id: O75155 gene_symbol: CAND2 product_type: PROTEIN status: COMPLETE taxon: id: NCBITaxon:9606 label: Homo sapiens description: >- CAND2 (Cullin-associated NEDD8-dissociated protein 2; also TIP120B, TBP-interacting protein of 120 kDa B) is a large (~1236 aa) HEAT-repeat (alpha-solenoid) protein and the paralog of CAND1. Like CAND1, it functions as a regulator of cullin-RING ubiquitin ligase (CRL/SCF) assembly rather than as a ligase itself: it binds unneddylated, substrate-receptor-free cullin-RBX cores (CAND-bound cullins cannot be neddylated, and neddylated cullins do not stably bind CAND) and acts as an F-box-protein exchange factor, sequestering cullins and accelerating the dissociation/exchange of F-box (and other substrate-recognition) modules to dynamically reshape the cellular repertoire of active CRL complexes within the NEDD8/COP9-signalosome remodeling cycle. CAND2 has no catalytic ubiquitin-ligase activity of its own. Biochemically it binds the CUL1-RBX1 core comparably to CAND1 but catalyzes SCF disassembly with lower efficiency (higher KM, faster koff), and CAND1 and CAND2 can act nonredundantly to support optimal activity of specific SCF ligases (e.g. SCF(FBXL5)-mediated IRP2 turnover). It was originally identified as a TBP-interacting protein expressed preferentially in muscle and induced during myogenesis, where it binds CUL1 and suppresses SCF-dependent ubiquitination of the differentiation factor myogenin to accelerate myogenic differentiation; it also interacts with the transcription-related factors TBP and CNOT3/NOT3, and is a specific in vitro substrate of the HECT E3 ligase UBE3C/KIAA10, which targets it for proteasomal degradation. CAND2 is enriched in skeletal and cardiac muscle and is also found in epididymis; subcellular pools are reported in both the nucleus and the cytosol. Common variation at the CAND2 locus (rs4642101) is associated with atrial fibrillation risk. existing_annotations: - term: id: GO:0005634 label: nucleus evidence_type: IBA original_reference_id: GO_REF:0000033 qualifier: is_active_in review: summary: Phylogenetic assignment of nuclear localization, consistent with UniProt's reported nuclear localization and the TBP-interaction history. action: KEEP_AS_NON_CORE reason: Plausible localization (UniProt lists Nucleus), but a cytosolic pool is also reported (IDA) and the CRL-assembly regulator function is not restricted to the nucleus; non-core. supported_by: - reference_id: file:human/CAND2/CAND2-uniprot.txt supporting_text: 'SUBCELLULAR LOCATION: Nucleus' - term: id: GO:0010265 label: SCF complex assembly evidence_type: IBA original_reference_id: GO_REF:0000033 qualifier: involved_in review: summary: Phylogenetic assignment of involvement in SCF complex assembly, the core function of the CAND family as CRL substrate-receptor exchange factors. Core process, now supported by direct human CAND2 biochemistry (F-box exchange-factor activity on the CUL1-RBX1 core) and the muscle myogenin/CUL1 study. action: ACCEPT reason: Core biological process; CAND2, like CAND1, promotes exchange of the F-box substrate-recognition subunit in SCF complexes, regulating CRL assembly. Human CAND2 binds the CUL1-RBX1 core and acts as an F-box exchange factor (less efficient than CAND1), and in muscle it binds CUL1 to modulate SCF-dependent turnover of myogenin. additional_reference_ids: - file:human/CAND2/CAND2-deep-research-falcon.md supported_by: - reference_id: file:human/CAND2/CAND2-uniprot.txt supporting_text: promotes the exchange of the substrate-recognition F-box subunit in SCF complexes, thereby playing a key role in the cellular repertoire of SCF complexes - reference_id: file:human/CAND2/CAND2-deep-research-falcon.md supporting_text: 'CAND-family proteins (classically CAND1, by inference also CAND2) preferentially associate with **unneddylated** cullins; structural/mechanistic work emphasizes that **CAND-bound cullins cannot be neddylated**, and **neddylated cullins do not stably bind CAND**' - term: id: GO:0016567 label: protein ubiquitination evidence_type: IBA original_reference_id: GO_REF:0000033 qualifier: