dgcB (PA5399) encodes the iron-sulfur, membrane-integral component of the DgcAB system of Pseudomonas aeruginosa, which converts N,N-dimethylglycine (DMG) to sarcosine in the catabolism of glycine betaine. Glycine betaine is both an osmolyte and a carbon and nitrogen source, and arises from choline and carnitine; it is demethylated to DMG by GbcAB, DMG to sarcosine by DgcAB, and sarcosine to glycine by the sarcosine oxidase SoxBDAG, with the AraC-family regulator GbdR inducing gbcAB and dgcAB in response to glycine betaine or DMG. DgcB has five predicted transmembrane helices, two 4Fe-4S ferredoxin-type domains and two cysteine-rich CCG domains, and belongs to a PANTHER family of electron-transfer iron-sulfur subunits that includes heterodisulfide reductase subunit C, Bacillus FadF and an EtfAB:quinone oxidoreductase. Its partner DgcA (PA5398) is a predicted FMN oxidoreductase resembling N-methylproline demethylases. The genetic requirement of both genes for the DMG to sarcosine step is established, by loss of growth on DMG and by accumulation of 13C-labelled DMG in mutants, but no activity of either protein has been measured: which subunit removes the methyl group, what the immediate electron acceptor is, and whether the one-carbon product is free formaldehyde or a tetrahydrofolate adduct are all unknown.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0016491 oxidoreductase activity | IBA GO_REF:0000033 | ACCEPT | Summary: IBA from PTN000229983, whose IBD is seeded only by this gene's two PseudoCAP rows and propagates to 33 leaves in 17 taxa. The PAINT curator generalised from the seeded dimethylglycine dehydrogenase and oxidase terms to the root oxidoreductase term. This gene appearing in its own WITH/FROM reflects that the seeding annotations are on the target, and is expected rather than circular. Reason: Retained as the most specific molecular function this gene can currently carry. The generic term is supported independently of the seed: PA5399 has two PROSITE-matched 4Fe-4S ferredoxin domains and two CCG domains, and its PANTHER family (PTHR43255) is a family of electron-transfer iron-sulfur oxidoreductase subunits whose members include heterodisulfide reductase subunit C, Bacillus FadF and an EtfAB:quinone oxidoreductase. Both the domain content and the chemistry of the DMG demethylation the gene is required for imply a redox role, so the claim is right even though no activity of this protein has been assayed. The curator's generalisation is also the correct call for a 33-leaf, 17-taxon clade of diverse electron-transfer proteins; it is the specific terms on the seed rows, not this one, that overreach. Propagation Review Root cause: SOURCE WEAK OR INFERRED Failure modes: SOURCE EVIDENCE WEAK Sources checked: PANTHER:PTN000229983 · PTN000229983 SUPPORTS TRANSFER Sole-seeded node, 33 leaf IBAs in 17 taxa, within PTHR43255 (Fe-S electron-transfer subunits: HdrC, FadF, EtfAB:quinone oxidoreductase, membrane-anchored ferredoxins). The node term is appropriate to that breadth and is independently plausible from domain content, but its only seed is an indirect mutant phenotype, so the IBD has no measured activity anywhere in its support chain. Reported to GO Central as a query rather than a correction. UniProtKB:Q9HTG5 · PA5399 dgcB SOURCE WEAK OR INFERRED The target's own PseudoCAP rows (GO:0047865 and GO:0047866, both citing PMID:17951379) are the seed, hence the self-reference in WITH/FROM. Those rows rest on transposon-mutant growth on DMG and 13C NMR metabolite accumulation; the paper measured no enzyme activity and states that the biochemistry of DgcA and DgcB remains for future work. They support a redox role for the DgcAB system, which is what the generalised node term claims, but not the specific activities they assert. Supporting Evidence: file:PSEAE/dgcB/dgcB-uniprot.txt /note="4Fe-4S ferredoxin-type" PMID:17951379 with mutations in the dgcAB (PA5398-PA5399) genes, which exhibit homology to file:PSEAE/dgcB/dgcB-notes.md This is a family of **electron-transfer Fe-S subunits** |
