CAF120 (systematic name YNL278W) is a large (1060-residue, ~118 kDa), non-essential Saccharomyces cerevisiae protein whose molecular function is essentially uncharacterized. It was originally identified as a "CCR4-associated factor" of 120 kDa (hence its name) and is reported to associate with the CCR4-NOT complex, the conserved machinery that combines a nuclear transcription-regulatory role with the cytoplasmic major mRNA deadenylase activity; however, CAF120 is not enumerated among the canonical CCR4-NOT core subunits and no gene-specific molecular activity has been demonstrated for it. Its only recognized folded domain is a divergent pleckstrin-homology (PH) domain near the N-terminus (a member, with its paralog SKG3, of the Skg3/CAF120-like PH subfamily); the remaining ~85% of the protein is largely intrinsically disordered and low-complexity. The protein localizes to the cytoplasm, nucleus, and the bud neck, and it relocalizes under DNA-damaging stress. It is a substrate of the cyclin-dependent kinase Cdk1 and is multiply phosphorylated on serine residues. CAF120 has a whole-genome-duplication paralog, SKG3 (YHR133C), which is likewise poorly characterized. In meiosis, CAF120 has been identified as a metaphase-I-enriched interactor of the MutLgamma (Mlh1-Mlh3)-Exo1 crossover machinery and acts as a regulator of meiotic crossing-over: its deletion reproducibly reduces crossover frequency, although the molecular mechanism by which it contributes to crossover formation is not yet established.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0004672 protein kinase activity | IBA GO_REF:0000033 | REMOVE | Summary: Phylogenetically-inferred (IBA) protein kinase activity propagated within PANTHER family PTN001969686. The reference members driving the inference are Arabidopsis MAP kinase kinase kinase loci (AT1G05100, AT3G50310, AT4G26890 in the WITH/FROM field), which are bona fide kinase-domain proteins. CAF120 has NO protein kinase domain: its only Pfam/ InterPro-recognized folded module is a PH domain (residues 75-204; PF00169/PF25381, IPR058155), with the rest of the protein intrinsically disordered. This is a family over-propagation in which a shared accessory PH module drew a non-kinase yeast protein into a plant-MAP3K-dominated cluster, causing it to inherit a catalytic activity it does not possess. Biologically indefensible. Reason: CAF120 lacks any protein kinase domain (UniProt/InterPro/Pfam show only a PH domain plus disordered regions); the IBA kinase activity is over-propagated from kinase-domain Arabidopsis MAP3K family members via a shared PH module and has no gene-specific support. OpenScientist independently refuted the assignment by domain and catalytic-motif analysis. Propagation Review Root cause: PROPAGATION BAD Failure modes: WRONG ORTHOLOG OR PARALOG FUNCTIONAL DIVERGENCE Sources checked: PANTHER:PTN001969686 · PANTHER family node driving the IBA transfer SUPPORTS SOURCE BUT NOT TARGET Family clustering appears PH-domain-driven; the kinase MF comes from the plant MAP3K members, not from CAF120, which has no kinase domain. AGI_LocusCode:AT1G05100 · Arabidopsis MAP kinase kinase kinase SUPPORTS SOURCE BUT NOT TARGET Bona fide kinase-domain protein; activity does not transfer to non-kinase CAF120. AGI_LocusCode:AT3G50310 · Arabidopsis MAP kinase kinase kinase (MKKK20) SUPPORTS SOURCE BUT NOT TARGET Kinase-domain reference member; not applicable to CAF120. AGI_LocusCode:AT4G26890 · Arabidopsis MAP kinase kinase kinase SUPPORTS SOURCE BUT NOT TARGET Kinase-domain reference member in the GOA WITH/FROM field; not applicable to PH-only CAF120. Supporting Evidence: file:yeast/CAF120/CAF120-goa.tsv UniProtKB P53836 CAF120 enables GO:0004672 protein kinase activity molecular_function ECO:0000318 IBA GO_REF:0000033 AGI_LocusCode:AT1G05100|AGI_LocusCode:AT3G50310|AGI_LocusCode:AT4G26890|PANTHER:PTN001969686 559292 Saccharomyces cerevisiae (strain ATCC 204508 / S288c) GO_Central CCR4-NOT transcriptional complex subunit CAF120 20260601 file:yeast/CAF120/CAF120-hypotheses/function-hypothesis-go-0004672/openscientist.md The seed hypothesis that CAF120 possesses **protein kinase activity (GO:0004672)** is **refuted**. |
