Catalytically inactive member of the alpha-carbonic anhydrase family (a carbonic anhydrase-related protein, CARP VIII) that acts as a cytosolic regulator of intracellular calcium release rather than as an enzyme. CA8 retains the carbonic anhydrase fold but carries arginine in place of one of the three histidines that coordinate the catalytic zinc, so it neither binds zinc stoichiometrically nor hydrates carbon dioxide; restoring the metal-binding centre by engineered substitution is what converts it into an active enzyme. The preserved active-site cavity is instead repurposed as a protein-interaction surface: CA8 binds the modulatory domain of the inositol 1,4,5-trisphosphate receptor type 1 (ITPR1/IP3R1), an endoplasmic reticulum calcium-release channel, and lowers the receptor's affinity for IP3 without changing the number of ligand-binding sites, thereby damping IP3-evoked calcium release. It is expressed most strongly in cerebellar Purkinje cells, where IP3R1 is abundant, and the interaction is required for normal cerebellar function: recessive loss-of-function variants cause congenital cerebellar ataxia with mild intellectual disability and a tendency to quadrupedal gait, and the orthologous mouse mutant waddles is ataxic and dystonic.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0004089 carbonate dehydratase activity | IEA GO_REF:0000002 | REMOVE | Summary: Fold-based electronic assignment to a protein that lacks a catalytic zinc site and has no measurable carbonic anhydrase activity. Reason: This is an IEA from InterPro signatures IPR018338/IPR023561 via GO_REF:0000002 - that is, the alpha-carbonic anhydrase fold is present, so the activity was inferred. The fold is present and the activity is not. CA8 carries Arg116 in place of one of the three conserved zinc-coordinating histidines, so the catalytic metal centre cannot assemble; UniProt states outright that the protein "Does not have a carbonic anhydrase catalytic activity", and the approved HGNC name is "carbonic anhydrase 8 (inactive)". Direct measurement agrees: Sjoblom et al. purified the recombinant protein and found it catalytically inactive, and had to engineer two substitutions (Arg117His, Glu115Gln, murine numbering) to obtain CO2 hydration. Every subsequent structural and review treatment classifies CA8 as acatalytic. There is no experimental annotation being second-guessed here - GOA holds no IDA/IMP/EXP row for this term - so this is precisely the over-propagated electronic inference that can be argued down on biological grounds. Supporting Evidence: PMID:8977131 While unmodified CARP is catalytically inactive, the mutant catalyzes CO2 hydration with a significantly higher efficiency than the mammalian low-activity carbonic anhydrase isozyme III. PMID:32316137 Though the protein cannot hydrate CO2, CA-VIII is essential for calcium (Ca2+) homeostasis within the body, and achieves this by allosterically inhibiting the binding of inositol 1,4,5-triphosphate (IP3) to the IP3 receptor type 1 (ITPR1) protein. PMID:42268453 However, carbonic anhydrase VIII (CA VIII) represents a catalytically inactive member of this family due to the absence of one of the three histidine residues required for enzymatic activity |
| GO:0008270 zinc ion binding | IEA GO_REF:0000002 | REMOVE | Summary: The zinc site is exactly what CA8 has lost; unmodified CARP is not isolated as a zinc-protein complex. Reason: Same provenance as the catalytic annotation - IEA from InterPro IPR018338/IPR023561 via GO_REF:0000002 - and it fails for the same reason, more directly. The alpha-CA zinc centre requires three histidines; CA8 substitutes Arg116 for one of them, and UniProt flags this in a CAUTION line. The defining experiment is a contrast: engineering the metal-binding centre back in yielded a 1:1 zinc-protein complex, whereas the unmodified protein did not. Retaining GO:0008270 asserts the one property the primary literature specifically excludes. Supporting Evidence: PMID:8977131 In contrast to unmodified CARP, this double mutant was isolated as a 1:1 zinc-protein complex. PMID:8977131 By introducing two mutations, Arg117 --> His and Glu115 --> Gln, we created a metal-binding center homologous to that in the carbonic anhydrases from the animal kingdom. |
