res2 (P41412, SPAC22F3.09c; synonym pct1) - curation notes

Working notes for the GO annotation review of Schizosaccharomyces pombe res2.
Companion file to res2-ai-review.yaml. Deep research: res2-deep-research-falcon.md
(Edison/falcon synthesis, generated for this review). Sibling reviews written in
parallel: genes/SCHPO/cdc10, genes/SCHPO/res1; finished comparators
genes/yeast/MBP1, genes/yeast/SWI4.

1. Identity check

UniProt P41412, "Cell division cycle-related protein res2/pct1", 657 aa. Domain
architecture from the UniProt record: HTH APSES-type domain 8-115 with the H-T-H
motif at 39-60, two ankyrin repeats (247-276, 368-397), a disordered basic/low
complexity region at 113-192.
Independently cloned twice in 1994: as res2+ (multicopy suppressor of a res1 null;
Miyamoto, Tanaka and Okayama) and as pct1+ (a p85cdc10 partner; Zhu, Takeda,
Nasmyth and Jones). PMID:7926774 PMID:7926774
Name trap: a second, unrelated S. pombe protein is also called Pct1, the
303-residue RNA 5'-triphosphatase of the mRNA capping machinery. The UniProt
SUBUNIT line on P41412 ("Interacts with cdc10 and cdk9", PMID:12475973) is the
capping-enzyme Pct1, not Res2; the deep-research file flags this explicitly and
the review does not use it. No GOA row on P41412 derives from that literature.

2. Biological overview

Res2 is one of the two APSES-domain DNA-binding subunits (with its paralog Res1) of
the fission yeast MBF/DSC1 complex, the G1/S (Start) transcription factor built
around the shared Swi6-like subunit Cdc10. The Res2-Cdc10 heteromer binds MCB
elements but not SCB or E2F sites PMID:7926774, and the DNA-binding
activity of the complex maps to the Res2 N-terminus PMID:24006488.
Res1 and Res2 together form the DNA-binding unit of the vegetative complex
PMID:24006488, and
Res2 is retained on Cdc10 even when the complex leaves chromatin after DNA damage
PMID:24006488.

Two parallel Start systems. res2+ is largely redundant with res1+ in the mitotic
cycle but is the dominant subunit in meiosis PMID:8168485
PMID:8168485.
res2+ is nitrogen-starvation/conjugation induced and depends on Cdc10 for activity
PMID:8168485; the pct1 deletion is viable with a
severe meiotic defect PMID:7926774. In premeiotic S phase a DSC1-like complex
drives the MCB genes rec8, rec11, cdc18 and cdc22 and requires res2 but not res1
PMID:14648198
PMID:14648198. The conjugation-specific cig2 (cyc17)
transcript is res2-dependent PMID:7909513.

Coactivator and complex. Rep2 is the activator subunit recruited by the Res2-Cdc10
DNA-binding unit PMID:7588609. Cdc10 purification recovers Res1 and Res2 as the top hits
PMID:21132016.

Activator and repressor. The Baum/Nurse work first showed that Res2 is needed to
shut S-phase transcription off in G2 PMID:9303312 PMID:9303312. The mechanism is docking of negative-feedback regulators on Res2: the
cyclin Cig2 PMID:11781565 and the Nrm1/Yox1
corepressor pair PMID:30635289 PMID:21132016. At the cnp1 (CENP-A) promoter, Res2 occupancy and
the MCB1 box are what confine transcription to G1 PMID:30635289
PMID:30635289. The complex is promoter-bound
throughout the cycle, so the switch is a change of state of DNA-bound MBF rather
than loss of binding PMID:21132016.

