Gene 30 of the temperate Mu-like Pseudomonas phage DMS3 encodes AcrIE3 (originally ACR3-30), a 68-residue anti-CRISPR protein that neutralises the type I-E CRISPR-Cas system of Pseudomonas aeruginosa. It lies in the conserved anti-CRISPR operon that these phages carry between the homologue of Mu gene G and the protease/scaffold gene, and is one of four unrelated families of type I-E anti-CRISPRs found there. Expressed inside the infected or lysogenised bacterium, it binds the Cas8e subunit of the type I-E surveillance complex Cascade and occludes the protospacer-adjacent-motif recognition site through an acidic surface on an all-helical fold, so that Cascade can no longer recognise its DNA target and cannot recruit Cas3 to destroy it. The consequence is that a phage genome bearing a matching protospacer survives host CRISPR-Cas immunity and replicates. Its activity is narrow in two directions. It does not inhibit the type I-F system of the same host, and it does not inhibit the distantly related type I-E system of Escherichia coli, so it is specific for the P. aeruginosa type I-E Cascade rather than for the subtype as such.
Definition: Binds to a CRISPR-Cas surveillance (effector) complex, or to a subunit of such a complex, and inhibits its activity, so that the complex no longer recognises or licenses its nucleic acid target or no longer recruits the effector nuclease. This is the molecular function of anti-CRISPR proteins that act on the surveillance complex rather than on the nuclease, such as AcrIE3 and AcrIE5 binding Cas8e of the Pseudomonas aeruginosa type I-E Cascade.
Justification: GO currently has no molecular-function term for anti-CRISPR activity. A QuickGO ontology search for "CRISPR" returns only biological-process terms (GO:0099048, GO:0043571, GO:0098672, plus two obsoletes), so a protein whose entire function is to inhibit a CRISPR-Cas system can be given the process term GO:0098672 and no activity at all. The nearest existing molecular function, GO:0140721 nuclease inhibitor activity, is a poor fit for this class. In a type I system the nuclease is Cas3, whereas AcrIE3 binds Cas8e within the Cascade surveillance complex and blocks target recognition upstream of Cas3 recruitment, and in GOA GO:0140721 and its children are in practice occupied by ribonuclease inhibitors of the RraA/RraB type. Proposing the term rather than forcing an ill-fitting id also keeps the two mechanistic classes of anti-CRISPR distinguishable, since inhibitors that do block catalysis, such as AcrIIA5 on Cas9, are already served by GO:0140721. The parent proposed below is the generic inhibitor term. This is an ontology gap rather than a biological unknown, so it is recorded here and not as a knowledge gap.
Parent term: molecular function inhibitor activity
Supporting Evidence:
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0098672 symbiont-mediated suppression of host CRISPR-cas system | IDA PMID:24736222 A new group of phage anti-CRISPR genes inhibits the type I-E... | ACCEPT | Summary: CACAO row, evidence ECO:0007040 (plaque assay evidence used in manual assertion), dated 2025-05-03, and the only GO annotation this gene product has from any source. What the cited paper shows for this gene is that DMS3 gene 30 encodes ACR3-30, one of four unrelated type I-E anti-CRISPR families the paper discovers. Cloned out of DMS3 and expressed from a plasmid in Pseudomonas aeruginosa SMC4386, whose native type I-E CRISPR-Cas system the same paper had just shown to be active, it let the CRISPR-targeted phage JBD8 plaque three to five orders of magnitude better than empty vector (Fig. 3C), with five type I-F anti-CRISPR genes and the conserved JBD30-36 gene as negative controls in the same figure. Independently, isogenic DMS3 phages in which gene 30 was replaced by a type I-F anti-CRISPR gene, or by a frameshifted copy of it, lost the ability to plaque on SMC4386 while plaquing normally on a CRISPR-Cas-negative strain (Fig. 4), and the authors state the conclusion for gene 30 in those words. The term matches the assay exactly and is maximally specific, since GO:0098672 has no children and the gain-of-function and loss-of-function plaque and transformation assays measure suppression of host CRISPR-Cas immunity itself rather than a downstream phenotype. IDA is the right code, because Fig. 3C assays the cloned gene product directly. Later structural work on this exact protein (PDB 8HEL and 8ZEY both carry A0SMP0's 68 residues) shows it binds the Cas8e subunit of Cascade at the PAM recognition site, confirming that the protein is the direct effector of the suppression. Reason: The claim is correct, maximally specific, and directly supported by two independent experiments