involved_in review: summary: Phylogenetic assignment of involvement in protein ubiquitination. CAND2 has no ligase activity; it regulates CRL assembly, indirectly affecting ubiquitination flux. action: MARK_AS_OVER_ANNOTATED reason: CAND2 is a CRL assembly regulator/exchange factor, not a ubiquitin ligase or conjugation enzyme. It does not itself ubiquitinate substrates; an involved_in protein ubiquitination annotation overstates a direct catalytic/process role. Its effect on ubiquitination is indirect via SCF complex assembly. supported_by: - reference_id: file:human/CAND2/CAND2-uniprot.txt supporting_text: Probable assembly factor of SCF (SKP1-CUL1-F-box protein) E3 ubiquitin ligase complexes - term: id: GO:0005634 label: nucleus evidence_type: IEA original_reference_id: GO_REF:0000120 qualifier: located_in review: summary: Combined automated electronic assignment of nuclear localization, transferred from the UniProt subcellular location / mouse ortholog. action: KEEP_AS_NON_CORE reason: Plausible localization redundant with the IBA nucleus annotation; a cytosolic pool also exists, and localization is non-core relative to the CRL-assembly function. supported_by: - reference_id: file:human/CAND2/CAND2-uniprot.txt supporting_text: 'SUBCELLULAR LOCATION: Nucleus' - term: id: GO:0010265 label: SCF complex assembly evidence_type: IEA original_reference_id: GO_REF:0000002 qualifier: involved_in review: summary: InterPro-based electronic assignment (CAND1/CAND2 family signature IPR039852) of involvement in SCF complex assembly. Core process. action: ACCEPT reason: Core biological process; redundant with the IBA assignment and consistent with the CAND family's exchange-factor role. supported_by: - reference_id: file:human/CAND2/CAND2-uniprot.txt supporting_text: promotes the exchange of the substrate-recognition F-box subunit in SCF complexes, thereby playing a key role in the cellular repertoire of SCF complexes - term: id: GO:0005515 label: protein binding evidence_type: IPI original_reference_id: PMID:23864651 qualifier: enables review: summary: Membrane yeast two-hybrid interaction with the GLP-1 receptor (GLP1R) from a screen for GLP-1R interactors. Bare protein binding is uninformative. action: KEEP_AS_NON_CORE reason: High-throughput interaction screen; bare protein binding is uninformative per curation guidelines and the GLP1R interaction is peripheral to CAND2's CRL-assembly function. supported_by: - reference_id: file:human/CAND2/CAND2-uniprot.txt supporting_text: 'O75155; P43220: GLP1R; NbExp=2; IntAct=EBI-5656182, EBI-7466542;' - term: id: GO:0005515 label: protein binding evidence_type: IPI original_reference_id: PMID:32296183 qualifier: enables review: summary: Binary interactome reference map interactions (e.g. SYP, FHL2, CIDEB). Bare protein binding is uninformative. action: KEEP_AS_NON_CORE reason: High-throughput interactome; bare protein binding is uninformative. supported_by: - reference_id: file:human/CAND2/CAND2-uniprot.txt supporting_text: 'O75155; P08247: SYP; NbExp=3; IntAct=EBI-5656182, EBI-9071725;' - term: id: GO:0017025 label: TBP-class protein binding evidence_type: IEA original_reference_id: GO_REF:0000107 qualifier: enables review: summary: Ortholog-based electronic assignment of TBP-class protein binding, reflecting the original identification of TIP120B as a TBP-interacting protein. action: KEEP_AS_NON_CORE reason: Consistent with the TIP120B naming and the TBP-affinity origin (UniProt SUBUNIT lists TBP binding), but this transcription-related interaction is a historical/secondary aspect distinct from the core CRL-assembly function and is only an electronic ortholog transfer. supported_by: - reference_id: file:human/CAND2/CAND2-uniprot.txt supporting_text: 'SUBUNIT: Binds TBP, CNOT3 and UBE3C.' - term: id: GO:0005829 label: cytosol evidence_type: IDA original_reference_id: PMID:12692129 qualifier: located_in review: summary: Direct (ARUK-UCL curated) evidence for cytosolic localization from the study of TIP120B targeting by the HECT E3 KIAA10/UBE3C. action: ACCEPT reason: IDA-supported localization; a cytosolic pool is consistent