CLN2 (P20438, Saccharomyces cerevisiae) - curation notes

Working journal for the GO annotation review in CLN2-ai-review.yaml.
Systematic name YPL256C; Complex Portal CPX-342 (CLN2-CDC28 kinase complex).
Not to be confused with human CLN2 (neuronal ceroid lipofuscinosis 2, now
TPP1, a lysosomal tripeptidyl peptidase), which is unrelated.

1. Biology in brief

Cln2 is one of the two closely related late-G1 cyclins (Cln1/Cln2) of budding
yeast and, with the upstream cyclin Cln3, forms the trio of Cln proteins of
which at least one is required for passage through Start. It has no catalytic
activity: it is a cyclin-box protein that binds the sole essential CDK Cdc28
and confers a strong, G1-restricted kinase activity on it.

Discovery and genetics. CLN1 and CLN2 were cloned as dosage suppressors of
cdc28-ts; the dominant CLN2-1 truncation removes the destabilising C
terminus and short-circuits G1 control
PMID:2569741
PMID:2569741.
The three CLN genes are redundant for an essential G1 function
PMID:2574633
PMID:2574633,
and the triple-mutant arrest is a genuine Start arrest independent of the
pheromone pathway
PMID:2147225
PMID:2147225.

Cdc28 holoenzyme and activator activity. Cln2 co-immunoprecipitates with
p34CDC28 as an active kinase
PMID:2142620.
Activation was reconstituted biochemically: a GST-Cln2 chimera added to a
cyclin-depleted G1 extract activates Cdc28 as an H1 kinase in a manner that
needs ATP, cytosol and the CAK site Thr169
PMID:7862657
PMID:7862657.
The active G1 holoenzyme is a trimer with Cks1
PMID:10913169
PMID:10913169.
Cln1/Cln2-associated kinase is far stronger than that of the rare Cln3, which
was proposed to act upstream
PMID:8387915
PMID:8387915.

SBF-driven transcription and the Whi5 feedback loop. CLN1 and CLN2 are
transcribed in late G1 by SBF (Swi4/Swi6)
PMID:1832338,
and the transcript and protein peak in G1 and collapse in S phase or on
pheromone exposure
PMID:2142620.
Cln3-Cdc28 begins the derepression of SBF by acting on the bound repressor
Whi5; Cln2-Cdc28 completes it. Recombinant Cln2-Cdc28 phosphorylates Whi5 and
strips it from a preassembled Whi5-Swi4-Swi6 complex, an activity that purified
Cln3-Cdc28 and Pcl9-Pho85 lack, and in vivo the Whi5 phosphoforms depend on
CLN1/CLN2
PMID:19823668
PMID:19823668
PMID:19823668
PMID:19823668
PMID:15210110.
Cln1/Cln2-Cdc28 are also the physiological kinases for the Swi6-binding Start
regulator Stb1
PMID:10409718
PMID:10409718.
Note the historical caveat that Cln3 alone suffices to switch SBF on
PMID:10409718;
the Cln1/Cln2 contribution is the positive feedback that makes the burst
complete and coherent
PMID:19823668.
The deep-research synthesis reaches the same picture
[file:yeast/CLN2/CLN2-deep-research-falcon.md "SBF induces CLN1 and CLN2, after which Cln1/2–Cdc28 reinforces the transition by phosphorylating the SBF repressor Whi5. This creates positive feedback that sharpens late-G1 transcription and promotes commitment."].

Sic1 and Far1 phosphorylation - releasing Clb-Cdc28. DNA replication needs
Clb5/6-Cdc28, which is held inactive by Sic1, an inhibitor that does not act on
Cln-Cdc28
PMID:7954792.
Cln2-Cdk1 initiates the multisite phosphorylation of Sic1 through a
Cln2-specific docking motif, building the platform that Clb5-Cdk1 then extends
to the Cdc4 degrons
PMID:21993622
PMID:21993622
PMID:21993622.
Far1, the pheromone-induced CDK inhibitor that binds Cln-Cdc28 after Fus3
phosphorylation, is itself phosphorylated on Ser87 by Cdc28-Cln2, which
licenses its SCF(Cdc4)-dependent ubiquitylation in the nucleus
PMID:11080155
PMID:11080155
PMID:11080155
PMID:11080155.