| GO:0005515 protein binding | IPI PMID:25416956 A proteome-scale map of the human interactome network. | REMOVE | Summary: Bare protein binding from a proteome-scale interactome screen; uninformative, and it misses the interaction that matters. Reason: Per project curation guidance, GO:0005515 should give way to an informative molecular function. All five GO:0005515 rows for CA8 come from systematic binary-interactome screens, and their partners (CRX, GGA2, HSD17B14, INTS7, KLHL8, LMNB2, LNX1, MAGED1, RAB34, SPDL1, TBX3) are unrelated to any characterised CA8 biology. Meanwhile ITPR1, the partner established by targeted biochemistry and validated reciprocally by disease mutations on both sides of the interface, appears nowhere in GOA. The informative replacement is the calcium channel inhibitor activity recorded in core_functions. No more informative molecular function is supportable from this evidence, so the row is removed rather than rewritten. Removal does not imply the reported interaction is false. |
| GO:0005515 protein binding | IPI PMID:25910212 Widespread macromolecular interaction perturbations in human... | REMOVE | Summary: Bare protein binding from a systematic interactome-perturbation screen; uninformative. Reason: 'Same judgment as the other GO:0005515 rows: a high-throughput binary interaction catalogue contributes partner identities but no molecular function, and none of the partners recorded here corresponds to a characterised CA8 activity.'. No more informative molecular function is supportable from this evidence, so the row is removed rather than rewritten. Removal does not imply the reported interaction is false. |
| GO:0005515 protein binding | IPI PMID:27107012 Pooled-matrix protein interaction screens using Barcode Fusi... | REMOVE | Summary: Bare protein binding from a pooled-matrix two-hybrid screen; uninformative. Reason: 'Same judgment as the other GO:0005515 rows: methodological interactome output, with no functional content for this gene.'. No more informative molecular function is supportable from this evidence, so the row is removed rather than rewritten. Removal does not imply the reported interaction is false. |
| GO:0005515 protein binding | IPI PMID:31515488 Extensive disruption of protein interactions by genetic vari... | REMOVE | Summary: Bare protein binding from a variant-effect interactome survey; uninformative. Reason: Same judgment as the other GO:0005515 rows. Of note, this class of study is where an informative CA8 annotation could have come from - disease variants do disrupt the CA8-IP3R1 interaction - but the partners captured in this row are not ITPR1. No more informative molecular function is supportable from this evidence, so the row is removed rather than rewritten. Removal does not imply the reported interaction is false. |
| GO:0005515 protein binding | IPI PMID:32296183 A reference map of the human binary protein interactome. | REMOVE | Summary: Bare protein binding from the human reference binary interactome; uninformative. Reason: 'Same judgment as the other GO:0005515 rows: a yeast-two-hybrid reference map entry carrying no functional information.'. No more informative molecular function is supportable from this evidence, so the row is removed rather than rewritten. Removal does not imply the reported interaction is false. |
| GO:0005737 cytoplasm | IEA GO_REF:0000107 | ACCEPT | Summary: Correct compartment: CA8 is a soluble protein acting on the cytoplasmic modulatory domain of an ER membrane channel. Reason: Orthology transfer from mouse Car8 (UniProtKB:P28651) via Ensembl Compara. CA8 has no signal peptide or transmembrane segment, and it engages the modulatory domain of IP3R1 (residues 1387-1647 of the receptor), which faces the cytosol - so a cytoplasmic location is both correct and consistent with the mechanism. Nothing more specific is supported by the available evidence. Supporting Evidence: PMID:12611586 Using deletion mutagenesis, we established that amino acids 45-291 of CARP are essential for its association with IP(3)R1, and that the CARP-binding site is located within the modulatory domain of IP(3)R1 amino acids 1387-1647. |