| GO:0007165 signal transduction | IBA GO_REF:0000033 | REMOVE | Summary: Phylogenetically-inferred (IBA) signal transduction from the same PANTHER family (PTN001969686), with the same Arabidopsis MAP3K reference set (AT1G05100, AT2G32510, AT3G50310) in the WITH/FROM field. "Signal transduction" is the generic process that the kinase-domain plant members carry; it was inherited by CAF120 through the same PH-domain- driven mis-clustering that produced the erroneous kinase-activity call. There is no experimental or domain-based evidence that CAF120 acts in a signaling cascade. Reason: Over-propagated from kinase-containing plant MAP3K family members; CAF120 has no kinase domain and no evidence of participation in a signaling pathway. Removing alongside the linked kinase-activity IBA that generated it. Propagation Review Root cause: PROPAGATION BAD Failure modes: WRONG ORTHOLOG OR PARALOG FUNCTIONAL DIVERGENCE Sources checked: PANTHER:PTN001969686 · PANTHER family node driving the IBA transfer SUPPORTS SOURCE BUT NOT TARGET Same PH-domain-driven family cluster; "signal transduction" is the generic BP of the plant MAP3K members and does not apply to the non-kinase CAF120. AGI_LocusCode:AT1G05100 · Arabidopsis MAP kinase kinase kinase SUPPORTS SOURCE BUT NOT TARGET Signaling role is kinase-cascade-dependent; not transferable to CAF120. |
| GO:0005634 nucleus | IEA GO_REF:0000044 | KEEP AS NON CORE | Summary: Electronic annotation derived from the UniProtKB-SubCell mapping (SL-0191), which is in turn backed by the experimental subcellular-location record in UniProt (Nucleus; ECO:0000269|PubMed:14562095, the Huh 2003 GFP localization atlas). Consistent with the multi-compartment (cytoplasm/nucleus/bud neck) localization of the protein. Correct but broad, and not the core function of this dark gene. Reason: Evidence-backed subcellular location (via SubCell mapping from a GFP-atlas experimental record), but a general localization term rather than a molecular function; retain as non-core. |
| GO:0005737 cytoplasm | IEA GO_REF:0000044 | KEEP AS NON CORE | Summary: Electronic annotation from the UniProtKB-SubCell mapping (SL-0086), backed by the experimental UniProt subcellular-location record (Cytoplasm; ECO:0000269|PubMed:14562095). A correct but very general location. Reason: Correct high-level cytoplasmic localization derived from an experimentally-backed SubCell mapping; general and non-core. |
| GO:0005935 cellular bud neck | IEA GO_REF:0000044 | ACCEPT | Summary: Electronic annotation from the UniProtKB-SubCell mapping (SL-0029), backed by the experimental UniProt record (Bud neck; ECO:0000269|PubMed:14562095). This duplicates the more direct IDA bud-neck annotation below (PMID:25028499) and is subsumed by it. Reason: Correct bud-neck localization, consistent with (and subsumed by) the direct-assay bud-neck annotation (GO:0005935, IDA, PMID:25028499); accepted for consistency with that annotation. |
| GO:0003674 molecular_function | ND GO_REF:0000015 | ACCEPT | Summary: Root-level "molecular function unknown" placeholder (ND). This honestly reflects the dark-gene status of CAF120: despite the PH domain and CCR4-NOT association, no specific molecular activity (catalytic, binding, adapter, or scaffold) has been experimentally demonstrated, and GO/SGD do not assign it one. Appropriate to retain. Reason: No defensible molecular function can be assigned; the ND annotation correctly records that the molecular function is unknown. |
