CEP152 review notes

2026-07-18: source collection and review scope

• just fetch-gene human CEP152 produced a 61-row GOA snapshot and UniProt record for O94986.
• just fetch-gene-pmids human CEP152 cached every PMID used by the GOA snapshot.
• Automated deep research was attempted first with Falcon/Edison and then with the Perplexity-lite fallback. The provider endpoints returned HTTP 402 and HTTP 401, respectively, so no provider-named deep-research file was created. Primary publications and the local PAINT source were reviewed directly instead; the synthesis is in CEP152-deep-research-manual.md.
• PAINT source inspection: PANTHER node PTN000818649 carries IBD assertions for centrosome and centriole replication and seeds the human O94986 and mouse Cep152 annotations. The node is CEP152-specific even though the broader PTHR10337 family also contains SHC-family proteins. The IBA calls are therefore biologically coherent and independently supported, but are not independent experimental evidence because human CEP152 is one of the source genes.

Functional synthesis

• CEP152 is a long, coiled-coil pericentriolar-material scaffold. Its N terminus binds PLK4 and its C-terminal region binds CPAP/CENPJ, providing a spatial platform for procentriole initiation PMID:20852615
• Human CEP152 is required to recruit PLK4 and CPAP and for faithful centrosome duplication PMID:21059844
• CEP63 and CEP152 form a self-assembling cylindrical architecture around the parent centriole that recruits PLK4 PMID:30858376
• The newer structural/cell-biological analysis confirms stepwise CEP63-CEP152 assembly and shows that assembly-defective mutants compromise CEP152 organization, PLK4 relocalization, and centriole duplication PMID:37433832
• CEP152 also has a temporally regulated mitotic role. It recruits APC/C to spindle poles and is an APC/C substrate; CEP152 ubiquitylation releases CEP57 so CEP57 can engage pericentrin and promote microtubule nucleation PMID:34878135 This supports adding mitotic spindle assembly but does not justify assigning catalytic ubiquitin-ligase or microtubule-nucleator activity to CEP152.
• In multiciliated-cell models, DEUP1 condensates recruit CEP152 and PLK4 PMID:33658185 Mouse airway multiciliated cells show Cep152 at parental centrioles and deuterosomes PMID:33627667 These annotations are credible specialized-context functions, but direct human evidence is not in the current source set.
• Biallelic CEP152 variants cause MCPH9 and Seckel syndrome 5. A truncating allele prevented centrosomal localization in transfected cells PMID:20598275 Disease phenotypes are consequences of centrosome/centriole dysfunction, not separate molecular functions.

Annotation decisions

• Retain centriole-replication, centrosome-duplication, centrosome, centriole, pericentriolar-material, procentriole, and procentriole-replication-complex annotations. They converge with direct depletion, localization, interaction, and structural evidence.
• Retain protein-macromolecule adaptor activity and protein kinase binding. Replace PLK4-supported generic protein binding records with protein kinase binding; replace the CENPJ/CPAP generic binding record with adaptor activity because the experiment establishes a two-ended scaffold.
• Mark generic protein-binding records for CEP131, MOV10, CCDC66, and CINP as over-annotated. The interactions may be real, but GO:0005515 does not convey CEP152's role and the available evidence does not support a more precise molecular activity for those individual contacts.
• Keep basal-body, deuterosome, and multiciliated epithelial centriole-amplification annotations as non-core specialized-context annotations.
• Keep the 22 Reactome-derived cytosol records as non-core. A soluble pool is plausible, but cytosol is broad and not CEP152's defining site of action; the record multiplicity reflects Reactome event membership rather than 22 independent localization experiments.
• Add mitotic spindle assembly from the direct APC/C-CEP152-CEP57-pericentrin study.

Boundaries and unresolved points

• The architecture and recruitment mechanism are strong, but the stoichiometry and dynamic handoff among CEP192, CEP63-CEP152, PLK4, STIL, and CENPJ across the cell cycle remain incompletely resolved.
• It is unclear how much of the deuterosome/multiciliated-cell mechanism is conserved in human epithelia and how essential deuterosomes are relative to parent-centriole-associated amplification routes.
• Isoform-specific functions are not established. UniProt lists four alternative products, but the reviewed functional studies do not provide a robust basis for assigning distinct GO activities to individual isoforms.