| GO:0004089 carbonate dehydratase activity | TAS PMID:8977131 Two point mutations convert a catalytically inactive carboni... | REMOVE | Summary: The cited paper states the opposite of the annotation: it reports that unmodified CARP is catalytically inactive and that two engineered mutations are required to create an active enzyme. Reason: This TAS (ECO:0000304, assigned by PINC) is a misannotation of its own source. PMID:8977131 is Sjoblom et al., FEBS Lett 1996, "Two point mutations convert a catalytically inactive carbonic anhydrase-related protein (CARP) to an active enzyme" - the title already carries the finding. The abstract reports that the recombinant wild-type protein is catalytically inactive and does not form a zinc complex, and that CO2 hydration appears only after Arg117His and Glu115Gln are introduced to rebuild the metal-binding centre. The paper is therefore evidence AGAINST carbonate dehydratase activity in CA8, not for it. A traceable author statement is not an experimental annotation, and here the traceable statement is traceably wrong. Removing it also makes GOA consistent with the UniProt FUNCTION line and with the approved gene name, "carbonic anhydrase 8 (inactive)". Same action as the IEA row for this term, as required. Supporting Evidence: PMID:8977131 While unmodified CARP is catalytically inactive, the mutant catalyzes CO2 hydration with a significantly higher efficiency than the mammalian low-activity carbonic anhydrase isozyme III. PMID:8977131 By introducing two mutations, Arg117 --> His and Glu115 --> Gln, we created a metal-binding center homologous to that in the carbonic anhydrases from the animal kingdom. |
| GO:0019855 calcium channel inhibitor activity | IDA PMID:12611586 Carbonic anhydrase-related protein is a novel binding protei... | NEW | Summary: Proposed replacement for the removed catalytic annotations: CA8 binds IP3R1 and reduces its activity, which is the molecular function the repurposed fold actually performs. Reason: With GO:0004089 and GO:0008270 removed, CA8 would carry no molecular function at all apart from bare protein binding, which is worse than the misannotation it replaces. GO:0019855 is defined as "Binds to and stops, prevents, or reduces the activity of a calcium channel", and that is what the primary literature shows: CA8 binds the cytosolic modulatory domain of the ER calcium-release channel IP3R1 (receptor residues 1387-1647, via CA8 residues 45-291) and lowers the receptor's affinity for IP3 without changing the number of ligand-binding sites, so IP3-evoked release is damped. The assignment is corroborated from the receptor side: the SCA29 variant V1538M in the CA8-binding region abolishes both the interaction and CA8-mediated inhibition. I checked the alternatives - GO:0005246 calcium channel regulator activity is correct but loses the unidirectional inhibition, GO:0044325 transmembrane transporter binding drops the functional consequence, and GO:0070679 inositol 1,4,5 trisphosphate binding is simply wrong because CA8 binds the receptor and not the ligand. No GO term exists for "IP3 receptor binding" or "IP3 receptor inhibitor activity". Supporting Evidence: PMID:12611586 CARP inhibits IP(3) binding to IP(3)R1 by reducing the affinity of the receptor for IP(3). PMID:32316137 CA-VIII allosterically inhibits ITPR1 by reducing the receptor’s affinity for IP3 without altering the maximum number of ligand binding sites. PMID:30429331 The SCA29 mutation V1538M within the CA8-binding site of IP3R1 completely eliminated its interaction with CA8 and CA8-mediated IP3R1 inhibition. |
| GO:0051280 negative regulation of release of sequestered calcium ion into cytosol | IDA PMID:30429331 Aberrant IP(3) receptor activities revealed by comprehensive... | NEW | Summary: The biological process that follows from CA8's inhibition of IP3R1: less calcium released from the ER store for a given IP3 stimulus. Reason: IP3R1 is the endoplasmic reticulum calcium-release channel, and CA8 suppresses its activity; the direct consequence is reduced release of sequestered calcium into the cytosol. This is asserted as a process CA8 carries out rather than as a downstream organismal phenotype: it is measured as CA8-mediated suppression of IP3R1 in calcium-flux assays, and it is lost when either partner carries an ataxia-associated variant at the interface. Reviews of the field describe the directionality the same way - CA8 as a negative regulator that restrains excessive calcium release. Supporting Evidence: PMID:30429331 Furthermore, pathological mutations in CA8 decreased CA8-mediated suppression of IP3R1 by reducing protein stability and the interaction with IP3R1. PMID:42268453 CA VIII is thought to act as a negative regulator of IP3R1, reducing excessive calcium release and maintaining intracellular calcium homeostasis |
Loading supporting content…
Download this section (compressed HTML)Q: Should GOA replace the removed catalytic annotation with an explicit negative one - NOT|enables GO:0004089 carbonate dehydratase activity, cited to PMID:8977131? The paper supports a negative assertion directly, and recording it would stop the InterPro alpha-CA signature from reinstating the activity at the next remapping.