| GO:0031457 glycine betaine catabolic process | IMP PMID:17951379 Identification of two gene clusters and a transcriptional re... | ACCEPT | Summary: PseudoCAP row from the genetic dissection of the glycine betaine route. dgcAB mutants grow on glycine betaine's upstream precursors but not on dimethylglycine, and accumulate 13C-labelled dimethylglycine. Reason: The core biological process, and the best-supported claim about this gene. DgcB is part of the system that carries out one step of the pathway, the dimethylglycine to sarcosine demethylation, so it does the work of the process rather than merely being required for it. IMP is the right evidence code here, and the step assignment comes from metabolite accumulation in the mutant as well as from the growth phenotype. Supporting Evidence: PMID:17951379 with mutations in the dgcAB (PA5398-PA5399) genes, which exhibit homology to PMID:17951379 conversion of DMG to sarcosine. PMID:17951379 growth on GB and DMG and for the induction of gbcA, gbcB, and dgcAB in response |
| GO:0042426 choline catabolic process | IMP PMID:17951379 Identification of two gene clusters and a transcriptional re... | KEEP AS NON CORE | Summary: PseudoCAP row reflecting that glycine betaine, and hence dimethylglycine, is a choline catabolite, so the DgcAB step lies on the route from choline to glycine. Reason: True and as precise as the evidence allows, but peripheral: the gene acts three steps downstream of choline, on dimethylglycine, and the same step serves glycine betaine arriving from carnitine or taken up directly as an osmolyte. The 13C-choline labelling experiment is exactly what ties it to choline catabolism, so the term is kept rather than refined or dropped; the core process is glycine betaine catabolism. Supporting Evidence: PMID:17951379 intermediate in the catabolism of choline and carnitine, can serve as a sole PMID:17951379 conversion of DMG to sarcosine. |
| GO:0047865 dimethylglycine dehydrogenase activity | IDA PMID:17951379 Identification of two gene clusters and a transcriptional re... | MARK AS OVER ANNOTATED | Summary: PseudoCAP row, one of the two seeds of the PTN000229983 oxidoreductase IBD. It asserts the electron-transfer-flavoprotein-dependent demethylation reaction for this subunit on the strength of a genetic experiment. Reason: The DgcAB system does convert dimethylglycine to sarcosine, so the row is not simply wrong, but three things overreach. First, the evidence is a transposon mutant's inability to grow on dimethylglycine plus 13C NMR accumulation of the substrate, not an assay: the paper's full text states that characterising the biochemical activities of DgcA and DgcB remains for future research, so IDA misdescribes it. Second, this term fixes the acceptor as an electron-transfer flavoprotein and the product as free formaldehyde, neither of which was tested; the companion row asserts the incompatible O2-dependent oxidase reaction for the same protein, and both cannot hold. Third, the activity is more likely to be DgcA's: PA5398 is the predicted FMN oxidoreductase resembling N-methylproline demethylases, while PA5399 has five transmembrane helices, two 4Fe-4S clusters and two CCG domains, the architecture of the membrane electron-transfer subunit in its family, so contributes_to would fit it better than enables. Kept as an over-annotation rather than removed because the subunit is genuinely part of the system that performs the reaction and a curator with the bench data may be able to assign the activity properly. Supporting Evidence: PMID:17951379 genes that encode other enzymes with demethylase activity, are required for the file:PSEAE/dgcB/dgcB-notes.md GO:0047865 dimethylglycine dehydrogenase activity is "N,N-dimethylglycine + electron-transfer flavoprotein + H2O = sarcosine + formaldehyde + reduced electron-transfer flavoprotein" |
