Endolysin of Staphylococcus phage K (genus Kayvirus), a 495-residue, 54 kDa modular peptidoglycan hydrolase encoded late in infection and acting on the murein sacculus of its staphylococcal host. LysK is built from three modules: an N-terminal CHAP (cysteine, histidine-dependent amidohydrolase/peptidase) domain with a papain-like fold and a Cys54/His117/Glu134 catalytic triad, which hydrolyses the D-alanyl-glycyl bond between the D-alanine of the tetrapeptide stem and the first glycine of the pentaglycine cross-bridge (EC 3.4.22.-); a central Amidase_2 domain that cleaves the amide bond between N-acetylmuramic acid and L-alanine of the stem peptide (N-acetylmuramoyl-L-alanine amidase, EC 3.5.1.28); and a C-terminal SH3b cell wall-binding domain that docks the enzyme on staphylococcal peptidoglycan and is needed for full lytic activity. The CHAP domain carries a buried structural Ca(2+) ion essential for activity and a more loosely bound Zn(2+) ion at the active site; a further Zn(2+) site is inferred in the amidase domain. Cutting two bonds in the same glycopeptide collapses the cell wall, and in the phage life cycle this is the terminal lytic step that ruptures the infected cell and releases progeny virions. Because staphylococcal peptidoglycan is accessible from the outside, LysK also lyses untreated cells when applied exogenously, killing coagulase-positive and coagulase-negative staphylococci including MRSA and vancomycin-resistant strains while sparing other Gram-positive genera - the basis of interest in it as an enzybiotic.
Definition: Catalysis of the hydrolysis of the peptide bond between the D-alanine of the peptidoglycan stem peptide and the glycine that begins the interpeptide cross-bridge.
Justification: The bond specificity of the CHAP domains of staphylococcal phage endolysins (LysK, phi11 LytA) and of related CHAP-family peptidoglycan hydrolases is established by mass spectrometry of digested peptidoglycan. GO has no term for it: the nearest available terms are GO:0004197 cysteine-type endopeptidase activity, which names the mechanism but not the substrate, and GO:0008745, which is the other, amidase activity of the same protein. A term for the D-Ala-Gly bond would let the two activities of bifunctional endolysins be distinguished without reaching for a mechanism-only parent, and would apply to a well-populated family (CHAP, MEROPS C51, Pfam PF05257). It is a sibling of the glycyl-glycine endopeptidase chemistry of lysostaphin, which is likewise unnamed in GO.
Parent term: cysteine-type endopeptidase activity
Supporting Evidence:
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0001897 symbiont-mediated cytolysis of host cell | IDA PMID:16199588 The recombinant phage lysin LysK has a broad spectrum of lyt... | ACCEPT | Summary: CACAO row. Recombinant LysK expressed in Lactococcus lactis forms a 54 kDa lytic zone on a zymogram containing MRSA cells (Fig. 1A, lane 5) and, in a kill curve, reduces live MRSA DPC5645 by 99% within one hour (Fig. 1B). The gene product lyses cells of its staphylococcal host, which is what the term asserts. Reason: CACAO assignment, experimental (IDA), full text available, and the term matches what the paper shows for this gene. PMID:16199588 demonstrates, on this protein, both cell-wall degradation (zymogram clearing at the predicted mass of LysK, absent from vector controls) and killing of live cells (Fig. 1B kill curve), i.e. cytolysis of host cells. Notably the student chose a term on the interspecies-interaction branch - GO:0001897 is under symbiont-mediated killing of host cell and cytolysis in another organism - which is the correct direction for a phage-encoded lysin acting on its host, and is exactly what the two electronic rows on this entry (GO:0009253, GO:0042742) get wrong. This is not a phenotype read as a process: LysK itself performs the peptidoglycan hydrolysis that ruptures the cell, so the molecular step and the process term coincide. The one caveat, which does not affect the action, is that the assay applies the enzyme from the outside rather than following an infection; the term speaks only of cytolysis of a host cell, which exogenous lysis of S. aureus by a phage K protein satisfies. Supporting Evidence: PMID:16199588 These results confirmed that recombinant LysK from lactococci is enzymatically active and capable of degrading staphylococcal cell walls. PMID:16199588 In kill curves using a human MRSA strain (DPC5645), a 99% reduction in staphylococcal cell numbers was observed 1 h after the addition of lysates containing LysK (Fig. 1B ), demonstrating that recombinant LysK is capable of killing live pathogenic staphylococci. |
