adg2 encodes a serine/threonine-rich, heavily N-glycosylated secretory-pathway glycoprotein of the fission yeast Schizosaccharomyces pombe. The protein carries an N-terminal signal peptide, a long internally disordered Ser/Thr-rich body (~50% Ser+Thr, 17 predicted N-glycosylation sites), and a C-terminal hydrophobic tail with the features of a glycosylphosphatidylinositol (GPI) anchor-addition signal, and it is one of the small set of predicted GPI-anchored proteins in the S. pombe genome. It has no recognizable catalytic domain. It transits the endoplasmic reticulum and is assigned to the fungal cell surface, consistent with the architecture of cell-wall/cell-surface glycoproteins. Its transcription is periodic and Ace2-dependent, peaking around cytokinesis, and it is induced during meiosis (hence the alias mug46), but no molecular activity or direct biological role has been demonstrated experimentally for the protein itself.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0005576 extracellular region | IEA GO_REF:0000044 | KEEP AS NON CORE | Summary: Broad extracellular localization mapped by UniProt from the Secreted subcellular-location keyword (SL-0243). Defensible for a signal-peptide-bearing, GPI/secretion-predicted glycoprotein, but it is a coarse location term inferred electronically, not a core function. Reason: Consistent with the protein's secretory architecture (N-terminal signal peptide, no internal TM domains) and its inclusion among predicted S. pombe GPI proteins (PMID:12845604). Retained as a broad, non-core cellular-component annotation; the more specific cell surface term (GO:0009986) is preferred where a surface location is intended. |
| GO:0005783 endoplasmic reticulum | IEA GO_REF:0000044 | KEEP AS NON CORE | Summary: ER localization, mapped electronically by UniProt (SL-0095) but ultimately grounded in an experimental YFP-tagging localization screen (PMID:16823372). The ER is the expected transit compartment for a secretory-pathway glycoprotein. Reason: Experimentally supported localization (PMID:16823372) that is fully consistent with a secreted/surface glycoprotein transiting the ER. Retained as a legitimate, non-core cellular component; it reflects where the protein is processed, not its molecular function. Supporting Evidence: PMID:16823372 we determined the localization of 4,431 proteins, corresponding to approximately 90% of the fission yeast proteome, by tagging each ORF with the yellow fluorescent protein |
| GO:0003674 molecular_function | ND GO_REF:0000015 | ACCEPT | Summary: Root molecular_function with ND (No biological Data). This is the correct, honest representation: adg2 has no recognizable catalytic domain and no experimentally determined molecular activity. The InterPro/PANTHER "GPI-anchor adhesion regulation" family name is a computational label, not evidence of adhesin activity. Reason: The molecular function of Adg2 is genuinely unknown; ND on the MF root faithfully captures this. No informative MF term (including adhesion or binding terms) is supportable from current evidence, so no replacement is proposed. See knowledge_gaps. |
| GO:0008150 biological_process | ND GO_REF:0000015 | ACCEPT | Summary: Root biological_process with ND. adg2 is co-regulated with cell-separation genes (Ace2/Eng1 cluster, PMID:15966770) and induced during meiosis (mug46), but no biological process has been experimentally demonstrated for the protein itself, so the ND root is appropriate. Reason: The cell-separation association is a cluster-level co-regulation inference, not a demonstrated adg2 function; there is no reported adg2 loss-of-function phenotype. Retaining ND on the BP root is the conservative, defensible choice. The candidate process is recorded in knowledge_gaps rather than asserted as an annotation. |
| GO:0009986 cell surface | TAS PMID:12845604 Genome-wide identification of fungal GPI proteins. | KEEP AS NON CORE | Summary: Cell surface localization curated by PomBase (TAS) from the genome-wide fungal GPI-protein prediction (PMID:12845604), consistent with the Ser/Thr-rich, heavily glycosylated, GPI-signal-bearing architecture and PomBase's "conserved fungal cell surface" product description. Reason: Best-supported surface localization for a predicted GPI-anchored glycoprotein; a legitimate cellular-component annotation. It is a location, not a molecular function, and GPI anchorage is predicted rather than experimentally confirmed, so it is retained as non-core. Supporting Evidence: PMID:12845604 only 33 GPI candidates were identified |
Loading supporting contentβ¦
Download this section (compressed HTML)Q: Is Adg2 an adhesin/flocculin, a passive structural cell-surface glycoprotein, or does it have another activity β and does the IPR052479 "GPI-anchor adhesion regulation" family designation reflect a real adhesion-related function?
Suggested experts: Frans M. Klis, Gordon Chua
Q: Does adg2 deletion produce a cell-separation, cell-wall-integrity, flocculation, or sporulation phenotype, and is any such role redundant with its paralog adg1?
