ESPL1 encodes separase (separin), a large caspase-like cysteine endopeptidase of clan CD whose C-terminal peptidase C50 domain (catalytic Cys2029) cleaves the kleisin subunit of cohesin after arginine in ExxR motifs. At the metaphase-to-anaphase transition it cuts RAD21/SCC1 in the residual centromeric cohesin, opening the cohesin ring and allowing sister chromatids to be pulled to opposite spindle poles; in meiosis it cleaves the kleisin REC8. Efficient cleavage depends on a docking (LPE) motif in RAD21 and is enhanced by PLK1 phosphorylation of the substrate. Through most of the cell cycle separase is held inactive in the separase-securin complex, in which securin (PTTG1) acts as a folding chaperone and pseudosubstrate inhibitor, or by mutually exclusive binding of CDK1-cyclin B1-CKS1 after phosphorylation of Ser1126; APC/C-Cdc20-mediated destruction of securin and cyclin B1 unleashes the enzyme, which then autocleaves at Arg1486/1506/1535 with its fragments remaining associated. Separase is predominantly cytoplasmic in interphase, concentrates at centrosomes during prometaphase and metaphase, and at the end of mitosis cleaves centrosomal cohesin and pericentrin-B/kendrin to disengage centrioles and license the next round of centrosome duplication. Loss of separase leaves sister chromatids unseparated and produces lagging chromosomes and polyploidy, while premature activation causes precocious sister separation and centriole disengagement.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0000070 mitotic sister chromatid segregation | IMP PMID:12672959 Processing, localization, and requirement of human separase ... | ACCEPT | Summary: Antisense depletion of endogenous separase in HeLa cells caused a >4-fold increase in anaphases with lagging chromosomes, directly showing that separase is required for faithful mitotic sister chromatid segregation. Reason: Direct human loss-of-function evidence, consistent with RNAi (PMID:12194817) and non-cleavable SCC1 (PMID:11509732) phenotypes. Separase performs the cohesin-cleavage step itself, so it is a genuine participant in this process; the more specific step is captured by GO:0051306. Supporting Evidence: PMID:12672959 Antisense oligonucleotide-mediated depletion of separase results in a >4-fold increase in the mitotic cells with lagging chromosomes PMID:12194817 Inhibition of separase expression in human cells by RNA interference |
| GO:0000280 nuclear division | IEA GO_REF:0000117 | MODIFY | Summary: ARBA infers nuclear division; separase contributes the specific cohesin-cleavage step that detaches sister chromatids at anaphase. Reason: The prediction is not wrong but is far too general: nuclear division covers mitosis and meiosis as a whole, whereas separase's contribution is one defined step, mitotic sister chromatid separation, which is already annotated from Reactome and supported by human depletion data. Proposed replacements: mitotic sister chromatid separation Supporting Evidence: PMID:12672959 separase activation at anaphase in mammalian cells is the primary trigger for sister chromatid separation Reactome:R-HSA-2467813 RAD21 cleavage promotes dissociation of cohesin complexes from sister centromeres, leading to separation of sister chromatids |
| GO:0000281 mitotic cytokinesis | NAS PMID:11509732 Cohesin cleavage by separase required for anaphase and cytok... | MARK AS OVER ANNOTATED | Summary: Cells expressing non-cleavable SCC1 fail to complete cytokinesis, but this is a downstream consequence of unseparated chromatids obstructing the cleavage plane rather than a cytokinetic activity of separase. Reason: The cited abstract states that cohesin cleavage is required for completion of cytokinesis, yet separase does none of the work of cytokinesis; the defect follows from failed sister separation. Human separase is lost from centrosomes at anaphase onset and, unlike budding-yeast Esp1, has no documented mitotic-exit role. The term overstates the evidence and should not be treated as a core process. Supporting Evidence: PMID:11509732 cohesin cleavage by separase is essential for sister chromatid separation and for the completion of cytokinesis PMID:12672959 separase activation at anaphase in mammalian cells is the primary trigger for sister chromatid separation |