| GO:0001897 symbiont-mediated cytolysis of host cell | IDA PMID:19493008 LysK CHAP endopeptidase domain is required for lysis of live... | ACCEPT | Summary: CACAO row on the second paper. Full-length LysK and a CHAP-SH3b fusion lyse untreated Staphylococcus aureus and coagulase-negative staphylococci; CHAP-domain truncations lyse S. aureus in zymogram, plate-lysis and turbidity-reduction assays. Same claim as the row above, from an independent set of assays. Reason: CACAO assignment, experimental (IDA), and the term matches the paper's assays for this gene. The cached record is abstract-only (full_text_available: false), but the abstract itself states the live-cell lysis result for LysK by name, so nothing here rests on evidence that could not be seen; the GONUTS wiki page for this PMID, which would record the student's figure-level reasoning, returned only "This site is temporarily offline while we perform upgrades" when queried on 2026-10-29. Rows sharing a term reinforce one another, so this row and the PMID:16199588 row are two independent demonstrations of one claim rather than a duplication to be pruned. No REMOVE is entertained for an experimental row whose full text has not been read. Supporting Evidence: PMID:19493008 Both LysK and the CHAP-SH3b fusion were shown to lyse untreated S. aureus and the coagulase-negative strains. PMID:19493008 Truncations of the LysK containing just the CHAP domain lyse Staphylococcus aureus cells in zymogram analysis, plate lysis, and turbidity reduction assays |
| GO:0001897 symbiont-mediated cytolysis of host cell | IEA GO_REF:0000117 | ACCEPT | Summary: ARBA machine-learning rule assigning the same host-cytolysis term that the two CACAO IDA rows carry. The prediction coincides with the experimentally demonstrated claim. Reason: The claim is correct for this gene and is experimentally established by PMID:16199588 and PMID:19493008, so this row reinforces the IDA rows rather than adding an unverified prediction; rows sharing a term must be given consistent actions, and the two IDA rows are accepted. The ARBA rule itself is unvetted and is not cited as support anywhere in this review - the support quoted below is the primary data. The only thing worth noting is that ARBA reached the right branch here while the keyword-mapping and InterPro2GO rules on the same entry did not. Supporting Evidence: PMID:16199588 In kill curves using a human MRSA strain (DPC5645), a 99% reduction in staphylococcal cell numbers was observed 1 h after the addition of lysates containing LysK (Fig. 1B ), demonstrating that recombinant LysK is capable of killing live pathogenic staphylococci. file:BPPGK/lysK/lysK-uniprot.txt Endolysin that degrades host peptidoglycans and participates in the sequential events which lead to the programmed host cell lysis releasing the mature viral particles |
| GO:0003824 catalytic activity | IEA GO_REF:0000043 | TOO GENERAL | Summary: Root molecular-function term from the Antimicrobial and Multifunctional enzyme keywords. True but carries no information beyond the specific activities already asserted on this entry. Reason: GO:0003824 is an ancestor of both GO:0008745 N-acetylmuramoyl-L-alanine amidase activity (accepted below) and GO:0004197 cysteine-type endopeptidase activity (proposed as the replacement for the peptidase rows), so asserting it adds nothing once those are asserted. This judges the term, not the keyword mapping: LysK is certainly catalytic, and indeed bifunctional, which is why KW-0511 is on the entry. Supporting Evidence: PMID:19493008 Both catalytic domains are active on purified peptidoglycan by positive-ion electrospray ionization MS. |
| GO:0006508 proteolysis | IEA GO_REF:0000043 | KEEP AS NON CORE | Summary: Generic proteolysis term from the Protease keyword. The CHAP domain does hydrolyse a peptide bond - the D-alanyl-glycyl bond of the peptidoglycan cross-bridge - so the claim holds, but the informative content of that chemistry sits in the molecular function rows and in the host-directed cell-wall terms. Reason: The claim is true and, within the proteolysis branch, as precise as GO allows: there is no child of GO:0006508 for peptidoglycan stem-peptide or cross-bridge cleavage, so there is no more specific term to MODIFY to. It is kept rather than removed, but it is not where this gene's biology lives - GO:0006508 is a substrate-agnostic metabolic term, whereas what LysK actually does is disrupt the cell wall of another organism (GO:0052009, proposed below) in the service of host cytolysis (GO:0001897, accepted above). Marked non-core rather than too general because neither of those terms is a descendant of proteolysis, so it is not formally entailed by anything else in this review. Supporting Evidence: PMID:25064136 the CHAP domain hydrolyzes it between the D-alanine of the tetra-peptide stem and the first glycine of the penta-glycine cross-bridge |