with CAND2's role with cytoplasmic CRL components. supported_by: - reference_id: file:human/CAND2/CAND2-uniprot.txt supporting_text: TBP-interacting protein of 120 kDa B - term: id: GO:0045893 label: positive regulation of DNA-templated transcription evidence_type: ISS original_reference_id: GO_REF:0000024 qualifier: involved_in review: summary: Sequence-similarity assignment of a transcriptional activator role, transferred from the mouse ortholog (P97536/TIP120A-related) and reflecting the TBP-interacting protein origin. action: MARK_AS_OVER_ANNOTATED reason: The transcriptional-activator characterization is best established for the paralog TIP120A/CAND1; CAND2/TIP120B's own function is CRL substrate-receptor exchange. The ISS transfer (from P97536) likely over-propagates a transcriptional-activation role; treat as over-annotated relative to the supported CRL-assembly function. supported_by: - reference_id: PMID:12207886 supporting_text: Classical TIP120, TIP120A, which functions as a transcriptional activator, is expressed ubiquitously whereas TIP120B is specifically expressed in muscle tissues - term: id: GO:0005515 label: protein binding evidence_type: IPI original_reference_id: PMID:12207886 qualifier: enables review: summary: Yeast two-hybrid / GST pull-down interaction with CNOT3/NOT3 (UniProtKB:O75175). Bare protein binding is uninformative. action: KEEP_AS_NON_CORE reason: Records a real interaction (CNOT3) from the myogenesis study, but bare protein binding is uninformative; this transcription-related interaction is secondary to the CRL-assembly function. supported_by: - reference_id: PMID:12207886 supporting_text: hNOT3L is associated with TIP120B but not with TIP120A - term: id: GO:0005515 label: protein binding evidence_type: IPI original_reference_id: PMID:12692129 qualifier: enables review: summary: Interaction with the HECT E3 ligase KIAA10/UBE3C (UniProtKB:Q15386), which targets TIP120B for proteasomal degradation. Bare protein binding is uninformative. action: KEEP_AS_NON_CORE reason: Records the functionally relevant UBE3C/KIAA10 interaction (CAND2 is its substrate), but bare protein binding is uninformative. supported_by: - reference_id: PMID:12692129 supporting_text: TIP120B, but not the closely related protein TIP120A, is a specific substrate of KIAA10 in vitro - term: id: GO:0097602 label: cullin family protein binding evidence_type: ISS original_reference_id: GO_REF:0000024 qualifier: enables review: summary: Proposed molecular function annotation. As a CAND-family protein, CAND2 binds unneddylated cullin-RBX cores; this cullin-binding activity is the molecular basis of its SCF substrate-receptor exchange-factor role and is the core molecular function absent from the existing GOA. action: NEW reason: CAND2's core molecular function is binding cullin-RBX cores to regulate SCF assembly. This is proposed by similarity to its paralog CAND1 (ISS; GO_REF:0000024 = manual curator-judgment transfer to an ortholog/paralog), which is the basis recorded in the evidence metadata rather than a GOA import. The orthology inference is corroborated by reported human structural/biochemical data (a CAND2-cullin cryo-EM structure, PDB 8VVY, and a reported F-box-protein exchange-factor activity on the CUL1-RBX1 core, less efficient than CAND1, preferentially engaging unneddylated cullins); that primary work is recorded here as a lead (see supported_by) and has not yet been read in full, so the annotation conservatively rests on the CAND-family orthology basis. The term is not in the existing GOA and is added here. additional_reference_ids: - file:human/CAND2/CAND2-deep-research-falcon.md supported_by: - reference_id: file:human/CAND2/CAND2-uniprot.txt supporting_text: Probable assembly factor of SCF (SKP1-CUL1-F-box protein) E3 ubiquitin ligase complexes - reference_id: file:human/CAND2/CAND2-deep-research-falcon.md supporting_text: a 2025 Nature Communications study (received Feb 2024) showing that human CAND2 can **promote SCF-mediated protein degradation** by functioning as an **F-box protein exchange factor** interacting with the CUL1·RBX1 