| GO:0003824 catalytic activity | NAS PMID:12672959 Processing, localization, and requirement of human separase ... | MODIFY | Summary: Separase is a catalytically active enzyme; the paper shows that the C2029A active-site mutant loses autocatalytic processing. Reason: Catalytic activity is the root of the molecular-function branch and conveys nothing about what separase does. The evidence supports the specific term cysteine-type endopeptidase activity, which is already annotated by IBA and InterPro2GO. Proposed replacements: cysteine-type endopeptidase activity Supporting Evidence: PMID:12672959 The cysteine at the position 2029 of separase is essential for proteolytic activity PMID:12672959 no cleavage products were detected even on prolonged induction up to 72 h |
| GO:0004197 cysteine-type endopeptidase activity | IBA GO_REF:0000033 | ACCEPT | Summary: The PAINT node PTN000300670 (seeded by budding-yeast Esp1 and separase orthologs) assigns cysteine-type endopeptidase activity to the separase family; human ESPL1 is a clan-CD caspase-like endopeptidase with catalytic Cys2029 that cleaves RAD21 after Arg172 and Arg450. Reason: Core molecular function, directly demonstrated for the human enzyme by in vitro cleavage of SCC1/RAD21 and loss of activity in the C2029A mutant; the phylogenetic placement is sound. Supporting Evidence: PMID:15737063 separase cleaves Scc1 at the same sites that are cleaved at anaphase onset in vivo, Arg172 and Arg450 PMID:12672959 The cysteine at the position 2029 of separase is essential for proteolytic activity file:human/ESPL1/ESPL1-uniprot.txt Caspase-like protease, which plays a central role in the |
| GO:0004197 cysteine-type endopeptidase activity | IEA GO_REF:0000002 | ACCEPT | Summary: InterPro2GO mapping from the Peptidase_C50 and separin core domains; matches the experimentally verified endopeptidase activity of human separase. Reason: Correct and specific domain-based inference, corroborated by direct human biochemical evidence. Supporting Evidence: PMID:15737063 separase cleaves Scc1 at the same sites that are cleaved at anaphase onset in vivo, Arg172 and Arg450 Reactome:R-HSA-2467809 ESPL1 (separin i.e. separase) cleaves RAD21 (SCC1) subunit of centromeric cohesin at two sites |
| GO:0005515 protein binding | IPI PMID:12194817 Regulation of human separase by securin binding and autoclea... | REMOVE | Summary: Inhibitor-binding experiments show that securin/PTTG1 bound to separase blocks access of substrate analogues to the active site. Reason: Generic protein binding is uninformative. The partner is securin/PTTG1, the pseudosubstrate inhibitor and chaperone of separase; GO has no informative molecular-function term for the enzyme side of an enzyme-inhibitor pair, and the biology is already captured by the separase-securin complex annotation (GO:1990520, IPI, retained) and by PTTG1's own cysteine-type endopeptidase inhibitor activity. Removal does not dispute the interaction, which is one of the best-established in mitosis. Supporting Evidence: PMID:12194817 the access of substrate analogs to the active site of separase. Reactome:R-HSA-2467798 Up to anaphase onset, ESPL1 (separase i.e. separin) forms a complex with PTTG1 |
| GO:0005515 protein binding | IPI PMID:12672959 Processing, localization, and requirement of human separase ... | MODIFY | Summary: IntAct-derived interaction with RAD21/SCC1; the cached paper treats RAD21 as the established separase substrate rather than reporting a binding assay in its main text. Reason: Generic protein binding is uninformative; RAD21/SCC1 is the substrate of separase and the informative molecular function is the endopeptidase activity that acts on it (GO:0004197, with RAD21 as input). Substrate binding is not annotated as a separate function. Proposed replacements: cysteine-type endopeptidase activity Supporting Evidence: PMID:15737063 separase cleaves Scc1 at the same sites that are cleaved at anaphase onset in vivo, Arg172 and Arg450 Reactome:R-HSA-2467809 ESPL1 (separin i.e. separase) cleaves RAD21 (SCC1) subunit of centromeric cohesin at two sites |
| GO:0005515 protein binding | IPI PMID:15737063 Dissociation of cohesin from chromosome arms and loss of arm... | MODIFY | Summary: Activated human separase cleaves recombinant SCC1/RAD21 at Arg172 and Arg450 in vitro; PLK1 phosphorylation of S454 enhances cleavage at the second site. Reason: Generic protein binding is uninformative; RAD21/SCC1 is the substrate of separase and the informative molecular function is the endopeptidase activity that acts on it (GO:0004197, with RAD21 as input). Substrate binding is not annotated as a separate function. Proposed replacements: cysteine-type endopeptidase activity Supporting Evidence: PMID:15737063 separase cleaves Scc1 at the same sites that are cleaved at anaphase onset in vivo, Arg172 and Arg450 |