| GO:0008233 peptidase activity | IEA GO_REF:0000043 | MODIFY | Summary: Keyword-derived peptidase term for the CHAP endopeptidase domain. Correct, but less specific than the structural and mutagenesis evidence supports: CHAP(K) is a papain-fold cysteine endopeptidase with a Cys54/His117/Glu134 triad. Reason: The 1.69 A structure of CHAP(K) shows a papain-like topology with the catalytic triad in a hydrophobic cleft, and alanine substitution of Cys54 or His117 eliminates staphylolytic activity entirely while E134A strongly reduces it (PMID:25064136), which identifies the mechanistic class. The domain is additionally classified as Peptidase C51 by PROSITE-ProRule and the entry's SUPFAM assignment is "Cysteine proteinases". GO:0004197 cysteine-type endopeptidase activity is a descendant of GO:0008233 and is the most specific term the ontology offers: GO has no term for the actual bond specificity (D-alanyl-glycyl endopeptidase, EC 3.4.22.-), which is filed as a proposed new term in this review. The same replacement is applied to the IMP row on this term, so the two rows stay consistent. Proposed replacements: cysteine-type endopeptidase activity Supporting Evidence: PMID:25064136 The structure of CHAP(K) revealed a papain-like topology with a hydrophobic cleft, where the catalytic triad is located. PMID:25064136 Mutation of the conserved Cys54 and His117 residues to alanine resulted in complete elimination of staphylolytic activity of the enzyme as demonstrated by zymographic analysis, indicating an essential role of these residues and supporting the hypothesis that they are part of the catalytic triad. file:BPPGK/lysK/lysK-uniprot.txt /note="Peptidase C51" |
| GO:0008233 peptidase activity | IMP PMID:19493008 LysK CHAP endopeptidase domain is required for lysis of live... | MODIFY | Summary: CACAO row. The paper's central result is a domain-dissection experiment: truncations retaining only the CHAP endopeptidase domain lyse live S. aureus in zymogram, plate lysis and turbidity assays, whereas amidase-only truncations are barely active. That shows an endopeptidase activity resident in the CHAP domain, and the abstract also names the cleaved bond. Reason: CACAO assignment, experimental (IMP), and the evidence code fits what was done: the inference comes from engineered truncations and their altered lytic phenotype, not from a direct assay of the full-length protein, so IMP is the right code and the student read the figure set correctly. The term is the weak point, and the failure is granularity rather than sense - peptidase activity is the top of the branch, while the paper (D-Ala/Gly cut site), the crystal structure (papain fold, Cys54/His117/Glu134) and the Peptidase C51 domain assignment together fix the activity as a cysteine-type endopeptidase. MODIFY to GO:0004197, a descendant of the annotated term. Two things this row is not: it is not a phenotype read as an activity, because the lytic readout is the direct consequence of the hydrolysis the domain performs; and it is not mis-attributed, since the paper is about this protein and names it in its title. No REMOVE: the cached record is abstract-only and the GONUTS page for the PMID was offline (2026-10-29), so the student's full-text reading cannot be checked, and the abstract already supports the substance. Proposed replacements: cysteine-type endopeptidase activity Supporting Evidence: PMID:19493008 The cut sites are identical to LytA (phi11 endolysin), with cleavage between d-alanine of the stem peptide and glycine of the cross-bridge peptide, and N-acetylmuramoyl-l-alanine amidase activity. PMID:19493008 In contrast, truncations harboring just the amidase lytic domain show faint activity in both the zymogram and turbidity reduction assays, but no detectable activity in either plate lysis or MIC assays. PMID:25064136 The structure of CHAP(K) revealed a papain-like topology with a hydrophobic cleft, where the catalytic triad is located. |
