CYFIP2 (PIR121) is a cytoplasmic component of the WAVE regulatory complex that couples Rac signaling to actin-dependent cellular organization. It associates with Rac1 and ABI/WAVE proteins, balances WAVE restraint and activation, and contributes to cell shape, neuronal projections and synaptic organization. CYFIP2 also forms RNA-sensitive complexes with FMRP. Soluble and particulate cytoplasmic pools coexist with synaptic and membrane-associated pools, and nuclear accumulation increases when nuclear export is inhibited. Human cells show context-dependent effects on fibronectin adhesion and apoptosis. Disease-associated variants disturb this regulatory balance and can also destabilize CYFIP2. The two splice products contain 1278 and 1253 residues; the shorter product lacks an internal 25-residue segment.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0000902 cell morphogenesis | IBA GO_REF:0000033 | ACCEPT | Reason: The actual PTHR12195 tree places human CYFIP2 below the cell-morphogenesis IBD at PTN000257565. Its broad descendant set is consistent with a conserved WAVE-complex contribution to actin-dependent cell shape. The literal donor list is not treated as a vote or as pairwise similarity evidence. Human cell morphology and the mouse neurite assays support this biological scope; this does not assign autonomous actin nucleation to CYFIP2. Propagation Review Root cause: NO FAILURE CORE |
| GO:0005515 protein binding | IPI PMID:11438699 A highly conserved protein family interacting with the fragi... | KEEP AS NON CORE | Reason: The 2001 study tests a human CYFIP2 fragment (residues 1–890) against GST-FMRP in vitro and also uses yeast two-hybrid assays. These support FMRP association, without establishing a CYFIP2 RNA-binding or translational-repressor activity. The deposited human clone AF160973.1 exactly matches current Q96F07-2; this is not a claim that the imported source asserted an isoform. Retain supported generic binding under the project instruction because no more informative partner-class activity was established. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:15048733 CYFIP2 is highly abundant in CD4+ cells from multiple sclero... | MODIFY | Reason: The original human CD4-cell experiment detects CYFIP2 with Rac1 by co-immunoprecipitation, including the reported reciprocal experiment. Rac1 is a small GTPase, so small GTPase binding is more informative than generic protein binding. This is a cellular association assay, not a purified binary affinity measurement, and does not make CYFIP2 a GTPase. Proposed replacements: small GTPase binding Supporting Evidence: PMID:15048733 a physical partner with CYFIP2 |
| GO:0005515 protein binding | IPI PMID:17101133 NESH (Abi-3) is present in the Abi/WAVE complex but does not... | KEEP AS NON CORE | Reason: The exact MINT record IM-10986-8 identifies human ABI3 (Q9P2A4) and CYFIP2 (Q96F07), anti-bait co-immunoprecipitation, HUVEC and Figure 4D; both recorded sequence ROGIDs match the official proteins. Together with the complete abstract and expert IPI curation this supports association. The full 2006 original remains inaccessible, so antibody/paralog resolution and a direct binary interface are not independently established. Retain the supported association without inventing an adapter or enzyme activity. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:24658146 Subcellular fractionation and localization studies reveal a ... | KEEP AS NON CORE | Reason: The original human HeLa experiment detects endogenous CYFIP2 in FMRP immunoprecipitates. RNase strongly disrupts this association, and the direct purified-binding experiments in this paper instead concern nucleolin. Thus the source supports an RNA-sensitive association, not direct CYFIP2 RNA binding or autonomous translational repression. Supported generic binding is retained as non-core. Supporting Evidence: PMID:24658146 FMRP co-precipitated with nucleolin, RPLP0, eIF5, and CYFIP2. Interaction with the latter two proteins was sensitive to RNase treatment. |
| GO:0005515 protein binding | IPI PMID:40205054 Multimodal cell maps as a foundation for structural and func... | KEEP AS NON CORE | Reason: The source is a human U2OS AP-MS screen, which measures co-complex association rather than necessarily direct binary binding. Its exact CYFIP2–ABI3 target table was not recovered, so this paper-specific pair remains unverified. Nevertheless the same human pair is supported by the independently reconstructed 2006 co-IP record, making contextual retention with explicit curator deference defensible. No more specific activity is inferred from this high-throughput association. |