| GO:0005515 protein binding | IPI PMID:20360068 Systematic analysis of human protein complexes identifies ch... | REMOVE | Summary: MitoCheck tandem-affinity purification recovered the separase-securin pair among ~100 human mitotic complexes. Reason: Generic protein binding is uninformative. The partner is securin/PTTG1, the pseudosubstrate inhibitor and chaperone of separase; GO has no informative molecular-function term for the enzyme side of an enzyme-inhibitor pair, and the biology is already captured by the separase-securin complex annotation (GO:1990520, IPI, retained) and by PTTG1's own cysteine-type endopeptidase inhibitor activity. Removal does not dispute the interaction, which is one of the best-established in mitosis. Supporting Evidence: PMID:20360068 tandem-affinity purification-mass spectrometry PMID:34290405 We determined the structures of a wild-type separase-securin complex |
| GO:0005515 protein binding | IPI PMID:24781523 PP2A delays APC/C-dependent degradation of separase-associat... | REMOVE | Summary: Separase-associated securin is stabilised by separase-bound PP2A; the study directly analyses the separase-securin pool before anaphase. Reason: Generic protein binding is uninformative. The partner is securin/PTTG1, the pseudosubstrate inhibitor and chaperone of separase; GO has no informative molecular-function term for the enzyme side of an enzyme-inhibitor pair, and the biology is already captured by the separase-securin complex annotation (GO:1990520, IPI, retained) and by PTTG1's own cysteine-type endopeptidase inhibitor activity. Removal does not dispute the interaction, which is one of the best-established in mitosis. Supporting Evidence: PMID:24781523 Prior to anaphase, most separase is kept inactive by association with securin |
| GO:0005515 protein binding | IPI PMID:26496610 A human interactome in three quantitative dimensions organiz... | REMOVE | Summary: Quantitative AP-MS interactome recovered the ESPL1-PTTG1 pair; the accessible abstract describes the platform, not this pair. Reason: Generic protein binding is uninformative. The partner is securin/PTTG1, the pseudosubstrate inhibitor and chaperone of separase; GO has no informative molecular-function term for the enzyme side of an enzyme-inhibitor pair, and the biology is already captured by the separase-securin complex annotation (GO:1990520, IPI, retained) and by PTTG1's own cysteine-type endopeptidase inhibitor activity. Removal does not dispute the interaction, which is one of the best-established in mitosis. Supporting Evidence: PMID:26496610 Using quantitative proteomics, we detect specific PMID:34290405 We determined the structures of a wild-type separase-securin complex |
| GO:0005515 protein binding | IPI PMID:33961781 Dual proteome-scale networks reveal cell-specific remodeling... | REMOVE | Summary: BioPlex proteome-scale AP-MS recovered the ESPL1-PTTG1 pair; the article text does not discuss separase specifically. Reason: Generic protein binding is uninformative. The partner is securin/PTTG1, the pseudosubstrate inhibitor and chaperone of separase; GO has no informative molecular-function term for the enzyme side of an enzyme-inhibitor pair, and the biology is already captured by the separase-securin complex annotation (GO:1990520, IPI, retained) and by PTTG1's own cysteine-type endopeptidase inhibitor activity. Removal does not dispute the interaction, which is one of the best-established in mitosis. Supporting Evidence: PMID:33961781 mass spectrometry, we have created two proteome-scale, cell-line-specific PMID:34290405 We determined the structures of a wild-type separase-securin complex |
| GO:0005515 protein binding | IPI PMID:34290405 Structural basis of human separase regulation by securin and... | MODIFY | Summary: Cryo-EM and cleavage assays define how SCC1/RAD21 engages separase: the cleavage-site motif, an NxLxΦE motif and the LPE docking motif bind adjacent sites, and deleting the linkers between them abolishes SCC1 cleavage. Reason: Generic protein binding is uninformative; RAD21/SCC1 is the substrate of separase and the informative molecular function is the endopeptidase activity that acts on it (GO:0004197, with RAD21 as input). Substrate binding is not annotated as a separate function. Proposed replacements: cysteine-type endopeptidase activity Supporting Evidence: PMID:34290405 Deletion of these linkers, or replacement with a random sequence, abolished Scc1 cleavage in vitro |
