PWL2 (Pyricularia oryzae, UniProt G5EI71) research notes

Identity

• Secreted effector PWL2 = "Pathogenicity/Prevents pathogenicity toward Weeping Lovegrass 2"; 145 aa precursor
  with a 21-aa signal peptide; ORFs MGG_04301 / MGG_13863 (two copies in strain 70-15).
• Originally defined genetically as a host-species-specificity gene: PMID:7549480 and PMID:7549480
• Deep research report: PWL2-deep-research-falcon.md (Edison/Falcon) — flags that Were et al. was a preprint
  at the time; it is now published as PMID:40341381.

Secretion and localization

• Secreted by invasive hyphae into the biotrophic interfacial complex (BIC), then translocated into the rice
  cytoplasm and moves cell-to-cell: PMID:20435900,
  PMID:20435900. BAS4, an apoplastic
  effector, is not translocated.
• Non-conventional secretion route, distinct from the ER-Golgi route used by apoplastic effectors:
  PMID:34236677, PMID:23774898.
• Punctate accumulation at the BIC: PMID:26472068; used as the standard symplastic-effector marker PMID:26472068.
• Host-cell entry confirmed with an NLS-tagged reporter against a rice nuclear marker line: PMID:31749824 — the nuclear signal is an artefact of the engineered NLS, so no
  native nuclear location should be annotated.
• Uptake appears to be clathrin-mediated: PMID:38653755;
  clathrin inhibitors distort BIC localization PMID:38722963
• Methods chapter devoted to imaging PWL2 translocation: PMID:30182232

Molecular activity: binding host HMA (HIPP43) proteins

• Crystal structure (1.8 A) of Pwl2 with the rice OsHIPP43 HMA domain; Pwl2 is a MAX effector with an extra
  C-terminal helix/loop: PMID:38968126, PMID:38968126
• Nanomolar affinity by ITC: PMID:38968126; paralogues Pwl1/Pwl4 bind similarly PMID:38968126.
• In planta the barley orthologue HvHIPP43 is bound by Pwl2 (IP-MS, Y2H, co-IP) and displaced from
  plasmodesmata: PMID:40341381.
• Direction of effect on the target is unresolved: Pwl2 may stabilise/accumulate HIPP43 rather than simply
  inactivate it — PMID:40341381
  — and HIPP43 overexpression alone suppresses immunity PMID:40341381. So "protein sequestering activity" (GO:0140311, "Binding to a protein to prevent it from
  interacting with other partners or to inhibit its localization to the area of the cell or complex where it is
  active") is plausible but not established; I do not assert an MF term.

Biological process: suppression of host PTI; virulence vs avirulence

• Transgenic plants expressing signal-peptide-less Pwl2 have attenuated PTI: PMID:40341381, PMID:40341381
• Binding is required for both target relocalisation and the virulence contribution: PMID:40341381
• Overall conclusion: PMID:40341381. PWL2 expression is Pmk1-dependent and occurs at plasmodesmata-containing
  pit fields PMID:40341381.
• Deletion of all three Guy11 copies: PMID:40341381 — i.e. one molecule, two opposite outcomes depending on host
  recognition.
• Recognition by a barley NLR: PMID:37820736.
• Host-range role also demonstrated in finger-millet blast isolates from eastern Africa: PMID:37245238 (PubMed abstract text
  is mangled by lost italic gene names; the point is that PWL1/PWL2 transformants of a Ugandan isolate gained
  avirulence on weeping lovegrass while remaining virulent on finger millet).
• Biotechnological consequence: an engineered rice NLR with the OsHIPP43 HMA domain replacing Pik-1's integrated
  HMA recognises Pwl2 and other PWL alleles PMID:38968126

Curation decisions

• All 15 GOA rows are cellular-component rows (extracellular region, host cell cytoplasm), each well supported
  for the secreted and host-translocated pools respectively; all accepted. Several of the cited papers are
  methods chapters or resource papers that use PWL2 as a reporter rather than studying it; these are weaker
  sources but not wrong, so ACCEPT with a note rather than REMOVE (CLAUDE.md: do not overrule curators).
• GOA has no process or function annotation. One NEW process annotation is proposed,
  GO:0052034 effector-mediated suppression of host pattern-triggered immunity (symbiont-side, as the project
  requires). Participation test: Pwl2 itself binds and relocalises the host target, and a binding-dead variant
  loses the virulence contribution — the effector does the work. Comparator check: other PTI-suppressing
  secreted effectors carry this branch (M. oryzae Slp1 and C. fulvum Ecp6 reviews use GO:0140423, a child of
  GO:0052034). The parent, not the "signaling" child GO:0140423, is used because whether HIPP43 acts in the
  signalling chain is explicitly left open by the authors.
• No MF term is asserted: binding to host HIPP43 is firmly established, but "protein binding" is uninformative
  per CLAUDE.md and the mechanistic interpretation (sequestration vs stabilisation/co-option) is unresolved.
  Raised in suggested_questions instead.
• Avirulence/host-range activity is the host's recognition of Pwl2; no symbiont-side GO term describes "being
  recognised", so nothing is annotated for it. Noted as a question.