| GO:0005634 nucleus | EXP PMID:17245118 CYFIP2, a direct p53 target, is leptomycin-B sensitive. | KEEP AS NON CORE | Reason: The recovered original shows partial nuclear accumulation of tagged CYFIP2 after leptomycin B treatment in COS-derived cells, against predominantly cytoplasmic localization at low expression. This supports a regulated or conditional nuclear pool, not a dominant constitutive nuclear function. Leptomycin B did not enhance the measured CYFIP2-associated cell-death response. Supporting Evidence: |
| GO:0005634 nucleus | IEA GO_REF:0000044 | KEEP AS NON CORE | Reason: The UniProt subcellular-location mapping refers to the same conditional nuclear accumulation described in the original leptomycin B experiment. Retain that nonexclusive localization as contextual; a normal nuclear enzymatic or transcriptional activity is not established. Propagation Review Root cause: NO FAILURE NON CORE Supporting Evidence: |
| GO:0005737 cytoplasm | EXP PMID:10449408 Increased apoptosis induction by 121F mutant p53. | ACCEPT | Reason: The 1999 original main text was not recovered, so its precise localization assay is not re-created from the abstract. Cytoplasmic CYFIP2 is independently established in the 2001 human-cell fractionation/localization work and the 2007 tagged-cell experiments. Retain the biologically correct broad source assertion with curator deference; cytoplasm is not collapsed to cytosol because particulate and perinuclear pools also occur. |
| GO:0005737 cytoplasm | EXP PMID:17245118 CYFIP2, a direct p53 target, is leptomycin-B sensitive. | ACCEPT | Reason: The original observes diffuse cytoplasmic tagged CYFIP2 at lower expression, with perinuclear aggregates at high expression and only partial nuclear accumulation after export inhibition. The broad cytoplasmic term fits this actual assay scope; it is not equivalent to exclusively soluble cytosol. Supporting Evidence: |
| GO:0005737 cytoplasm | IDA PMID:11438699 A highly conserved protein family interacting with the fragi... | ACCEPT | Reason: Human CYFIP2-specific fractionation detects protein in both soluble and particulate cytoplasmic fractions, with cytoplasmic/perinuclear localization in the transfected-cell experiments. The broad cytoplasm annotation preserves these observed pools rather than narrowing every pool to cytosol. Supporting Evidence: |
| GO:0005737 cytoplasm | IEA GO_REF:0000044 | ACCEPT | Reason: The UniProt cellular-location mapping is consistent with the directly observed soluble and particulate cytoplasmic pools. Retain the broad location alongside the separate, independently supported cytosol assertions. Propagation Review Root cause: NO FAILURE CORE |
| GO:0005829 cytosol | IDA GO_REF:0000052 | ACCEPT | Reason: Current HPA immunofluorescence metadata reports approved cytosolic staining in HeLa cells, with additional plasma-membrane and cell-line-dependent ER localization. This supports the cytosol assertion without making it exclusive or claiming that every HPA compartment is a core site. The underlying images were not reread. |
| GO:0005829 cytosol | TAS Reactome:R-HSA-2029465 | ACCEPT | Reason: The actual Reactome graph has a target-bearing path from R-HSA-2029465 through a CYFIP1/CYFIP2 set to CYFIP2 [cytosol] (1671642) and its Q96F07 reference entity (412747). This supports cytosolic complex membership. CYFIP2 is not assigned the kinase, GTPase, actin-nucleation or other catalytic activities of neighboring participants. The bounded graph capture resolves this target path without claiming every unrelated branch was downloaded. |
| GO:0005829 cytosol | TAS Reactome:R-HSA-2029466 | ACCEPT | Reason: The actual Reactome graph has a target-bearing path from R-HSA-2029466 through a CYFIP1/CYFIP2 set to CYFIP2 [cytosol] (1671642) and its Q96F07 reference entity (412747). This supports cytosolic complex membership. CYFIP2 is not assigned the kinase, GTPase, actin-nucleation or other catalytic activities of neighboring participants. The bounded graph capture resolves this target path without claiming every unrelated branch was downloaded. |
