gp63 (leishmanolysin, MSP) — Leishmania major — curation notes

Note on provenance: automated deep research was NOT available in this environment
(no working provider credentials), so no -deep-research-*.md file exists for this gene.
What follows is a manual literature review done by reading the UniProt record, the
cached publications under publications/, and primary papers located via PubMed
(E-utilities) and cached with ai-gene-review fetch-pmid. Every PMID cited below was
verified against PubMed before caching. Quotes are verbatim from the cached records.

1. Identity and structure

• UniProt P08148, GP63_LEIMA, "Leishmanolysin", EC 3.4.24.36, 602 aa precursor;
  signal peptide 1–39, activation propeptide 40–100, mature chain 101–577, C-terminal
  propeptide 578–602 removed on GPI attachment (UniProt FT lines).
• Zinc metalloendopeptidase of the peptidase M8 family (PANTHER:PTHR10942,
  Pfam PF01457, MEROPS M08.001). Crystal structure 1LML at 1.86 Å
  PMID:9739094;
  the fold places it in the metzincins PMID:9739094.
• Catalytic HEXXH motif; Glu-265 is the general base
  PMID:8626468.
  UniProt COFACTOR: "Binds 1 zinc ion per subunit." (ECO:0000269|PubMed:9739094).
• Membrane attachment is by a GPI anchor on Asn-577
  PMID:2145267, and
  mutating that site releases the protein
  PMID:8626468.
  So the protein is an anchored component of the outer (external) face of the parasite
  plasma membrane, with a released/secreted pool as well.
• Gene family: in L. major the msp/gp63 genes are a tandem array on chromosome 10 plus
  copies on chromosomes 28 and 31, with stage-specific expression (genes 1–6 promastigote,
  gene 7 metacyclic/amastigote) PMID:9489664. P08148 is one member of this family;
  paralog-level over-annotation is therefore a real risk, but the biochemistry
  (zinc endopeptidase, surface GPI anchor) is shared across the family and is directly
  established for the L. major protein whose structure was solved.

2. Localization

Cell surface of the promastigote, GPI-anchored, catalytically active in situ
PMID:9739094. A soluble
pool is released, and released enzyme is active
PMID:12540585.
GP63 is also exported in exosomes
PMID:22442494,
and reaches the host macrophage cytosol via lipid rafts, where it meets cytoplasmic
substrates PMID:19797268.

Protist-specific check: no kinetoplastid-specific compartment term (glycosome,
kinetoplast, paraflagellar rod, flagellar pocket) is warranted from the primary data for
P08148. The amastigote pool is reported as partly flagellar-pocket associated in the
review literature, but that is secondary and not species/accession-specific enough to
annotate here.

3. Catalytic activity and substrates

EC 3.4.24.36, "Preference for hydrophobic residues at P1 and P1' and basic residues at
P2' and P3'" (UniProt CATALYTIC ACTIVITY). Documented substrates span host proteins:

• Complement C3b → iC3b: surface GP63 accelerates C3b conversion, limiting assembly
  of the terminal complement complex
  PMID:7673725,
  PMID:7673725,
  PMID:22919663.
  Loss-of-function mutants are complement-sensitive
  PMID:9489664,
  PMID:9489664,
  and the whole-locus (gp63 1–7) deletion is complement-sensitive and delayed in lesion
  formation PMID:11849703.
  Importantly, complement fixation itself does not need GP63 proteolysis
  PMID:7673725 — the proteolysis-dependent
  part is the C3b→iC3b conversion and the escape from lysis. The iC3b coat then engages
  CR3/Mac-1 PMID:7673725,
  which is how GP63 contributes to attachment/uptake by macrophages
  PMID:11849703.

• Host protein tyrosine phosphatases SHP-1, PTP1B, TCPTP: cleaved by GP63, which
  activates them and thereby shuts down JAK/STAT (IFN-γ) and MAPK signaling
  PMID:19797268,
  PMID:19797268,
  PMID:22919663.

• mTOR: GP63 cleaves mTOR, inhibiting mTORC1 and activating 4E-BP1, repressing
  cap-dependent translation in the infected macrophage
  PMID:21501832,
  PMID:21501832.

• Synaptotagmin XI (host): directly degraded, de-repressing TNF/IL-6 secretion
  PMID:25063865,
  PMID:25063865,
  PMID:25063865.
  Note the direction: here GP63 increases a proinflammatory output, i.e. it perturbs
  (not simply suppresses) the host inflammatory response.

• Antimicrobial peptides: degraded by surface GP63; knockouts are AMP-hypersensitive
  PMID:17074074,
  PMID:17074074.

• Host extracellular matrix (collagen IV, fibronectin), promoting migration
  PMID:12540585.

4. Curation decisions taken

• GO:0004222 metalloendopeptidase activity and GO:0008237 metallopeptidase activity:
  both correct; the endopeptidase term is the core MF, the generic metallopeptidase row
  is kept as the same claim at lower resolution (MODIFY → GO:0004222).
• GO:0005886 plasma membrane: correct but less specific than the GPI data support →
  MODIFY to GO:0009897 external side of plasma membrane. (The anchored-component terms
  that would state the GPI attachment directly, GO:0031362 and GO:0046658, are obsolete
  in the ontology version used by validation, so they cannot be used here.)
• GO:0016020 membrane: TOO_GENERAL (entailed by the plasma-membrane claim).
• GO:0007155 cell adhesion: an InterPro-family electronic transfer of a generic
  (animal-flavoured) adhesion term. The real, documented adhesion claim is
  parasite-to-host-cell attachment → MODIFY to GO:0044650 adhesion of symbiont to host cell.
• GO:0006508 proteolysis (IEA and IMP rows): accepted, core.
• Host-interaction terms (GO:0039514, GO:0052028, GO:0052032, GO:0141115 ×2):
  all supported by the cited primary work and used in the right direction
  (suppression of JAK-STAT; activation of host signaling and perturbation — here
  augmentation — of the inflammatory response; inactivation of complement proteins).
• NEW proposed: GO:0008270 zinc ion binding (structure/cofactor, PMID:9739094) and
  GO:0039604 symbiont-mediated suppression of host translation (GP63 itself cleaves
  mTOR, so it performs the step; PMID:21501832).
  AMP degradation and ECM degradation are real but I did not find a GO term that fits
  without stretching, so they are recorded as questions rather than asserted annotations.

5. Revision (review-fix round, PR #5029)

• Truncated quotations repaired. Two supporting_text values had been cut at the cached
  file's hard line wrap and ended mid-clause ("the Leishmania protease GP63 cleaves"), so they
  no longer stated the claim they were attached to. Both are now full clauses:
  PMID:21501832. The companion quote in the same row was completed likewise:
  PMID:21501832.
• GO:0008270 zinc ion binding (NEW) re-anchored. It had been supported by a metzincin
  class statement ("a member of the metzincin class of zinc proteinases"), which is a
  family-level property, not an observation on P08148. It is now anchored to the three
  explicit UniProt BINDING features for the catalytic Zn(2+) in P08148 itself -- residues
  264, 268 and 334, each /ligand="Zn(2+)" /ligand_id="ChEBI:CHEBI:29105" /ligand_note="catalytic" and evidenced to PubMed:9739094 -- cited as
  file:LEIMA/gp63/gp63-uniprot.txt, the form other reviews use for UniProt features
  (cf. genes/DANRE/asah2, genes/human/AGTPBP1). Residue identities were read off the
  sequence in the flat file: 264 H, 265 E (catalytic base), 268 H, 334 H, i.e. the
  HEXXH pair plus the third metzincin histidine, not a Met-turn methionine.