| GO:0008745 N-acetylmuramoyl-L-alanine amidase activity | IDA PMID:19493008 LysK CHAP endopeptidase domain is required for lysis of live... | ACCEPT | Summary: CACAO row. Positive-ion electrospray ionisation MS of purified peptidoglycan digested by LysK identifies N-acetylmuramoyl-L-alanine amidase activity alongside the endopeptidase cut site - a direct assay of the reaction the term names. Reason: CACAO assignment, experimental (IDA), and the term matches the assay exactly, at exactly the right granularity: the MS experiment identifies the cleaved bond, which is what distinguishes EC 3.5.1.28 from the neighbouring muralytic chemistries, and GO:0008745 is the EC-equivalent term. No downstream phenotype stands in for the activity here; the substrate is purified peptidoglycan and the readout is the reaction product. UniProt independently curates this as the entry's catalytic activity with ECO:0000269 evidence from this same paper, so the curated record and the student row agree. This is the best of the four CACAO rows on this gene. Supporting Evidence: PMID:19493008 Both catalytic domains are active on purified peptidoglycan by positive-ion electrospray ionization MS. PMID:19493008 The cut sites are identical to LytA (phi11 endolysin), with cleavage between d-alanine of the stem peptide and glycine of the cross-bridge peptide, and N-acetylmuramoyl-l-alanine amidase activity. file:BPPGK/lysK/lysK-uniprot.txt Reaction=Hydrolyzes the link between N-acetylmuramoyl residues and L-amino acid residues in certain cell-wall glycopeptides. |
| GO:0008745 N-acetylmuramoyl-L-alanine amidase activity | IEA GO_REF:0000120 | ACCEPT | Summary: Combined automated annotation from the Amidase_2 domain signatures and EC 3.5.1.28. Unusually for an IEA, its input is experimentally grounded: the EC number on this entry carries ECO:0000269 evidence from PMID:19493008. Reason: The domain-to-function inference is correct and independently confirmed by the MS assay in PMID:19493008, so this row reinforces the IDA row for the same term. The central Amidase_2 domain (residues 205-334) is the structural basis, and family and domain classification is legitimate support for membership. Consistent with the IDA row, which is accepted. Supporting Evidence: PMID:19493008 Both catalytic domains are active on purified peptidoglycan by positive-ion electrospray ionization MS. file:BPPGK/lysK/lysK-uniprot.txt InterPro; IPR002502; Amidase_domain. |
| GO:0009253 peptidoglycan catabolic process | IEA GO_REF:0000002 | MODIFY | Summary: InterPro2GO mapping from the Amidase_2 signatures to an own-process metabolic term. The chemistry is right and the direction of the interaction is wrong: GO:0009253 describes an organism breaking down its own peptidoglycan, and a phage has none. Reason: Checked in OLS, GO:0009253 sits wholly inside the metabolic branch - peptidoglycan metabolic process, aminoglycan catabolic process, macromolecule catabolic process, all under GO:0009987 cellular process - with no interspecies parent anywhere in its ancestry. It therefore asserts that Staphylococcus phage K turns over peptidoglycan of its own, which is false: LysK hydrolyses the sacculus of its staphylococcal host. GO has the matching host-directed term, GO:0052009 symbiont-mediated disruption of host cell wall, on the interspecies-interaction branch, and replacing the term keeps the substance while putting it in the right branch. This is the same error the CACAO batch-1 review found on PRD1 endolysin P15 with GO:0044277 cell wall disassembly, and it is instructive that here it arrives by InterPro2GO rather than from a student: the domain signature cannot know whose wall is being degraded, so an own-process mapping from a muralytic domain will mis-direct every phage, type-VI-secretion and bacteriocin protein that carries it. The general fix is a taxon-aware mapping, or a taxon constraint on GO:0009253 excluding viruses. Proposed replacements: symbiont-mediated disruption of host cell wall Supporting Evidence: PMID:16199588 These results confirmed that recombinant LysK from lactococci is enzymatically active and capable of degrading staphylococcal cell walls. file:BPPGK/lysK/lysK-uniprot.txt Endolysin that degrades host peptidoglycans and participates in the sequential events which lead to the programmed host cell lysis releasing the mature viral particles |
| GO:0016787 hydrolase activity | IEA GO_REF:0000043 | TOO GENERAL | Summary: Root hydrolase term from KW-0378. True of both catalytic domains, and subsumed by the specific activities asserted elsewhere in this review. Reason: GO:0016787 is an ancestor of GO:0008745 (accepted) and of GO:0004197 (the proposed replacement for the peptidase rows). The claim is correct - LysK hydrolyses an amide bond and a peptide bond - but uninformative once those are asserted. The term is being judged here, not the keyword mapping, which is accurate. Supporting Evidence: PMID:19493008 Both catalytic domains are active on purified peptidoglycan by positive-ion electrospray ionization MS. |
