AIGR Focused Hypothesis Report — eno (Q88MF9, PSEPK) extracellular region (GO:0005576)

Gene: eno (enolase / phosphopyruvate hydratase), Pseudomonas putida KT2440
UniProt: Q88MF9 (ENO_PSEPK, reviewed Swiss-Prot), EC 4.2.1.11, 429 aa
Focus: function-assignment — "eno has extracellular region (GO:0005576)"
Annotation under review: GO:0005576, IEA, GO_REF:0000044

Executive Judgment

Verdict: Over-annotated (weakly supported for this organism).

The extracellular-region assignment is a purely computational propagation, not an
observation about P. putida enolase. QuickGO confirms GO:0005576 on Q88MF9 is
IEA / GO_REF:0000044 / ECO:0007322, mechanically mapped from the UniProtKB
"Secreted" subcellular-location term. That "Secreted"/"Cell surface" location
is itself asserted by HAMAP rule MF_00318 (ECO:0000255, by sequence
similarity) for every bacterial enolase matching the rule, carrying the generic
note "Fractions of enolase are present in both the cytoplasm and on the cell
surface." There is no experimental (non-IEA) evidence for extracellular or
cell-surface enolase in P. putida.

The moonlighting surface/secreted enolase phenomenon is real biology, but the
primary literature that demonstrates it is confined to pathogens and parasites
(M. tuberculosis, Mycoplasma, Leptospira, Streptococcus, Bifidobacterium, Giardia),
where surface enolase acts as a plasminogen receptor contributing to virulence.
P. putida KT2440 is a non-pathogenic soil saprophyte; extending a virulence-linked
moonlighting location to it by rule is a frequency/paralog-style over-annotation.

The gene product's direct, primary function is a cytoplasmic Mg²⁺-dependent
glycolytic lyase (2-phosphoglycerate ⇌ phosphoenolpyruvate). Extracellular
localization is at best a context-specific secondary role and, for this organism,
unsupported by evidence.

A cross-ortholog audit (Iteration 3) makes the over-annotation explicit: the same
IEA:UniProtKB-SubCell extracellular term is attached to E. coli enolase (a
canonical cytoplasmic protein with no surface evidence) but is absent from
S. pneumoniae and S. aureus enolases, where surface localization is
experimentally established. The annotation is thus anti-correlated with real
biology and carries no organism-specific information for P. putida. No
experimental GO:0005576 annotation exists on any enolase examined.

Evidence Matrix

GO Curation Implications

Lead (requires curator verification): do NOT accept GO:0005576 as a supported/core
annotation for eno in this review; treat it as non-core / candidate for removal or
downgrade.

• The annotation is a mechanical IEA / GO_REF:0000044 SubCell mapping seeded by a
  pan-bacterial HAMAP rule, not evidence that P. putida enolase directly resides in
  the extracellular region. Under AIGR guidance, a rule-propagated CC term with no
  organism-specific support and a plausible over-annotation mechanism should not be
  marked as a demonstrated function.
• Cellular component that IS supported: GO:0005737 (cytoplasm) — consistent with
  the enzyme's characterized function and the HAMAP cytoplasm assertion.
• The related IEA companions GO:0009986 (cell surface) and the "Secreted" keyword ride
  on the same rule and inherit the same weakness; if the curator removes/downgrades
  GO:0005576, the cell-surface term should be handled identically.
• Molecular function / Biological process (the true core): phosphopyruvate hydratase
  activity (GO:0000287 Mg binding; GO:0004634 phosphopyruvate hydratase activity) and
  glycolytic process (GO:0006096) — these are the informative, defensible annotations.

Recommended action label: NOT_SUPPORTED / mark as over-annotation for GO:0005576
(pending curator confirmation), with rationale "IEA rule propagation of a
pathogen-associated moonlighting location to a non-pathogen; no organism-specific evidence."

Mechanistic Scope

• Direct molecular activity being annotated by CC term: physical presence of the
  enolase polypeptide in the extracellular region.
• Established direct activity of the gene product: cytoplasmic, homodimeric,
  Mg²⁺-dependent lyase catalyzing 2-PG ⇌ PEP in glycolysis/gluconeogenesis.
• The extracellular claim is downstream/secondary at best: in organisms where it
  is real, surface enolase is a moonlighting plasminogen receptor reached by a
  non-classical (signal-peptide-independent) export route and re-association with the
  envelope — a virulence-linked secondary role, not the enzyme's primary catalytic job.
• For P. putida there is no demonstrated export, no demonstrated surface display, and
  no functional consequence reported; the CC term reflects rule inference only.

Conflicts and Alternatives

• Rule/database carry-over (most likely explanation): HAMAP MF_00318 applies
  Secreted/Cell surface uniformly, so the annotation reflects the rule, not the strain.
• Organism mismatch: all supporting experimental data are from pathogens/parasites
  that exploit host plasminogen; P. putida is an environmental non-pathogen with no
  such host interface, weakening biological plausibility.
• Mechanistic requirement not evidenced: surface plasminogen binding depends on a
  C-terminal lysine and specific internal Lys sites (e.g., PMID 27569900, 28770921);
  whether Q88MF9 has the relevant C-terminal/internal Lys determinants was not tested
  here and would still not establish localization in P. putida.
• No competing experimental annotation places enolase extracellularly in Pseudomonas;
  the only conflict is between an automatic prediction and the absence of evidence.

