PA4088 (Q9HWU0) review notes

No deep-research provider was available; notes compiled from UniProt (unreviewed), the cached abstract of PMID:29684343 (full text behind subscription), a Europe PMC search, and a sequence analysis in PA4088-bioinformatics/.

Structure: PMID:29684343; PMID:29684343; PMID:29684343. No activity assay in the abstract.
Europe PMC search "(PA4088 OR glutamate-1-semialdehyde) AND aeruginosa" (2026-11-05) found no paper reporting an activity for PA4088.
PAO1 has a reviewed canonical HemL, PA3977 (P48247, 427 aa).
PANTHER node PTN000241400 (15 leaves) contains canonical HemLs plus PA4088, PA5523 and B. japonicum blr6632 and blr3552.
Sequence identity (file:PSEAE/PA4088/PA4088-bioinformatics/RESULTS.md): E. coli HemL vs PAO1 HemL 65.7%; HemL vs PA4088 about 27%; PA4088 vs PA5523 49.9%; PA4088 vs other PAO1 class-III aminotransferases 21-24%.

PseudoCAP IBA-seed assessment

PTN node	node term	n_seeds	leaf IBAs	seed annotation (term, evidence, PMID)	other seeds	IBD stands without this seed?	verdict	rationale
PTN000241400	GO:0042286 glutamate-1-semialdehyde 2,1-aminomutase activity	2	15 (12 taxa)	GO:0042286, EXP, PMID:29684343 (PseudoCAP)	E. coli HemL P23893	For the canonical HemL subclade yes; for the PA4088/PA5523/blr paralog lineage no	OVER-SPECIFIC	Structure-only paper supports PLP binding and GSAM-like family membership, not GSAM activity; PA4088 is a distant paralog in a genome with its own HemL.
PTN000241365	GO:0016853 isomerase activity	3	100 (69 taxa)	same EXP row (propagated to the isomerase parent)	E. coli HemL P23893; Arabidopsis GSA (AT5G63570)	Yes, on E. coli and plant GSAM	OVER-SPECIFIC	IBD stands on the two GSAM seeds; the broad term at the family root is acceptable for GSAMs. No leaf IBA of this term appears on PA4088 in GOA.

Defects: PSEUDOCAP WRONG_TERM on the EXP row (PROBABLE; abstract read only, plus sequence data). GO_CENTRAL on PTN000241400 (now PROBABLE; see the tree check below). No GO_CENTRAL defect on PTN000241365.

PANTHER tree check (2026-11-05)

Topology of PTN000241400 in PTHR43713 fetched with projects/PSEUDOCAP/panther_tree.py node PTN000241400 --depth 4 and --contains. The node is a SPECIATION node (Proteobacteria), 15 leaves, all PTHR43713:SF3, 13 of them named hemL.

Child node	Event	Leaves	Subfamilies	Contents	Verdict
PTN002885326 / PTN002885327 / PTN000241401	LEAF	1 each	SF3	hemL of H. pylori, N. meningitidis, G. sulfurreducens	GSAM sound
PTN000241402	DUPLICATION	2	SF3	Bradyrhizobium blr6632, blr3552 (no hemL-named copy in the family)	GSAM unverified, separate duplication
PTN000241405	SPECIATION (Gammaproteobacteria)	10	SF3	hemL of Coxiella, Shewanella, Vibrio, Xanthomonas; PTN002463962; PTN001736628	GSAM sound
PTN002463962	DUPLICATION	3	SF3	PA4088 (PTN002463963), PA5523 (PTN002463964), PAO1 HemL P48247 (PTN002463965)	HemL retained in P48247; PA4088/PA5523 diverged
PTN001736628	SPECIATION (Enterobacteriaceae)	3	SF3	Salmonella, Yersinia, E. coli hemL (P23893, the seed)	GSAM sound

Two things change the earlier call:

The premise that PA4088, PA5523, blr6632 and blr3552 form a single divergent lineage is wrong. The Bradyrhizobium pair is its own duplication near the base of the node; PA4088 and PA5523 are Pseudomonas-specific, and their closest relative in the tree is PAO1 HemL itself. --contains Q9HWU0,Q9HT50 returns PTN002463962, a three-leaf duplication that also holds P48247; --contains Q9HWU0,P23893 returns PTN000241405.
Consequently the IBD cannot be moved to a HemL-only subclade — no such clade exists, because the paralogs are nested inside the HemL radiation. The node placement for the GSAM term is sound; the over-annotation is the post-duplication propagation to PA4088 and PA5523. The right fix is a NOT/loss on those two leaf branches (PTN002463963, PTN002463964), and the defect explanation was rewritten to ask for that instead of relocation.

The tree raises confidence from SUSPECTED to PROBABLE: it confirms a recent lineage-specific duplication in which one copy keeps the hemL name and annotation while the other two are named only "probable aminotransferase" at ~27% identity — the classic paralog over-annotation shape. It does not establish that PA4088 has lost GSAM activity; only an assay (or the full text of PMID:29684343) would do that, so the call is not CONFIRMED.