Protein p1 (gp1) is an 86-residue early protein of Bacillus phage phi29 that builds the membrane platform on which the phage replicates its genome. It carries a predicted coiled coil (residues 38-67) and a hydrophobic C-terminus, and it self-polymerises into long protofilaments that pack side by side into two-dimensional sheets resembling tubulin and bacterial FtsZ polymers; the coiled-coil residues Leu-46, Met-53 and Leu-60 are each required for sheet formation. The hydrophobic C-terminus attaches these polymers peripherally to the inner face of the Bacillus subtilis membrane - p1 has no transmembrane segment and behaves as an amphiphilic peripheral membrane protein - and the multimers assemble in infected cells whether or not viral DNA synthesis is running. Through its N-terminal 42 residues p1 binds the free primer terminal protein, so the terminal protein/DNA polymerase heterodimer that initiates protein-primed replication is recruited to the p1 sheet, giving a membrane-associated multiprotein initiation complex; p1 also concentrates at midcell by associating with the host cell-division protein FtsZ. Loss of p1 severely impairs phage DNA synthesis in vivo, more strongly at 37 degrees C than at 30 degrees C, while an excess of p1 inhibits replication in a dosage-dependent way, as expected of a stoichiometric scaffold. Two products arise from the same open reading frame by alternative in-phase initiation: a 10 kDa form and a 6 kDa form (delta-1) that lacks the first 34 residues and therefore the terminal-protein-binding region. A separate activity reported by one laboratory is binding to the polycistronic mRNA of the phi29 dna operon, through which p1 represses translation of the DNA polymerase and terminal protein genes, coupling the scaffold's accumulation to the supply of the replication proteins it organises.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0003723 RNA binding | IEA GO_REF:0000043 | MODIFY | Summary: Keyword-mapped row for RNA binding. The UniProt RNA-binding keyword on this entry is curated from two experimental papers (entry references 6 and 10), which report that gp1 binds RNA in vitro and represses the upstream phi29 dna genes post-transcriptionally. The claim is sound, but the demonstrated ligand is the polycistronic mRNA of the dna operon, so a more specific term is available. Reason: The row is true and rests on real experiments, not on an unvetted prediction: the keyword traces to Takeuchi et al., who showed that the amount, size and integrity of the dna-operon mRNA are unchanged by gp1 while gp1 itself binds RNA in vitro, and to Hashiyama et al., who tied self-assembly to RNA binding and translational repression. Because the characterised ligand is an mRNA, GO:0003729 mRNA binding - an is_a descendant of GO:0003723 - states the same claim at the granularity the evidence supports, so this is a refinement rather than a correction. The caveat worth keeping in view is provenance rather than quality: both papers come from one laboratory and the activity has not been reproduced elsewhere, which is why the derived regulatory process row below is proposed rather than placed among the core functions. Proposed replacements: mRNA binding Supporting Evidence: PMID:9480849 the RNA binding ability of gp1 was demonstrated by in vitro system PMID:9480849 the dna genes was affected by gp1 through binding to mRNA. PMID:15883018 to be necessary for the efficient binding to RNA and the translational |
| GO:0006260 DNA replication | IEA GO_REF:0000120 | MODIFY | Summary: Combined-IEA row from the phi29 gp1 InterPro family and the DNA replication keyword. p1 is genuinely a replication protein, but the genome it helps replicate is the viral one, and GO separates viral genome replication from cellular DNA replication. Reason: This is a lateral correction rather than a refinement. Checked against the QuickGO ancestors endpoint, GO:0039693 viral DNA genome replication is not a descendant of GO:0006260 DNA replication - its ancestors run through GO:0071897 DNA biosynthetic process and GO:0019079 viral genome replication - so the two rows make different claims, and GO:0006260 asserts participation in replication of cellular DNA, which p1 does not do. The underlying biology is the viral term, which this entry already carries by IDA and IMP, so the row should be re-pointed to GO:0039693 rather than retained as a harmless generalisation or removed as false. The keyword and the family signature are both correct about p1 being a replication protein; only the branch is wrong. Proposed replacements: viral DNA genome replication Supporting Evidence: PMID:9774353 membrane-associated protein required for in vivo viral DNA replication. file:BPPH2/1/1-uniprot.txt provides a specific site for viral DNA replication. |
