pop2 (Sud1) — S. pombe — Review notes

UniProt: O14170 (POP2_SCHPO), 703 AA. PomBase SPAC4D7.03. Synonym sud1 ("proteolysis factor sud1").
F-box domain (UniProt 236-283) plus C-terminal WD40 repeats (UniProt lists six repeats spanning 389-654;
the primary papers describe a seven-bladed propeller). Paralogue: pop1/Ste16 (P87060). Budding yeast
comparator: Cdc4 (P07834); human comparators: FBXW7 (sequence) and SKP2 (role as the CKI-degrading
F-box adaptor). Do not confuse with the CCR4-NOT deadenylase Pop2/Caf1 of S. cerevisiae (S. pombe
caf1), which shares only the name.

Inputs used: pop2-uniprot.txt, pop2-goa.tsv, pop2-deep-research-falcon.md (Edison/Falcon synthesis of the
same primary papers plus Uchiyama 2015 on Rev1, which is not in the publication cache and was not used),
and the cached publications. Full text is cached for PMID:9653157, PMID:12167173 and PMID:17016471;
PMID:9990507, PMID:10209119, PMID:14970237 and PMID:15147268 are abstract-only, which is recorded in
each reference_review.

Core biology

Pop2 is the second Cdc4/Fbw7-type F-box/WD40 substrate receptor of the fission yeast SCF ubiquitin
ligase. It has no catalytic activity of its own. Its N-terminal region (residues 1-241) mediates
heterooligomerisation with Pop1 and homooligomerisation, its F-box binds Psh1 (Skp1), and its WD40
propeller is the presumed phosphodegron-reading surface. At endogenous levels Pop2 co-purifies with
Pop1, Pcu1 (cullin-1), Psh1 and Pip1 (Rbx1) in a ~500 kDa SCF(Pop1-Pop2) complex, and it also forms a
Pop1-independent SCF(Pop2) complex that carries ubiquitin ligase activity in vitro. The two paralogues are
not interchangeable: they fail to cross-complement, Cdc18 binding to Pop2 needs Pop1, Pop2 alone does not
bind Rum1, and within the heteromer the Pop1 F-box is essential while the Pop2 F-box is dispensable, so
Pop2 is thought to be tethered to the SCF core largely through Pop1.

Substrates with direct evidence:
- Cdc18 (replication initiator, Cdc6 orthologue): sud1/pop2 deletion hyperaccumulates Cdc18; Sud1 binds
Cdc18 in vivo only after CDK phosphorylation at consensus sites; Cdc18 binding to Pop2 requires Pop1.
- Rum1 (CDK inhibitor): accumulates in pop2 mutants (half-life ~20 min -> >100 min); the immunopurified
Pop1-Pop2 complex binds CDK-phosphorylated Rum1 and polyubiquitylates it with E1, UBC3, ATP and
ubiquitin; Pip1 complexes from pop2-deleted lysate have no Rum1 ubiquitylation activity.
- Cig2 (S-phase cyclin): Pop1 and Pop2 are responsible for SCF-dependent Cig2 instability in G2 and M
(the abstract reports Pop1 binding Cig2; the PomBase IPI for Pop2 rests on the full text).

Phenotypes: pop2/sud1 deletants are viable but undergo spontaneous re-replication and diploidise;
hyperaccumulated Rum1 contributes to the re-replication but does not explain the Cdc18 proteolysis defect.
The phenotype is milder than that of pop1 deletion. Pop2 lacks the N-terminal NLS of Pop1 and is found
in both the nucleus and the cytoplasm; the complete SCF(Pop1-Pop2) complex is nuclear, where the
substrates are, and the cytoplasmic pool has been proposed to act on unidentified substrates.

Key evidence (verbatim quotes)

Decision table (19 GOA rows; 20 in the TSV, the two PMID:15147268 Skp1 rows collapse to one)