| GO:0008150 biological_process | ND GO_REF:0000015 | ACCEPT | Summary: Root-level "biological process unknown" placeholder (ND) reflecting the state of GOA at the time of annotation. Since then, a meiotic role has been reported (CAF120 is a metaphase-I-enriched MutLgamma-Exo1 interactor whose deletion reduces crossover frequency; PMID:31351878), which is captured as a proposed NEW annotation below (GO:0010520). The ND is retained as-is (it is a trusted GOA record and this review does not rewrite GOA ids), but it no longer reflects the best available knowledge of the gene's biological role. Reason: GOA-sourced ND placeholder correctly recording that GOA holds no specific biological process; complemented (not contradicted) by the proposed meiotic-crossover-regulation annotation below (GO:0010520; PMID:31351878). |
| GO:0005935 cellular bud neck | IDA PMID:25028499 Proteome-wide remodeling of protein location and function by... | ACCEPT | Summary: Direct-assay (IDA) localization of Caf120 to the cellular bud neck, from a high-resolution cellular-imaging study of proteome-wide, condition-dependent protein localization. The same study reports that Caf120 changes subcellular location under DNA-damaging conditions. This is the best-supported experimental annotation for the gene and is consistent with the UniProt bud-neck subcellular-location record (from the Huh 2003 GFP atlas). Reason: Well-supported direct-assay localization to the bud neck; the strongest experimental annotation available for this otherwise uncharacterized protein. |
| GO:0010520 regulation of reciprocal meiotic recombination | IMP PMID:31351878 Network Rewiring of Homologous Recombination Enzymes during ... | NEW | Summary: Proposed new annotation. A proteomic and genetic dissection of the recombination- intermediate-processing enzymes found that CAF120 is a meiosis-specific (metaphase-I- enriched) interactor of the MutLgamma (Mlh1-Mlh3)-Exo1 crossover machinery, and that deletion of CAF120 reproducibly reduces meiotic crossover frequency (genetic distance at the CEN8-THR1 interval), placing it among the identified regulators of crossing-over. This is a specific, experimentally-supported biological role for an otherwise dark gene. The general "regulation of reciprocal meiotic recombination" term (GO:0010520) is chosen rather than a directional (positive/negative) child because the authors explicitly flag the molecular mechanism as uncharacterized ("interesting candidates for further functional analyses"). Reason: Not present in current GOA, but supported by direct genetic evidence (reproducible reduction in crossover frequency in caf120-null meiosis) plus a meiosis-specific physical association with the MutLgamma-Exo1 crossover machinery (PMID:31351878). Conservatively annotated to the general regulation term because the direction/mechanism is not resolved. Supporting Evidence: PMID:31351878 Rtk1, Caf120, and Chd1 as regulators of crossing-over PMID:31351878 chd1Δ, rtk1Δ, and caf120Δ |
Loading supporting content…
Download this section (compressed HTML)Q: Does the divergent PH domain of CAF120 bind a phosphoinositide, a protein partner, or neither, and does this binding determine its localization to the bud neck?
Q: Is CAF120 a stable stoichiometric subunit of the CCR4-NOT complex or a transient associated factor, and does it contribute measurably to CCR4-NOT mRNA-deadenylase or transcriptional activity?
Q: Are CAF120 and its whole-genome-duplication paralog SKG3 functionally redundant, and does a double deletion reveal a phenotype masked in the single mutants?
Q: By what molecular mechanism does CAF120 promote meiotic crossover formation — does it act on recruitment, stability, or activity of the MutLgamma-Exo1 machinery at recombination intermediates, and is this role connected to its CCR4-NOT association or PH domain?
Suggested experts: Joao Matos
Experiment: Express and purify the recombinant CAF120 PH domain (and full-length protein) and assay phosphoinositide binding by lipid-strip and liposome-flotation assays; in parallel, perform affinity-purification / proximity-labeling (e.g. BioID or TurboID) from yeast to identify protein interactors, and test whether mutating the predicted ligand-binding surface disrupts bud-neck localization of a GFP fusion.
Hypothesis: The CAF120 PH domain targets the protein to a specific membrane/lipid or protein partner at the bud neck.