| GO:0016020 membrane | IEA GO_REF:0000002 | MODIFY | Summary: InterPro2GO row placing p1 at a membrane, from the phi29 gp1 family signature. The location is right in substance - p1 is an amphiphilic peripheral membrane protein whose hydrophobic C-terminus attaches it to the Bacillus subtilis membrane - but for a phage protein the informative term is the host-scoped one, which this entry already carries with direct evidence. Reason: GO:0016020 membrane is a generic component term, not a root organelle term, so the rule that a correct but less specific term should be re-pointed applies. The replacement is lateral rather than a descendant relation: GO:0033644 host cell membrane sits under GO:0033643 host cell part and GO:0018995 host cellular component and, as checked in QuickGO, does not have GO:0016020 among its ancestors. Re-pointing the row therefore removes a claim that could be read as p1 residing in a membrane of its own and replaces it with the one the experiments actually establish. The membrane association itself is well mapped, to the hydrophobic C-terminal 43 residues, so nothing about the row is doubtful except its scoping. Proposed replacements: host cell membrane Supporting Evidence: PMID:9193003 association of protein p1 with membranes was abolished when the C-terminal 43 PMID:12904294 immunoelectron microscopy studies reveal that p1 has a peripheral file:BPPH2/1/1-uniprot.txt SUBCELLULAR LOCATION: Host membrane |
| GO:0016032 viral process | IEA GO_REF:0000002 | TOO GENERAL | Summary: InterPro2GO row asserting that p1 takes part in a viral process. True of every protein encoded by a phage, and in this case already entailed by the specific term this review accepts. Reason: GO:0039693 viral DNA genome replication, accepted on three rows of this entry, is a descendant of GO:0016032 through GO:0019079 viral genome replication and GO:0019058 viral life cycle, so this row adds nothing that the specific claim does not already state. The judgement is about the term, not about the evidence or the truth of the claim: p1 is indeed a participant in a viral process, and the InterPro family signature is a correct mapping. It is marked too general rather than modified because the specific descendant is already asserted elsewhere in this review, so there is no replacement to propose that would not duplicate an existing row. Supporting Evidence: PMID:12904294 protein p1 enhances the rate of in vivo viral DNA replication. |
| GO:0033644 host cell membrane | EXP PMID:11032825 Compartmentalization of phage phi29 DNA replication: interac... | ACCEPT | Summary: Experimental row from Bravo, Illana and Salas, who state in the first sentence of the paper that p1 is membrane associated in infected Bacillus subtilis cells and map the requirement to the C-terminal region: the truncated p1DeltaC43 neither associates with membranes in vivo nor self-interacts in vitro. Reason: The location is directly established in the host during infection, which is exactly what GO:0033644 asserts, and the term is as precise as GO allows for a phage protein on the inner face of the bacterial membrane. There is no sub-compartment term to add: the protein has no transmembrane segment and is peripheral, and GO has no host-scoped term for the cytoplasmic face of the host plasma membrane or for the host division-site region where p1 actually concentrates. This is the location at which the protein's core scaffolding function is performed, so the row is core rather than peripheral. Supporting Evidence: PMID:11032825 associated in Bacillus subtilis-infected cells. PMID:11032825 C-terminally truncated p1 protein that neither associates with membranes in vivo |
| GO:0033644 host cell membrane | IDA PMID:12904294 In vivo assembly of phage phi 29 replication protein p1 into... | ACCEPT | Summary: IDA row from Serrano-Heras, Salas and Bravo, who showed by immunoelectron microscopy that p1 has a peripheral subcellular location and, by in vivo cross-linking combined with cell fractionation, that it assembles into multimers associated with the bacterial membrane. Reason: This is the strongest of the three host-membrane rows and is consistent with them: two independent assays in infected cells, plus the demonstration that p1 is amphiphilic like an integral membrane protein despite having no transmembrane segment. The paper also shows the structures form when DNA synthesis is blocked for lack of other replisome components, so the location is a property of p1 itself and not a consequence of being carried there by the replication machinery. Accepted as core, at the same granularity as the EXP row above. Supporting Evidence: PMID:12904294 immunoelectron microscopy studies reveal that p1 has a peripheral PMID:12904294 cells into multimeric structures that are associated with the bacterial PMID:12904294 protein p1, like integral membrane proteins, has an amphiphilic nature. |