| GO:0005515 protein binding | IPI PMID:34290405 Structural basis of human separase regulation by securin and... | REMOVE | Summary: Cryo-EM structure of the human separase-securin complex at ~2.9 Γ
resolves securin as an extended pseudosubstrate along separase. Reason: Generic protein binding is uninformative. The partner is securin/PTTG1, the pseudosubstrate inhibitor and chaperone of separase; GO has no informative molecular-function term for the enzyme side of an enzyme-inhibitor pair, and the biology is already captured by the separase-securin complex annotation (GO:1990520, IPI, retained) and by PTTG1's own cysteine-type endopeptidase inhibitor activity. Removal does not dispute the interaction, which is one of the best-established in mitosis. Supporting Evidence: PMID:34290405 We determined the structures of a wild-type separase-securin complex PMID:34290405 Consistent with previous studies, we find that securin acts as an extended pseudosubstrate that blocks substrate binding at the catalytic site and at nearby docking sites. |
| GO:0005634 nucleus | IBA GO_REF:0000033 | ACCEPT | Summary: PAINT infers nuclear activity for the separase family (human ESPL1 is among the descendant sources); separase cleaves chromosome-bound cohesin, and UniProt records nuclear localisation. Reason: Separase's substrate is chromosomal cohesin, so the site of action is nuclear chromatin (exposed after nuclear envelope breakdown in open mitosis). Interphase separase is largely exported from the nucleus, but the annotation describes where the enzyme acts, and the phylogenetic placement is consistent with all orthologs. Supporting Evidence: file:human/ESPL1/ESPL1-uniprot.txt SUBCELLULAR LOCATION: Cytoplasm. Nucleus. PMID:24781523 Because separase is excluded from the nucleus |
| GO:0005634 nucleus | IEA GO_REF:0000120 | ACCEPT | Summary: Automated nuclear localisation from UniProt subcellular-location keywords and InterPro; consistent with the reviewed UniProt record. Reason: Correct compartment, independently represented by IBA and by the HPA nucleoplasm annotation. Supporting Evidence: file:human/ESPL1/ESPL1-uniprot.txt SUBCELLULAR LOCATION: Cytoplasm. Nucleus. |
| GO:0005654 nucleoplasm | IDA GO_REF:0000052 | ACCEPT | Summary: Human Protein Atlas immunofluorescence places ESPL1 in the nucleoplasm. Reason: Consistent with the UniProt nuclear localisation and with a nuclear pool that accumulates on DNA damage; not in conflict with the predominantly cytoplasmic interphase distribution. Supporting Evidence: file:human/ESPL1/ESPL1-uniprot.txt SUBCELLULAR LOCATION: Cytoplasm. Nucleus. |
| GO:0005737 cytoplasm | IBA GO_REF:0000033 | ACCEPT | Summary: PAINT infers cytoplasmic activity for the separase family; human separase is predominantly cytoplasmic in interphase owing to a nuclear-export sequence and acts on centrosomes. Reason: Directly supported for human separase (cytoplasmic interphase pool, centrosomal activity) and by the UniProt record; the phylogenetic inference is sound. Supporting Evidence: file:human/ESPL1/ESPL1-uniprot.txt SUBCELLULAR LOCATION: Cytoplasm. Nucleus. file:human/ESPL1/ESPL1-deep-research-falcon.md predominantly cytoplasmic in HeLa cells, partly because of a nuclear-export sequence |
| GO:0005737 cytoplasm | IEA GO_REF:0000044 | ACCEPT | Summary: UniProt subcellular-location keyword mapping to cytoplasm; agrees with direct human localisation data. Reason: Correct compartment, consistent with IBA and with the cytoplasmic interphase pool of separase. Supporting Evidence: file:human/ESPL1/ESPL1-uniprot.txt SUBCELLULAR LOCATION: Cytoplasm. Nucleus. |
| GO:0005813 centrosome | IBA GO_REF:0000033 | ACCEPT | Summary: PAINT node PTN000980515 (sources: human ESPL1 itself and C. elegans SEP-1) infers centrosomal activity; human separase co-localises with gamma-tubulin at centrosomes in prometaphase/metaphase and cleaves pericentrin-B/kendrin to promote centriole disengagement. Reason: Human ESPL1 appearing in its own WITH/FROM reflects the IDA on which the node was placed. The location is functionally meaningful: separase acts at centrosomes to license centriole duplication. Supporting Evidence: PMID:12672959 revealed colocalization of separase and Ξ³-tubulin in centrosomes during the early stages of mitosis PMID:24781523 Separase also cleaves centrosomal cohesin and pericentrin-B (kendrin), which promotes centriole disengagement and licensing of subsequent centrosome duplication |