3. Decision table

37 GOA rows reviewed. Counts: ACCEPT 22, MODIFY 13, KEEP_AS_NON_CORE 2, REMOVE 1.

GO term	Evidence / ref	Action	One-line reason
GO:0000082 G1/S transition of mitotic cell cycle	IBA GO_REF:0000033	ACCEPT	Res2-Cdc10 is one of the two parallel Start systems; node placement sound.
GO:0000122 negative regulation of transcription by Pol II	EXP PMID:30635289	ACCEPT	Res2 on cnp1 promoter; Nrm1/Yox1 repress cnp1 through Res2 (double mutants = res2 alone).
GO:0000122 negative regulation of transcription by Pol II	IMP PMID:9303312	ACCEPT	res2p required to switch off S-phase transcription in G2.
GO:0000785 chromatin	IDA PMID:17936710	ACCEPT	Abstract-only (Ctp1 paper); claim correct for the gene, curator's full-text read not second-guessed.
GO:0000785 chromatin	IDA PMID:24006488	ACCEPT	Res2-HA ChIP on cdc22/cdc18 promoters.
GO:0000978 Pol II cis-regulatory region seq-specific DNA binding	HDA PMID:40015273	ACCEPT	TF ChIP-seq atlas; agrees with targeted MCB-binding evidence.
GO:0000978	IBA GO_REF:0000033	ACCEPT	APSES node; Res2 in own WITH list is expected (own IDA seeds the IBD).
GO:0000978	IDA PMID:30635289	ACCEPT	ChIP at cnp1 MCB1 plus box mutagenesis abolishing res2-dependent effect.
GO:0000978	IDA PMID:7926774	ACCEPT	p73pct1/p85cdc10 binds MCB, not SCB/E2F; contributes_to appropriate.
GO:0001227 DNA-binding transcription repressor activity	EXP PMID:9303312	ACCEPT	Res2 is the DNA-bound subunit through which Cig2 and Nrm1/Yox1 repress.
GO:0001228 DNA-binding transcription activator activity	EXP PMID:14648198	ACCEPT	res2 (not res1) required for meiotic MCB-driven transcription.
GO:0001228	IBA GO_REF:0000033	ACCEPT	All donors are G1/S activator DNA-binding subunits.
GO:0001228	IEA GO_REF:0000117	ACCEPT	ARBA reproduces the experimental MF.
GO:0001228	IMP PMID:7909513	ACCEPT	Conjugation-induced cig2 transcript lost in res2 cells.
GO:0003677 DNA binding	IEA GO_REF:0000002	ACCEPT	Generic but true: APSES domain carries MBF DNA-binding activity.
GO:0005515 protein binding (Cig2, PomBase)	IPI PMID:11781565	MODIFY -> GO:0030332 cyclin binding	Abstract attributes direct cyclin contact to Res2.
GO:0005515 protein binding (Cdc10, PomBase)	IPI PMID:11781565	MODIFY -> GO:0046982 protein heterodimerization activity	Obligate Res2-Cdc10 unit of MBF.
GO:0005515 protein binding (Res1, PomBase)	IPI PMID:11781565	MODIFY -> GO:0046982	Res1-Res2 heterodimeric DNA-binding domain.
GO:0005515 protein binding (Cdc10, IntAct P01129)	IPI PMID:11781565	MODIFY -> GO:0046982	IntAct duplicate of the row above.
GO:0005515 protein binding (Res1, IntAct P33520)	IPI PMID:11781565	MODIFY -> GO:0046982	IntAct duplicate.
GO:0005515 protein binding (Cig2, IntAct P36630)	IPI PMID:11781565	MODIFY -> GO:0030332	IntAct duplicate.
GO:0005515 protein binding (Nrm1)	IPI PMID:18682565	MODIFY -> GO:0001222 transcription corepressor binding	Nrm1 is the MBF corepressor docking on Res2.
GO:0005515 protein binding (Cdc10)	IPI PMID:21132016	MODIFY -> GO:0046982	Cdc10 IP-MS recovers Res2; heterodimeric unit.
GO:0005515 protein binding (Rep2)	IPI PMID:7588609	MODIFY -> GO:0001223 transcription coactivator binding	Rep2 is the activator subunit recruited by Res2-Cdc10.
GO:0005515 protein binding (Cdc10)	IPI PMID:7588609	MODIFY -> GO:0046982	Same heterodimer.
GO:0005515 protein binding (Cdc10)	IPI PMID:7926774	MODIFY -> GO:0046982	Discovery of the p73pct1/p85cdc10 heteromer.
GO:0005634 nucleus	HDA PMID:16823372	ACCEPT	Site of the core function (promoter chromatin).
GO:0005737 cytoplasm	HDA PMID:16823372	KEEP_AS_NON_CORE	Single ORFeome YFP observation from an over-expressed construct.
GO:0005737 cytoplasm	IBA GO_REF:0000033	REMOVE	Swi6-seeded is_active_in propagation; Res2 has no cytoplasmic activity and stays promoter-bound.
GO:0005829 cytosol	HDA PMID:16823372	KEEP_AS_NON_CORE	Same dataset as the cytoplasm HDA row.
GO:0006357 regulation of transcription by Pol II	IDA PMID:8168485	MODIFY -> GO:0045944 positive regulation of transcription by Pol II	Founding paper establishes a positive Start function; term too general.
GO:0030907 MBF transcription complex	EXP PMID:9303312	ACCEPT	DSC1 bandshift requires res2p.
GO:0030907	IBA GO_REF:0000033	ACCEPT	Restates Res2's own experimental complex membership.
GO:0030907	IDA PMID:24006488	ACCEPT	Co-IP with Cdc10-HA; released from chromatin as a complex.
GO:0045944 positive regulation of transcription by Pol II	IBA GO_REF:0000033	ACCEPT	Family-wide G1/S activator function.
GO:0045944	IMP PMID:7909513	ACCEPT	Loss of cig2 induction in res2 cells.
GO:0051445 regulation of meiotic cell cycle	IMP PMID:7926774	MODIFY -> GO:0051446 positive regulation of meiotic cell cycle	Direction is clearly positive (loss blocks premeiotic S and meiotic progression).
GO:0090575 Pol II transcription regulator complex	IEA GO_REF:0000117	ACCEPT	Correct parent of GO:0030907.