| GO:0005829 cytosol | TAS Reactome:R-HSA-2029469 | ACCEPT | Reason: The actual Reactome graph has a target-bearing path from R-HSA-2029469 through a CYFIP1/CYFIP2 set to CYFIP2 [cytosol] (1671642) and its Q96F07 reference entity (412747). This supports cytosolic complex membership. CYFIP2 is not assigned the kinase, GTPase, actin-nucleation or other catalytic activities of neighboring participants. The bounded graph capture resolves this target path without claiming every unrelated branch was downloaded. |
| GO:0005829 cytosol | TAS Reactome:R-HSA-2130194 | ACCEPT | Reason: The actual Reactome graph has a target-bearing path from R-HSA-2130194 through a CYFIP1/CYFIP2 set to CYFIP2 [cytosol] (1671642) and its Q96F07 reference entity (412747). This supports cytosolic complex membership. CYFIP2 is not assigned the kinase, GTPase, actin-nucleation or other catalytic activities of neighboring participants. The bounded graph capture resolves this target path without claiming every unrelated branch was downloaded. |
| GO:0005829 cytosol | TAS Reactome:R-HSA-2197690 | ACCEPT | Reason: The actual Reactome graph has a target-bearing path from R-HSA-2197690 through a CYFIP1/CYFIP2 set to CYFIP2 [cytosol] (1671642) and its Q96F07 reference entity (412747). This supports cytosolic complex membership. CYFIP2 is not assigned the kinase, GTPase, actin-nucleation or other catalytic activities of neighboring participants. The bounded graph capture resolves this target path without claiming every unrelated branch was downloaded. |
| GO:0005829 cytosol | TAS Reactome:R-HSA-9013145 | ACCEPT | Reason: The actual Reactome graph has a target-bearing path from R-HSA-9013145 through a CYFIP1/CYFIP2 set to CYFIP2 [cytosol] (1671642) and its Q96F07 reference entity (412747). This supports cytosolic complex membership. CYFIP2 is not assigned the kinase, GTPase, actin-nucleation or other catalytic activities of neighboring participants. The bounded graph capture resolves this target path without claiming every unrelated branch was downloaded. |
| GO:0006915 apoptotic process | IBA GO_REF:0000033 | KEEP AS NON CORE | Reason: The actual apoptosis IBD is on the human CYFIP2 path at PTN002638922, with human Q96F07 itself among the descendant evidence; that self-evidence is legitimate, not circular. The 2007 human CYFIP2 expression experiments independently show apoptosis readouts in DLD1 cells, supporting a contextual phenotype. CYFIP2 depletion did not abolish p53-mediated apoptosis, so this is not a universal required apoptotic mechanism. The older source experiment remains separately uncertain in its own row. Propagation Review Root cause: NO FAILURE NON CORE |
| GO:0006915 apoptotic process | IDA PMID:10449408 Increased apoptosis induction by 121F mutant p53. | UNDECIDED | Reason: The cached 1999 abstract establishes induction of the gene in a p53 setting, but the original main text remains inaccessible. The recovered 2007 Discussion explicitly reports that the earlier adenoviral CYFIP2 experiment did not alter cell-cycle distribution or the sub-G1 fraction; this is attributed later reporting, not an independent reread of the 1999 assay. The 2007 paper itself demonstrates a positive context-dependent apoptosis phenotype, so apoptosis biology is not rejected. The exact older IDA support is unresolved rather than confidently wrong or fully verified. |
| GO:0007411 axon guidance | IBA GO_REF:0000033 | KEEP AS NON CORE | Reason: Human CYFIP2 descends from the actual axon-guidance IBD at PTN000257567. The literal Drosophila and zebrafish donor identities were resolved against official records and their experimental annotations; this is an ancestral inference, not a direct human axon-guidance assay. CYFIP2-dependent neuronal morphogenesis makes the transfer biologically plausible. Retain the specialized developmental outcome as contextual without treating donor number as evidence strength. Propagation Review Root cause: NO FAILURE NON CORE |
| GO:0016020 membrane | HDA PMID:19946888 Defining the membrane proteome of NK cells. | KEEP AS NON CORE | Reason: The original HDA assertion concerns an NK-cell membrane-enrichment proteome. Its target supplemental entry was not recovered, so the particular membrane pool cannot be resolved from the inspected source. Retain the broad, biologically plausible membrane association as contextual with curator deference. Independent HPA staining and WRC pathway observations concern other sources and do not narrow this NK-cell assertion to plasma membrane; no integral membrane topology is inferred. Supporting Evidence: |