| GO:0033644 host cell membrane | IEA GO_REF:0000044 | ACCEPT | Summary: Subcellular-location keyword mapping of the same claim the two experimental rows above make. The SL-0380 host membrane annotation on this entry is itself evidenced by the two papers behind those rows, so the row is a faithful restatement rather than an independent inference. Reason: Rows sharing a term must be judged consistently, and this one makes the same claim as the EXP and IDA rows, which are accepted. The mapping is sound and the underlying UniProt subcellular-location statement carries experimental evidence tags for PubMed:11032825 and PubMed:12904294 rather than resting on an association rule, so nothing here is a prediction. It is kept as a reinforcing row; this is not a case for removal as a duplicate. Supporting Evidence: file:BPPH2/1/1-uniprot.txt SUBCELLULAR LOCATION: Host membrane file:BPPH2/1/1-uniprot.txt Peripheral membrane protein |
| GO:0039693 viral DNA genome replication | IDA PMID:11032825 Compartmentalization of phage phi29 DNA replication: interac... | ACCEPT | Summary: IDA row for participation in replication of the phage genome. The paper demonstrates that p1 binds the primer terminal protein, that the N-terminal 42 residues are what interfere with formation of the TP-dAMP initiation complex when supplied in excess, and that p1 alters the rate of phi29 DNA replication in vivo in a dosage-dependent manner, and concludes with the model in which the viral replisome attaches to a membrane-associated p1 structure. Reason: p1 does some of the work of this process rather than merely being required for it: it is the structural element that recruits and holds the initiation machinery, and the interaction with the primer terminal protein is shown directly by cross-linking. The dosage-dependent inhibition by excess p1 and by the truncated p1DeltaC43 is the signature of a stoichiometric scaffold, not of a substrate or a bystander. The qualifier involved_in is right for a participant, and the term is at the correct granularity - the viral genome, not cellular DNA replication, and not a downstream outcome such as burst size. Supporting Evidence: PMID:11032825 can interact with the primer terminal protein (TP) PMID:11032825 of phi29 DNA replication in vivo in a dosage-dependent manner. PMID:11032825 attaches to a membrane-associated p1-based structure. |
| GO:0039693 viral DNA genome replication | IEA GO_REF:0000043 | ACCEPT | Summary: Keyword-mapped row from the viral DNA replication keyword, restating the claim the IDA and IMP rows establish. The UniProt FUNCTION statement behind the keyword is curated from four experimental papers. Reason: Consistent with the other two rows on this term, which are accepted on direct and mutant evidence. The keyword is not an unvetted prediction: the FUNCTION comment it summarises carries ECO:0000269 experimental tags for PubMed:9193003, PubMed:9774353, PubMed:11032825 and PubMed:12904294. Kept as a reinforcing row rather than removed as redundant. Supporting Evidence: file:BPPH2/1/1-uniprot.txt provides a specific site for viral DNA replication. PMID:9774353 membrane-associated protein required for in vivo viral DNA replication. |
| GO:0039693 viral DNA genome replication | IMP PMID:820555 Bacillus subtilis phage phi29. Characterization of gene prod... | ACCEPT | Summary: CACAO row from the 1976 Carrascosa et al. survey of phi29 gene products and functions, which scored suppressor-sensitive and temperature-sensitive mutants of each cistron for phage DNA synthesis under restrictive conditions. The paper names gene 1 specifically: sus1(629) is listed among the mutants that did not synthesise phage DNA, and the Discussion states that cistrons 1, 2, 3, 5, 6 and 17 are involved in phage-specific DNA synthesis. The term and the mutant-based evidence code match. Reason: This row was selected to test a predicted one-paper-many-genes failure, on the suspicion that a single multi-gene mutant survey had been spread across the phi29 entries without individually establishing the term for each. The prediction does not hold here. The cached record is abstract-only and its summary sentence says only that mutants in six genes do not synthesise phage DNA, but the full text (Eur. J. Biochem. 