TgpA (PA2873) is a 668-residue polytopic inner-membrane protein of P. aeruginosa built from an N-terminal DUF3488 region of six predicted transmembrane helices and a large periplasmic domain (residues 180-544) that contains a transglutaminase-like (papain-fold) catalytic core with a conserved Cys-His-Asp triad (Cys404, His448, Asp464); a crystal structure of this domain is available (PDB 6G49, 6G4H). The purified periplasmic domain incorporates an amine donor into a glutamine-containing acyl donor in a colorimetric transglutaminase assay, at about 45% of the guinea-pig transglutaminase reference, and unlike the eukaryotic enzyme its activity is not lost on adding EDTA, so it is calcium-independent like other microbial transglutaminases. The protein is detected in PAO1 membrane fractions with its periplasmic domain exposed to trypsin shaving, and it is expressed in exponential and stationary phase from an operon with PA2874-PA2875. Antisense knockdown and conditional-promoter depletion show that tgpA is essential for viability; its physiological substrate is unknown, and the authors propose a cell-envelope role.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0003810 protein-glutamine gamma-glutamyltransferase activity | IBA GO_REF:0000033 | ACCEPT | Summary: IBA from PTN002126683 (PTHR42736), whose only seed is TgpA itself. The claim is defensible for TgpA, whose purified catalytic domain is active in a transglutaminase assay, but the node carries a very precise EC-level term across a clade of transglutaminase-like membrane proteins that have not been characterised. Reason: For this gene the term rests on the gene's own experiment, and TgpA in its own With/From is the expected marker of that grounding, not circularity. I do not challenge the row on the target. The node placement is the problem and is recorded as a GO_CENTRAL source defect: the transglutaminase-like superfamily is a papain-like fold used for several different chemistries (proteases, deamidases, cysteine peptidases), the only characterised close relative cited by the same authors, Bordetella WbmE, is an LPS O-antigen deamidase rather than a cross-linker, and the seed itself is a surrogate-substrate assay on an isolated domain with no physiological substrate identified. Propagation Review Root cause: TERM SCOPING PROBLEM Sources checked: PANTHER:PTN002126683 · PTN002126683 (PTHR42736) SOURCE WEAK OR INFERRED 17 leaf IBAs in 15 taxa in the PAINT leaf file, spanning Chloroflexi, Aquificae, Thermodesulfovibrio, Dictyoglomus, cyanobacteria, Deinococcus, Planctomycetes (three Rhodopirellula paralogues), Firmicutes, Streptomyces, Shewanella, Geobacter and Xanthomonas. Every leaf is named only by its domain content ("transglutaminase domain protein", "transglutaminase-like domain-containing protein"), so the clade is uncharacterised apart from TgpA, and a precise EC 2.3.2.13 cross-linking term is more than descent from this node supports. UniProtKB:Q9HZX3 · tgpA (self; sole IBD seed) SUPPORTS TRANSFER The in vitro assay on the purified periplasmic domain is real experimental grounding for the chemistry, but it was done with the kit's dipeptide acyl donor and biotin-cadaverine amine donor, so it constrains the catalytic chemistry rather than the physiological reaction. Supporting Evidence: PMID:23209712 In this study we show that the periplasmic portion of the PA2873 protein does have transglutaminase activity in vitro. |
| GO:0003810 protein-glutamine gamma-glutamyltransferase activity | IDA PMID:23209712 TgpA, a protein with a eukaryotic-like transglutaminase doma... | ACCEPT | Summary: PseudoCAP IDA and, with the identical UniProt row, the seed of the IBD at PTN002126683. The purified His-tagged periplasmic domain (aa 180-544) was positive in a transglutaminase colorimetric assay at about 45% of the guinea-pig reference, and its activity was EDTA-insensitive. Reason: This is the gene product's own activity measured on purified protein, not a mutant phenotype, and the assayed reaction (gamma-glutamyl transfer from a glutamine donor to a primary amine) is the reaction EC 2.3.2.13 names, to which GO:0003810 is an exact match; UniProt curators independently made the same call. The caveat, which I record rather than act on, is that the assay used the kit's surrogate substrates, so the protein-to-protein isopeptide cross-linking in GO's definition has not been shown for TgpA and no physiological substrate is known; that is the ground for questioning the node term, not this row. Supporting Evidence: PMID:23209712 TG180–544 was positive to the TGase test. PMID:23209712 In actual fact, 2.5 µg of TG180–544, equal to reference gpTGase, showed TGase activity that differed significantly from the negative control and was about 45% of gpTGase activity in the absence of EDTA. |