GO term Evidence / reference / with Action One-line reason
GO:0005515 protein binding IPI PMID:10209119 / Cdc18 (P41411) MODIFY -> GO:1990756 Substrate capture by the Pop1-Pop2 receptor; same paper as the IDA adaptor-activity annotation
GO:0005515 protein binding IPI PMID:10209119 / Pop1 (P87060) MODIFY -> GO:1990756 Heterooligomerisation is how Pop2 is built into the substrate receptor; N-terminus fusion rescues the double deletion
GO:0005515 protein binding IPI PMID:12167173 / Pcu1 (O13790) MODIFY -> GO:1990756 Endogenous cullin association within an active, pop2-dependent ligase; complex membership kept under GO:0019005/GO:0043224
GO:0005515 protein binding IPI PMID:12167173 / Pop1 (P87060) MODIFY -> GO:1990756 Heteromer, not either protein alone, binds phospho-Rum1; Pop2 F-box dispensable, so the Pop1 contact is the mechanism
GO:0005515 protein binding IPI PMID:12167173 / Psh1 (Q9Y709) MODIFY -> GO:1990756 Canonical F-box/Skp1 attachment demonstrated at endogenous levels in a reconstituted ligase
GO:0005515 protein binding IPI PMID:15147268 / Psh1 (Q9Y709) REMOVE Twelve-F-box survey of Skp1; adds nothing to the focused SCF(Pop) evidence; interaction not disputed
GO:0005515 protein binding IPI PMID:17016471 / Psh1 (Q9Y709) REMOVE Pof14/ergosterol paper; Pop2 only a reference F-box; cached full text does not mention Pop2
GO:0005515 protein binding IPI PMID:9653157 / Cdc18 (P41411) MODIFY -> GO:1990756 Founding demonstration of phosphorylation-dependent substrate recognition (full text)
GO:0005515 protein binding IPI PMID:9990507 / Pcu1 (O13790) REMOVE Same paper and partner underlie the GO:0019005 IPI; the binding row is a duplicate without function
GO:0005515 protein binding IPI PMID:9990507 / Pop1 (P87060) MODIFY -> GO:1990756 Hetero-/homo-dimers are the substrate-receptor units controlling Rum1 and Cdc18
GO:0005634 nucleus IEA GO_REF:0000044 ACCEPT Endogenous-tag immunofluorescence and fractionation; the nuclear SCF acts on nuclear Cdc18/Rum1
GO:0005737 cytoplasm IEA GO_REF:0000044 ACCEPT Experimentally solid, distinguishes Pop2 from nuclear-restricted Pop1; substrates of the cytoplasmic pool unknown
GO:0019005 SCF ubiquitin ligase complex IPI PMID:9990507 / cul1 ACCEPT Core CC; five-subunit ~500 kDa complex confirmed at endogenous levels with ligase activity
GO:0031146 SCF-dependent proteasomal ubiquitin-dependent protein catabolic process ISO GO_REF:0000024 / CDC4 ACCEPT Core BP; transfer from Cdc4 fully corroborated by S. pombe genetics and in vitro ubiquitylation
GO:0043224 nuclear SCF ubiquitin ligase complex ISO GO_REF:0000024 / CDC4 ACCEPT Transfer from Cdc4 corroborated by nuclear coexistence of all five SCF(Pop) subunits
GO:1903467 negative regulation of mitotic DNA replication initiation IMP PMID:9653157 ACCEPT Core BP; Pop2 performs the phospho-Cdc18 recognition step; specific, correctly signed term
GO:1990756 ubiquitin-like ligase-substrate adaptor activity IDA PMID:10209119 ACCEPT Core MF, correctly typed for a non-catalytic F-box receptor
GO:1990756 ubiquitin-like ligase-substrate adaptor activity IEA GO_REF:0000117 / ARBA ACCEPT Electronic inference lands exactly on the experimentally established MF
GO:1990756 ubiquitin-like ligase-substrate adaptor activity IPI PMID:14970237 / Cig2 ACCEPT Extends substrate range to Cig2; abstract reports Pop1 binding, deferring to the curator who read the full text

Action counts: ACCEPT 9, MODIFY 7, REMOVE 3 (no UNDECIDED, no NEW).

Notable curation decisions

  1. Protein binding rows were split by information content rather than treated uniformly. The seven
    rows whose partner is a substrate (Cdc18) or the co-receptor Pop1, or that come from the paper
    demonstrating an active endogenous ligase (PMID:12167173), are MODIFY to GO:1990756 because each
    interaction is a step of the adaptor function. The three rows that duplicate a complex annotation
    from the same paper (cullin-1 in PMID:9990507) or come from surveys whose subject is another protein
    (Skp1 in PMID:15147268; Pof14 in PMID:17016471) are REMOVE as uninformative, with the explicit note
    that the interactions themselves are not disputed. This matches the pop1 review, where every
    protein-binding row was MODIFY to GO:1990756; the only divergence is that pop2 also carries survey-
    and duplicate-type rows that pop1 does not, and those are removed here.
  2. Both localisation IEAs (nucleus and cytoplasm) are ACCEPT with core status. The cytoplasm row is
    kept as core because PMID:12167173 shows a Pop1-independent SCF(Pop2) complex with ligase activity
    and Pop2 lacks Pop1's NLS; this is modelled as a second core function with an explicit knowledge gap
    (no cytoplasmic substrate identified; the in vitro activity was a substrate-independent assay).
    The pop1 review removed the pop1 cytoplasm IBA because Pop1 is exclusively nuclear, so the two
    paralogues are deliberately treated differently on this term.
  3. The two ISO transfers from S. cerevisiae Cdc4 (GO:0031146, GO:0043224) are ACCEPT with
    propagation_review.root_cause: NO_FAILURE_CORE, on the grounds that the fission yeast evidence
    independently establishes each claim.
  4. GO:1903467 (IMP, PMID:9653157) passes the participation test: Pop2 itself performs the
    phospho-Cdc18 recognition step, so it is a core BP rather than a necessity-only phenotype term.
  5. The Cig2 IPI for GO:1990756 (PMID:14970237) is ACCEPT despite the abstract naming Pop1, not Pop2,
    as the Cig2-binding protein, following the "do not overrule curators from incomplete evidence" rule.
  6. No NEW terms were proposed. In particular GO:1900087 (positive regulation of G1/S transition), which
    pop1 carries by IMP, was not added to pop2 even though Rum1 half-life rises >5-fold in pop2 mutants;
    the omission is raised as a suggested question rather than filled in.
  7. The deep-research Rev1 claim (Uchiyama 2015, PLoS One) was not used: the paper is not in the
    publication cache and no GOA row cites it.

Open questions

Housekeeping