Type: biochemical and cell-biological interaction mapping
Experiment: Perform quantitative affinity purification of tagged core CCR4-NOT subunits (e.g. NOT1, CCR4) with stoichiometry measurement to determine whether CAF120 co-purifies stably; and measure global and reporter-specific mRNA poly(A) tail lengths / deadenylation kinetics and reporter transcription in caf120-null versus wild-type cells.
Hypothesis: CAF120 is only transiently/substoichiometrically associated with CCR4-NOT and does not materially affect complex activity.
Type: quantitative interaction proteomics and mRNA-turnover assay
Experiment: Construct caf120 skg3 single and double deletion strains and profile them for growth, budding/cytokinesis morphology, stress (including DNA-damage) sensitivity, and amino-acid metabolism, looking for synthetic phenotypes absent from the single mutants.
Hypothesis: CAF120 and SKG3 are functionally redundant paralogs.
Type: genetic redundancy / synthetic phenotype analysis
Experiment: Dissect the CAF120-MutLgamma-Exo1 relationship in meiosis: map the meiosis-specific interaction interface (co-IP / crosslinking-MS of tagged Caf120 with Mlh1/Mlh3/Exo1), test whether caf120Δ alters MutLgamma-Exo1 chromatin recruitment or joint-molecule resolution (physical recombination assays, ChIP), and determine epistasis of caf120Δ with mlh3Δ/exo1Δ for crossover frequency and spore viability.
Hypothesis: CAF120 promotes meiotic crossovers by modulating the MutLgamma-Exo1 crossover machinery.
Type: meiotic recombination mechanism / genetic-interaction analysis
What is not known — curated, literature-grounded statements of the open unknowns (the inverse of core functions).
Gap: The molecular function of CAF120 is unknown. No catalytic, nucleic-acid-binding, adapter, or scaffolding activity has been demonstrated, and the ligand and role of its divergent N-terminal PH domain (Skg3/CAF120-like PH subfamily) are undetermined — it is not known whether this PH domain binds phosphoinositides (as canonical PH domains do), a protein partner, or acts as a non-binding structural module.
OPEN BIOLOGY MF_DARK
What is known: It is firmly established that CAF120 is a 1060-residue, ~118 kDa non-essential protein whose only recognized folded domain is a PH domain (residues 75-204; Pfam PF00169/PF25381, InterPro IPR058155 Skg3/CAF120-like PH), with the remainder largely intrinsically disordered and low-complexity. It localizes to the cytoplasm, nucleus, and bud neck, is a Cdk1 phosphosubstrate, is expressed at ~1380 molecules/cell in log phase, and has a whole-genome-duplication paralog, SKG3. GO and SGD record its molecular function as unknown (ND).
Significance: CAF120 is a conserved-fold, phospho-regulated protein that sits at the interface of two heavily-studied systems (the CCR4-NOT machinery and the bud-neck cytokinesis apparatus) yet has no assigned mechanism. Determining what its PH domain binds would likely reveal both its molecular activity and where it acts, and would clarify whether the electronic kinase/ signaling annotations it currently inherits are truly spurious.
What would resolve it: Biochemical characterization of the recombinant PH domain (phosphoinositide-strip / liposome binding, and pulldown/proximity proteomics for protein ligands); structural determination of the PH domain to assess whether its ligand-binding pocket is competent; and identification of interaction partners of the full-length protein.
Provenance (the field's own admissions):
Gap: It is unresolved whether CAF120 is a bona fide, stable core subunit of the CCR4-NOT complex or merely a transient / substoichiometric associated factor, and — if it does associate — whether it makes any functional contribution to CCR4-NOT-dependent mRNA deadenylation/ turnover or transcriptional regulation.
OPEN BIOLOGYCURATION BP_DARK
What is known: CAF120 is named "CCR4-associated factor 120 kDa", and UniProt lists it as a subunit of the 1.0 MDa CCR4-NOT core complex interacting with NOT1, attributing this to the 2001 complex- purification study. However, that study's own report of the 1.0 MDa complex composition names CCR4, CAF1, NOT1-5 and the two newly identified proteins CAF40 and CAF130, and the GO ontology's definition of the Saccharomyces "CCR4-NOT core complex" (GO:0030015) enumerates Ccr4p, Caf1p, Caf40p, Caf130p and Not1-5p — CAF120 is absent from both lists. Correspondingly, GOA carries no CCR4-NOT complex (GO:0030014/GO:0030015) annotation for CAF120, and SGD/GO record its biological process as unknown.