4. Notable curation decisions

Both activator and repressor MFs kept as core. Unlike Res1, Res2 is the subunit
on which the negative-feedback regulators dock, and its loss derepresses most of
the regulon in G2. The 2019 paper states the dual nature outright PMID:30635289. GO:0001227 (repressor) and GO:0001228 (activator) are
therefore both accepted and each anchors its own core function. Whether GO:0001227
overstates an intrinsic repressor activity (versus a docking function) is left as
a suggested question.
All eleven protein binding IPI rows MODIFIED, never removed. Each partner has
a well-defined role: Cdc10 (x5, PomBase and IntAct duplicates) and Res1 (x2) ->
GO:0046982 protein heterodimerization activity, mirroring the treatment of
Mbp1-Swi6, Swi4-Swi6 and Res1-Cdc10 in the sibling reviews; Cig2 (x2) -> GO:0030332
cyclin binding, since the abstract attributes the direct cyclin contact to Res2;
Nrm1 -> GO:0001222 transcription corepressor binding PMID:18682565; Rep2 -> GO:0001223 transcription coactivator
binding PMID:7588609.
The single REMOVE is the Swi6-seeded is_active_in cytoplasm IBA. The PAINT
node PTN000917496 groups the DNA-binding subunits with the shared regulatory
subunits, and the sole donor for cytoplasm is budding yeast Swi6 (SGD:S000004172),
whose regulated nucleocytoplasmic shuttling is a Swi6-specific property. Res2 acts
on promoter chromatin, stays promoter-bound through the cycle, and no cytoplasmic
function has been described; the same call was made for Res1, Cdc10, Swi4 and
Mbp1. The HDA cytoplasm/cytosol rows from the ORFeome screen PMID:16823372 are kept as non-core rather than removed, because they are genuine
observations, if from over-expressed plasmid constructs.
Two generality MODIFYs. GO:0006357 (IDA, founding paper) -> GO:0045944, and
GO:0051445 -> GO:0051446, both because the direction of the effect is unambiguous
in the cited evidence. No process term for meiosis itself was proposed: Res2 does
the work as a transcription factor, so a regulation term is the right level.
Deference on abstract-only rows. Eleven of the fourteen cited papers are
abstract-only in the cache. The GO:0000785 chromatin IDA from the Ctp1 paper
PMID:17936710 does
not mention Res2 in the abstract; it is ACCEPTed with deference to the PomBase
curator, and the same term is independently supported by full-text ChIP data
PMID:24006488. Likewise the HDA rows from PMID:16823372 and
PMID:40015273 PMID:40015273 rest on datasets in which Res2 is not named in the abstract.
IBA rows with Res2 in its own WITH list (GO:0000978, GO:0001228, GO:0030907,
GO:0045944) are accepted as expected: Res2's own experimental rows seeded the IBD.
Not marked circular.
No NEW terms. The meiotic role is fully represented by GO:0051446 plus the
activator MF; a meiosis-specific transcription term was not added (comparator
check: budding yeast Mbp1/Swi4 do not carry meiotic process terms, and the res1
review does not either).

5. Core functions (as written in the YAML)

MCB-binding activator subunit of MBF at the mitotic Start (GO:0001228; contributes
to GO:0000978; GO:0045944, GO:0000082; chromatin/nucleus; GO:0030907).
Docking subunit through which MBF is repressed outside G1 (GO:0001227; GO:0000122;
chromatin/nucleus; GO:0030907).
Res2-Cdc10 as the meiotic Start complex (GO:0001228; GO:0045944, GO:0051446).

GO:0051446 (core function 3) is not itself among the existing_annotations; it is the
proposed replacement for the GO:0051445 IMP row. just validate SCHPO res2 passes
with no validator warnings.

6. Open questions

Is the G2 repression by Res2 purely a docking function for Nrm1/Yox1 and Cig2, or
does Res2 make a promoter-specific contribution of its own? Genome-wide, most MBF
targets are derepressed in res2 cells but yox1, cig2 and mik1 are not (deep-research
summary of Eshaghi et al. 2011, not cached), which hints at promoter-specific
behaviour.
Which meiotic outputs (rec8, rec11, cdc18, cdc22, cig2 poly(A)- transcript, and the
older mat1/mei3 link) are direct Res2 targets by ChIP in meiotic cells?
Should the PAINT node be split so that DNA-binding subunit terms (MCB binding,
chromatin) and regulatory-subunit terms (Swi6 cytoplasmic shuttling) propagate to
the right clades? This would remove the cytoplasm IBA at source for Res1, Res2,
Swi4 and Mbp1.
The cdc10-129 rescue but not cdc10-null rescue by pct1+ overexpression
PMID:7926774 shows the heterodimer, not Res2 alone,
is the functional unit; whether Res2 has any Cdc10-independent DNA-bound role
remains untested in vivo.

7. Provenance

Papers with full text cached: PMID:18682565 (abstract + discussion), PMID:21132016,
PMID:24006488, PMID:30635289. All others abstract-only.
Deep-research file quoted only for summaries of papers not in the cache (Zhu 1997,
Eshaghi 2011, Dutta 2008); those claims were cross-checked against the cached primary
abstracts before use.
Review completed 2026-09-26/27; notes written after the YAML.