Protein 114 (gp29/ORF114a) of Sulfolobus islandicus rod-shaped virus 1 is a small homodimeric DUF1874-family cyclic tetra-adenylate ring nuclease. Its dimer-interface pocket recognizes cA4, and a conserved catalytic site cleaves the signal to linear di-adenylate with a 2',3'-cyclic phosphate. The enzyme prefers cA4 over cA6. Ectopic plasmid expression can suppress cA4-dependent type III CRISPR immunity and permit infection by another virus. Its native anti-CRISPR role is disputed: a later SIRV2 infection study found no protection from native gp37 or SIRV1 gp29 transplanted into SIRV2 under the tested expression conditions, whereas earlier induced gp29 expression restored protection. This did not test infection by SIRV1 itself. The established enzymatic activity therefore does not demonstrate that all native infections or DUF1874 homologs use it as an effective counterdefence.
Definition: Catalysis of the endonucleolytic cleavage of an internal 3'-5' phosphodiester linkage within a cyclic oligoadenylate second messenger, opening the ring to yield linear oligoadenylate products bearing a 2',3'-cyclic phosphate.
Justification: GO has no term for cleavage of a cyclic oligoadenylate messenger, and each available alternative is either too general or simply wrong. GO:0004521 RNA endonuclease activity, asserted on this entry as the nearest existing term, is defensible because cyclic tetra-adenylate is a short ribonucleotide and ring opening is an internal phosphodiester cleavage, but it loses the substrate entirely, and the substrate is what the enzyme is for. GO:0001730 2'-5'-oligoadenylate synthetase activity is the metazoan OAS enzyme, which synthesises a different, linear 2'-5'-linked molecule, so it is the wrong term in both direction and chemistry, and it must not be used for this family or for the Cas10 cyclase that makes the substrate. The cyclic-nucleotide phosphodiesterase term GO:0004112 is about cAMP and cGMP second messengers and about generating a linear nucleoside monophosphate, which is neither the substrate nor the product here. The term is not a single-gene request. It is needed for experimentally established ring nucleases in viral and cellular lineages, and it would equally serve the host-encoded Crn1 and Crn2 ring nucleases, for example UniProtKB:Q97YD2, which perform the same chemistry as self-regulation of the host response rather than as counter-defence. It is also the mirror image of a second gap, the absence of a term for the cyclic oligoadenylate synthase activity of the Cas10 Palm domain, recorded in MODULE:crispr_cas_adaptive_immunity, and the two should be requested together so that the ontology can represent both ends of the signalling step.
Parent term: RNA endonuclease activity
Supporting Evidence:
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0098672 symbiont-mediated suppression of host CRISPR-cas system | IMP PMID:31942067 An anti-CRISPR viral ring nuclease subverts type III CRISPR ... | UNDECIDED | Summary: Previous NEW process proposal withdrawn pending resolution of expression-context-dependent native counterdefence; biochemical and plasmid-expression results remain established. Reason: This was a reviewer-proposed NEW row, not an imported GOA assertion. PMID:31942067 establishes cA4 destruction and plasmid-expression rescue, so gp29 performs the relevant chemistry rather than merely being necessary. PMID:40240860 compares viral-genome and plasmid expression; its publisher Figure 2 caption explicitly identifies the SIRV2 Para-gp29 construct. The tested SIRV2 context must not be presented as a direct native SIRV1 infection assay. The Comment and Reply leave physiological deployment dependent on expression and infection context. Canonical publication identities are retained under their PMIDs; reproducibly extracted publisher previews and captions provide the accessible evidence, while complete Comment/Reply text is not available in the normal PMID caches. Prior main-text inspection is documented in notes. Retain the retired UNDECIDED proposal pending native-scope adjudication, not because the demonstrated ectopic phenotype or nuclease activity is disproved. A context-qualified process proposal remains an explicit curator question. Supporting Evidence: PMID:31942067 However, the same cells expressing the SIRV1 gp29 gene from a plasmid were readily infected, giving rise to plaque formation. These data are consistent with the hypothesis that SIRV1 gp29 is functioning as an Acr specific for type III CRISPR defence. file:SIRV1/orf114/orf114-source-analysis/publisher-preview-excerpts.md cannot inhibit subtype III-B CRISPR immunity when expressed from the viral genome, but exhibit anti-CRISPR activity when expressed from a plasmid. file:SIRV1/orf114/orf114-source-analysis/publisher-preview-excerpts.md AcrIII-1 and Crn2 are cA4-specific ring nucleases in vitro file:SIRV1/orf114/orf114-source-analysis/publisher-preview-excerpts.md c, Schematic of SIRV2 Para-gp29 virus. |