Suggested experts: Janet Leatherwood
Experiment: Endogenously tag adg2 at the C-terminus internal to the predicted GPI signal (or N-terminally after the signal peptide), image steady-state localization by fluorescence microscopy, and test anchor type biochemically (PI-PLC release, and alkali/glucanase extraction to distinguish plasma-membrane GPI from covalent cell-wall linkage).
Hypothesis: Adg2 is a GPI-anchored, cell-wall-associated surface glycoprotein rather than a soluble secreted protein.
Type: protein localization and biochemical anchor characterization
Experiment: Construct adg2Ξ and adg1Ξ adg2Ξ strains and assay cell-separation defects (cell chaining, calcofluor septum staining), cell-wall-integrity stress sensitivity, flocculation/aggregation, and sporulation efficiency, comparing to eng1Ξ/agn1Ξ separation phenotypes.
Hypothesis: Adg2 contributes directly to daughter-cell separation and/or cell-surface adhesion, beyond its co-regulation with the Ace2/Eng1 cluster.
Type: gene deletion and phenotypic analysis
What is not known β curated, literature-grounded statements of the open unknowns (the inverse of core functions).
Gap: The molecular function of Adg2 is undetermined. No catalytic activity, ligand, adhesion target, or binding partner has been identified, and the protein has no recognizable enzymatic domain.
OPEN BIOLOGY MF_DARK
What is known: Sequence and family evidence firmly establish that Adg2 is a secretory-pathway serine/threonine-rich glycoprotein (N-terminal signal peptide; ~50% Ser+Thr; 17 predicted N-glycosylation sites; long disordered stalk) with a C-terminal GPI-anchor-addition-signal architecture, placing it in InterPro IPR052479 ("GPI-anchor_Adhesion_Reg") and PANTHER PTHR35185. What that family name implies mechanistically β whether Adg2 is a genuine adhesin, a structural cell-wall/surface component, or something else β is not experimentally supported.
Significance: Adg2 is a conserved fungal cell-surface protein of unknown activity (UniProt PE 3, inferred from homology). Assigning its molecular role would convert a coarse "cell surface glycoprotein" into a defined surface function and clarify what the IPR052479 "GPI-anchor adhesion regulation" family actually does.
What would resolve it: Biochemical/cell-biological characterization: confirm GPI anchorage and cell-wall vs plasma-membrane residence, test for adhesion/flocculation or aggregation activity, and search for interaction partners of the mature glycoprotein.
Provenance (the field's own admissions):
Gap: The direct biological role of Adg2 is not experimentally established. Its assignment to cell separation rests on transcriptional co-regulation with characterized separation genes, not on any measured adg2 loss-of-function phenotype.
OPEN BIOLOGY BP_DARK
What is known: adg2 is a member of the nine-gene, Ace2-regulated, cell-cycle-periodic "Eng1 cluster" that peaks around cytokinesis and is described as involved in cell separation, where it is grouped with the glucan hydrolases eng1, agn1 and adg3; it is also induced during meiosis (alias mug46) and modulated within the Adn2/Adn3 flocculation network. All of this is regulatory/expression evidence about when adg2 is transcribed, not about what the protein does.
Significance: Distinguishing a genuine, direct contribution to daughter-cell separation (or to meiotic surface remodeling) from mere co-regulation is required before a specific biological-process term could be annotated for adg2 itself.
What would resolve it: Phenotypic analysis of an adg2 deletion (and adg1/adg2 double deletion) for cell-separation, cell-wall-integrity, flocculation and sporulation defects, with localization of tagged Adg2 relative to the division septum.
Provenance (the field's own admissions):
Gap: Whether Adg2 is actually GPI-anchored, and whether it resides at the plasma membrane or is cross-linked into the cell wall, has not been experimentally verified; the C-terminal omega (GPI-attachment) site has not been mapped.
OPEN BIOLOGY CC_DARK
What is known: Adg2 has a C-terminal hydrophobic tail consistent with a GPI-anchor-addition signal and was computationally predicted to be one of 33 S. pombe GPI proteins (PMID:12845604); the only experimental localization reported is endoplasmic reticulum from a genome-scale YFP screen (PMID:16823372), a transit compartment rather than a mature surface/wall location.
Significance: GPI anchorage versus soluble secretion, and plasma-membrane versus covalent cell-wall residence, change how the surface annotation should be interpreted and constrain the protein's possible function.
What would resolve it: Direct biochemistry (e.g. PI-PLC sensitivity, GPI-anchor mapping, alkali/glucanase extraction of the wall fraction) and steady-state surface imaging of endogenously tagged Adg2.
Provenance (the field's own admissions):
Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)