PSEPK ppu00785: Endogenous protein lipoylation
This batch separates cofactor installation from the many lipoate-dependent
enzymes grouped under KEGG ppu00785. The reusable module is
Endogenous protein lipoylation:
it models the conserved octanoyl-transfer and sulfur-insertion chemistry as a
direct LipB-LipA route plus separately characterized Bacillus and human
GcvH-relay variants.
The KEGG map contains 19 KT2440 genes, but only lipB and lipA catalyze these
two installation reactions. The remaining map members are lipoylated carrier
proteins or components of enzyme systems that use the installed cofactor. A
third selected locus, PP_4797, is not a KEGG member; it was promoted as a
diagnostic because its automated UniProt name says "Lipoyl synthase" even
though its 54-residue sequence is an ArfA-family ribosome-rescue protein. The
taxon-level research identified a fourth selected locus outside the KEGG
bucket, PP_0423, whose BPL/LPL-family assignment makes it an uncertain
salvage-or-GcvH-relay candidate.
Workflow Status
- [x] Partition all 19 KEGG candidates by role.
- [x] Fetch
lipA,lipB,PP_4797, andPP_0423with current GOA records. - [x] Build a species-neutral module with direct and GcvH-relay variants.
- [x] Complete and assess generic OpenScientist module research.
- [x] Complete and assess PSEPK module + pathway + taxon research.
- [x] Complete and assess all four gene-level OpenScientist reports.
- [x] Complete the initial manual review of every fetched GOA row.
- [x] Reconcile the reviews and module against the completed research.
- [x] Validate and render all affected artifacts.
- [x] Open draft pull request #2181 for this module.
- [x] Finish checks and move pull request #2181 out of draft.
- [x] Merge pull request #2181 after review and CI (2026-07-25).
Pathway Satisfiability
| Order | Reaction | KT2440 implementation | Review status | Decision |
|---|---|---|---|---|
| 1 | Octanoyl-ACP + apo lipoyl-domain protein to octanoylated protein | lipB / PP_4801 / Q88DM4 |
Curated, research assessed | covered |
| 2 | Octanoylated protein to lipoylated protein by sulfur insertion | lipA / PP_4800 / Q88DM5 |
Curated, research assessed | covered |
Module result: covered. KT2440 has one high-confidence LipB and one
high-confidence LipA. Their adjacent genomic placement is consistent with the
two-step route, but module inclusion is grounded in the exact catalytic
functions rather than neighborhood alone.
PP_0423 / Q88QR5 is assigned to the LipM/LipL family, but the same PANTHER
subfamily contains both characterized LipM and LipL proteins. A gene-level
report proposed an LplA-type salvage hypothesis, but exact InterPro and PANTHER
classification does not distinguish the two relay-transfer directions or
establish ATP-dependent ligation. PP_0423 therefore does not yet satisfy a
relay implementation or establish salvage. The direct route remains covered
independently of this unresolved candidate.
Octanoyl-ACP production is an upstream fatty-acid-synthesis dependency. The
pyruvate, 2-oxoglutarate, branched-chain 2-oxoacid, acetoin, and glycine-cleavage
systems are downstream clients. Exogenous-lipoate salvage would require a
lipoate--protein ligase route and is outside this direct endogenous branch.
Candidate Partition
| Bucket | Genes | Treatment |
|---|---|---|
| Core installation route | lipB, lipA |
Curate here and use as exact PSEPK module exemplars |
| Glycine-cleavage carrier and clients | gcvH1, gcvH2, gcvP1, gcvP2, gcvT-I, gcvT |
Lipoylated carrier proteins or client enzymes; not installation catalysts |
| 2-oxoacid/acetoin dehydrogenase clients | aceE, aceF, acoC, sucA, sucB, bkdAA, bkdAB, bkdB |
Cofactor-dependent complexes; not installation catalysts |
| Shared E3 client components | lpd, lpdG, lpdV |
Dihydrolipoyl dehydrogenases that recycle client-bound lipoamide; not biosynthetic steps |
| Uncertain non-bucket BPL/LPL candidate | PP_0423 |
Review salvage-ligase versus relay-transferase hypotheses; do not assign a specific reaction without discriminating evidence |
| Diagnostic nonmember | PP_4797 |
Curate as ArfA-family ribosome rescue and exclude from lipoylation |
The machine-generated 19-gene source table is retained at
ppu00785_lipoate_installation.tsv.
