GPM3 (systematic name YOL056W) is one of three phosphoglycerate-mutase-family paralogs in Saccharomyces cerevisiae, alongside the sole functional glycolytic mutase GPM1 and the second homolog GPM2. It is a cytoplasmic protein of the histidine-phosphatase superfamily (cofactor/2,3-bisphosphoglycerate-dependent phosphoglycerate mutase, dPGM, subfamily) and retains the family's catalytic histidine residues. Despite this, GPM3 has no detectable phosphoglycerate mutase activity even when strongly overexpressed, and its deletion (alone or together with GPM2) causes no growth defect on diverse carbon sources and no change in key glycolytic-intermediate levels. It arose by gene duplication and is regarded as a probable non-functional homolog; it is expressed as a low-abundance protein (thousands of molecules per cell) whose molecular activity and biological role remain undetermined.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0004619 phosphoglycerate mutase activity | IBA GO_REF:0000033 | MARK AS OVER ANNOTATED | Summary: Phylogenetic (IBA) transfer of the family's phosphoglycerate mutase activity to GPM3. GPM3 was experimentally assayed and shown to lack detectable mutase activity even when strongly overexpressed (Heinisch et al. 1998), so this is an over-propagation of the ancestral family activity onto a probable non-functional paralog. GPM1 is the sole functional yeast phosphoglycerate mutase. Marked as over-annotated rather than removed because the evidence is contradictory rather than merely thin: UniProt itself hedges ("Could be non-functional"), the dPGM catalytic histidines are conserved in sequence, and weak, expression-limited residual activity below the original assay's detection cannot be fully excluded (forced overexpression did partially complement gpm1). The term should not, however, be retained as a functional annotation for this paralog. Propagation Review Root cause: PROPAGATION BAD Failure modes: WRONG ORTHOLOG OR PARALOG FUNCTIONAL DIVERGENCE Sources checked: PANTHER:PTN002630707 · PGAM ancestral node (PTHR11931) SUPPORTS SOURCE BUT NOT TARGET Ancestral PGAM activity is real for the family and for GPM1, but GPM3 has direct experimental evidence of no detectable mutase activity; catalytic histidines are present in sequence, so this is functional divergence rather than an obvious active-site lesion. Supporting Evidence: PMID:9544241 Higher level expression under the control of the yeast PFK2 promoter partially complemented the gpm1 defects, without restoring detectable enzymatic activity. |
| GO:0005829 cytosol | IBA GO_REF:0000033 | KEEP AS NON CORE | Summary: Phylogenetic assignment of cytosolic localization. This is consistent with the independent experimental (HDA) cytoplasm localization of GPM3 and with the cytoplasmic nature of the family, so the location is reasonable. Kept as non-core because GPM3 has no demonstrated activity for which the location would represent a core function; the "is_active_in" qualifier overstates a demonstrated activity at this site. This IBA cytosol call is independently corroborated by the experimental (HDA) cytoplasm localization of GPM3 (GO:0005737, PMID:14562095), which is why the location itself is accepted even though the accompanying activity is not. Supporting Evidence: PMID:14562095 Global analysis of protein localization in budding yeast. |
| GO:0061621 canonical glycolysis | IBA GO_REF:0000033 | MARK AS OVER ANNOTATED | Summary: Phylogenetic transfer of a glycolytic role. GPM3 is not a functional glycolytic enzyme: it has no detectable mutase activity, and deletion of GPM3 (with or without GPM2) does not perturb growth on diverse carbon sources or the levels of key glycolytic intermediates (Heinisch et al. 1998). This is an over-annotation of the family process onto a non-functional paralog. Propagation Review Root cause: PROPAGATION BAD Failure modes: WRONG ORTHOLOG OR PARALOG FUNCTIONAL DIVERGENCE Sources checked: PANTHER:PTN002630707 · PGAM ancestral node (PTHR11931) SUPPORTS SOURCE BUT NOT TARGET Canonical glycolysis is correct for GPM1 but propagates to a paralog whose deletion is metabolically silent and which lacks the enabling enzymatic activity. Supporting Evidence: PMID:9544241 deletion of either GPM2 or GPM3, or the two deletions in concert, did not produce any obvious lesions for growth on a variety of different carbon sources, nor did they change the levels of key intermediary metabolites. |