core, analogous to CAND1 but less efficient references: - id: GO_REF:0000002 title: Gene Ontology annotation through association of InterPro records with GO terms findings: [] - id: GO_REF:0000024 title: Manual transfer of experimentally-verified manual GO annotation data to orthologs by curator judgment of sequence similarity findings: [] - id: GO_REF:0000033 title: Annotation inferences using phylogenetic trees findings: [] - id: GO_REF:0000107 title: Automatic transfer of experimentally verified manual GO annotation data to orthologs using Ensembl Compara findings: [] - id: GO_REF:0000120 title: Combined Automated Annotation using Multiple IEA Methods findings: [] - id: PMID:12207886 title: TBP-interacting protein 120B, which is induced in relation to myogenesis, binds to NOT3. findings: - statement: TIP120B is specifically expressed in muscle and induced during myogenesis (unlike ubiquitously expressed TIP120A/CAND1, a transcriptional activator); TIP120B binds the CCR4-NOT complex subunit NOT3/CNOT3. reference_section_type: ABSTRACT reference_review: relevance: MEDIUM correctness: VERIFIED review_notes: Establishes TIP120B muscle-specific expression and the CNOT3 interaction; contrasts TIP120B with the transcriptional-activator paralog TIP120A. Abstract-only in cache. - id: PMID:12692129 title: Proteolytic targeting of transcriptional regulator TIP120B by a HECT domain E3 ligase. findings: - statement: TIP120B (but not TIP120A/CAND1) is a specific in vitro substrate of the HECT E3 ligase KIAA10/UBE3C, which associates with TIP120B via its N-domain and targets it for proteasomal degradation; localizes to the cytosol. reference_section_type: ABSTRACT reference_review: relevance: MEDIUM correctness: VERIFIED review_notes: Source of the cytosol (IDA) localization and the UBE3C/KIAA10 interaction; establishes CAND2 as a degradation substrate of a HECT E3. Abstract-only in cache. - id: PMID:23864651 title: The identification of novel proteins that interact with the GLP-1 receptor and restrain its activity. findings: - statement: A membrane yeast two-hybrid screen for GLP-1 receptor interactors identified CAND2 among candidate interactors that can modulate GLP-1R signaling. reference_section_type: ABSTRACT reference_review: relevance: LOW correctness: VERIFIED review_notes: High-throughput GLP-1R interaction screen; source of a bare protein binding (GLP1R) annotation peripheral to CAND2's core function. - id: PMID:32296183 title: A reference map of the human binary protein interactome. findings: [] reference_review: relevance: LOW correctness: VERIFIED review_notes: Binary interactome reference map; source of several bare protein binding annotations (SYP, FHL2, CIDEB). - id: file:human/CAND2/CAND2-deep-research-falcon.md title: Falcon deep research report for human CAND2 findings: - statement: CAND2/TIP120B is a large HEAT/ARM-repeat cullin-binding protein of the CAND family that regulates cullin-RING ligases (CRLs) and is not itself an enzyme; it is the muscle-enriched paralog of CAND1. supporting_text: Primary literature explicitly equates TIP120B with CAND2 and places it in the **CAND family** of large HEAT/ARM-repeat cullin-binding proteins involved in **cullin-RING ligase (CRL)** regulation (not an enzyme itself). - statement: CAND-family proteins preferentially bind unneddylated cullins; CAND-bound cullins cannot be neddylated and neddylated cullins do not stably bind CAND, placing CAND2 in the NEDD8/CSN CRL-remodeling cycle. supporting_text: 'CAND-family proteins (classically CAND1, by inference also CAND2) preferentially associate with **unneddylated** cullins; structural/mechanistic work emphasizes that **CAND-bound cullins cannot be neddylated**, and **neddylated cullins do not stably bind CAND**' - statement: Human CAND2 binds the CUL1·RBX1 core and functions as an F-box-protein exchange factor that promotes SCF-mediated degradation, analogous to CAND1 but less efficient (higher KM / faster koff). supporting_text: a 2025 Nature Communications study (received Feb 2024) showing that human CAND2 can **promote SCF-mediated protein degradation** by