Q: Why is ITPR1 absent from CA8's GO:0005515 protein binding annotations when it is the only functionally characterised partner, while eleven partners from proteome-scale two-hybrid screens are present? Is there a curation gap for the targeted biochemistry in PMID:12611586 and PMID:30429331?
Q: Is GO:0019855 calcium channel inhibitor activity the right molecular function, or does CA8's effect on IP3 affinity warrant a more specific term for an allosteric modulator of a ligand-gated intracellular calcium-release channel? No existing GO term captures "IP3 receptor binding" or "IP3 receptor inhibitor activity".
Q: Does CA8 retain any catalytic activity at all, on a substrate other than CO2? "Cannot hydrate CO2" is well established; "has no enzymatic activity of any kind" has not actually been tested, and other pseudoenzymes have turned out to have neofunctionalised chemistry.
Q: How should the dissenting result in PMID:26399641 be weighed - CA8 overexpression stunted Purkinje dendritic growth with no evidence of direct binding to IP3R1 in that system? Does CA8 have an IP3R1-independent role in dendritic development, or is this a system-specific failure to detect the interaction?
Q: Do the paralogous acatalytic carbonic anhydrase-related proteins CA10 and CA11 carry the same InterPro-derived carbonate dehydratase and zinc-binding annotations, and do they also act through protein-interaction surfaces built on the disused active-site cavity?
Experiment: Reconstitute the inhibition in a defined system: purified recombinant CA8 added to IP3R1 in planar lipid bilayers or in permeabilised-cell calcium-flux assays, measuring single-channel open probability and the IP3 dose-response as a function of CA8 concentration. This would convert the "reduces affinity for IP3" result into a quantitative affinity and stoichiometry, which no study has yet provided.
Experiment: Structure of the CA8-IP3R1 modulatory-domain complex by cryo-EM. This would show whether the vestigial active-site cavity is the binding surface, as comparative modelling suggests, and would explain how binding at residues 1387-1647 propagates to the distant IP3-binding domain - possibly, as proposed, to the adjacent subunit of the tetramer.
Experiment: Test for residual catalysis: screen purified human CA8 against esterase substrates and a panel of small-molecule hydrolysis/hydration reactions alongside CA2 as a positive control and the Arg116His revertant as an internal comparator, to establish whether "acatalytic" means genuinely inert or merely inactive toward CO2.
Experiment: Separation-of-function mouse: knock in a CA8 variant that is stable but cannot bind IP3R1 (guided by the residue 45-291 mapping and by the structure above), and compare with a Ca8 null for cerebellar calcium handling in Purkinje cells, dendritic morphology and motor phenotype. This asks directly whether the ataxia is entirely attributable to loss of IP3R1 regulation, and would also adjudicate the dendritic-growth result of PMID:26399641.
Experiment: Cell-type-resolved calcium imaging in Purkinje neurons from CA8 patient-derived or isogenic iPSC models carrying S100P and other reported variants, measuring IP3-evoked release with and without proteasome inhibition, to test whether restoring CA8 protein levels restores normal calcium dynamics.
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)