on this phage's own gene in the cited paper, so the annotation is retained as the gene's core process. The only defect is the GAF qualifier. acts_upstream_of_or_within_positive_effect describes a gene product that influences a process from outside it, and positively, whereas AcrIE3 itself executes the suppression by binding host Cas8e and blocking target recognition by Cascade. The correct qualifier is involved_in, which is what UniProt's own row for the same term on the sibling protein from the same figure of the same paper (Q6TM71, GO_REF:0000043) uses. That is a source-record field error, recorded in source_defects, and not a reason to reject the claim. Noted for completeness because it was looked for, this annotation is not a cross-isolate evidence transfer. The paper assays one representative gene from each of four different phages, DMS3 gene 30 among them, and the phage used in Fig. 4, "DMS3m", is DMS3 itself carrying five point mutations in gene 42, not a sibling isolate. Supporting Evidence: PMID:24736222 As shown in Fig. 3C, expression of putative anti-CRISPR genes from phages JBD88a, DMS3, D3112, and JBD5 allowed CRISPR-sensitive phage JBD8 to form plaques with 103- to 105-fold-higher efficiency than on cells containing empty vector. PMID:24736222 We designate the proteins encoded by these anti-CRISPR genes as ACR88a-32, ACR3-30, ACR3112-31, and ACR5-34 PMID:24736222 In contrast, a mutant phage in which the DMS3m type I-E anti-CRISPR gene was replaced with a type I-F anti-CRISPR gene (DMS3m30-35 [9]) was not able to form plaques on SMC4386. PMID:24736222 These results demonstrate that gene 30, the type I-E anti-CRISPR gene of phage DMS3m, functions during lytic infection to inhibit the type I-E CRISPR-Cas system of SMC4386 and, thus, allow robust phage replication within this strain. PMID:39541974 Our results reveal that AcrIE3 selectively binds to the Cas8e subunit of the Cascade complex. |
Loading supporting content…
Download this section (compressed HTML)Q: Should GO have a molecular-function term for anti-CRISPR activity? At present a protein whose entire function is to inhibit a CRISPR-Cas system can carry only the process term GO:0098672, and the nearest activity term, GO:0140721 nuclease inhibitor activity, does not describe an inhibitor that binds a surveillance complex subunit and blocks target recognition rather than catalysis.
Q: Should the qualifier on the nine CACAO rows for GO:0098672 that use acts_upstream_of_or_within_positive_effect be corrected in bulk? Every one is on a protein that executes the suppression itself, and GOA already holds involved_in rows for the same term on some of the same proteins.
Q: Which residues of Cas8e distinguish the Pseudomonas aeruginosa Cascade, which AcrIE3 inhibits, from the Escherichia coli Cascade, which it does not, given that the two systems already differ in PAM specificity and AcrIE3 binds the PAM recognition site?
Q: Do AcrIE3, AcrIE4 and AcrIE5, which are unrelated in sequence and fold but all reported to converge on the PAM recognition site of Cas8e, inhibit by the same physical mechanism, and does that convergence mark the PAM site as the most evolutionarily accessible target in a type I-E Cascade?
Q: Phage DMS3 carries a type I-E anti-CRISPR but no type I-F anti-CRISPR, while most phages in this family carry one or both in a mixed-and-matched arrangement. What selects for retaining only the I-E gene in DMS3, and does a DMS3 prophage leave its host's type I-F immunity intact as a consequence?
Experiment: Determine a cryo-EM structure of AcrIE3 bound to reconstituted Pseudomonas aeruginosa type I-E Cascade, and in parallel titrate target DNA against AcrIE3 in electrophoretic mobility shift and Cas3-dependent cleavage assays to test whether the block is relieved by target excess.
Hypothesis: AcrIE3 inhibits Cascade competitively, by occupying the PAM recognition site of Cas8e in the manner of target DNA.
Experiment: Measure AcrIE3 binding to purified E. coli and P. aeruginosa Cas8e and to each assembled Cascade, then build reciprocal chimaeras exchanging the PAM-recognition-site regions of the two Cas8e proteins and assay inhibition of each chimaeric Cascade.
Hypothesis: The failure of AcrIE3 against the Escherichia coli type I-E system is due to divergence in the Cas8e PAM recognition site rather than to any other subunit or to protein stability.
Experiment: Follow gene 30 transcript and protein through a single synchronised infection of a type I-E-immune host by quantitative RT-PCR and quantitative immunoblot, alongside a fluorescent reporter of phage genome integrity in single cells, and compare phages in which gene 30 is placed under early, middle and late promoters.
