GND1 (Candida albicans SC5314, A0A1D8PFS4) curation notes

2026-09-04 - enzyme specificity review (projects/ENZYME_SPECIFICITY.md)

GND1 was the last pending gene in the Enzyme Specificity project. It was chosen as a
check on cofactor specificity (project category 3): GO:0004616 is defined as the
NADP+-dependent reaction (EC 1.1.1.44), whereas 6PGDH enzymes also occur as NAD+-specific
and dual-specificity forms PMID:35234135.

Identity and reaction

• UniProt A0A1D8PFS4 = GND1 / orf19.12491 / CAALFM_C113860CA, 495 aa, 6PGDH family,
  EC 1.1.1.44, RHEA:10116 (from the UniProt record).
• Gnd1 is one of the two NADPH-producing dehydrogenases of the oxidative PPP in
  C. albicans PMID:22094058. The deep-research report states that the Strijbis 2012 assay
  used 6-phosphogluconate + NADP+ with NADPH monitored at 340 nm; that paper is
  abstract-only in our cache so the assay text itself could not be quoted.
• Garg et al. 2025 (full text cached) measure oxidative-PPP NADPH production in C. albicans
  lysates with 0.4 mM NADP+ and glucose-6-phosphate, noting that "ZWF1 catalysis provides
  the 6-phosphogluconate substrate for GND1" PMID:40183578; GND1 is markedly induced
  under iron starvation.

Cofactor-specificity check (GND1-bioinformatics/)

• Hanau & Helliwell 2022 describe the determinants of NADP+ specificity: the
  Gly-X-Ala-X-Met-Gly fingerprint and the Asn-Arg-Thr turn whose Asn/Arg bind the
  2'-phosphate; NAD+-preferring enzymes carry Asp-Arg-Asp PMID:35234135.
• C. albicans Gnd1 has GLAVMG at 13-18 and NRT at 36-38, coinciding with the UniProt
  NADP(+) BINDING features; no Asp-Arg-Asp. Verdict: NADP+-specific determinants present.
  See GND1-bioinformatics/RESULTS.md.
• Conclusion: GO:0004616 is the correct cofactor-specific MF; no specificity error.

Substrate over-generalisation caught (project category 1)

• GO:0019521 D-gluconate metabolic process (IEA, GO_REF:0000043) comes from the UniProt
  keyword "Gluconate utilization" (ARBA00023064 / RuleBase RU000485). The keyword is applied
  by UniProt to 6-phosphogluconate-metabolising enzymes regardless of pathway context, so it
  does not by itself imply that free D-gluconate is metabolised. On this entry the keyword is
  attached by an automated family rule [genes/CANAL/GND1/GND1-uniprot.txt:116 "KW Gluconate
  utilization {ECO:0000256|ARBA:ARBA00023064,"], i.e. it travels with 6PGDH sequence
  membership rather than with pathway context; the same keyword is carried by human PGD
  [genes/human/PGD/PGD-uniprot.txt:307 "KW Gluconate utilization"] and by the bacterial
  ortholog gntZ [genes/PSEPK/gntZ/gntZ-uniprot.txt:89 "KW Gluconate utilization
  {ECO:0000256|ARBA:ARBA00023064};"] alike. The substrate argument stands on its own: Gnd1's
  substrate is 6-phospho-D-gluconate, and in the yeast oxidative PPP it is
  supplied by Zwf1/6-phosphogluconolactonase, not from free D-gluconate. No C. albicans
  evidence for gluconate assimilation via Gnd1 was retrieved. Changed from
  KEEP_AS_NON_CORE to MARK_AS_OVER_ANNOTATED.
• Divergence noted: the human ortholog review keeps the identical keyword-derived annotation
  as KEEP_AS_NON_CORE [genes/human/PGD/PGD-ai-review.yaml:217]. Both reviews call the term a
  keyword-driven substrate generalisation; they differ on whether to record that as
  acceptable-but-peripheral or as over-annotation. genes/PSEPK/gntZ/gntZ-uniprot.txt:89
  carries the keyword in an organism that genuinely does catabolise gluconate, so the
  human/fungal/bacterial trio is the discriminating case if this is revisited.

Localization

• Predominantly cytosolic, with a minor PTS2 splice isoform in peroxisomes (about 5%)
  [PMID:22094058 "the majority is cytosolic, but a small fraction is peroxisome
  associated"; PMID:34065948 "approximately 10% and 5% of the proteins, respectively,
  were localized in peroxisomes"]. Cytosol ACCEPT; peroxisome KEEP_AS_NON_CORE.
• Biofilm-matrix detection (PMID:27609602) is a proteomic survey hit; KEEP_AS_NON_CORE.

Outcome

• Status set to COMPLETE. Two references added (PMID:34065948, PMID:35234135, both
  verified via PubMed) plus the bioinformatics RESULTS file.

2026-10-01 re-review after GOA refresh

GOA was refreshed from remote. Two previously reviewed UniProt-keyword (GO_REF:0000043) rows are no
longer present in the current GOA snapshot and were marked retired: true (reviews kept for
provenance; retirement is not a biological REMOVE judgment):

• GO:0016491 oxidoreductase activity | IEA | GO_REF:0000043 (was MARK_AS_OVER_ANNOTATED)
• GO:0019521 D-gluconate metabolic process | IEA | GO_REF:0000043 (was MARK_AS_OVER_ANNOTATED; the
  "Gluconate utilization" keyword mapping no longer yields this GO row)

New GOA row reviewed: GO:0016614 oxidoreductase activity, acting on CH-OH group of donors (IEA,
ARBA:ARBA00028437) -> MARK_AS_OVER_ANNOTATED, a generic ancestor of the annotated GO:0004616.
Other judgments unchanged.