| GO:0004521 RNA endonuclease activity | IDA PMID:31942067 An anti-CRISPR viral ring nuclease subverts type III CRISPR ... | NEW | Summary: The purified recombinant protein rapidly degrades cyclic tetra-adenylate to linear di-adenylate bearing a 2',3'-cyclic phosphate, with an observed single-turnover cleavage rate exceeding five per minute at 50 degrees Celsius and a strong preference for the tetrameric over the hexameric cyclic oligoadenylate. Opening the ring is an internal phosphodiester cleavage within a short ribonucleotide, so GO:0004521 applies, and the co-crystal structure with substrate plus the H47A loss-of-activity variant with partial chemical rescue by imidazole together establish the active site. Reason: Asserted as the nearest existing term, since GO has no term for cleavage of a cyclic oligoadenylate second messenger. QuickGO searches for "ring nuclease" return only generic nuclease and nuclease-regulator terms, and searches for "cyclic oligoadenylate" return GO:0001730 2'-5'-oligoadenylate synthetase activity, which is the metazoan OAS enzyme acting in the opposite direction on a different molecule and is the wrong term, plus cyclic-nucleotide binding and phosphodiesterase terms that do not fit. GO:0004521 is correct but loses the substrate, which is the whole point of the enzyme, so a specific term is proposed alongside it and the core function carries the proposed name rather than this identifier. Evidence is IDA on purified protein. Supporting Evidence: PMID:31942067 Both proteins possess a potent ring nuclease activity, rapidly degrading cA4 to generate linear di-adenylate (ApA>P) with a cyclic 2’,3’ phosphate PMID:31942067 With a catalytic rate exceeding 5 min-1, the Acr enzyme is at least 60-fold more active than the cellular ring nuclease Crn1 from S. solfataricus. PMID:31942067 Both SIRV1 gp29 and YddF enzymes show a strong preference for cA4 over cA6 |
| GO:0030551 cyclic nucleotide binding | IDA PMID:31942067 An anti-CRISPR viral ring nuclease subverts type III CRISPR ... | NEW | Summary: The 1.55 Angstrom co-crystal structure of the catalytically inactive H47A variant with cyclic tetra-adenylate shows one molecule of the cyclic nucleotide bound in a pocket at the dimer interface, buried by a loop that closes over it. The fold is unrelated to the CRISPR-associated Rossmann fold domain, which was the only protein module previously known to bind cyclic oligoadenylates. Reason: Asserted because substrate binding is established independently of the catalytic step, by a structure of a catalytically dead variant in complex with intact substrate, and because the specificity is a measured property rather than an inference, the enzyme preferring cyclic tetra-adenylate over the hexameric form. GO:0030551 cyclic nucleotide binding is the correct existing parent; there is no GO term for cyclic tetra-adenylate specifically, although GO:0180001 cyclic-di-AMP binding exists for the shorter relative, which suggests a sibling term would be accepted. Raised in suggested_questions rather than proposed here, to keep the single proposed term in this review focused on the activity. Supporting Evidence: PMID:31942067 The complex reveals a molecule of cA4 bound at the dimer interface. PMID:31942067 The structure of AcrIII-1 is unrelated to the CARF domain, which is the only protein family thus far known to bind cOA PMID:31942067 we co-crystallised an inactive variant (H47A) of SIRV1 gp29 with cA4 and solved the structure to 1.55 Å resolution |
| GO:0030430 host cell cytoplasm | IC PMID:31942067 An anti-CRISPR viral ring nuclease subverts type III CRISPR ... | NEW | Summary: Host-cell cytoplasmic action is inferred from the intracellular substrate and plasmid-expression rescue experiments; no microscopy localization is claimed. Reason: The viral protein acts on a soluble cyclic oligoadenylate signal produced by intracellular CRISPR machinery. Host-qualified cytoplasm is the appropriate inferred location for that experimental activity. This does not establish effective suppression during native infection, which is separately disputed, and no sequence-family occurrence argument is used as localization evidence. Supporting Evidence: PMID:31942067 However, the same cells expressing the SIRV1 gp29 gene from a plasmid were readily infected, giving rise to plaque formation. These data are consistent with the hypothesis that SIRV1 gp29 is functioning as an Acr specific for type III CRISPR defence. |
Loading supporting content…
Download this section (compressed HTML)Q: The viral enzyme is more than sixty-fold faster than the host's own Crn1 ring nuclease on the same substrate. Is that because the host must keep its messenger around long enough for immunity to work, so Crn1 is under selection to be slow, while the virus is under selection only to be fast?