| GO:0047866 dimethylglycine oxidase activity | IDA PMID:17951379 Identification of two gene clusters and a transcriptional re... | MARK AS OVER ANNOTATED | Summary: The second PseudoCAP molecular-function row from the same genetic paper, and the second seed of the PTN000229983 IBD. It asserts the O2-dependent, H2O2-releasing demethylation for this subunit. Reason: Same assessment as the companion dehydrogenase row, and the two cannot both be correct: this term requires molecular oxygen as acceptor and hydrogen peroxide as product, neither of which was tested, while the companion row requires an electron-transfer flavoprotein instead. The genetic requirement of dgcB for the dimethylglycine to sarcosine step is not in doubt, which is why the row is marked as an over-annotation rather than removed, but the reaction asserted is unmeasured and is more plausibly performed by the FMN oxidoreductase subunit DgcA. Supporting Evidence: PMID:17951379 genes that encode other enzymes with demethylase activity, are required for the file:PSEAE/dgcB/dgcB-notes.md GO:0047866 dimethylglycine oxidase activity is "N,N-dimethylglycine + H2O + O2 = formaldehyde + H2O2 + sarcosine" |
| GO:0051536 iron-sulfur cluster binding | IEA GO_REF:0000002 | MODIFY | Summary: InterPro2GO from the alpha-helical ferredoxin signature (IPR009051), the generic iron-sulfur cluster binding term. Reason: Correct but less specific than the evidence supports. The entry carries two PROSITE 4Fe-4S ferredoxin-type domain matches (PS51379 at 236-266 and 311-342, with the PS00198 4Fe-4S consensus), which identify the cluster type, so the is_a descendant GO:0051539 4 iron, 4 sulfur cluster binding is the right level. Nothing in the record suggests a 2Fe-2S or 3Fe-4S cluster instead. Proposed replacements: 4 iron, 4 sulfur cluster binding Supporting Evidence: file:PSEAE/dgcB/dgcB-uniprot.txt /note="4Fe-4S ferredoxin-type" file:PSEAE/dgcB/dgcB-uniprot.txt PROSITE; PS00198; 4FE4S_FER_1; 1. |
Loading supporting content…
Download this section (compressed HTML)Q: Which of DgcA and DgcB removes the methyl group from dimethylglycine, and what is the immediate electron acceptor of DgcB's 4Fe-4S clusters: a membrane quinone, an electron-transfer flavoprotein, or DgcA's flavin?
Q: Is the one-carbon product released as formaldehyde, as the GO dimethylglycine dehydrogenase and oxidase definitions require, or transferred to tetrahydrofolate as in the mammalian enzyme?
Q: Do DgcA and DgcB form a stable membrane complex, as the failure of multiple dgcA rescue constructs to complement the dgcA deletion suggests?
Experiment: Express and purify DgcA and DgcB (the latter from membranes, in detergent or nanodiscs), confirm the iron-sulfur content by EPR and iron/sulfide analysis, and assay dimethylglycine-dependent reduction of candidate acceptors (electron-transfer flavoprotein, menaquinone analogues, artificial dyes, O2 with peroxide detection) for each protein alone and for the pair, to settle which activity term applies and to which subunit.
Experiment: Trace the removed methyl carbon with 13C-labelled dimethylglycine in cell-free extracts, testing for free formaldehyde versus 5,10-methylenetetrahydrofolate, and compare wild type with dgcA and dgcB deletions.
What is not known — curated, literature-grounded statements of the open unknowns (the inverse of core functions).
Gap: No activity of DgcB or DgcA has ever been measured, so the reaction that converts dimethylglycine to sarcosine in Pseudomonas aeruginosa is uncharacterised: which subunit removes the methyl group, what accepts the electrons (an electron-transfer flavoprotein, a membrane quinone, or DgcA's flavin), and whether the one-carbon product is free formaldehyde or a tetrahydrofolate adduct.
BIOLOGY MF_DARK
What is known: Firmly established: dgcA and dgcB are both genetically required for the dimethylglycine to sarcosine step, mutants accumulate 13C-labelled dimethylglycine, and both genes are induced by glycine betaine and dimethylglycine through GbdR. DgcB's two 4Fe-4S ferredoxin domains and five transmembrane helices are signature-level predictions, not measurements.
Significance: The two molecular functions currently annotated to dgcB assert mutually exclusive chemistries, and a PAINT IBD propagates an oxidoreductase annotation to 33 leaves from this single unverified seed, so the gap has consequences well beyond this gene.
What would resolve it: Purify DgcA and DgcB, separately and together, and assay dimethylglycine consumption with candidate acceptors, measuring the one-carbon product.
Provenance (the field's own admissions):
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)