| GO:0005813 centrosome | IDA PMID:12672959 Processing, localization, and requirement of human separase ... | ACCEPT | Summary: Polyclonal and monoclonal antibodies show separase foci co-localising with gamma-tubulin at centrosomes in prometaphase and metaphase, with the signal lost at anaphase onset. Reason: Direct human immunofluorescence with antibody-specificity controls (antigen competition, antisense depletion); a functionally relevant location given separase-dependent centriole disengagement. Supporting Evidence: PMID:12672959 revealed colocalization of separase and Ξ³-tubulin in centrosomes during the early stages of mitosis PMID:24781523 Separase also cleaves centrosomal cohesin and pericentrin-B (kendrin), which promotes centriole disengagement and licensing of subsequent centrosome duplication |
| GO:0005829 cytosol | TAS Reactome:R-HSA-2467775 | ACCEPT | Summary: Reactome places this separase reaction (autocleavage, securin sequestration or centromeric cohesin cleavage) in the cytosol. Reason: Compatible with the predominantly cytoplasmic interphase pool of separase and with the cytoplasmic environment of the mitotic cell after nuclear envelope breakdown; consistent with the UniProt record. Supporting Evidence: Reactome:R-HSA-2467775 is rapidly autocatalytically cleaved after arginine residues at positions 1506 and 1535 file:human/ESPL1/ESPL1-uniprot.txt SUBCELLULAR LOCATION: Cytoplasm. Nucleus. |
| GO:0005829 cytosol | TAS Reactome:R-HSA-2467798 | ACCEPT | Summary: Reactome places this separase reaction (autocleavage, securin sequestration or centromeric cohesin cleavage) in the cytosol. Reason: Compatible with the predominantly cytoplasmic interphase pool of separase and with the cytoplasmic environment of the mitotic cell after nuclear envelope breakdown; consistent with the UniProt record. Supporting Evidence: Reactome:R-HSA-2467798 Up to anaphase onset, ESPL1 (separase i.e. separin) forms a complex with PTTG1 file:human/ESPL1/ESPL1-uniprot.txt SUBCELLULAR LOCATION: Cytoplasm. Nucleus. |
| GO:0005829 cytosol | TAS Reactome:R-HSA-2467809 | ACCEPT | Summary: Reactome places this separase reaction (autocleavage, securin sequestration or centromeric cohesin cleavage) in the cytosol. Reason: Compatible with the predominantly cytoplasmic interphase pool of separase and with the cytoplasmic environment of the mitotic cell after nuclear envelope breakdown; consistent with the UniProt record. Supporting Evidence: Reactome:R-HSA-2467809 ESPL1 (separin i.e. separase) cleaves RAD21 (SCC1) subunit of centromeric cohesin at two sites file:human/ESPL1/ESPL1-uniprot.txt SUBCELLULAR LOCATION: Cytoplasm. Nucleus. |
| GO:0006508 proteolysis | IEA GO_REF:0000002 | ACCEPT | Summary: InterPro2GO infers proteolysis from the peptidase C50 domain; separase proteolyses RAD21, itself and pericentrin. Reason: Correct though general; the specific processes (mitotic sister chromatid separation, protein autoprocessing) are captured elsewhere in this review. Supporting Evidence: Reactome:R-HSA-2467809 ESPL1 (separin i.e. separase) cleaves RAD21 (SCC1) subunit of centromeric cohesin at two sites |
| GO:0006915 apoptotic process | TAS PMID:11875078 Caspase proteolysis of the cohesin component RAD21 promotes ... | REMOVE | Summary: The cited paper shows that RAD21 is cleaved at Asp279 by caspases-3 and -7 during apoptosis; separase is mentioned only as the enzyme that cleaves RAD21 at anaphase. Reason: No evidence links separase to the apoptotic process: the apoptotic RAD21 cleavage is performed by caspases, and the abstract names separase purely as background. This is a traceable-author-statement annotation resting on a substrate shared with caspases, not on any separase activity in cell death. Supporting Evidence: PMID:11875078 preferentially cleaved at Asp(279) by caspases-3 and -7 in vitro PMID:11875078 cleaved by a caspase-like Esp1/separase at the onset of anaphase to trigger |
| GO:0008233 peptidase activity | IEA GO_REF:0000107 | ACCEPT | Summary: Ensembl Compara transfer of peptidase activity from mouse Espl1; human separase is a verified endopeptidase. Reason: Correct but general; the specific cysteine-type endopeptidase activity is already annotated by IBA and IEA. Supporting Evidence: PMID:15737063 separase cleaves Scc1 at the same sites that are cleaved at anaphase onset in vivo, Arg172 and Arg450 |