66:229-241, retrieved externally for this review and quoted in the gene notes rather than here, since supporting text must be verbatim from the cached source) names this locus three times: in the Results, among the mutants that did not synthesise phage DNA alongside ts6, sus6 and sus17; in the Discussion, in the explicit list of the six DNA-synthesis cistrons as 1, 2, 3, 5, 6 and 17; and again among the cistrons whose products were characterised as early proteins. The identification of sus1 with this gene is independently secure, because Prieto et al. later sequenced the same sus1(629) allele, found an ochre codon in ORF-6, and concluded that ORF-6 is gene 1 - so this is not an adjacent-locus mis-attribution. The qualifier acts_upstream_of_or_within_positive_effect is also the appropriate level of caution for a 1976 mutant survey that scored the outcome rather than the mechanism, and it is the more honest reading of a phenotype that is severe at 37 degrees C but mild at 30 degrees C. Supporting Evidence: PMID:820555 Mutants in six genes do not synthesize phage DNA. PMID:2526779 29 protein p1 is needed for the viral DNA replication. PMID:2526779 indicating that this ORF corresponds to gene 1. |
| GO:0005198 structural molecule activity | IDA PMID:9774353 Polymerization of bacteriophage phi 29 replication protein p... | NEW | Summary: Proposed new molecular function row. p1 has no catalytic activity; what it does is polymerise into protofilaments that pack into highly ordered two-dimensional sheets, shown directly by sedimentation assays and negative-stain electron microscopy, and those sheets are the structure that holds the replication machinery at the host membrane. The assembly determinants are mapped to individual coiled-coil residues, and single substitutions there abolish sheet formation. Reason: GOA records what p1 is involved in and where it sits but gives it no molecular function other than RNA binding, which is not the activity through which it supports replication. GO:0005198 structural molecule activity is the term for a molecule that contributes to the structural integrity of an assembly, which is precisely what the direct evidence shows: self-association into ordered polymers whose formation depends on identified residues, with loss of assembly and loss of replication support tracking together. The term is proposed rather than a more specific child because GO has no structural-constituent term for a viral replication platform, and rather than left implicit because the molecular-function slot is where this review's core function has to be expressed. It is not a duplicate of any existing row, being in a different aspect from the process and component rows and not an ancestor or descendant of GO:0003723. Supporting Evidence: PMID:9774353 propose that protein p1 is a component of a viral-encoded structure that PMID:9774353 protofilaments that associated in a highly ordered, parallel array forming large PMID:11283004 demonstrate that residues Leu(46), Met(53), and Leu(60), but not Leu(39), are |
| GO:0017148 negative regulation of translation | IDA PMID:9480849 Gene 1, one of the dna genes of bacteriophage phi 29, repres... | NEW | Summary: Proposed new row for the regulatory activity that the mRNA-binding term above enables. Takeuchi et al. showed that gp1 represses expression of the upstream dna genes for the DNA polymerase and the primer terminal protein while leaving the amount, size and integrity of the polycistronic mRNA unchanged, and that gp1 binds RNA in vitro, concluding that gp1 acts on translation of those genes by binding their mRNA. Hashiyama et al. reproduced the repression and tied it to gp1 self-assembly. Reason: The entity that performs the step is p1 itself: it binds the transcript and the repression survives with the transcript intact, so this is participation rather than necessity, and the existing RNA-binding row supplies the molecular function without the process it serves being recorded anywhere in GOA. The term is proposed at the regulation-of-translation level rather than a more specific child because the mechanism - occlusion of initiation, altered ribosome loading, or something else - was not resolved. One caveat is recorded deliberately: both papers come from a single laboratory, the Salas group's extensive work on p1 never examined this activity, and no structural or in vivo reporter follow-up exists, so the row is offered as a proposal for a curator rather than carried into core_functions, and a corroboration experiment is suggested below. Supporting Evidence: PMID:9480849 of the polycistronic mRNA were not influenced by the presence of gene 1. PMID:9480849 the dna genes was affected by gp1 through binding to mRNA. PMID:15883018 suppresses the synthesis of phi29 DNA polymerase and primer protein in the |
Loading supporting content…
Download this section (compressed HTML)Q: Does the p1 sheet contribute anything to initiation beyond concentrating the terminal protein and DNA polymerase at the membrane - for example activating the heterodimer or juxtaposing the two genome ends - or is proximity the whole mechanism? The distinction decides whether a scaffolding molecular function is the right description or whether a regulatory activity should also be annotated.
