AIGR Gene Hypothesis Deep Research — Dictyostelium discoideum ACG/acgA (UniProt Q03101)

Focus: function_assignment — does ACG directly have peptide receptor activity (GO:0001653)?
Hypothesis slug: function-hypothesis-go-0001653
Source: genes/DICDI/acgA/acgA-ai-review.yaml (free-text)

Summary

The GO:0001653 "peptide receptor activity" annotation on Dictyostelium discoideum ACG/acgA is over-annotated and should be flagged by a curator for removal or NOT-qualification (or at minimum demoted to non-core, low-confidence status). The annotation is supported only by phylogenetic inference (IBA:GO_Central), whereas ACG's experimentally characterized molecular functions — adenylate cyclase activity, osmosensor activity, and dimerization — rest on direct assays. Decisively, the QuickGO provenance shows the peptide-receptor term was propagated within PANTHER family PTN000229249 from mammalian natriuretic peptide receptors NPR1 (P16066) and NPR2 (P20594) — receptor guanylyl cyclases that are the genuine peptide receptors — simply because ACG shares their "extracellular sensor + single transmembrane + class-III cyclase" architecture.

ACG's real molecular biology contradicts a direct peptide-receptor role. It is a primary osmosensor: expressed in a yeast adenylyl-cyclase-null background, it reconstitutes osmotic activation with no Dictyostelium-specific ligand or auxiliary sensor, proving the sensor is intramolecular and ligand-independent. Its extracellular module is a CHASE small-molecule sensory domain (Pfam PF03924), a domain family characterized as binding small molecules such as cytokinins — not peptides — and in Dictyostelium cytokinins are sensed by the histidine kinase DhkB, not ACG.

On the specific SDF-1 sub-question, the answer is clear: there is no evidence that the phosphopeptide SDF-1 is sensed by ACG, directly or as an upstream dependency. SDF-1 is a PKA-substrate phosphopeptide that acts through an intracellular PKA-dependent pathway at a different developmental stage (prespore→spore encapsulation) than ACG (osmolarity-dependent inhibition of mature-spore germination). SDF-1's receptor has never been identified, but no primary study links it to ACG. The most important caveat is that "refuted" reflects an evidence-of-absence plus positive-provenance argument — no published assay has tested and excluded ACG peptide binding directly — so the recommendation is stated conservatively.

Executive Judgment

Verdict: Over-annotated / Refuted for the seed hypothesis as stated (direct peptide receptor activity).

Three converging lines of evidence support removal or NOT-qualification of GO:0001653 on ACG:

1. The annotation is phylogenetic, not experimental. In UniProt Q03101, GO:0001653 carries evidence code IBA:GO_Central (Inferred from Biological Ancestor, GO_REF:0000033). ACG's experimentally supported functions carry IDA/IPI codes. No IDA/IMP/IPI annotation supports peptide receptor activity.

2. The provenance is a family carry-over from bona fide peptide receptors. QuickGO shows the IBA term was propagated within PANTHER family PTN000229249 from human/mouse/rat natriuretic peptide receptors NPR1/NPR2 — receptor guanylyl cyclases that truly bind natriuretic peptides.

3. ACG's real sensory mechanism is intramolecular and ligand-independent. ACG reconstitutes osmotic activation in yeast with no partner, and its extracellular module is a CHASE small-molecule domain, not a peptide-binding fold.

Regarding SDF-1: there is no evidence it is sensed by ACG; it acts intracellularly via PKA at a different stage, and its receptor is unidentified but explicitly not ACG in any primary study.

Most important caveat: "Refuted" means the direct-receptor claim is unsupported and the GO annotation is family carry-over. It does not mean ACG was experimentally tested and shown unable to bind any peptide — no such negative binding assay exists. The osmosensor role does not logically exclude peptide sensing, but the hypothesis requires positive evidence that is simply absent.

Key Findings

Finding 1 — GO:0001653 rests only on phylogenetic inference (IBA), not experiment

The UniProt Q03101 GO cross-references show that GO:0001653 (peptide receptor activity) carries evidence code IBA:GO_Central — an annotation inferred phylogenetically from an ancestral protein via the GO_Central/PAINT pipeline (GO_REF:0000033). This is a computational/curatorial inference, not a wet-lab result. By contrast, ACG's molecular-function annotations with experimental backing are:

The domain architecture of ACG is fully consistent with a receptor-cyclase and provides no dedicated peptide-binding module: an N-terminal signal-anchor transmembrane segment (residues 19–41), an extracellular CHASE sensory domain (~86–317; Pfam PF03924 / InterPro IPR006189), and a C-terminal class-III adenylyl/guanylyl-cyclase catalytic domain (~396–526; Pfam PF00211). An IBA-only molecular-function term that conflicts with the experimentally characterized function should be treated as low-confidence and reviewed for removal or NOT-qualification.