| GO:0031640 killing of cells of another organism | IEA GO_REF:0000043 | TOO GENERAL | Summary: Keyword-derived host-killing term. LysK does kill cells of another organism, but the specific term for how and why - symbiont-mediated cytolysis of the host cell - is accepted above and sits below this one. Reason: GO:0001897, accepted on two experimental rows, is a descendant of GO:0031640 via both symbiont-mediated killing of host cell and cytolysis in another organism (checked in OLS). The broad term is true but loses the two things that matter: that the killing is lysis of the cell wall, and that the organism killed is this phage's own host rather than an incidental target. Judging the term, not the row's evidence; the Antimicrobial and Cytolysis keywords are both appropriate for this entry. Supporting Evidence: PMID:16199588 In kill curves using a human MRSA strain (DPC5645), a 99% reduction in staphylococcal cell numbers was observed 1 h after the addition of lysates containing LysK (Fig. 1B ), demonstrating that recombinant LysK is capable of killing live pathogenic staphylococci. |
| GO:0042742 defense response to bacterium | IEA GO_REF:0000043 | REMOVE | Summary: KW-0081 (Antimicrobial) mapped to a host-defence process term. The direction of the interspecies interaction is inverted: phage K attacks the bacterium, it does not defend itself against one. Reason: GO:0042742 denotes a reaction mounted to protect the annotated organism from a bacterium. LysK degrades the peptidoglycan of its own host in order to release progeny virions, and is of interest as an antibacterial precisely because it attacks staphylococci; there is no defence and no bacterium being resisted by phage K. This is a demonstrably wrong electronic keyword-to-GO mapping - the Antimicrobial keyword covers both host antimicrobial peptides and phage lytic enzymes, and only the former fits the defence framing - which the guidelines list as legitimate ground for REMOVE. The real content of the keyword is already carried, correctly, by GO:0031640 and GO:0001897. The identical row was removed on PRD1 endolysin P15 in CACAO batch 1 and reported upstream; this is a second instance, so the defect is systematic across phage-encoded lytic enzymes rather than a one-off. Supporting Evidence: file:BPPGK/lysK/lysK-uniprot.txt Endolysin that degrades host peptidoglycans and participates in the sequential events which lead to the programmed host cell lysis releasing the mature viral particles PMID:16199588 In kill curves using a human MRSA strain (DPC5645), a 99% reduction in staphylococcal cell numbers was observed 1 h after the addition of lysates containing LysK (Fig. 1B ), demonstrating that recombinant LysK is capable of killing live pathogenic staphylococci. |
| GO:0046872 metal ion binding | IEA GO_REF:0000043 | MODIFY | Summary: Generic metal-binding term from KW-0479. The identities of the metals are known: the CHAP domain crystal structures resolve a buried structural Ca(2+) ion whose ligands are individually required for activity, and a more loosely bound Zn(2+) at the active site, both confirmed by X-ray fluorescence; UniProt infers a further Zn(2+) site in the amidase domain. Reason: Correct but less specific than the evidence supports, so MODIFY with descendant terms rather than ACCEPT. GO:0005509 calcium ion binding is experimentally established: the 1.69-1.79 A structures (PDB 4CSH, 4CT3) show Ca(2+) coordinated by Asp45, Asp47, Tyr49, His51 and Asp56, and alanine substitution of Asp45, Asp47 or Asp56 abolishes staphylolytic activity (PMID:25064136). GO:0008270 zinc ion binding is also supported, though less firmly: the structures place a Zn(2+) at the catalytic Cys54 but describe it as loosely bound, and UniProt adds a by-similarity Zn(2+) site at 214/324/332 in the amidase domain with a note that Zn(2+) is required for amidase activity. Both are listed, as the guidelines direct when several specific descendants are supported; calcium is the one carried into core_functions, since it is the metal shown to be required. Proposed replacements: calcium ion binding zinc ion binding Supporting Evidence: PMID:25064136 CHAP(K) contains an additional lobe in its N-terminal domain, with a structural calcium ion, coordinated by residues Asp45, Asp47, Tyr49, His51 and Asp56. PMID:25064136 Zymogram analysis demonstrated that mutation of residues Asp45, Asp47 and Asp56 resulted in the complete abolishment of the staphylolytic activity of the enzyme PMID:25064136 The presence of a zinc ion in the active site was also apparent, coordinated by the catalytic residue Cys54 and a possible substrate analogue. file:BPPGK/lysK/lysK-uniprot.txt Note=Zn(2+) is required for amidase activity. |