Suggested experts: Daniel Munoz-Espin
Q: Has the mRNA-binding and translational-repression activity of gp1 been reproduced outside the laboratory that reported it, and is it a property of the full-length 10 kDa form, the 6 kDa delta-1 form, or both? The RNA-binding keyword on P03679, and the GOA row derived from it, rest on two papers from one group.
Q: GO has no host-scoped component term for the bacterial division site or the host FtsZ ring, which is where p1 actually concentrates, nor for the cytoplasmic face of the host plasma membrane. Should such terms exist under GO:0033643 host cell part, or is GO:0033644 host cell membrane the intended stopping point for phage proteins of this kind?
Q: Is the 6 kDa delta-1 form a distinct functional product or a translation artefact? It lacks the N-terminal 34 residues, and therefore the region shown to bind the primer terminal protein, while retaining the coiled coil and the membrane-anchoring C-terminus, so it is predicted to polymerise and sit on the membrane without recruiting the replisome - a dominant-negative architecture that has never been tested as such.
Experiment: Express the 10 kDa and 6 kDa forms separately and together from plasmids in Bacillus subtilis, measure sheet formation by negative-stain electron microscopy and sedimentation and membrane association by fractionation, assay terminal-protein binding by cross-linking, and score complementation of a sus1 phage and inhibition of wild-type phage DNA synthesis. An internal-initiation-dead variant that preserves the p1 coding sequence would separate the two products genetically.
Hypothesis: The 6 kDa delta-1 product of gene 1 assembles into membrane-associated sheets but cannot recruit the primer terminal protein, and so acts as a dominant-negative regulator of phi29 replication rather than as a redundant copy of p1.
Type: structure-function and complementation analysis
Experiment: Repeat the repression independently with modern methods: ribosome profiling of phi29-infected Bacillus subtilis comparing wild-type with sus1 infection at equal mRNA levels, CLIP or RIP-seq from infected cells to locate the gp1 binding site on the polycistronic transcript, and in vitro translation of the dna-operon mRNA with purified gp1 titrated against the assembly-deficient coiled-coil substitutions. Separating transcript level from ribosome occupancy is the key control the original agarose-gel and S1-protection assays could only approximate.
Hypothesis: gp1 represses translation of the phi29 dna operon by binding its mRNA, and this is a genuine second function of the protein rather than an artefact of the overexpression systems used in the original reports.
Type: ribosome profiling, RNA-protein crosslinking and in vitro translation
Experiment: Isolate or design p1 variants that still polymerise and bind membranes but no longer co-localise with FtsZ, then follow replication in single cells by time-lapse imaging of a labelled phage genome alongside FtsZ-ring assembly, in wild-type cells and in cells where division is blocked at defined steps. Pairing this with quantitative measurement of phi29 DNA accumulation would show whether the FtsZ association sets the timing, the site, or only the efficiency of replication.
Hypothesis: Association with the host FtsZ ring times phi29 replication to the host division cycle, so that uncoupling p1 from FtsZ delays or desynchronises genome replication without abolishing it.
Type: live-cell imaging with separation-of-function mutants
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)