| GO:0030833 regulation of actin filament polymerization | IEA GO_REF:0000002 | ACCEPT | Reason: The current InterPro IPR008081 family entry actually matches human CYFIP2 and maps to regulation of actin filament polymerization. Human Rac1 association, WAVE-complex association and the neuronal cytoskeletal experiments support the regulatory role. WAVE/ARP2/3 supplies the downstream nucleation machinery; this mapping does not give CYFIP2 an autonomous actin-polymerizing catalytic activity. Propagation Review Root cause: NO FAILURE CORE |
| GO:0031175 neuron projection development | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: The exact Ensembl/UniProt source is mouse Cyfip2 (Q5SQX6), whose current experimental record includes PMID27605705. The original mouse-neuron experiments show reduced dendritic branching after individual CYFIP2 depletion, with different effects from CYFIP1 and combined depletion. This supports a conserved neuron-projection-development role while preserving the transferred evidence scope; it is not a direct human-neuron loss-of-function assay. Propagation Review Root cause: NO FAILURE NON CORE |
| GO:0031209 SCAR complex | IDA PMID:17101133 NESH (Abi-3) is present in the Abi/WAVE complex but does not... | ACCEPT | Reason: The complete 2006 abstract and the literal HUVEC co-IP record support CYFIP2 association with an ABI3/WAVE complex. Full original access and exact antibody discrimination remain limited. Retain the curated SCAR/WAVE complex membership in light of that evidence and later independent WRC studies, without claiming that the family-level PAINT negative record invalidates the human experimental assertion. |
| GO:0031209 SCAR complex | IDA PMID:26428302 The NESH/Abi-3-based WAVE2 complex is functionally distinct ... | ACCEPT | Reason: The original HUVEC experiments immunoprecipitate ABI1- or ABI3-containing WAVE2 complexes and detect the PIR121/Sra1 component. The antibody readout is not CYFIP2-exclusive, so this is not treated as a paralog-resolved purified CYFIP2 structure. With the curated target assertion and independent CYFIP2 association evidence, SCAR/WAVE complex membership is well supported. Supporting Evidence: PMID:26428302 both Abi-1 and NESH/Abi-3 co-precipitated WAVE2 and PIR121/Sra1 |
| GO:0031209 SCAR complex | IEA GO_REF:0000107 | ACCEPT | Reason: The actual mouse Cyfip2 donor and Ensembl alias are resolved, and current donor annotations include the WAVE-complex evidence. Human membership is independently supported by curated cellular assays. The PAINT target-path SCAR NOT record is recorded as an unresolved reconciliation issue; it does not automatically override these separate experimental and orthology-based sources. Propagation Review Root cause: NO FAILURE CORE |
| GO:0031267 small GTPase binding | IEA GO_REF:0000002 | ACCEPT | Reason: Both current InterPro entries actually match CYFIP2: the CYFIP family and the N-terminal Rac1-binding domain (displayed-product residues 64–273). Human Rac1 co-immunoprecipitation corroborates the small-GTPase-binding class. This mapping assigns binding, not GTP hydrolysis or a mechanism unique to every CYFIP paralog. Propagation Review Root cause: NO FAILURE CORE Supporting Evidence: |
| GO:0043005 neuron projection | IBA GO_REF:0000033 | ACCEPT | Reason: The actual PTN000257567 IBD lies on the target path and is grounded in the resolved mouse Cyfip1 source; a current rat Cyfip1 donor is an additive database change, not silently inserted into the historical source object. CYFIP2 itself is observed in neuronal processes and supports neuronal cytoskeletal organization. Retain the conserved location without calling the donor a direct human CYFIP2 assay. Propagation Review Root cause: NO FAILURE CORE |
| GO:0045202 synapse | IBA GO_REF:0000033 | ACCEPT | Reason: The actual synapse IBD at PTN000257567 is ancestral to the target. Both human paralogs and resolved mouse donors contribute evidence; human CYFIP2 self-evidence is expected and not circular. The original synaptosomal fractionation and later CYFIP2 neuronal work support synaptic association, without requiring every product to have identical distribution. Propagation Review Root cause: NO FAILURE CORE |