| GO:0008234 cysteine-type peptidase activity | IEA GO_REF:0000107 | ACCEPT | Summary: Ensembl Compara transfer of cysteine-type peptidase activity from mouse Espl1. Reason: Correct; slightly less specific than the endopeptidase term already present. Supporting Evidence: PMID:12672959 The cysteine at the position 2029 of separase is essential for proteolytic activity |
| GO:0008234 cysteine-type peptidase activity | NAS PMID:12672959 Processing, localization, and requirement of human separase ... | ACCEPT | Summary: The paper identifies separase as a CD-clan cysteine protease whose Cys2029 is essential for activity. Reason: Correct; the endopeptidase child term is annotated separately and represents the precise activity. Supporting Evidence: PMID:12672959 The cysteine at the position 2029 of separase is essential for proteolytic activity |
| GO:0008234 cysteine-type peptidase activity | TAS Reactome:R-HSA-2467775 | ACCEPT | Summary: Reactome autocleavage reaction assigns cysteine-type peptidase activity to ESPL1. Reason: Correct; autocleavage after Arg1506/Arg1535 is catalysed by separase's own cysteine protease domain. Supporting Evidence: Reactome:R-HSA-2467775 is rapidly autocatalytically cleaved after arginine residues at positions 1506 and 1535 |
| GO:0034088 maintenance of mitotic sister chromatid cohesion | NAS PMID:34290405 Structural basis of human separase regulation by securin and... | REMOVE | Summary: The structural paper shows how securin and CDK1-cyclin B1 keep separase inactive before anaphase; maintaining cohesion is the work of cohesin, shugoshin-PP2A and the separase inhibitors, not of separase. Reason: Separase is the enzyme that dissolves cohesion; it is the object of inhibition, and being inhibited is not participating in cohesion maintenance. Even the autoinhibitory loops structured by CDK1-cyclin B1 are regulation of separase activity rather than a contribution to holding sisters together. The term is correctly carried by PTTG1 and is the mirror image of separase's role. Supporting Evidence: PMID:34290405 Consistent with previous studies, we find that securin acts as an extended pseudosubstrate that blocks substrate binding at the catalytic site and at nearby docking sites. PMID:34290405 Unexpectedly, Cdk1-cyclin B1-Cks1 suppresses separase activity by activating autoinhibitory sequences in loops of separase itself |
| GO:0034090 maintenance of meiotic sister chromatid cohesion | NAS PMID:34290405 Structural basis of human separase regulation by securin and... | REMOVE | Summary: Meiotic counterpart of the preceding row; the cited cryo-EM study contains no meiotic experiments. Reason: Same logic as for GO:0034088: separase cleaves REC8 to release meiotic cohesion and does not maintain it; in addition the cited paper provides no meiotic data. Supporting Evidence: PMID:34290405 Consistent with previous studies, we find that securin acts as an extended pseudosubstrate that blocks substrate binding at the catalytic site and at nearby docking sites. |
| GO:0040001 establishment of mitotic spindle localization | NAS PMID:12672959 Processing, localization, and requirement of human separase ... | REMOVE | Summary: The paper's discussion notes that budding-yeast separase is needed for spindle position and C. elegans separase for paternal centrosome migration; no human spindle-positioning data are presented. Reason: This author statement is an extrapolation from yeast and worm; human separase was not tested for spindle localisation and was not even detected on spindle microtubules. Without human evidence the process should not be annotated. Supporting Evidence: PMID:12672959 separase is also required in budding yeast for establishing normal spindle position PMID:12672959 We have not observed separase staining along spindle microtubules |
| GO:0045842 positive regulation of mitotic metaphase/anaphase transition | NAS PMID:12672959 Processing, localization, and requirement of human separase ... | MODIFY | Summary: Separase activation at anaphase is the primary trigger for sister chromatid separation. Reason: Separase does not regulate the metaphase/anaphase transition; it executes its defining step downstream of APC/C-Cdc20-mediated securin destruction. Annotating an effector to a regulation term misplaces it upstream of the process it carries out; the correct term is the executed process, mitotic sister chromatid separation. Proposed replacements: mitotic sister chromatid separation Supporting Evidence: PMID:12672959 separase activation at anaphase in mammalian cells is the primary trigger for sister chromatid separation PMID:24781523 The universal triggering event of eukaryotic chromosome segregation is cleavage of centromeric cohesin by separase |