Selected Reviews
| Gene | Locus / UniProt | Selection reason | Current decision |
|---|---|---|---|
lipB |
PP_4801 / Q88DM4 | Core octanoyl-transfer step | Specific MF and protein lipoylation accepted; broad acyltransferase marked over-annotated |
lipA |
PP_4800 / Q88DM5 | Core sulfur-insertion step | Specific MF and lipoylation processes accepted; broad catalytic and Fe-S parents de-duplicated |
PP_0423 |
PP_0423 / Q88QR5 | BPL/LPL-family candidate found by taxon-level research | Salvage-ligase versus relay-transferase role left unresolved; no GO term inferred from computational evidence alone |
PP_4797 |
PP_4797 / Q88DM8 | False lipoyl-synthase name adjacent to lipA/lipB |
ArfA-associated ribosome rescue accepted; excluded from module |
Curation Decisions
- The module is defined by shared chemistry rather than by one organism's
topology. It contains a two-reaction direct route and separate three-reaction
Bacillus and human GcvH-relay variants. - Molecular functions occur only on leaf reaction annotons. The module has no
redundant module-level cytoplasm/cytosol assertions. - LipL and LIPT1 are not represented by broad
acyltransferase activityin the
repaired module. Their exact RHEA:20213 and RHEA:16413 reactions plus reviewed
UniProt exemplars carry the physiological specificity. Current QuickGO has
GO:7770091 for the LIPT1 reaction, but that term is not yet available in the
repository's configured ontology and therefore is not asserted here. - The P39648 and Q9Y234 UniProt flat-file records are cached in-repo. They
independently confirm the current exemplar names and the RHEA:20213 and
RHEA:16413 catalytic-reaction assignments; PMID:23960015 and PMID:28757203
are cached as literature support for the two relay steps. - LipB is the physiological octanoyl-ACP:protein transferase in this route.
The genericacyltransferase activityparent adds no information beside
lipoyl(octanoyl) transferase activity. - LipA is a radical-SAM enzyme with two 4Fe-4S clusters. Its core annotation is
lipoate synthase activity; cluster binding and cytoplasmic localization are
retained only as non-core mechanistic context. PP_4797is only 54 residues and carries InterProIPR005589and Pfam
PF03889, both ArfA. It lacks the radical-SAM and LipA family architecture
found in the 338-residue Q88DM5 protein.PP_0423carries InterProIPR050664and PANTHERPTHR43679:SF2, but those
identifiers group LipM and LipL activities together. The OpenScientist
salvage-ligase hypothesis does not override that unresolved exact-family
evidence. It remains a candidate, not a second satisfied route.- Client complexes are not counted as additional pathway steps simply because
KEGG groups lipoate use and lipoate installation on the same map.
Research
- Generic module report target:
modules/endogenous_protein_lipoylation-deep-research-openscientist.md - Species-aware report target:
projects/P_PUTIDA/deep-research/PSEPK__endogenous_protein_lipoylation__ppu00785-deep-research-openscientist.md - Gene-level report targets:
genes/PSEPK/{lipA,lipB,PP_4797,PP_0423}/*-deep-research-openscientist.md
Validation
Checkpoint on 2026-07-18:
- All 13 fetched GOA rows across
lipA,lipB, andPP_4797have manual actions;
noPENDINGvalues remain. - The
PP_0423review records its empty GOA set and deliberately leaves its
exact core activity unresolved after assessing the salvage-ligase and
relay-transferase hypotheses. - The revised three-variant module passes
ModuleReviewLinkML validation. The
custom semantic validator reports only its expected inability to validate
InterPro labels through the current OAK prefix configuration. - All four selected gene reviews validate without warnings and all affected
gene, module, and project pages have been regenerated. The sole module
warning is the documented OAK configuration gap for InterPro label lookup.
Wave 114 Repair Audit (2026-09-01)
The reusable boundary remains endogenous protein lipoylation: the direct
LipB-LipA route and the separately characterized Bacillus subtilis and human
GcvH/GCSH relay routes are modeled as topological variants. Free-lipoate
salvage, octanoyl-ACP production, and downstream lipoate-dependent client
complexes remain outside. PSEPK-specific satisfiability and the unresolved
PP_0423 candidate are documented in this batch rather than in generic variant
descriptions.
ONE_OR_MORE is retained as relaxed modeling, not as a coexistence claim. The
cited work generally presents lineage-associated alternatives; neither strict
route exclusivity nor coexistence in one genome is asserted without
organism-specific evidence.
All eight variant leaves now carry reviewed UniProt exemplars with current
names. The PSEPK LipB and LipA reviews currently carry only electronic GO
evidence, so they are not described as experimentally characterized. PANTHER
and InterPro identifiers are kept
only where labels and member containment were verified. PTN000107427 is used
for bacterial LipB activity and PTN000101947 for eukaryotic lipoyl synthase
activity because the local PAINT tables directly establish those IBD claims;
uncertain relay-family PTNs are omitted. The human LIPT1 descriptor is narrowed
from the broad PTHR12561 parent to exact PTHR12561:SF3.
The mandatory annotation-reviewer pass covered every existing GOA row for
lipA, lipB, PP_0423, and PP_4797. No gene-review YAML changes were required.
The broad GO:0016746 annotation on PSEPK lipB is already correctly marked
over-annotated beside GO:0033819. PP_0423 still has no GOA rows and no asserted
MF, while PP_4797 remains an ArfA-family ribosome-rescue protein excluded from
lipoylation.
Fresh generic OpenScientist research completed with the configured 7200-second
job allowance in 1236.73 seconds. It independently recovered the direct
LipB-LipA ordering, the Bacillus octanoyl-GcvH relay before LipA sulfur
insertion, and the human LIAS-before-LIPT1 mature-lipoyl relay. The report warns
against generalizing these lineage examples and does not establish coexistence
of the modeled routes in one organism; it also retains free-lipoate salvage and
downstream client complexes outside this module.
Wave 114 is submitted for review in pull request #2890.