PA4521 (G3XD69, AmpE) notes

Identity

UniProt G3XD69, unreviewed, 278 aa; locus PA4521, no gene name in the entry
(name "AmpE protein" is ProtNLM-generated). Five TM helices (Phobius),
Pfam PF17113 AmpE, InterPro IPR031347 AmpE, IPR052966
Beta-lactamase_Reg; PANTHER PTHR38684 (PROTEIN AMPE). Second gene of the
ampDE operon (ampD = PA4522).

Literature

Langaee et al. 1998 (PMID:9835532; abstract only, PMC full text not
retrievable): cloned and characterised P. aeruginosa ampD and ampE as an
operon; the only functional result in the abstract is for ampD
[PMID:9835532 "These genes are transcribed in the same orientation and form an
operon."; "The ampD product transcomplemented Escherichia coli ampD mutants to
wild-type beta-lactamase expression."]. No ampE mutant or ampE-specific
phenotype is mentioned in the abstract.
P. aeruginosa clinical isolates (Juan et al. 2005; abstract): some ampD
mutants were complemented only by ampDE, and the authors infer an indirect
role for AmpE PMID:16251318.
E. coli/Enterobacter AmpE: the two 1989 papers disagree. Honoré et al.:
AmpE is a cytoplasmic membrane protein needed for induction
PMID:2607970; they also note E. coli lacks ampR
PMID:2607970. Lindquist et al.: AmpE is an integral membrane
protein but not needed for induction PMID:2691840.

Participation test

If AmpE relays a peptidoglycan-recycling signal to AmpR it would take part in
the beta-lactam induction response, but whether it does is contested in the
enterobacteria and only indirectly supported in P. aeruginosa. The cited
paper's abstract shows no ampE phenotype, so the PseudoCAP IMP cannot be
checked. "Response to antibiotic" is a broad phenotype-level term; for AmpE it
would hold only where an inducible AmpC/AmpR system exists, and E. coli (a
leaf) has no ampR. The membrane location is solid (5 TM helices; E. coli
AmpE a cytoplasmic membrane protein).

Clade check: PTN002214697 leaves are E. coli AmpE (P0AE14), Salmonella,
Yersinia, Coxiella, Shewanella and Xanthomonas AmpE homologues and PA4521.
None of the leaves has experimental support for a response-to-antibiotic role
of AmpE itself in this record set.

PseudoCAP IBA-seed assessment

PTN node	node term	n_seeds	leaf IBAs	seed annotation (term, evidence, PMID)	verdict	rationale
PTN002214697	GO:0046677 response to antibiotic	1 (sole)	7 (7 taxa)	GO:0046677, IMP, PMID:9835532 (PseudoCAP)	UNVERIFIABLE	Abstract only; it reports ampD complementation, not an ampE mutant. AmpE's role in beta-lactamase induction is contested (PMID:2607970 vs PMID:2691840) and only indirect in P. aeruginosa (PMID:16251318). A broad antibiotic-response phenotype is a poor clade term, and E. coli, a leaf, has no AmpR-inducible AmpC.

Decisions summary

ACCEPT: plasma membrane (IBA).
UNDECIDED: response to antibiotic (IMP, paper not readable; and the IBA,
kept consistent), with a GO_CENTRAL CHECK defect on PTN002214697 arguing the
phenotype term is a poor clade term.

PANTHER tree check (2026-11-05)

Topology of PTN002214697 fetched with projects/PSEUDOCAP/panther_tree.py node PTN002214697 --depth 4 / --leaves (PANTHER treeinfo for PTHR38684).

PTN002214697: SPECIATION, Gammaproteobacteria, 7 leaves, all PTHR38684:SF1.

child node	event / taxon	leaves	subfamilies	content	verdict
PTN002214698	LEAF Coxiella burnetii	1	SF1	ampE, "beta-lactamase induction protein"	AmpE ortholog
PTN002214699	LEAF Shewanella oneidensis	1	SF1	ampE, "AmpD signaling modulator AmpE"	AmpE ortholog
PTN002214701	LEAF Xanthomonas campestris	1	SF1	XCC0490, transmembrane protein	AmpE ortholog
PTN002446049	LEAF P. aeruginosa	1	SF1	PA4521, AmpE	the seed leaf
PTN002214702	SPECIATION Enterobacteriaceae	3	SF1	Salmonella, Yersinia, E. coli ampE	AmpE orthologs

Conclusion: the clade is homogeneous -- one AmpE ortholog per organism, a single
subfamily, no duplication or horizontal-transfer event anywhere below the node, and the
node is exactly the gammaproteobacterial AmpE orthology group. The earlier worry that
"a broad antibiotic-response phenotype is a poor clade-level term" because the clade is
too broad or mixed does not hold, and that part of the concern is withdrawn.

The defect is therefore left at SUSPECTED, with the explanation rewritten to say
what the tree does and does not settle. What remains in doubt is the seed itself: the
PseudoCAP IMP rests on PMID:9835532, whose abstract describes only AmpD, and AmpE's
role in beta-lactamase induction is contested (PMID:2607970 versus PMID:2691840) with
P. aeruginosa data suggesting an indirect role (PMID:16251318). Because the clade is a
clean ortholog set, the term is either right for all 7 leaves or wrong for all 7 --
there is no relocation or partial negation available. Only the full text of
PMID:9835532 (whether an ampE mutant was assayed) would settle it; topology cannot.
No annotation actions changed (the GO:0046677 IBA and IMP rows stay UNDECIDED).