Significance: Whether CAF120 is a real CCR4-NOT subunit or an associated factor determines whether it can legitimately inherit any CCR4-NOT function and shapes how this dark gene should be curated. The discrepancy between the gene's name / UniProt subunit statement and the canonical complex composition (in the primary paper and in GO) is exactly the kind of unresolved association that should not be propagated as fact without confirmation.
What would resolve it: Quantitative interaction proteomics (e.g. affinity purification with stoichiometry determination, or reciprocal tagging of core subunits) to test stable versus transient CCR4-NOT association; and functional assays (mRNA deadenylation / poly(A) tail length, reporter-gene transcription) in caf120-null versus wild-type cells to test any contribution to CCR4-NOT activity.
Provenance (the field's own admissions):
Gap: The molecular mechanism by which CAF120 promotes meiotic crossing-over is unknown: it is not established how (or whether directly) CAF120 acts on the MutLgamma-Exo1 crossover machinery, what its meiosis-specific binding is to within that network, or why loss of CAF120 lowers crossover frequency.
OPEN BIOLOGY RESIDUAL_SUBGAP
What is known: It is established (by affinity-proteomics plus a genetic crossover assay) that CAF120 is a metaphase-I-enriched interactor of the MutLgamma (Mlh1-Mlh3)-Exo1 crossover machinery and that caf120-null meiosis shows a reproducible reduction in crossover frequency, so CAF120 is a bona fide regulator of crossing-over. However, the study characterized the mechanism only for a co-identified factor (Chd1, via its DNA-binding activity) and explicitly left CAF120 (and Rtk1) as candidates for further functional analysis; CAF120 has no catalytic domain, so any contribution is presumed non-catalytic/regulatory but is otherwise undefined.
Significance: This is the one sharp mechanistic hole in an otherwise newly-illuminated function: CAF120 is now known to matter for crossover formation, but the causal step it controls is undefined. Closing it would convert a "regulator of crossing-over" statement into a mechanistic model and would clarify whether CAF120's role is exerted through, or independently of, its reported CCR4-NOT / PH-domain-mediated interactions.
What would resolve it: Mechanistic dissection of the CAF120-MutLgamma-Exo1 relationship: mapping the meiosis- specific interaction interface, testing whether CAF120 affects MutLgamma-Exo1 recruitment, stability, or nuclease activity at recombination intermediates, and epistasis of caf120 with mlh1/mlh3/exo1 in crossover assays.
Provenance (the field's own admissions):
Gap: The biological role and physiological significance of CAF120's bud-neck localization and of its DNA-damage-induced relocalization are unknown, as is the basis of its subtle deletion phenotypes (reduced competitive fitness, altered free amino-acid profile, stress sensitivity) and its functional relationship to (or redundancy with) its paralog SKG3.
OPEN BIOLOGY BP_DARK
What is known: CAF120 is experimentally localized to the bud neck (IDA) in addition to the cytoplasm and nucleus, and it changes subcellular location under DNA-damaging conditions. Its deletion is viable, with only diffuse high-throughput phenotypes reported, and it has a WGD paralog, SKG3, that is itself uncharacterized. No process-level mechanism connects these observations.
Significance: Bud-neck enrichment and stress-triggered relocalization hint at a cytokinesis- and/or stress-response-related role, but without a mechanism these remain descriptive. Testing whether CAF120 and SKG3 are functionally redundant is essential, because paralog redundancy is a common reason a single-gene deletion shows only subtle phenotypes and would explain the gene's apparent dispensability.
What would resolve it: Phenotypic analysis of caf120 skg3 double mutants (and stress/DNA-damage challenges) for synthetic defects in growth, budding/cytokinesis and stress survival; live-cell imaging of the relocalization dynamics; and genetic-interaction / co-expression profiling to place the gene in a pathway.
Provenance (the field's own admissions):
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)