| GO:0042834 peptidoglycan binding | IMP PMID:19493008 LysK CHAP endopeptidase domain is required for lysis of live... | NEW | Summary: Proposed new annotation for the C-terminal SH3b domain, the module that docks LysK on staphylococcal peptidoglycan. Fusing the SH3b domain back onto the isolated CHAP domain restores near full-length lytic activity, which the authors read as a requirement for a C-terminal binding domain. Reason: This is a distinct molecular function of the protein, missing from GOA, and LysK itself performs it: the SH3b module binds the substrate on which the two catalytic domains act, which is why its removal costs activity and its restoration recovers it. The structural basis is explicit in the entry - an SH3b domain at residues 412-481 with a PROSITE-ProRule assignment and a solution NMR structure of that fragment (PDB 5O1Q) - and SH3b is the canonical staphylococcal cell-wall-binding module. The evidence code is IMP because the claim rests on engineered domain deletion and reconstruction rather than a binding assay; a direct binding measurement on the isolated SH3b domain would be the way to upgrade it to IDA, and is listed in suggested_experiments. Conservative in scope: no process term is proposed from this, only the binding function. Supporting Evidence: PMID:19493008 A fusion of the CHAP domain to the SH3b domain has near full-length LysK lytic activity, suggesting the need for a C-terminal binding domain. file:BPPGK/lysK/lysK-uniprot.txt The SH3b domain increases the CHAP domain activity file:BPPGK/lysK/lysK-uniprot.txt /note="SH3b" |
| GO:0044659 viral release from host cell by cytolysis | IC PMID:16199588 The recombinant phage lysin LysK has a broad spectrum of lyt... | NEW | Summary: Proposed new annotation for the life-cycle role: LysK is the lytic enzyme of a virulent staphylococcal phage, expressed late in infection, whose peptidoglycan hydrolysis ruptures the infected cell and liberates progeny virions. This is the term under GO:0001897 that names why the cytolysis happens. Reason: The entry carries both the Host cell lysis by virus and Viral release from host cell keywords, and UniProt curates the role explicitly, yet no GO:0044659 row reached GOA - a gap in the keyword mapping rather than a disagreement. LysK passes the participation test: it performs the peptidoglycan hydrolysis step that ruptures the cell, it is not merely required for or consumed by the process. The evidence code is deliberately IC, an inference this review authors: neither cached paper follows an infection, both apply the enzyme from outside, so the release role rests on the enzyme's demonstrated muralytic activity plus the late timing of the lysK transcript (10-20 min post-infection, reported as data not shown) and on UniProt's own "(Probable)" assessment. The row is therefore proposed, and the missing holin genetics for phage K is recorded as a knowledge gap rather than papered over. Kept to the one term: no separate GO:0019076 or holin-related claim is proposed. Supporting Evidence: file:BPPGK/lysK/lysK-uniprot.txt Endolysin that degrades host peptidoglycans and participates in the sequential events which lead to the programmed host cell lysis releasing the mature viral particles PMID:16199588 Reverse transcription-PCR results demonstrated that the lysK transcript appears between 10 and 20 min after phage infection (data not shown). PMID:25064136 Los bacteriófagos codifican endolisinas para lisar sus bacterias hospedadoras y permitir la liberación de su progenie. |
Loading supporting content…
Download this section (compressed HTML)Q: Should GO carry a taxon constraint excluding viruses from GO:0009253 peptidoglycan catabolic process and the other own-process cell-wall terms, so that domain-based mappings from muralytic signatures cannot place phage endolysins, bacteriocins and secreted effectors on the metabolic branch?
Suggested experts: GO ontology editors
Q: Should the UniProtKB KW-0081 (Antimicrobial) to GO:0042742 mapping be restricted by taxon, or the keyword split, given that it is now documented to invert the direction of the interaction on at least two phage endolysins (LysK and PRD1 P15)?
