV cofactor studies while distinguishing target experiments from paralog-specific binding/localization assays?
Suggested experts: UniProt curators
Experiment: Measure purified AGFG2-dependent GTP hydrolysis for ARF substrates with appropriate lipids and cofactors, comparing wild type, R75Q and T89D with active ARFGAP1, ASAP3 and GIT proteins. Validate each mutant biochemically rather than naming it inactive from the secretion phenotype.
Hypothesis: AGFG2 may retain altered or conditional GAP activity despite divergence at the ASAP3 Asp site; a noncatalytic effector role is an alternative.
Type: biochemical assay
Experiment: Image endogenous AGFG2 during stimulated Weibel-Palade body release and test domain/NPF mutants in perturbation-rescue assays with secretion and granule maturation measured separately.
Hypothesis: AGFG2 contributes a late trafficking step through specific protein interactions that need not depend on GAP catalysis.
Type: structure-function mapping
Experiment: Test target-specific germ-cell localization and perturbation phenotypes for AGFG2, including acrosomal vesicle traffic, keratin distribution and nuclear shaping.
Hypothesis: The inherited PAINT capacities are implemented by conserved germ-cell trafficking machinery.
Type: loss of function
What is not known — curated, literature-grounded statements of the open unknowns (the inverse of core functions).
Gap: The catalytic activity and substrate specificity of the AGFG2 ArfGAP-type domain remain unresolved.
OPEN BIOLOGY MF_DARK
What is known: Arg75 is retained and the ASAP3 Asp484 position maps to Thr89, but active GIT proteins lack the homologous Asp. The existing sequence panel and R75Q secretion phenotype cannot establish universal catalytic inactivity.
What would resolve it: Direct hydrolysis and ARF-binding assays, interpreted with lipid/cofactor dependence and full-length protein controls.
Provenance (the field's own admissions):
Gap: The domain and interaction requirements of endothelial secretion, inherited germ-cell functions and viral cofactor activity remain incompletely resolved.
OPEN BIOLOGY
What is known: These functions can coexist. Functional perturbation supports participation without identifying a purified catalytic or adaptor mechanism; human germ-cell capacity remains phylogenetic.
Provenance (the field's own admissions):
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)