PP_0076 (Q88RQ3) is a 307-aa signal-peptide-bearing periplasmic binding protein
with choline/glycine-betaine transporter substrate-binding domains
[PP_0076-uniprot.txt, "FT SIGNAL 1..20";
"InterPro; IPR017783; ABC_choline_sub-bd."]. It has no transmembrane helix and
therefore cannot independently provide transmembrane transporter activity.
The adjacent KT2440 betC region was implicated in choline-O-sulfate
utilization, but the available paper does not individually assay PP_0076 and
shows that intact choline-O-sulfate still accumulates when betC is deleted
[PMID:17116241, "This mutant still accumulated intact COS but failed to use
this compound as carbon or nitrogen source."]. The adjacent partner PP_0075 is
an SLC26A/SulP-STAS protein rather than a canonical ABC permease plus
nucleotide-binding subunit. A syntenic Pseudomonas BetD homolog is nevertheless
described by in-silico analysis as a "substrate-binding ABC transporter protein"
PMID:21369825. Because orphan periplasmic binding proteins can use transporter
components encoded elsewhere, ABC-complex membership cannot currently be
confirmed or excluded.
The report retrieves useful ChoX/OpuAC-family context and explicitly notes the
lack of direct target characterization. It nevertheless constructs a complete
PP_0074/PP_0075/PP_0076 ABC importer by misidentifying PP_0074/Q88RQ5 as an ABC
ATPase. PP_0074 is shikimate dehydrogenase AroE, while PP_0075 is an
SLC26A/SulP-STAS protein. The definitive complex, partner, and choline-specific
conclusions are therefore not adopted.
The three UNDECIDED rows were resolved on biological grounds (all are InterPro2GO
IEA rows; there is no target-specific literature to wait for):
Core function now uses GO:0050997 (binding) with transport processes. Status set to
COMPLETE.
Retained all six original row payloads and generic quaternary-ammonium binding/transport inference. Periplasmic location now ACCEPT as the core binding compartment. Clarified that unqualified transmembrane transporter MF is unsupported; a demonstrated functional complex could justify contributes_to, but none is established here. ABC-complex membership remains over-annotated, not disproven by absence of a local ATPase. ATCC19151 evidence is explicitly comparative and in silico.
Reviewed the existing OpenScientist provider report and primary/cached sources; preserved the provider report and its PDF/HTML artifacts. A provider report is already present, so no replacement research run was needed. All machine annotation payloads and raw UniProt/GOA files are unchanged. No new annotation rows were added. Generic family labels and locus proximity do not establish COS transport, a BetD-BetE complex, or substrate promiscuity in unassayed homologs.
Normal force-fetches of PMID:17116241 and PMID:21369825 still returned abstract-only caches; their assay limitations remain explicit. PMID:9736747 also remains abstract-only in the cache after the root fetch. PMC methods were independently accessible at https://pmc.ncbi.nlm.nih.gov/articles/PMC21653/ and explicitly name 102F34R; direct HTML retrieval encountered a challenge, Europe PMC XML returned HTTP500, and NCBI BioC returned no result. No publication cache was hand-edited or invented. Existing abstract quotations remain used as such. Full-text PMID:29458126 and PMID:35503214 distinguish recombinant source strains and assay conditions.
Focused gene validation passed. Preservation is reproducibly checked by .cache/module-audit-logs/choline-sulfate-20261106/check-existing-four.py, against the frozen original baseline: 16 total rows, one inherited NEW, zero added rows, eight raw and twelve provider/artifact hashes unchanged. PSEPK/betC retains the advisory about not citing its low-quality provider in annotation support; primary and curated family evidence are used instead.