Suggested experts: UniProt keyword curators
Q: Do the two catalytic domains of LysK act cooperatively or independently on intact staphylococcal peptidoglycan, and does cutting both bonds in the same glycopeptide matter for lysis, or is the endopeptidase cut sufficient?
Suggested experts: peptidoglycan biochemists
Q: Is the active-site zinc of CHAP(K) a catalytic cofactor, a regulator of substrate access to Cys54, or a crystallisation artefact?
Suggested experts: structural enzymologists
Experiment: Measure binding of the isolated SH3b domain (residues 412-481, for which a solution NMR structure exists, PDB 5O1Q) to purified S. aureus peptidoglycan and to peptidoglycan stripped of wall teichoic acid, by cell-wall pulldown and by surface plasmon resonance or isothermal titration calorimetry against soluble muropeptides. A measured affinity would upgrade the proposed GO:0042834 annotation from IMP to IDA and would show whether the determinant recognised is the glycan, the pentaglycine cross-bridge, or the teichoic acid.
Hypothesis: The SH3b domain of LysK binds staphylococcal peptidoglycan directly, which is why it raises the activity of the CHAP domain.
Type: BINDING_ASSAY
Experiment: Identify candidate holin genes in the phage K genome by membrane-topology prediction, then test a plasmid-borne lysis cassette in S. aureus as was done for PRD1: express the candidate holin alone, LysK alone, and both together, and follow optical density, potassium efflux and membrane potential. In parallel, construct a lysK-defective phage K (or suppress it with an antisense construct) and compare lysis timing and burst size with the wild type. This is the experiment that would convert the proposed GO:0044659 row from IC to IDA.
Hypothesis: Access of LysK to the host sacculus during infection requires a phage K holin, and LysK is the enzyme that performs the terminal lysis step in vivo.
Type: GENETIC_SCREEN
Experiment: Determine the specific activity of apo-CHAP(K), prepared by chelation and desalting, in a turbidimetric lysis assay, and titrate back Zn(2+) and Ca(2+) separately and together over a concentration range, with inductively coupled plasma MS to confirm the metal content of each preparation. The earlier metal-dependence assay did not strip metals before testing, so a true apoenzyme baseline is what is missing. A stimulatory-then-inhibitory zinc profile would support the regulatory model; a strict requirement would make it catalytic.
Hypothesis: The zinc ion seen beside Cys54 in the CHAP(K) structure is regulatory rather than catalytic.
Type: ENZYME_ASSAY
Experiment: Digest purified S. aureus peptidoglycan with full-length LysK, with the active-site mutants C54A (endopeptidase-dead) and with an Amidase_2 active-site mutant, and resolve the muropeptide products by reversed-phase HPLC and mass spectrometry; in parallel measure turbidity reduction and MIC for each enzyme on live cells. Comparing the product profiles of the single mutants with the double digest shows whether the two activities are independent, and comparing their lytic potency tests the sufficiency claim that the truncation experiments imply.
Hypothesis: The CHAP endopeptidase and the Amidase_2 domain cut the same glycopeptide, and the endopeptidase cut alone is sufficient for lysis of live cells.
Type: ENZYME_ASSAY
What is not known — curated, literature-grounded statements of the open unknowns (the inverse of core functions).
Gap: How LysK reaches the sacculus during infection is unknown for phage K. Endolysins of other phages depend on a holin to permeabilise the cytoplasmic membrane, but no holin has been identified or tested in phage K, and no lysis-timing experiment has been done: every published assay applies LysK from the outside, so the terminal lysis step has never been observed in an infected cell.
OPEN BIOLOGY
Significance: This is the evidential basis on which GO:0044659 viral release from host cell by cytolysis rests, which is why that row is proposed as an IC inference rather than an experimental annotation.
Gap: Why the CHAP-only truncation lyses cells in zymogram, plate-lysis and turbidity assays yet has no detectable activity in a minimal inhibitory concentration assay is not understood, while the CHAP-SH3b fusion is active in all of them. Whether this reflects potency, stability in growth medium, or substrate accessibility in growing versus stationary cells is untested.
OPEN BIOLOGY
Significance: Determines whether the SH3b domain contributes only substrate affinity or something further, such as protection from proteolysis or processivity on intact cells, and bears directly on the design of truncated enzybiotics.
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)