| GO:0045202 synapse | IDA PMID:11438699 A highly conserved protein family interacting with the fragi... | ACCEPT | Reason: The original synaptosome experiment uses mouse brain and CYFIP2-sensitive detection, with fraction-purity controls. It supports conserved synaptic association, while the imported human IDA assertion is retained with curator deference rather than recast as an independently performed human-brain assay. The separate human-cDNA neuronal experiments provide additional relevant context. |
| GO:0045202 synapse | IEA GO_REF:0000120 | ACCEPT | Reason: The combined mapping joins a resolved mouse Cyfip2/Ensembl source and the UniProt synapse-location vocabulary. It is concordant with actual synaptosomal CYFIP2 detection and later neuronal work. The inference is not evidence of human-specific or isoform-exclusive synaptic localization. Propagation Review Root cause: NO FAILURE CORE |
| GO:0045862 positive regulation of proteolysis | IDA PMID:18387192 A pathway sensor for genome-wide screens of intracellular pr... | KEEP AS NON CORE | Reason: The original screen reports PIR121/CYFIP2 as a positive human-ORF hit: ectopic expression increased normalized Actin-DEVDG reporter release 12.1-fold in 293ET cells, measured in triplicate. GO:0045862 covers processes that increase peptide-bond hydrolysis, and this experimentally observed cleavage-reporter outcome supports contextual positive regulation without requiring CYFIP2 itself to be a protease or its downstream mediator to be identified. The reporter was validated with caspase and cleavage-site controls at assay level; the later hit-specific alternative-motif validation concerns TBC1D10A, not CYFIP2. Retain the curated positive as non-core, limited to this overexpression screen; endogenous necessity and a universal proteolysis mechanism are not established. Supporting Evidence: PMID:18387192 ASPH, PIR121, PERP and TBC1D10A, which induced 14.2-, 12.1-, 10.4- and 5.5-fold increases |
| GO:0048471 perinuclear region of cytoplasm | IDA PMID:10449408 Increased apoptosis induction by 121F mutant p53. | KEEP AS NON CORE | Reason: The original 1999 main localization assay is unavailable. Perinuclear cytoplasmic CYFIP2 is independently observed in the 2001 and 2007 tagged-protein experiments, including expression-dependent aggregates in the latter. Retain the original location with curator deference as a contextual pool, not a constitutive organelle-specific activity. |
| GO:0048471 perinuclear region of cytoplasm | IEA GO_REF:0000044 | KEEP AS NON CORE | Reason: The UniProt perinuclear mapping is consistent with observed tagged-CYFIP2 distributions, including expression-dependent accumulation. It is a nonexclusive contextual location and does not establish a distinct perinuclear molecular mechanism. Propagation Review Root cause: NO FAILURE NON CORE |
| GO:0051388 positive regulation of neurotrophin TRK receptor signaling pathway | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: The actual donor is mouse Cyfip2, with experimental support from the retrolinkin/WAVE1 study. Its mechanism involves WRC recruitment and BDNF-dependent trafficking: combined CYFIP1/CYFIP2 depletion impairs TrkB internalization, whereas single depletion does not produce that same endpoint. Signaling experiments also perturb retrolinkin or WAVE1. Retain a conserved contextual contribution to TRK signaling with these redundancy limits; CYFIP2 is not the receptor kinase and this is not a direct human catalytic assay. Propagation Review Root cause: NO FAILURE NON CORE Supporting Evidence: |
| GO:0070062 extracellular exosome | HDA PMID:19056867 Large-scale proteomics and phosphoproteomics of urinary exos... | KEEP AS NON CORE | Reason: The original author-maintained urinary-exosome database explicitly lists CYFIP2/NP_055191 with one peptide and reference 2 (Gonzales 2009). This is positive fraction-detection evidence with limited peptide depth, suitable for contextual exosome retention. It does not establish a secreted catalytic role, constitutive secretion or an extracellular WRC mechanism. Supporting Evidence: |