Finding 2 — ACG's extracellular sensor is a CHASE (small-molecule) domain, and activation is intramolecular/osmotic, not peptide-mediated

Two primary papers establish ACG as a self-contained osmosensor:

• Saran & Schaap, 2004 PMID: 14718564: When ACG cDNA is expressed in a yeast adenylyl-cyclase-null mutant, cAMP production is activated by high osmolality "similarly … as in Dictyostelium." Because yeast lacks any Dictyostelium-specific ligand or auxiliary sensor, this demonstrates the osmosensor is intramolecular. The authors conclude ACG is "the first characterized primary osmosensor in eukaryotes." They further show dimerization (via a region outside the catalytic domain) is required for catalytic activity, but the dimer/monomer ratio does not change with osmolality — osmotic activation is not mediated by ligand-induced dimerization.

• van Es et al., 1996 PMID: 8798577: acg⁻ spores lose osmolarity-dependent inhibition of germination; ACG activity (in aca⁻/ACG cells) is "strongly stimulated by high osmolarity with optimal stimulation occurring at 200 milliosmolar."

Mechanistically, ACG's extracellular module is a CHASE domain (Pfam PF03924). Characterized CHASE domains bind small molecules — most notably cytokinins in plant hybrid histidine-kinase receptors (PMID: 17439640, PMID: 15498549) — not peptides. In Dictyostelium itself, cytokinins act through the histidine kinase DhkB, not ACG (PMID: 18216168), underscoring that ACG's CHASE domain is not the organism's peptide/hormone receptor. ACG activation is therefore an osmotically driven intramolecular conformational change, with no evidence of peptide-ligand binding.

Finding 3 — No evidence links phosphopeptide SDF-1 to ACG; SDF-1 acts intracellularly via PKA at a different developmental stage

The seed asks specifically whether the phosphopeptide SDF-1 is sensed by ACG. Primary literature answers in the negative:

• Anjard, van Bemmelen, Véron & Reymond, 1997 PMID: 9373946: SDF-1 is "a small, thermostable phospho-polypeptide." In-vitro dephosphorylation reduces its spore-differentiation activity, and PKA re-phosphorylation restores it — SDF-1 is a PKA substrate, and "the response of prespore cells to SDF involves an intracellular pathway dependent on PKA."

• Anjard, Zeng, Loomis & Nellen, 1998 PMID: 9473320: tagC⁻, dhkA⁻, and regA⁻ null strains still sporulate in response to SDF-1, showing SDF-1 does not use the DhkA two-component pathway (which transduces the distinct peptide SDF-2). SDF-1's receptor was not identified.

Crucially, SDF-1 and ACG operate at different developmental stages: SDF-1 governs prespore-cell encapsulation (sporulation), whereas ACG governs osmolarity-dependent inhibition of mature-spore germination (PMID: 8798577). No primary study reports SDF-1 binding to or signaling through ACG. The seed's "distinguish dependence from direct receptor action" framing collapses: there is neither a demonstrated dependence of SDF-1 signaling on ACG nor any direct SDF-1–ACG interaction.

Finding 4 — The IBA "peptide receptor activity" was propagated from natriuretic peptide receptors NPR1/NPR2

The decisive provenance comes from the QuickGO annotation record for Q03101 / GO:0001653 (evidence ECO:0000318/IBA, GO_REF:0000033, assignedBy GO_Central, PANTHER family PTN000229249). Its withFrom field lists the seed proteins:

P16066 and P20594 both carry GO:0016941 (natriuretic peptide receptor activity) and GO:0004383 (guanylate cyclase activity) — bona fide peptide (natriuretic) receptors. ACG shares the receptor-cyclase architecture (extracellular sensor + single TM + class-III cyclase catalytic domain, annotated PF00211 "Guanylate_cyc"), causing it to cluster in the same PANTHER family and to inherit "peptide receptor activity" by descent. This is a textbook case of family-level over-annotation: a genuine molecular function of the mammalian members is projected onto a divergent social-amoeba enzyme whose actual ligand is osmotic pressure, not a peptide.

Mechanistic Model / Interpretation

The findings assemble into an internally consistent picture that separates what ACG actually does from what the GO term claims.

