xcpQ (PA3105, P35818) review notes

Review session 2026-10-28 (PseudoCAP IBA-seed protocol)

Fetched with just fetch-gene PSEAE xcpQ; accession P35818 matches the worklist entry
(rank 52, PA3105, reviewed Swiss-Prot entry GSPD_PSEAE). No deep-research provider key was
available, so these notes are manual, from the cached publications, the UniProt record and
the cached PAINT table interpro/panther/PTHR30332/PTHR30332-paint.tsv.

What XcpQ is

XcpQ is the GspD-family secretin, the outer-membrane channel of the Xcp type II secretion
system (T2SS) of P. aeruginosa. UniProt: "Belongs to the bacterial secretin family. GSP D
subfamily"; NCBIfam TIGR02517 (type_II_gspD), Pfam PF00263 Secretin + PF03958 Secretin_N,
PANTHER PTHR30332.
It is an outer-membrane protein that oligomerises into a ring-shaped channel large enough to
pass folded exoproteins [PMID:9466268 "C-terminal part of XcpQ is sufficient for the formation
of the multimer."; "the XcpQ multimers revealed large, ring-shaped structures with an apparent"
central cavity of 95 Å]. Cryo-EM of the purified ~1-MDa complex shows a pentadecamer
PMID:29042496.
UniProt places it in the cell outer membrane with most of the mass in the periplasm, and
divides it into periplasmic N0-N3 domains plus the membrane-embedded secretin and S domains
(CC:SUBCELLULAR LOCATION, CC:DOMAIN, both ECO:0000305 from PubMed:29042493/29042496).
Loss of XcpQ abolishes Xcp-dependent export: UniProt CC:DISRUPTION PHENOTYPE (from
PubMed:29042493) records "Loss of export of PrpL, elastase LasB, chitin binding protein D",
and the nanobody-interference work compares its secretome to "an xcpQ deletion mutant"
PMID:29042493. The Xcp machine with XcpQ exports many hydrolases PMID:29042496.
XcpQ is unusual among P. aeruginosa secretins in inserting into the outer membrane without
the Bam/Lol machineries and in assembling spontaneously into liposomes PMID:22147293.
The homo-oligomer is the functional unit, and a point substitution traps an assembly
intermediate in both PulD and XcpQ PMID:24657091.

PseudoCAP row details

GO:0015627 IDA PMID:9466268 (Bitter et al. 1998): primary paper, abstract-only in the cache.
It purifies XcpQ as a stable homomultimer and images ring structures; this is direct evidence
for the secretin channel, and later cryo-EM (PMID:29042496) confirms the pentadecameric T2SS
secretin complex. Term correct, evidence code appropriate.
GO:0015628 IDA PMID:9818381: Filloux, Michel & Bally 1998 is a review (PubMed
publication type Review; publication_type: REVIEW in the cached file). The same row and the
same mis-coding was found in the xcpX review (genes/PSEAE/xcpX), so this is a systematic
PseudoCAP pattern across the xcp cluster, not a one-off. The term itself is correct for XcpQ —
it is the outer-membrane channel of the machine, and deletion abolishes export of the Xcp
substrates. Recorded as a WRONG_EVIDENCE_CODE source defect (re-key to a primary paper, e.g.
PMID:29042493 for the deletion phenotype, or downgrade to TAS for the review).

PAINT node context

interpro/panther/PTHR30332/PTHR30332-paint.tsv shows the family carries two IBD-bearing nodes:
PTN000768388 (T2SS: GO:0015627 seeded by P35818/Q9I2M7/Q9I5P0, GO:0015628 sole-seeded by
P35818) and PTN009110345 (T3SS: GO:0030257 seeded by P35672, GO:0030254 sole-seeded by pscC
Q9I319), with an IRD at PTN009110348 removing the T3SS terms in a plant-associated subclade.
The family therefore keeps the T2SS and T3SS secretin assertions on separate nodes, which is the
right shape: a shared "bacterial secretin" node would propagate T2SS terms onto T3SS members.

PseudoCAP IBA-seed assessment

PTN node	node term	n_seeds	leaf IBAs	seed annotation (term, evidence, PMID)	verdict	rationale
PTN000768388	GO:0015627 type II protein secretion system complex	3 (not sole)	23 (13 taxa)	GO:0015627, IDA, PMID:9466268	SOUND	XcpQ is the GspD secretin of the Xcp T2SS, purified as a ring-forming homomultimer (PMID:9466268) and resolved as a pentadecameric secretin complex by cryo-EM (PMID:29042496). The node is the GspD-specific subnode of PTHR30332 (the T3SS secretins sit on PTN009110345), so a T2SS-complex term fits the clade.
PTN000768388	GO:0015628 protein secretion by the type II secretion system	1 (sole seed)	23 (13 taxa)	GO:0015628, IDA, PMID:9818381	SOUND (term); evidence code WRONG (cited paper is a review with no original secretion data; full text checked)	The claim is correct — XcpQ is the outer-membrane conduit of the Xcp T2SS and its deletion abolishes export of the Xcp substrates (UniProt disruption phenotype from PMID:29042493). But PMID:9818381 is a review, so IDA cannot stand; the IBD conclusion is unaffected and the seed should be re-keyed rather than the node dropped. Same defect as the xcpX seed on the identical reference.
PTN001251149	GO:0009306 protein secretion	2 (not sole)	0	GO:0015628, IDA, PMID:9818381	SOUND (term)	The general node carries no leaf IBAs, so nothing propagates; the parent term is true of XcpQ. The same wrong evidence code applies to the underlying row.

Open questions

Whether the GO:0042802 identical protein binding row (IPI PMID:24657091, WITH P35818) is
worth keeping now that the homo-oligomer stoichiometry is known (15-mer, PMID:29042496): the
self-association is the channel, so a complex-membership statement arguably says more.