Suggested experts: Januka S. Athukoralage, Malcolm F. White
Q: Does GO want a cyclic tetra-adenylate binding term as a sibling of GO:0180001 cyclic-di-AMP binding, given that specificity for the tetramer over the hexamer is a measured property of this enzyme and of the host CARF-domain receptors alike?
Q: Which controlled native-expression and host/target comparisons would resolve the physiological scope disputed by the 2025 Comment and Reply, while preserving the established biochemical and ectopic-expression results?
Q: SIRV1 carries this gene while the lytic virus SSeV used as the reporter does not. Is the presence or absence of a duf1874 gene a good predictor of which archaeal viruses can persist in type III-competent Sulfolobus populations?
Suggested experts: Rachel J. Whitaker, Mart Krupovic
Q: Should the demonstrated plasmid-expression anti-CRISPR phenotype receive a context-qualified process annotation despite the disputed native infection role?
Suggested experts: GO microbial and viral annotation curators
Q: Should the reviewed UniProt entry be updated with a function statement and gp29/ORF114a nomenclature using the demonstrated Q8QL27–NP_666617–PDB 6SCF identity, while keeping native anti-CRISPR deployment explicitly disputed?
Experiment: If a SIRV1 isolate becomes available, compare native-promoter wild type, catalytic-site mutant and complemented virus across defined host CRISPR and target-transcript backgrounds, measuring expression timing and flanking-gene expression. In the available SIRV2 system, distinguish native gp37 from transplanted gp29 and include early-expression controls. Interpret a CRISPR-dependent phenotype together with direct cA4 cleavage and signal measurements; knockout necessity alone does not establish process participation or automatically justify a new annotation.
Hypothesis: If SIRV1 gp29 contributes to counterdefence in an appropriate native infection context, catalytic disruption should selectively impair protection under conditions in which the protein is expressed before the relevant CRISPR response.
Type: matched viral-expression and catalytic-mutant infection assays
Experiment: Measure full Michaelis-Menten parameters for cyclic tetra-adenylate turnover side by side for this enzyme and for the host Crn1 under identical conditions, separating the contributions of substrate affinity and maximum rate. A difference concentrated in the maximum rate rather than in affinity would support the selection argument, and would also give the kinetic parameters that the GO term request and any future quantitative model of the signalling step need.
Hypothesis: Catalytic speed, not substrate affinity, is what the virus is selected for, which would explain the sixty-fold rate difference from the host's own ring nuclease.
Type: comparative steady-state enzyme kinetics
Experiment: Identify interface residues from the 1.55 Angstrom structure, build substitutions designed to shift the monomer-dimer equilibrium without perturbing the monomer fold, verify the oligomeric state by size-exclusion chromatography with inline light scattering and the fold by circular dichroism, then assay cyclic tetra-adenylate cleavage. A quantitative correlation between dimer fraction and activity would establish that the functional unit is the dimer, which would in turn justify annotating protein homodimerization activity on this entry, something the current evidence supports structurally but has not tested functionally.
Hypothesis: Dimerisation is obligatory for activity, because the substrate pocket is formed at the dimer interface.
Type: interface mutagenesis with paired oligomeric-state and activity readouts
What is not known — curated, literature-grounded statements of the open unknowns (the inverse of core functions).
Gap: Under which native virus, host, target-transcript and expression-timing conditions does gp29 provide effective anti-CRISPR protection?
OPEN BIOLOGY
What is known: Plasmid-expression sufficiency and nuclease chemistry are established. The Comment publisher captions identify SIRV2 infection and a SIRV2 Para-gp29 construct; they do not establish native SIRV1 infection behavior. The normal PMID caches lack full text. Publisher excerpts are reproducibly available; the prior main-text inspection and more detailed Reply objections remain documented in notes with their access limits.
Significance: An experimentally effective ectopic inhibitor need not protect its native virus at the time CRISPR immunity acts. Conversely, failure in one infection context does not prove absence of every conditional viral counterdefence role.
What would resolve it: Resolve expression timing, host/target and multiplicity-of-infection dependence with matched constructs and controls for flanking-gene effects. Distinguish native SIRV2 gp37, transplanted SIRV1 gp29 and native SIRV1 infection. A negative result in one context does not settle all contexts.
Provenance (the field's own admissions):
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)