Knowledge Gaps

1. Organism-specific localization. Checked: QuickGO/UniProt evidence codes — all IEA.
   Gap: no P. putida surface-proteome / secretome / OMV proteomics located enolase.
   Resolution: strain-specific surface biotinylation, secretome LC-MS/MS, or
   immuno-EM/flow cytometry.
2. Export mechanism. Enolase is leaderless; whether P. putida has any route to
   externalize it is unknown. Resolution: fractionation + membrane-integrity controls to
   exclude lysis artifacts.
3. Sequence determinants. Not analyzed here: presence/absence of the C-terminal and
   internal lysine plasminogen-binding motifs in Q88MF9. Resolution: alignment to
   experimentally validated surface enolases; but positive motif ≠ localization proof.

Discriminating Tests

• Surface-shaving / biotinylation proteomics of intact P. putida KT2440 cells with
  strict cytoplasmic-contamination controls (e.g., absence of other abundant cytoplasmic
  markers) — the single most decisive experiment.
• Secretome and OMV LC-MS/MS under relevant growth conditions to see if enolase is
  externalized at all.
• Flow cytometry / immuno-EM with anti-enolase antibody on non-permeabilized cells.
• Comparative bioinformatics: align Q88MF9 to validated surface enolases (Mtb, Strep)
  to check C-terminal Lys / plasminogen-binding motif conservation — supportive context,
  not proof.

Curation Leads (require curator verification)

• Action change: downgrade/remove GO:0005576 (extracellular region) as a supported
  annotation for eno; classify as over-annotation from IEA rule propagation. Apply the
  same treatment to GO:0009986 (cell surface) and the "Secreted" keyword.
• Retain / promote: GO:0005737 (cytoplasm, CC), GO:0004634 (phosphopyruvate hydratase
  activity, MF), GO:0000287 (magnesium ion binding, MF), GO:0006096 (glycolytic process, BP).
• Provenance to verify in the record:
• QuickGO: GO:0005576 | IEA | GO_REF:0000044 | ECO:0007322 | assignedBy UniProt.
• HAMAP MF_00318 subcellular-location note: "Fractions of enolase are present in both the
  cytoplasm and on the cell surface" (rule-level, ECO:0000255) — the origin of the term.
• Candidate reference snippets (pathogen context, to justify "not transferable"):
• PMID 27569900: "Enolase, a glycolytic enzyme, has long been studied as an anchorless
  protein present on the surface of many pathogenic bacteria that aids in tissue remodeling
  and invasion by binding to host plasminogen."
• PMID 23872606: "They localise to the bacterial surface to take on additional activities,
  which have been hypothesised to contribute to bacterial virulence or bacterial benefit."
• Suggested curator question: Is there any P. putida (or close-relative) surface/secretome
  proteomic dataset placing enolase extracellularly? If not, the term stands only on rule inference.
• Suggested experiment: surface biotinylation + LC-MS/MS on intact KT2440 cells with
  cytoplasmic-leakage controls.

Cross-Ortholog Evidence-Code Audit (Iteration 3, computed)

Question: is the extracellular/cell-surface GO term on bacterial enolases ever
experimental, and does its presence track real surface biology?

Reading: the extracellular/cell-surface term is applied by rule to enolases where
surface display is not experimentally supported (E. coli), yet is absent from
several enolases where it is experimentally proven (S. pneumoniae, S. aureus,
one Mtb entry). The annotation is therefore anti-correlated with real biology —
classic frequency-bias/rule over-annotation that carries no organism-specific
information for P. putida. Every occurrence found is IEA:UniProtKB-SubCell; no
experimental GO:0005576 annotation exists on any enolase examined.

Sequence-Feature Provenance (Iteration 2, computed)

Direct check of Q88MF9's physical plausibility for extracellular localization (public FASTA):

Reading: the sequence is that of a classical cytoplasmic glycolytic enzyme with no
export signal and without the C-terminal lysine motif linked to plasminogen-binding
surface enolases. Any extracellular presence would require a non-classical export route
that has not been demonstrated in P. putida. This is supportive computational context,
reported conservatively — absence of a motif is not proof of non-localization, but it
provides no positive support for the GO:0005576 hypothesis and is consistent with the
over-annotation conclusion.

Limitations

This assessment relied on public UniProt/QuickGO/HAMAP records and PubMed; local
*-bioinformatics analyses were intentionally withheld. Targeted PubMed queries for
P. putida enolase surface/secretome returned no hits, which is itself informative
(absence of organism-specific evidence) but not absolute proof of non-localization.
No sequence-motif analysis of Q88MF9's lysine determinants was performed.

Artifacts

• OpenScientist final report
• OpenScientist final report