| GO:0003810 protein-glutamine gamma-glutamyltransferase activity | IDA PMID:23209712 TgpA, a protein with a eukaryotic-like transglutaminase doma... | ACCEPT | Summary: The UniProt row for the same experiment, which also supports the EC 2.3.2.13 and CATALYTIC ACTIVITY statements in the entry. Reason: Same claim and same evidence as the PseudoCAP row; two curation groups independently reading the full text reinforce each other, and deduplication is not a reason to remove either row. Supporting Evidence: PMID:23209712 Unlike gpTGase, EDTA addition did not affect TG180–544 activity. file:PSEAE/tgpA/tgpA-uniprot.txt Reaction=L-glutaminyl-[protein] + L-lysyl-[protein] = [protein]-L- |
| GO:0003810 protein-glutamine gamma-glutamyltransferase activity | IEA GO_REF:0000120 | ACCEPT | Summary: Electronic row derived from the EC number and Rhea reaction recorded in the UniProt entry, which themselves come from the same 2012 experiment. Reason: Consistent with the experimental rows for the same term; it adds no independent evidence (the EC assignment is the curated conclusion of PMID:23209712), but the claim is the same and must not be removed as a duplicate. Supporting Evidence: PMID:23209712 In this study we show that the periplasmic portion of the PA2873 protein does have transglutaminase activity in vitro. |
| GO:0005886 plasma membrane | IDA PMID:23209712 TgpA, a protein with a eukaryotic-like transglutaminase doma... | ACCEPT | Summary: TgpA peptides were identified by MudPIT in washed PAO1 total-membrane fractions, and all five peptides recovered by trypsin shaving of intact membranes lay in the predicted periplasmic domain. Reason: Direct detection of the protein in the membrane fraction, with the shaving result independently showing that the catalytic domain is exposed on the periplasmic face; six predicted transmembrane helices and a PhoA fusion screen agree. This is where the protein acts, so it belongs in the core function. The assay cannot distinguish inner from outer membrane on its own, but a polytopic helical protein with a periplasmic catalytic domain is an inner-membrane protein, and UniProt records Cell inner membrane. Supporting Evidence: PMID:23209712 Since all five PA2873 peptides “shaved” by trypsin belonged to the predicted outside membrane domain, this strongly supports the notion that it actually emerges from membrane itself. file:PSEAE/tgpA/tgpA-uniprot.txt SUBCELLULAR LOCATION: Cell inner membrane |
| GO:0005886 plasma membrane | IEA GO_REF:0000044 | ACCEPT | Summary: Electronic mapping of the UniProt Cell inner membrane location, which rests on the experimental row above. Reason: Agrees with the IDA row for the same term. Supporting Evidence: file:PSEAE/tgpA/tgpA-uniprot.txt Multi-pass membrane protein |
Loading supporting content…
Download this section (compressed HTML)Q: What does TgpA modify in the periplasm, and is the reaction a protein-protein isopeptide cross-link, an amine incorporation, or a deamidation?
Q: Is the catalytic triad (Cys404, His448, Asp464) required for viability, or is the essential role of TgpA structural and separable from its catalytic activity?
Experiment: Assay the purified periplasmic domain against candidate envelope substrates (peptidoglycan precursors and sacculi, LPS O-antigen, outer-membrane and periplasmic proteins) with and without labelled amine donors, and detect isopeptide or deamidation products by mass spectrometry, to replace the surrogate-substrate annotation with the physiological reaction.
Experiment: Complement a conditional tgpA depletion strain with catalytic-triad substitutions (C404A, H448A, D464A) expressed from the chromosome, to test whether the essential function requires catalysis; combined with envelope phenotyping (sacculus composition, LPS profile, sensitivity to envelope stress) this would identify the pathway TgpA acts in.
Experiment: Characterise two or three uncharacterised leaves of PTN002126683 from different phyla (for example Synechocystis sll1608, B. subtilis YebA, Deinococcus DR_0620) in the same assay, which would settle whether the node-level transglutaminase annotation is justified or should be generalised.
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)