functioning as an **F-box protein exchange factor** interacting with the CUL1·RBX1 core, analogous to CAND1 but less efficient - statement: In skeletal muscle (C2C12), CAND2/TIP120B is induced during myogenic differentiation, binds CUL1, suppresses SCF-dependent ubiquitination and degradation of myogenin, and accelerates differentiation. supporting_text: 'is induced during myogenic differentiation,' - statement: CAND1 and CAND2 can contribute nonredundantly to optimal SCF activity; in the SCF(FBXL5)-IRP2 pathway loss of either slows IRP2 degradation, and ectopic CAND2 rescues an SCF substrate (p-IkBalpha) in CAND1/CAND2 double-knockout cells. supporting_text: 'in the **SCFFBXL5–IRP2** pathway, IRP2 half-life increased **2.8-fold in DKO**, **1.7-fold in CAND1-KO**, and **1.8-fold in CAND2-KO**, indicating both proteins contribute to optimal SCF function in that context' reference_review: relevance: HIGH correctness: UNVERIFIED review_notes: Falcon deep-research synthesis; anchors CAND2 as a non-catalytic CAND-family CRL/SCF exchange factor that binds unneddylated CUL1-RBX cores (Wang 2025 kinetics; Shiraishi 2007 myogenin/CUL1; Liu 2002 family identity). Treated as leads cross-checked against UniProt; the muscle/myogenin and F-box-exchange roles are experimentally supported, while AF-locus and cardiac-remodeling associations are genetic/indirect. core_functions: - description: Cullin-RING ligase (CRL/SCF) substrate-receptor exchange factor that binds unneddylated cullin-RBX (CUL1-RBX1) cores and promotes dissociation/exchange of the F-box (substrate-recognition) subunit, regulating the dynamic assembly and cellular repertoire of SCF complexes within the NEDD8/CSN cycle. Human CAND2 acts comparably to CAND1 but with lower exchange efficiency, and the two paralogs can contribute nonredundantly to optimal activity of specific SCF ligases. CAND2 is a regulator/assembly factor, not a ligase. molecular_function: id: GO:0097602 label: cullin family protein binding supported_by: - reference_id: file:human/CAND2/CAND2-uniprot.txt supporting_text: Probable assembly factor of SCF (SKP1-CUL1-F-box protein) E3 ubiquitin ligase complexes that promotes the exchange of the substrate-recognition F-box subunit in SCF complexes - reference_id: file:human/CAND2/CAND2-deep-research-falcon.md supporting_text: a 2025 Nature Communications study (received Feb 2024) showing that human CAND2 can **promote SCF-mediated protein degradation** by functioning as an **F-box protein exchange factor** interacting with the CUL1·RBX1 core, analogous to CAND1 but less efficient directly_involved_in: - id: GO:0010265 label: SCF complex assembly proposed_new_terms: [] suggested_questions: - question: Does human CAND2 recapitulate the CAND1 biochemical mechanism (binding unneddylated CUL1-RBX1 and accelerating SCF substrate-receptor exchange), and does it act preferentially on muscle-enriched CRL clients given its tissue-restricted expression? - question: How are CAND2's apparently distinct activities (CRL assembly regulation versus its historical TBP/CNOT3 transcription-related interactions) related, and is the transcriptional role a genuine function or an artifact of the TBP-affinity discovery method? - question: In skeletal/cardiac muscle, does CAND2 act predominantly to inhibit specific SCF complexes (e.g. stabilizing myogenin or GRK5 by disrupting their SCF targeting) or as a general F-box exchange factor, and which muscle-enriched CRL clients explain its tissue-restricted expression and atrial-fibrillation association? suggested_experiments: - description: Reconstitute SCF dynamics in vitro with purified CAND2, neddylated and unneddylated CUL1-RBX1, SKP1 and F-box proteins, and measure CAND2-stimulated F-box exchange rates by FRET or pulldown, directly comparing to CAND1. - description: Solve or analyze the cryo-EM structure of CAND2 bound to a cullin-RBX core (PDB 8VVY corresponds to CAND2) to confirm the CAND1-like binding mode, and perform structure-guided mutagenesis to test which contacts are required for exchange-factor activity versus the reported TBP/CNOT3 interactions.