Hypothesis: Protection of an incoming phage genome depends on how quickly gene 30 is expressed relative to Cascade-directed cleavage, so the gene should behave as an early gene.
Experiment: Make charge-reversal and charge-neutralisation variants of each acidic patch separately and in combination, and score each for Cas8e binding by isothermal titration calorimetry and for inhibition in the JBD8 plaquing assay in SMC4386, to separate the binding determinant from the rest of the surface.
Hypothesis: The three acidic surface patches of AcrIE3 are not equivalent; one carries the Cas8e interaction and the others are structural.
Experiment: Introduce gene 30 alone into a type I-E CRISPR-sensitive phage that lacks any anti-CRISPR gene, such as JBD8, by homologous recombination, and measure plaquing on SMC4386 and on a CRISPR-Cas-negative strain, completing the sufficiency half of the Fig. 4 experiment in the phage genome rather than from a plasmid.
Hypothesis: AcrIE3 is sufficient to protect an otherwise unprotected phage from a type I-E system, as the reciprocal gene-replacement experiment showed it is necessary.
What is not known — curated, literature-grounded statements of the open unknowns (the inverse of core functions).
Gap: Why AcrIE3 inhibits the Pseudomonas aeruginosa type I-E system but not the Escherichia coli type I-E system is unknown. Both systems use a Cas8e-containing Cascade, and AcrIE3 was shown to be robustly expressed in E. coli, so the failure is a property of the interaction rather than of protein supply, but the determinants on either side have not been identified.
OPEN BIOLOGY
What is known: Established, all four Pseudomonas type I-E anti-CRISPRs, AcrIE3 among them, fail to protect phage M13 from an induced E. coli K-12 type I-E system while being expressed at levels visible by Coomassie staining; AcrIE3 binds P. aeruginosa Cas8e at the PAM recognition site; and the P. aeruginosa and E. coli systems are already known to differ in PAM specificity. Unknown, which Cas8e residues or which conformational features of the two Cascades account for the discrimination.
Significance: Anti-CRISPR proteins are proposed as off-switches for CRISPR-based tools, and a species-level specificity this sharp between two Cascades of the same subtype bears directly on whether such an off-switch can be transplanted. It also tests whether the PAM recognition site is the general target of divergent type I-E anti-CRISPRs, as the structural work proposes.
What would resolve it: Assay AcrIE3 against reconstituted E. coli and P. aeruginosa Cascade in vitro, measure binding to each Cas8e, and swap the divergent PAM-recognition-site regions between the two Cas8e proteins to find the determinants.
Provenance (the field's own admissions):
Gap: No structure of AcrIE3 bound to Cascade, or to Cas8e, has been determined, so whether the acidic surface of AcrIE3 mimics the phosphate backbone of target DNA at the PAM recognition site or instead locks Cas8e in a non-permissive conformation is not established.
OPEN BIOLOGY
What is known: Established, two independent crystal structures of free AcrIE3, a helical fold with three acidic surface patches; selective binding to Cas8e rather than to other Cascade subunits; and mutations in the acidic surface abolish inhibition. Unknown, the complex itself, and therefore which of the two mechanisms operates.
Significance: DNA mimicry and allosteric locking are the two paradigms for anti-CRISPRs acting on surveillance complexes, and the distinction determines whether inhibition is competitive with target DNA and therefore escapable by target excess. It also decides whether AcrIE3, AcrIE4 and AcrIE5, which all converge on the same site, do so by the same chemistry.
What would resolve it: A cryo-EM structure of the AcrIE3-Cascade complex, with competition experiments titrating target DNA against AcrIE3 to test whether the block is competitive.
Provenance (the field's own admissions):
Gap: It is not known when during infection, or at what abundance, AcrIE3 is produced, and therefore how a phage carrying a perfectly matched protospacer manages to express an inhibitor fast enough to outrun Cascade-directed destruction of its own genome.
OPEN BIOLOGY
What is known: Established, gene 30 lies in an operon with a conserved putative promoter upstream, it is sufficient for protection when supplied in trans before infection, and it is necessary during lytic infection of a targeted host. Unknown, the kinetics of expression from the incoming genome, and whether a sub-population of genomes is lost before enough inhibitor accumulates.
Significance: The race between anti-CRISPR expression and CRISPR-Cas targeting sets whether anti-CRISPRs protect individual phage genomes or only populations, and the same question has been shown to matter for type I-F anti-CRISPRs in this phage family.
What would resolve it: Time-resolved measurement of gene 30 transcript and protein from a single round of infection of a type I-E-immune host, with single-cell reporters of phage genome survival.
Provenance (the field's own admissions):
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)