coVIN (CAOG_05123) is one of two vinculin-family proteins of the amoeboid filasterean Capsaspora owczarzaki, a close unicellular relative of animals with an extensive integrin adhesome. The 834-residue protein carries the vinculin/alpha-catenin helical bundle architecture: an N-terminal vinculin head region (the talin-binding D1 domain in animal vinculins), central helix bundles, and a C-terminal bundle at the position of the vinculin tail, which in animal vinculins binds actin filaments. It is similar in length to sponge vinculin, which lacks the D2 domain of bilaterian vinculins. An antibody raised against its C-terminal region (residues 335-834) stains the cell body and distinct patches along the actin-dependent filopodia with which adherent Capsaspora cells attach to and move over the substrate. Some of these patches also contain integrin beta2, the major Capsaspora integrin beta subunit, whose ligand-binding domain is needed for adhesion to fibronectin. Vinculin staining in filopodia increases when cells adhere to fibronectin. These observations place coVIN at putative integrin-containing substrate contacts. Its own biochemical activities (binding to talin or actin filaments) and its requirement for adhesion have not been tested; in sponge and vertebrate vinculins, talin binding relieves autoinhibition and allows actin filament binding. The second Capsaspora vinculin paralog (CAOG_03344) is uncharacterized.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0003779 actin binding | IEA GO_REF:0000002 | ACCEPT | Summary: InterPro2GO mapping from the vinculin family signature (IPR017997, residues 1-257 of coVIN). The protein has the full vinculin/alpha-catenin bundle architecture, including a C-terminal bundle at the position of the F-actin-binding vinculin tail. Actin binding has not been measured for the Capsaspora protein; the antibody localisation places it in actin-rich filopodia [DOI:10.1101/2020.02.27.967653]. Reason: A general MF inference from family membership that is consistent with the conserved vinculin mechanism, shown experimentally in a sponge vinculin [PMID:29880641] and in vertebrate vinculins. Accepted at IEA level only. It must not be upgraded to an experimental annotation for Capsaspora: no binding assay was done with coVIN, and in sponge and vertebrate vinculins actin binding depends on talin-mediated release of autoinhibition, which has not been tested here. Supporting Evidence: PMID:29880641 Thus, binding of talin to Op vinculin D1 releases autoinhibition for F-actin binding, similar to the behavior of vertebrate vinculin. DOI:10.1101/2020.02.27.967653 Distinct patches of vinculin are observed in the filopodia. |
| GO:0005737 cytoplasm | IEA GO_REF:0000044 | ACCEPT | Summary: UniProt subcellular-location mapping (ARBA rule). Immunostaining of endogenous coVIN in fixed adherent Capsaspora cells shows staining of the cell body as well as of filopodial patches [DOI:10.1101/2020.02.27.967653]. Reason: Vinculin is a cytoplasmic protein in all organisms studied, and the cell-body staining in Capsaspora agrees. The more informative location is the filopodium, added below as a NEW IDA row. Supporting Evidence: DOI:10.1101/2020.02.27.967653 We observed that both antibodies stain the cell body and multiple distinct patches along filopodia |
| GO:0007155 cell adhesion | IEA GO_REF:0000002 | ACCEPT | Summary: InterPro2GO mapping from the vinculin/alpha-catenin domain. In Capsaspora, adherent cells attach to the substrate through Arp2/3-dependent filopodia, and an antibody against the ligand-binding domain of integrin beta2 inhibits adhesion to fibronectin. coVIN localises in patches along these filopodia, partly together with integrin beta2, and its filopodial staining increases on fibronectin [DOI:10.1101/2020.02.27.967653]. Reason: Vinculin-family proteins do part of the work of adhesion as the actin-anchoring linkers of adhesion complexes, so this general process term is a reasonable family-level inference, and it is consistent with where coVIN is found. It is not an animal-tissue or developmental term. Accepted only at IEA level: no vinculin-specific perturbation exists in Capsaspora (the adhesion perturbations used drugs against actin polymerisation and Arp2/3 and a function-blocking integrin antibody), so neither this term nor a more specific cell-substrate adhesion term is supported experimentally for coVIN. The authors themselves present the patches only as possible anchorage sites. SebΓ©-PedrΓ³s et al. argued that vinculin-family scaffolds predate integrins and had other ancestral roles [PMID:20479219], so a role in Capsaspora adhesion should not be assumed from homology alone. Supporting Evidence: DOI:10.1101/2020.02.27.967653 these results suggest that the integrin and vinculin patches may represent anchorage sites of adhesion DOI:10.1101/2020.02.27.967653 These results suggest that Fibronectin promotes cell adhesion by increasing the number and length of filopodia as well as by promoting recruitment of integrin and vinculin to the filopodia. PMID:20479219 suggesting that an ancient scaffolding machinery was coopted to the integrin adhesion system |