| GO:0003824 catalytic activity | IEA GO_REF:0000002 | MARK AS OVER ANNOTATED | Summary: Generic catalytic-activity term applied electronically from InterPro membership. No catalytic activity of any kind has been demonstrated for GPM3; the assayed mutase activity was undetectable. This is uninformative and, given the negative functional evidence, an over-annotation from family membership. Supporting Evidence: PMID:9544241 Higher level expression under the control of the yeast PFK2 promoter partially complemented the gpm1 defects, without restoring detectable enzymatic activity. |
| GO:0004619 phosphoglycerate mutase activity | IEA GO_REF:0000120 | MARK AS OVER ANNOTATED | Summary: Electronic assignment of phosphoglycerate mutase activity from the UniProt EC/Rhea mapping (EC 5.4.2.11) attached to the BPG-dependent PGAM subfamily. As with the IBA copy of this term, this is a family-level inference contradicted by the direct experimental finding of no detectable mutase activity for GPM3; the same UniProt entry states the protein "Could be non-functional". Marked as over-annotated rather than removed for the same reason as the IBA copy: the catalytic histidines are conserved and expression-limited residual activity cannot be fully excluded, so the call is deliberately conservative. Supporting Evidence: PMID:9544241 Higher level expression under the control of the yeast PFK2 promoter partially complemented the gpm1 defects, without restoring detectable enzymatic activity. |
| GO:0006096 glycolytic process | IEA GO_REF:0000120 | MARK AS OVER ANNOTATED | Summary: Electronic transfer of the glycolytic-process role from family/pathway mapping. Over-annotation for the same reasons as the canonical-glycolysis IBA term: GPM3 deletion is metabolically and phenotypically silent and the enzyme lacks detectable activity. Supporting Evidence: PMID:9544241 deletion of either GPM2 or GPM3, or the two deletions in concert, did not produce any obvious lesions for growth on a variety of different carbon sources, nor did they change the levels of key intermediary metabolites. |
| GO:0016868 intramolecular phosphotransferase activity | IEA GO_REF:0000002 | MARK AS OVER ANNOTATED | Summary: The specific mechanistic MF term for the dPGM reaction (intramolecular phosphotransfer), applied electronically from InterPro. No such activity has been demonstrated for GPM3; it is an over-annotation of the family mechanism onto a non-functional paralog. Supporting Evidence: PMID:9544241 Higher level expression under the control of the yeast PFK2 promoter partially complemented the gpm1 defects, without restoring detectable enzymatic activity. |
| GO:0005737 cytoplasm | HDA PMID:14562095 Global analysis of protein localization in budding yeast. | ACCEPT | Summary: Experimental (high-throughput direct assay) localization of GPM3 to the cytoplasm from the genome-wide GFP-fusion study. This is a well-supported, correct location annotation and the only positively demonstrated attribute of GPM3 besides its expression. Retained; it is a location rather than a core molecular/biological function. Supporting Evidence: PMID:14562095 Global analysis of protein localization in budding yeast. |
| GO:0003674 molecular_function | ND GO_REF:0000015 | ACCEPT | Summary: Root molecular_function with the ND (no biological data) evidence code, assigned by SGD. This honestly reflects the true state of knowledge: no molecular activity has been positively demonstrated for GPM3, and the assayed mutase activity was undetectable. Retained as an accurate no-data placeholder. |
| GO:0008150 biological_process | ND GO_REF:0000015 | ACCEPT | Summary: Root biological_process with the ND evidence code. Consistent with the absence of any demonstrated in-vivo role: GPM3 deletion is phenotypically silent under the conditions tested. Retained as an accurate no-data placeholder. |
Loading supporting content…
Download this section (compressed HTML)Q: Does GPM3 have any residual or alternative catalytic activity (e.g. phosphatase or a non-glycolytic phosphotransfer) despite lacking detectable phosphoglycerate mutase activity?