| GO:0070062 extracellular exosome | HDA PMID:20458337 MHC class II-associated proteins in B-cell exosomes and pote... | KEEP AS NON CORE | Reason: The original abstract describes B-cell exosome proteomics, but its CYFIP2 target table was not recovered. Independent urinary-exosome detection supports this broad compartment, so the curated HDA is retained as contextual with explicit paper-specific uncertainty. This does not verify CYFIP2 as an MHC-II binding partner or claim that urinary and B-cell vesicles have identical cargo mechanisms. |
| GO:0070062 extracellular exosome | HDA PMID:23533145 In-depth proteomic analyses of exosomes isolated from expres... | KEEP AS NON CORE | Reason: The original EPS-urine study uses pooled vesicle preparations and LC-MS/MS, with acknowledged fluid-protein/contamination limitations. Its exact CYFIP2 supplemental target entry was not recovered. Independent positive urinary-exosome detection makes contextual retention with curator deference reasonable, without claiming source-table verification or an extracellular CYFIP2 activity. |
| GO:0097484 dendrite extension | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: The resolved mouse Cyfip2 donor carries experimental evidence from the retrolinkin/WAVE1 study. Individual CYFIP2 depletion affects dendritic branching, while total-length and trafficking phenotypes depend on the assay and combined CYFIP depletion. Retain the conserved dendrite-extension annotation as a contextual developmental contribution, without equating it with direct autonomous neurite elongation by purified protein. Propagation Review Root cause: NO FAILURE NON CORE Supporting Evidence: PMID:27605705 Double but not single knockdown of CYFIP1 and CYFIP2 also impaired BDNF-induced dendrite outgrowth |
| GO:0098609 cell-cell adhesion | IDA PMID:15048733 CYFIP2 is highly abundant in CD4+ cells from multiple sclero... | MODIFY | Reason: The actual original assay measures adhesion of Jurkat and primary human CD4 cells to fibronectin-coated plates after CYFIP2 overexpression or antisense depletion. This is cell-matrix adhesion, not the cell-cell adhesion stated by the source term. Refine to the actual measured process while preserving the human IDA source. The data do not show CYFIP2 binding fibronectin directly. Proposed replacements: cell-matrix adhesion Supporting Evidence: PMID:15048733 overexpression of CYFIP2 in Jurkat cells increased fibronectin-mediated adhesion. |
| GO:0150052 regulation of postsynapse assembly | IDA PMID:28692454 Phosphorylation of CYFIP2, a component of the WAVE-regulator... | KEEP AS NON CORE | Reason: The original expresses human CYFIP2 cDNA in rat E18 hippocampal neurons and measures dendritic-spine classes/density. Its 1253-residue numbering and S582/S968/T1067 sites correspond to current Q96F07-2, but this is not full historical clone sequencing. The cellular structural phenotype supports contextual postsynapse-assembly regulation, without proving a particular assembly reaction or equal activity of both official products. Supporting Evidence: |
| GO:0150052 regulation of postsynapse assembly | IMP PMID:28692454 Phosphorylation of CYFIP2, a component of the WAVE-regulator... | KEEP AS NON CORE | Reason: This source assertion reflects the same human-CYFIP2 phosphosite-mutant experiment in rat hippocampal neurons, not a paper restricted to a nonhuman gene product. Spine and neurite outcomes support contextual regulation. Mutant effects and the reported unchanged Nap1 co-IP for T1067A do not establish a universal complex-disassembly mechanism; retain the distinct IMP source without merging it into the IDA record. Supporting Evidence: |
Loading supporting content…
Download this section (compressed HTML)Q: How should the current CYFIP2-path SCAR-complex NOT record be reconciled with curated human complex-association evidence and the separate positive node records?
Q: Do the 1278- and 1253-residue products differ in Rac-dependent WRC regulation, localization or partner interactions when compared at matched expression?
Q: Which CYFIP2-dependent step accounts for the context-sensitive apoptosis phenotype, and does it differ between endogenous loss and ectopic expression?
Experiment: Compare sequence-verified 1278- and 1253-residue human CYFIP2 in reconstituted WRCs and matched-expression cellular rescue. Measure Rac binding, complex integrity and actin assembly separately, with CYFIP1-containing complexes as controls.
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)