Bud emergence and the cytoplasmic pool. Cln2 is the cyclin thought to
trigger bud emergence, and unlike Cln3 it is mostly cytoplasmic, concentrated at
sites of polarized growth
PMID:11509671
PMID:10611233
PMID:10611233
PMID:11792824.
A nuclear pool is also functional: an appended NES cripples some Cln2
functions, Cln2-GFP is seen in both compartments, and forced-localization
cassettes split the functions between compartments
PMID:10611233
PMID:11080155
PMID:11792824.
The steady-state cytoplasmic residence is an actively maintained, Cdc28-dependent
state: unphosphorylated Cln2 is nuclear, and Cdc28-dependent phosphorylation of
the C terminus drives nuclear exclusion
PMID:11509671
PMID:11509671.
The specific cytoplasmic substrates remain incompletely defined
PMID:10611233;
the deep-research file highlights the cyclin-replacement work of Ercan et al.
2021 (not cached) showing that a mitotic cyclin cannot substitute for Cln
function in polarization and budding.

Grr1-dependent instability. Cln2 protein is short-lived; the deep-research
file gives a half-life of about 10 min and maps the instability to PEST-rich
C-terminal sequences
[file:yeast/CLN2/CLN2-deep-research-falcon.md "The protein is highly unstable, with an estimated half-life of approximately 10 minutes, and older measurements generally place it below 15 minutes."].
Turnover is mainly through SCF(Grr1), reconstituted for the paralog Cln1
PMID:11080155
PMID:10213692,
with a context-dependent SCF(Cdc4) contribution that the deep-research file
flags as unresolved
[file:yeast/CLN2/CLN2-deep-research-falcon.md "SCF^Grr1-mediated ubiquitylation and proteasomal destruction are strongly established. Loss of SCF^Grr1 activity impairs normal Cln2 turnover and causes abnormal accumulation."].

Pheromone response and recovery. Beyond degrading Far1, Cln2-Cdc28 shuts the
mating MAPK pathway off by phosphorylating the scaffold Ste5 (and Ste20), which
is why cells past Start are refractory to pheromone
PMID:17289571
PMID:17289571.
POG1 and MSG5 promote recovery from pheromone arrest through CLN2
PMID:9927449.

2. Review decisions

34 GOA rows plus one proposed NEW row. Summary of calls:

ACCEPT (core)

• GO:0000082 G1/S transition of mitotic cell cycle (IBA). The IBD at node
  PTN000019791 (G1 cyclins) is sound; Cln2's Whi5/Sic1/Far1 biology is exactly
  the term definition (build-up of G1 CDK, transcription of G1 cyclins,
  positive feedback committing the cell to S phase). Cln2 is not in this row's
  WITH/FROM, but the sister members Cln3, Clb5 and Clb6 are.
• GO:0007089 traversing start control point (IEA, IGI x2, IMP). The
  classic CLN genetics (Hadwiger 1989, Richardson 1989, Cross 1990) and an ARBA
  mapping that happens to be specific and correct.
• GO:0000307 cyclin-dependent protein kinase holoenzyme complex (IBA, IDA,
  IEA, IPI). Directly demonstrated (Wittenberg 1990, Reynard 2000). The target
  appearing in its own IBA WITH/FROM is expected, not circular.
• GO:0005634 nucleus (IBA is_active_in, IDA x3). Two labs and three
  methods (NES effect, GFP, fractionation/forced localization) plus nuclear
  substrates (Whi5, Stb1, Sic1, Far1).
• GO:0005737 cytoplasm (IBA is_active_in, IDA x4, IMP). The point of
  divergence from the CLN3 review: for Cln3 the IBA cytoplasm row is
  over-stated (the ER pool is sequestered), but for Cln2 cytoplasmic activity
  is a documented, Cln2-specific property confirmed by forced-localization
  genetics, so the same node-level IBA is accepted here. The IMP row (Miller &
  Cross 2001) records that phospho-site, cdc28-4 and cyclin-box mutants shift
  Cln2 into the nucleus, i.e. the cytoplasm is the actively maintained default.

MODIFY

• GO:0016538 CDK regulator activity -> GO:0061575 CDK activator activity
  (IBA, IEA, IDA x3). Same treatment as CLN3: every experiment measures Cln2
  conferring kinase activity on Cdc28 (Deshaies & Kirschner 1995 in vitro;
  Tyers 1993 and Ho 1999 immunoprecipitates), the PAINT node contains no CDK
  inhibitors, and the parent is merely less informative. All five rows carry
  the same action so the validator's consistency check is satisfied. IBA
  propagation_review: TERM_SCOPING_PROBLEM / GRANULARITY_MISMATCH, node
  SUPPORTS_TRANSFER.
• GO:0005515 protein binding (IPI x6, all with Cdc28). Repository policy:
  generic protein binding is uninformative. The four high-throughput AP-MS
  rows (Gavin 2002, Ho 2002, Gavin 2006, Breitkreutz 2010) -> GO:0019901
  protein kinase binding. Reynard 2000 (Cln2-Cdc28 kinase assays) and
  Strickfaden 2007 (Cln2/CDK phosphorylation of Ste5) -> GO:0061575, since
  both measure Cln2 acting as the activating, substrate-directing subunit.
• GO:1902806 regulation of cell cycle G1/S phase transition -> GO:1900087
  positive regulation of G1/S transition of mitotic cell cycle (NAS from
  Complex Portal on Hadwiger 1989). The definition fits and the sign is
  positive (CLN2-1 advances G1/S); mirrors the CLN3 call.
• GO:2000045 regulation of G1/S transition of mitotic cell cycle ->
  GO:1900087 (IEA, InterPro2GO from the Cyclin_CLN family IPR014399).
  Correct but direction-less family mapping.