| GO:0051015 actin filament binding | IEA GO_REF:0000002 | ACCEPT | Summary: InterPro2GO mapping from the vinculin/alpha-catenin domain. The C-terminal bundle region of coVIN (Gene3D 655-832, with the PRINTS vinculin fingerprint motifs at 691-780) corresponds to the actin-binding vinculin tail of animal vinculins. Sponge vinculin binds actin filaments once a talin peptide relieves autoinhibition [PMID:29880641]. Reason: The most likely core molecular activity of coVIN, inferred from family membership and domain architecture. Accepted at IEA level. Actin filament binding, and its talin dependence, have been shown for sponge and vertebrate vinculins but not for the Capsaspora protein, so they should not be transferred as experimental. coVIN may lack domain D2, as sponge vinculin does; this affects autoinhibition but not the presence of the actin-binding tail. Supporting Evidence: PMID:29880641 In the presence of Op talin peptide, full-length Op vinculin bound to F-actin filaments PMID:29880641 VIN proteins link actin filaments to membrane proteins at the plasma membrane and are found in all animals and their close outgroups |
| GO:0030175 filopodium | IDA DOI:10.1101/2020.02.27.967653 | NEW | Summary: Immunostaining of endogenous coVIN in fixed adherent Capsaspora cells, with an antibody raised against residues 335-834 of CAOG_05123, shows distinct patches along the phalloidin-stained filopodia. Some patches also contain integrin beta2. Filopodial staining intensity increases on fibronectin-coated compared with BSA-coated coverslips. The antibody was checked by Western blot against the antigen and Capsaspora extracts (major band at the predicted size), by immunoprecipitation with mass spectrometry, and with no-primary-antibody imaging controls [DOI:10.1101/2020.02.27.967653; PMID:32857975]. Reason: No experimental annotation exists for this protein. The localisation is a direct observation of the endogenous protein in Capsaspora, so IDA is the right evidence type. Filopodium is chosen rather than focal adhesion or cell-substrate junction: the patches have not been characterised as adhesion structures beyond partial overlap with integrin beta2, and the authors present them only as possible anchorage sites. Caveats: the full methods are in the bioRxiv preprint (the cached Curr Biol record is abstract only); there is no knockout or antigen-competition control, and cross-reaction with the second vinculin paralog (CAOG_03344) is not explicitly excluded in the text. Only the adherent stage was examined. Supporting Evidence: DOI:10.1101/2020.02.27.967653 Distinct patches of vinculin are observed in the filopodia. DOI:10.1101/2020.02.27.967653 Co-immunostaining showed a partial overlap, some patches contain both, whereas others have just one of the proteins DOI:10.1101/2020.02.27.967653 The vinculin-antigen is a polypeptide corresponding to the C-terminus region of one of the vinculin homologs (CAOG_05123, amino acid region 335-834). PMID:32857975 We show that integrin Ξ²2 and its associated protein vinculin localize as distinct patches in the filopodia. |
Loading supporting contentβ¦
Download this section (compressed HTML)Q: Does coVIN bind Capsaspora talin and, like sponge and vertebrate vinculins, become able to bind actin filaments only when talin relieves head-tail autoinhibition?
Q: Does coVIN lack domain D2, as sponge vinculin does, and does the longer paralog CAOG_03344 (1038 aa) retain it? Are the two paralogs expressed in different life stages or localised differently?
Q: Are the coVIN-positive filopodial patches adhesion sites? Do they also contain talin, and are they stable contacts with the substrate, given that animal filopodia are mostly exploratory?
Q: Where is coVIN in the aggregative stage, when adhesome genes are up-regulated? Is it at cell-cell contacts, as sponge vinculin is?
Experiment: Knock out coVIN (and separately CAOG_03344) by the homologous-recombination method now available for Capsaspora, then measure adhesion to uncoated and fibronectin-coated surfaces, filopodium number and length, and integrin beta2 patch formation.
Hypothesis: coVIN is required for filopodial substrate adhesion in Capsaspora.
Type: genetic knockout / adhesion assay
Experiment: Express recombinant full-length coVIN and its head domain. Measure binding to vinculin-binding-site peptides of Capsaspora talin by ITC, and F-actin co-sedimentation with and without talin peptide.
Hypothesis: coVIN is a talin-activated actin filament binding protein.
Type: in vitro binding
Experiment: Stain coVIN knockout cells with the anti-CAOG_05123 antibody, and image a fluorescently tagged coVIN in live cells to follow patch dynamics during filopodial movement.
Hypothesis: The anti-vinculin staining reflects coVIN specifically.
Type: localization / imaging
Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)