The selected PB protein differs from reviewed PA by two residues beyond the MPN/JAMM region. PMID:12183637(https://pubmed.ncbi.nlm.nih.gov/12183637/) directly tests the JAMM motif and links it to cullin deneddylation. This supports the catalytic role within COP9, represented with contributes_to_molecular_function. No autonomous CSN5 protease activity is asserted. Neuronal developmental effects do not establish the external synaptic-vesicle claim, which remains UNC. Three seeded annotations are reviewed; exact prediction metadata are retained.
The completed Falcon synthesis was inspected. Its interpretation of cullin deneddylation within COP9 agrees with the conserved MPN/JAMM analysis and PMID:12183637. Nuclear enrichment in fly immune cells and cullin-dependent neuronal branching provide context, but neither establishes a synaptic-vesicle location. The core function remains complex-dependent deNEDDylation; neuronal phenotypes do not identify an additional direct protease substrate.
COP9 membership is proposed as NEW with ISS for exact PB, supported by the primary eight-subunit isolation study PMID:10531038 and same-gene sequence conservation. This is the biological context for the deneddylase contribution.
No ProtNLM2 files were opened or used.
Identity. A0A0B4KHM2 is the unreviewed TrEMBL entry for FlyBase CSN5-PB (FBpp0306602, transcript CSN5-RB FBtr0334535, NP_001262639; 325 aa, PE 3). It is an alternative isoform of the same gene as the reviewed canonical Swiss-Prot entry Q9XZ58 (CSN5-PA, FBpp0082743, 327 aa). The existing alignment (CSN5-bioinformatics/results/results.json) gives 325/325 identical aligned residues. PB lacks only Gly278-Arg279, which lie outside the MPN domain (52-189). The JAMM motif (135-148) and the metal ligands His135, His137 and Asp148 are retained. It is neither a fragment nor a broken gene model; no domain or catalytic residue is lost.
Evolutionary grounding. Both accessions are in PANTHER PTHR10410. InterPro does not report a subfamily for either accession. The PANTHER subfamily named for this protein is PTHR10410:SF6, "COP9 SIGNALOSOME COMPLEX SUBUNIT 5", but I could not confirm the assignment programmatically, so I do not assert it in the review. QuickGO IBA rows on Q9XZ58 identify the PAINT node: PTN000045369, the CSN5 clade, which propagates COP9 signalosome, deNEDDylase activity, cytoplasm (is_active_in) and regulation of cell cycle. A deeper node, PTN000045368, propagates metal-dependent deubiquitinase activity from PSMD14/BRCC3-type donors. That node is not used here, because CSN5's established substrate is NEDD8-cullin, not ubiquitin. The function transfer to PB is a same-gene transfer (identical sequence over the catalytic region), not an inference across orthologs.
Decisions (3 GOA rows plus 2 NEW):
- nucleus IEA: ACCEPT. Q9XZ58 records cytoplasm and nucleus, and FlyBase has an IDA for nuclear periphery from hemocytes (PMID:17295838); nucleus is the right breadth.
- cytoplasm IEA: ACCEPT. FlyBase IDA on PA (PMID:11779478, PMID:17295838) and the PAINT IBA from PTN000045369 support it.
- peptidase activity IEA: MODIFY to GO:0140758 metal-dependent deNEDDylase activity. FlyBase has this IDA on PA (PMID:12737805); JAMM mutants abolish deneddylation (PMID:12183637). The core function uses contributes_to because the activity requires the holo-complex.
- COP9 signalosome: NEW (ISS), kept from the prior review and re-justified (PMID:10531038).
- protein deneddylation GO:0000338: NEW (ISS). CSN5's JAMM site performs the cleavage, so this is participation, not substrate necessity; PA has IMP annotations (PMID:12737805).
- The developmental IMP terms on PA (oocyte axis, GSC maintenance, R-cell differentiation, glial migration, hemocyte proliferation, protein stabilization, and others) are deliberately not proposed for PB. They are downstream consequences of altered CRL activity.
Prior review corrections. The earlier UniProt SUBCELLULAR LOCATION quote dropped the inline evidence tag; it now quotes the line verbatim. Added PMID:12737805, PMID:11779478 and PMID:17295838 as references. Added nucleus and cytoplasm as core-function locations and protein deneddylation as a directly_involved_in process. The description was rewritten as a standalone biological summary.
Family review. interpro/panther/PTHR10410/PTHR10410-review.yaml lists A0A0B4KHM2 under "Benchmark members without resolved subfamily placement". This does not conflict with the CSN5-clade placement from PAINT (PTN000045369). The family review does not assign a subfamily. A future revision could place CSN5 in the COP9 SIGNALOSOME COMPLEX SUBUNIT 5 subfamily once the member index confirms it.
Modules. modules/dmel_cop9_signalosome.yaml (CSN5 grounded as UniProtKB:Q9XZ58 in the CSN5-CSN6 MPN dimer).
Next relative. The canonical same-gene product Q9XZ58 (CSN5-PA) has no complete review in the repo (genes/DROME/CSN5 is keyed to A0A0B4KHM2), so it is the next target: genes/DROME/CSN5__Q9XZ58. It carries the experimental FlyBase/UniProt annotations (IDA, IMP, IPI, IBA).
The re-review proposed NEW rows (GO:0008180, GO:0000338) on this accession. GOA already carries the same terms on
another product of the same gene, reviewed in genes/DROME/CSN5__Q9XZ58/. GOA propagates gene-level
annotations to one representative product per gene, so adding them to a second product fills a
convention rather than a curation gap. The rows were withdrawn, following the rule applied across the
ProtNLM2 re-review cohorts (first raised in the review of PR #5303). The gene-level function is unchanged.