┌──────────────────────────────────────────────────────────┐
│  ACG (Q03101) domain architecture                        │
│                                                          │
│   TM(19–41)   CHASE (86–317)        Class-III cyclase     │
│   signal-     small-molecule/       catalytic (~396–526)  │
│   anchor      osmo-sensory          PF00211               │
│     │         PF03924                    │                │
│  ───┴─────────[   sensor   ]────TM────[ ATP → cAMP ]      │
└──────────────────────────────────────────────────────────┘
      ▲                                        │
      HIGH OSMOLALITY (intramolecular)                 ▼
      optimal ~200 mOsm  ──► conformational ──►  cAMP ↑ ──► PKA
      (reconstitutes in yeast, no ligand)                    │
                                                     ▼
                                  INHIBITION of spore germination

ACG's verified role (spore germination, mature spore): high external osmolality is sensed intramolecularly by ACG, stimulating adenylyl cyclase activity, raising cAMP, activating PKA, and keeping spores dormant until dispersal to low-osmolality environments. No peptide and no partner receptor is required — the yeast reconstitution proves it.

The SDF-1 pathway is a separate module (sporulation, prespore cells): SDF-1 is an intracellular PKA-substrate phosphopeptide promoting prespore→spore encapsulation. It does not use DhkA/TagC/RegA, and its surface receptor (if any) is unknown. It has no demonstrated connection to ACG.

  Peptide/ligand signaling in Dictyostelium sporulation (for contrast — NOT ACG):
    SDF-2 peptide  ──►  DhkA (histidine kinase receptor) ──┤ RegA PDE ──► cAMP↑ ──► PKA
    SDF-1 phospho- ──►  (unknown receptor) ── intracellular PKA-dependent pathway
      peptide
    GABA           ──►  GrlE (GPCR) ──► AcbA release ──► TagC cleaves ──► SDF-2
    cytokinins     ──►  DhkB (histidine kinase)

The seed hypothesis conflates two things: (a) ACG's architectural resemblance to mammalian peptide-receptor cyclases, and (b) the existence of peptide signals (SDF-1/SDF-2) elsewhere in Dictyostelium sporulation. Neither establishes that ACG is a peptide receptor. The genuine peptide receptor in this system is DhkA (for SDF-2), a histidine kinase — architecturally and mechanistically distinct from ACG.

GO decision summary:

Evidence Base / Evidence Matrix

GO Curation Implications

Lead (requires curator verification): Flag GO:0001653 (peptide receptor activity) on Q03101 for removal or NOT-qualification, or at minimum reclassify as non-core, low-confidence propagated (IBA) annotation.

• Aspect: Molecular Function.
• Rationale: The term is IBA-only and its withFrom provenance traces to natriuretic peptide receptors NPR1/NPR2. ACG's experimentally supported MF terms (adenylate cyclase, osmosensor) and its intramolecular, ligand-independent activation contradict a direct peptide-receptor function. This is family-level over-annotation driven by shared receptor-cyclase architecture and the PF00211 "Guanylate_cyc" catalytic-domain label.
• Retain as core: GO:0004016 (adenylate cyclase activity, IDA) and GO:0005034 (osmosensor activity, IDA).
• Do not substitute "protein binding": The informative, supported terms are the adenylate-cyclase and osmosensor activities; GO:0042802 (identical protein binding, IPI) captures dimerization but is not a substitute for core catalytic/sensory functions.
• If NOT-qualification is preferred over deletion: annotate GO:0001653 with a NOT qualifier and an ECO code reflecting curatorial review, citing the intramolecular osmosensor reconstitution (PMID: 14718564) and the absence of any peptide-binding assay.

Mechanistic Scope

The molecular function under test is direct peptide receptor activity — ACG binding an extracellular peptide ligand at its own extracellular domain and transducing that binding into a signaling output.

• Direct gene-product activity (supported): osmolality sensing (intramolecular) and ATP→cAMP catalysis — properties of the ACG polypeptide itself, demonstrated by heterologous reconstitution and enzyme assays.
• Downstream / pathway consequences (not the MF being tested): cAMP→PKA activation→inhibition of spore germination — a developmental output, not evidence of a receptor–peptide interaction.
• Separate signaling modules (not ACG): SDF-1 (intracellular PKA substrate), SDF-2 (peptide sensed by DhkA), GABA (via GrlE GPCR), cytokinins (via DhkB). None implicates ACG as a peptide receptor.

There is no experimental measurement of ACG at the level required to support GO:0001653 — a binding assay or a functional readout demonstrating peptide-dependent activation of ACG.