Q: Is GPM3 conditionally functional or important under a stress or nutrient state not tested in the original characterization?
Q: Given that the dPGM catalytic histidines are conserved in sequence, is the folded GPM3 active site competent, and does the GPM3-specific disordered insertion (residues ~168-198) disrupt folding or substrate binding?
Experiment: Express and purify recombinant GPM3 and assay it across a panel of phosphoglycerate mutase (with and without 2,3-bisphosphoglycerate priming) and histidine-phosphatase substrate reactions to test for any residual or alternative activity.
Hypothesis: GPM3 has no residual mutase activity but may retain a weak alternative phosphotransfer/phosphatase activity typical of the histidine-phosphatase superfamily.
Type: enzymatic assay
Experiment: Perform condition/stress phenotyping (e.g. carbon-source shifts, oxidative, osmotic, and stationary-phase stresses) of gpm3, gpm2 gpm3, and gpm1(ts) gpm3 strains to detect any conditional requirement for GPM3.
Hypothesis: GPM3 is required only under a specific, previously untested condition.
Type: genetic phenotyping
Experiment: Determine the GPM3 interactome by affinity purification / proximity labeling to test whether it acts as a partner or scaffold rather than an enzyme.
Hypothesis: GPM3 has a non-catalytic role mediated by protein partners.
Type: interaction proteomics
What is not known — curated, literature-grounded statements of the open unknowns (the inverse of core functions).
Gap: It is unknown whether GPM3 has any catalytic or other molecular activity, what (if any) its in-vivo biological role is, and why S. cerevisiae retains a stably expressed third phosphoglycerate-mutase homolog that is dispensable and lacks detectable mutase activity.
OPEN BIOLOGY WHOLLY_DARK
What is known: GPM3 is firmly established as a phosphoglycerate-mutase-family protein (histidine- phosphatase superfamily, BPG-dependent PGAM subfamily) that retains the dPGM catalytic histidine residues, is expressed as a low-abundance protein (~3730 molecules/cell) and localizes to the cytoplasm. It has been directly assayed for phosphoglycerate mutase activity and none was detectable even under forced (PFK2- promoter) overexpression, and deletion of GPM3 alone or together with GPM2 causes no growth defect on diverse carbon sources and no change in key glycolytic- intermediate levels. GPM1 is the sole functional yeast phosphoglycerate mutase.
Significance: GPM3 is a clean example of a conserved-looking paralog whose active-site residues are intact yet which has no directly measured ancestral activity. It is unresolved whether the protein is truly catalytically dead or retains weak residual activity that is masked physiologically by very low native expression: the same PFK2-driven overexpression that partially complements a gpm1 deletion on non-fermentable carbon also failed to restore detectable enzymatic activity in the original assay. Determining whether GPM3 has been pseudogenized at the protein level, retains weak residual activity limited by expression, acquired a new (non-mutase) activity, or is retained for a condition-specific role would clarify the fate of duplicated metabolic-enzyme genes and prevent electronic propagation of functional glycolytic-mutase annotations onto physiologically non-functional homologs.
What would resolve it: Broad enzymatic screening of purified recombinant GPM3 (including 2,3-BPG-primed mutase and phosphatase assays) to test for any residual or alternative activity; condition/stress-specific phenotyping of gpm3 and gpm1 gpm3 / gpm2 gpm3 mutants; affinity/proximity proteomics to identify partners; structural characterization to test whether the family fold and active site are competent despite the conserved residues, given the ~30-residue disordered insertion unique to GPM3.
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)