| GO:0045875 negative regulation of sister chromatid cohesion | NAS PMID:12194817 Regulation of human separase by securin binding and autoclea... | MODIFY | Summary: Separase cleaves cohesin and thereby dissolves sister chromatid cohesion; RNAi depletion leaves sisters unseparated. Reason: Cleaving cohesin is the act of separating sisters, not a regulation of cohesion. The process separase performs is mitotic sister chromatid separation; the regulation term should be reserved for factors that modulate cohesion (e.g. WAPL, shugoshin, securin). Proposed replacements: mitotic sister chromatid separation Supporting Evidence: PMID:12194817 Inhibition of separase expression in human cells by RNA interference Reactome:R-HSA-2467813 RAD21 cleavage promotes dissociation of cohesin complexes from sister centromeres, leading to separation of sister chromatids |
| GO:0051306 mitotic sister chromatid separation | TAS Reactome:R-HSA-2467813 | ACCEPT | Summary: Reactome: after APC/C destroys securin, ESPL1 cleaves centromeric RAD21, dissociating cohesin from sister centromeres and separating sister chromatids. Reason: This is the core biological process of separase and is directly executed by its endopeptidase activity; supported by human depletion, non-cleavable SCC1 and in vitro cleavage data. Supporting Evidence: Reactome:R-HSA-2467813 RAD21 cleavage promotes dissociation of cohesin complexes from sister centromeres, leading to separation of sister chromatids PMID:11509732 cohesin cleavage by separase is essential for sister chromatid separation and for the completion of cytokinesis PMID:12672959 Antisense oligonucleotide-mediated depletion of separase results in a >4-fold increase in the mitotic cells with lagging chromosomes |
| GO:0051307 meiotic chromosome separation | IBA GO_REF:0000033 | ACCEPT | Summary: PAINT node PTN000300670 (sources: A. thaliana AESP, S. pombe cut1, C. elegans sep-1) infers meiotic chromosome separation; mammalian separase cleaves the meiotic kleisin REC8, with direct evidence from mouse oocytes. Reason: The node placement is sound: separase-dependent kleisin cleavage drives homologue and sister separation in every lineage examined, and human ESPL1 is inside the clade with no evidence of loss. Human-specific meiotic data are lacking, so this is retained on orthology and mammalian evidence rather than as a primary mitotic function. Supporting Evidence: file:human/ESPL1/ESPL1-deep-research-falcon.md In mammalian meiosis, separase cleaves the meiosis-specific kleisin REC8. |
| GO:0072686 mitotic spindle | IBA GO_REF:0000033 | KEEP AS NON CORE | Summary: PAINT infers mitotic-spindle activity from fission-yeast cut1 and C. elegans sep-1; human separase localises to spindle poles/centrosomes until anaphase but was not detected along spindle microtubules. Reason: Spindle poles are part of the mitotic spindle, so the inference is not contradicted, but the human data show pole/centrosome rather than spindle-fibre localisation and no spindle-elongation role like that of yeast separase. Retained as a plausible but non-core location. Supporting Evidence: PMID:12672959 revealed colocalization of separase and Ξ³-tubulin in centrosomes during the early stages of mitosis PMID:12672959 We have not observed separase staining along spindle microtubules |
| GO:0098813 nuclear chromosome segregation | IEA GO_REF:0000117 | ACCEPT | Summary: ARBA infers nuclear chromosome segregation; separase triggers sister chromatid separation. Reason: Correct though general; the specific mitotic sister chromatid separation term is present and reviewed separately. Supporting Evidence: PMID:24781523 The universal triggering event of eukaryotic chromosome segregation is cleavage of centromeric cohesin by separase PMID:12672959 Antisense oligonucleotide-mediated depletion of separase results in a >4-fold increase in the mitotic cells with lagging chromosomes |
| GO:1990520 separase-securin complex | IPI PMID:34290405 Structural basis of human separase regulation by securin and... | ACCEPT | Summary: Cryo-EM at ~2.9 Γ
resolves the human separase-securin complex, in which securin lies along separase and blocks the catalytic and docking sites; the complex persists until APC/C-mediated securin destruction. Reason: Direct structural evidence for the complex that defines separase regulation before anaphase; a core component annotation. Supporting Evidence: PMID:34290405 We determined the structures of a wild-type separase-securin complex Reactome:R-HSA-2467798 Up to anaphase onset, ESPL1 (separase i.e. separin) forms a complex with PTTG1 PMID:24781523 Prior to anaphase, most separase is kept inactive by association with securin |