Conflicts and Alternatives

1. Paralog/family confusion (primary explanation). ACG co-clusters in PANTHER PTN000229249 with mammalian natriuretic peptide receptors, which are peptide receptors; ACG is not. IBA cannot distinguish the divergence of the extracellular sensor (CHASE small-molecule/osmo-sensor in ACG vs. natriuretic-peptide-binding ectodomain in NPR1/2), so it over-projects the term.
2. Architecture-driven false positive. ACG's catalytic domain is annotated PF00211 ("Guanylate_cyc"), reinforcing the guanylyl-cyclase-receptor analogy even though ACG functions as an adenylyl cyclase — likely amplifying the mis-annotation.
3. Organism-specific divergence. Dictyostelium uses histidine-kinase receptors (DhkA, DhkB) and GPCRs (GrlE) for peptide/hormone signaling — not receptor guanylyl/adenylyl cyclases. The mammalian receptor-cyclase paradigm does not map onto ACG's biology.
4. Alternative reading of the seed's "dependence" clause. One could hypothesize SDF-1 signaling depends on ACG-generated cAMP. But SDF-1 acts in prespore cells during sporulation, whereas ACG functions in mature spores during germination; the stages and cell types differ, and no genetic epistasis links them. The dependence interpretation is unsupported.
5. What would genuinely conflict with the refutation: a direct SDF-1 (or other peptide) binding assay on ACG, or an acg⁻ phenotype in SDF-1-driven sporulation. Neither exists in the literature reviewed.

Limitations and Knowledge Gaps

Discriminating Tests

1. Direct peptide-binding assay on ACG. Purify the ACG extracellular/CHASE domain (~86–317) and test binding to SDF-1 and other candidate peptides by SPR, ITC, or MST. No binding → refutes conclusively; specific binding → reopens the hypothesis.
2. Genetic epistasis of SDF-1 vs. ACG. Test whether acg⁻ cells retain a normal SDF-1 sporulation response. If yes, SDF-1 signaling is ACG-independent.
3. Yeast reconstitution + peptide challenge. Extend the Saran & Schaap yeast system (PMID: 14718564) by adding candidate peptides to test for peptide-dependent ACG activation beyond osmotic stimulation.
4. Structural pocket analysis. Build an AlphaFold model of the ACG ectodomain and computationally dock SDF-1; compare its pocket geometry to the NPR1/NPR2 natriuretic-peptide-binding ectodomain to confirm fold divergence.
5. PANTHER tree re-inspection. Verify that GO:0001653 propagation in PTN000229249 originates solely from natriuretic peptide receptors and that no experimentally annotated non-mammalian member independently supports peptide binding.

Proposed Follow-up Experiments / Actions (Curation Leads)

All items are leads requiring curator verification.

Candidate action change:
- Set GO:0001653 (peptide receptor activity) to removed or NOT-qualified for Q03101; document as IBA family carry-over from natriuretic peptide receptors. Retain GO:0004016 and GO:0005034 as core MF terms; retain GO:0042802 for dimerization.

Candidate references + snippets to verify:
- PMID: 14718564 — "This strongly suggests that the ACG osmosensor is intramolecular, which would define ACG as the first characterized primary osmosensor in eukaryotes."
- PMID: 8798577 — "ACG is an osmosensor controlling spore germination through activation of protein kinase A."
- PMID: 9373946 — "the response of prespore cells to SDF involves an intracellular pathway dependent on PKA."
- PMID: 9473320 — tagC⁻/dhkA⁻/regA⁻ strains "all sporulated efficiently when SDF-1 was added."
- PMID: 18216168 — cytokinins act "through a different histidine kinase, DhkB."

Candidate retained GO terms:
- MF: GO:0004016 (adenylate cyclase activity) — retain, core.
- MF: GO:0005034 (osmosensor activity) — retain, core.
- MF: GO:0042802 (identical protein binding) — retain (dimerization).
- BP: consider osmosensory-signaling / regulation of spore germination terms consistent with PMID: 8798577.

Suggested curator questions:
- Does any experimentally annotated member of PANTHER PTN000229249 support GO:0001653 independently of natriuretic peptide receptors?
- Is there any post-2004 primary study assigning a peptide ligand to ACG? (None found here.)

Suggested experiments: Items 1–3 under Discriminating Tests (direct binding assay; acg⁻ SDF-1 epistasis; yeast reconstitution + peptide challenge).

Notes on Reproducibility and Provenance

The conclusions rest on (a) the evidence-code contrast in UniProt Q03101 (IBA for GO:0001653 vs. IDA/IPI for characterized functions), (b) the QuickGO withFrom provenance listing NPR1/NPR2 as the IBA seed proteins, and (c) primary experimental papers on ACG osmosensing and SDF-1 signaling. Where a specific value (e.g., PANTHER family ID PTN000229249) is quoted, it should be re-verified by the curator at the current database version, since IBA propagations can change over time. No peptide-binding assay for ACG exists in the literature reviewed; the refutation is therefore an evidence-of-absence plus positive-provenance argument, stated conservatively.

Prepared for AI Gene Review — hypothesis-level curation support. All annotations flagged here are leads requiring curator verification.

Artifacts

• OpenScientist artifact GO decision table
• OpenScientist artifact IBA provenance GO0001653
• OpenScientist final report
• OpenScientist final report