KEEP_AS_NON_CORE

• GO:0000321 re-entry into mitotic cell cycle after pheromone arrest
  (IGI x2 with POG1 and MSG5, Leza & Elion 1999). The participation test is
  met - Cln2-Cdc28 phosphorylates Far1 and Ste5/Ste20, which is the work of
  recovery - so the annotation is correct, but it is a context-specific
  downstream application of the same G1 CDK activity that executes Start,
  hence non-core. Abstract-only cache; the epistasis details are deferred to
  the SGD curator.

NEW

• GO:0045944 positive regulation of transcription by RNA polymerase II
  (involved_in, IDA, PMID:19823668). Added to close the gap between
  core_functions (which lists the term for the Whi5/Stb1 feedback function)
  and existing_annotations, where CLN2 carried nothing on the transcription
  branch. Participation: Cln2-Cdc28 is the kinase that phosphorylates Whi5 and
  Stb1 and dissociates Whi5 from SBF - it performs the step, it is not merely
  required for it. Comparator: SGD annotates the sister cyclin Cln3 to
  GO:0006357 (IGI/IMP) from the same Huang et al. 2009 paper, so this branch is
  one curators use for the Whi5 kinase; the CLN3 review modifies those rows to
  GO:0045944, and using the positive term here keeps the two reviews
  consistent. GO:0045944 is on a different branch from GO:0000082 /
  GO:0007089, so it is not redundant with them. The "Cln1/Cln2 not required
  for gene activation" caveat (Ho 1999) is acknowledged in the row: it concerns
  the initial switch-on, which Cln3 can achieve alone, not the feedback that
  completes it.

No REMOVE, MARK_AS_OVER_ANNOTATED or UNDECIDED calls

Every experimental row is consistent with the synthesized picture. The
abstract-only caches (Hadwiger 1989, Richardson 1989, Cross 1990, Wittenberg
1990, Tyers 1993, Deshaies & Kirschner 1995, Leza & Elion 1999, Reynard 2000,
Edgington & Futcher 2001, de Bruin 2004, and the four AP-MS surveys) each state
the relevant result in the abstract; full text is cached for Miller & Cross
2000/2001, Ho 1999, Blondel 2000, Huang 2009, Koivomagi 2011, and the
abstract-plus-introduction of Strickfaden 2007.

3. Core functions

1. Cdc28 activator that executes Start - MF GO:0061575; BP GO:0000082,
   GO:0007089, GO:0045944; CC nucleus; complex GO:0000307. The Whi5/Stb1
   positive-feedback arm.
2. Inhibitor-clearing and morphogenetic arm - the same MF/complex; BP
   GO:1900087; locations nucleus and cytoplasm. Sic1 priming, Far1 Ser87
   phosphorylation, Ste5 inhibition, and the cytoplasmic pool at sites of
   polarized growth.

Open questions recorded in suggested_questions: the identity of the
cytoplasmic substrates behind the Cln2-specific budding role; whether the
transient nuclear pool of hypophosphorylated Cln2 in newborn cells is the one
that acts on Whi5/Sic1; and the physiological split between SCF(Grr1) and
SCF(Cdc4) in Cln2 turnover.

4. Validation log

• Initial state (previous reviewer, interrupted before notes): valid with 2
  warnings - no annotation cited the Falcon deep-research file, and the
  core-function term GO:0045944 was absent from existing_annotations.
• Fixes: NEW GO:0045944 row added (IDA, Huang et al. 2009 full text, plus de
  Bruin 2004 abstract, Ho 1999 and the deep-research file); the deep-research
  file also cited on the cytoplasm IBA row; file: and PMID:15210110
  reference entries added with reference_review.
• Coverage check: all 34 GOA rows (term x evidence x reference) have a
  matching existing_annotations entry with complete review blocks; no
  placeholder text remains. Final tally: 19 ACCEPT, 13 MODIFY, 2
  KEEP_AS_NON_CORE, 1 NEW.
• Final state: 0 errors, 0 warnings.