| GO:0016540 protein autoprocessing | IDA PMID:12672959 Processing, localization, and requirement of human separase ... | NEW | Summary: Human separase cleaves itself at Arg1486, Arg1506 and Arg1535 on release from securin; the catalytically inactive C2029A mutant is not processed, and the N- and C-terminal fragments remain associated. Reason: Autoprocessing is performed by separase's own endopeptidase activity, so participation is unambiguous, and the process is well documented in human cells by three independent groups (PMID:12672959, PMID:12194817, Reactome R-HSA-2467775). Comparator check: GO:0016540 carries IDA/IMP annotations on self-processing proteases such as CASP1, CASP4, CASP6, CTSL, PCSK9 and TMPRSS2, so annotating an autocleaving protease here follows convention. The term's parent, protein processing, names the entity that cleaves, which is separase itself. Supporting Evidence: PMID:12672959 no cleavage products were detected even on prolonged induction up to 72 h PMID:12194817 allowing rapid autocatalytic cleavage of separase at one of three sites Reactome:R-HSA-2467775 is rapidly autocatalytically cleaved after arginine residues at positions 1506 and 1535 |
| GO:0007098 centrosome cycle | TAS PMID:24781523 PP2A delays APC/C-dependent degradation of separase-associat... | NEW | Summary: Separase cleaves centrosomal cohesin and pericentrin-B/kendrin in late mitosis, promoting mother-daughter centriole disengagement that licenses the next round of centrosome duplication; in the cited study, replacing securin with a phosphomimetic variant that lets separase activate prematurely causes premature centriole disengagement in G2-arrested cells. Reason: Separase performs the proteolytic step (pericentrin-B/kendrin and centrosomal cohesin cleavage) that disengages centrioles, so it participates directly rather than merely being required. GO has no term for centriole disengagement (QuickGO search returns none); centrosome cycle is the nearest process and is carried by other executors of centrosome-cycle steps such as PLK1 (IMP), NEK2 and CDK1 in human GOA, so the comparator check supports the convention. Evidence in the cached paper is an author statement of the established role plus a securin-manipulation experiment, hence TAS; the primary demonstrations (Tsou & Stearns 2006; Schockel et al. 2011; Matsuo et al. 2012) are cited there but not cached here. Supporting Evidence: PMID:24781523 Separase also cleaves centrosomal cohesin and pericentrin-B (kendrin), which promotes centriole disengagement and licensing of subsequent centrosome duplication PMID:24781523 This property of the phosphorylation-mimicking variant suffices to cause premature centriole disengagement and precocious sister separation, respectively, in G2- and prometaphase-arrested cells. file:human/ESPL1/ESPL1-deep-research-falcon.md Separase promotes motherβdaughter centriole disengagement, an event that licenses centriole duplication in the following cell cycle. |
Loading supporting contentβ¦
Download this section (compressed HTML)Q: Which centrosomal substrate (pericentrin-B/kendrin, centrosomal RAD21, or another) is rate-limiting for separase-dependent centriole disengagement in human cells, and is a GO process term for centriole disengagement warranted?
Q: Does human separase cleave REC8 in human oocytes/spermatocytes as it does in mouse, and how is centromeric REC8 protected from it at meiosis I in human germ cells?
Q: What is the physiological purpose of separase autocleavage in human cells, given that non-cleavable separase retains cohesin-cleaving activity but perturbs the timing of activation and PP2A/cyclin B1 binding?
Experiment: N-terminomics (TAILS) on synchronized anaphase cells expressing protease-dead (C2029A) versus wild-type separase from the endogenous locus, to define the complete in vivo separase cleavage repertoire (RAD21, pericentrin-B, self, others) in human cells.
Experiment: Live-cell imaging of endogenously tagged separase together with securin and RAD21 after acute degron-mediated depletion of securin or cyclin B1, to quantify the kinetics of separase release, autocleavage and cohesin cleavage relative to anaphase onset.
Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)