G protein-coupled receptors (GPCRs): families, modules and functions
Bottom line: GPCRs are the largest receptor superfamily in the human genome
and the targets of a large share of approved drugs, yet only a handful have a
review here (human ADRB2, ADGRV1, FZD7, GPR101, GPR161, CASR and OPRM1, plus the transducer GNAS). This
project curates the superfamily on three levels: families (whether
PANTHER/PAINT and InterPro propagation of GPCR terms is safe), modules
(the transducer pathways downstream of the receptors), and gene functions
(per-gene reviews, starting with receptors that are drug targets). It is
paired with the dismech project
GPCR drug mechanisms and disease modules,
which uses the same receptors as the molecular anchors for drug-mechanism and
disease modules.
Why GPCRs
- Annotation volume, uneven depth.
GO:0004930G protein-coupled receptor
activity andGO:0007186G protein-coupled receptor signaling pathway are
among the most propagated terms in the corpus, mostly via IBA and
InterPro2GO. The ligand-specific MF terms (e.g.GO:0004941
beta2-adrenergic receptor activity) are much thinner, and the coupling
terms (GO:0007189adenylate cyclase-activating,GO:0007193
adenylate cyclase-inhibiting,GO:0007200phospholipase C-activating) are
inconsistently applied. - Families are large and functionally heterogeneous. Class A alone spans
aminergic, peptide, lipid, nucleotide, olfactory and orphan receptors; a
family-wide term is only safe at the right node. - Direct bridge to disease and pharmacology. The receptors carrying the
best experimental evidence are also the drug targets dismech needs as
anchors (beta-blockers, opioids, GLP-1 agonists, calcimimetics,
antipsychotics, triptans, antihistamines).
Workstream 1 — Families
Review PANTHER families for the coherence of GPCR terms, following the
Protein Families and IBA review conventions.
Candidate families (ids and names read from interpro/panther/panther.obo;
membership still to be checked against the PANTHER member index before any
module uses them):
| PANTHER id | PANTHER name | Question |
|---|---|---|
| PTHR24248 | ADRENERGIC RECEPTOR-RELATED G-PROTEIN COUPLED RECEPTOR | Which aminergic subfamilies does it contain, and at what node does Gs vs Gi/o vs Gq coupling diverge? |
| PTHR24247 | 5-HYDROXYTRYPTAMINE RECEPTOR | Same coupling question across 5-HT1/2/4/6/7. |
| PTHR24249 | HISTAMINE RECEPTOR-RELATED G-PROTEIN COUPLED RECEPTOR | H1 (Gq) vs H2 (Gs) vs H3/H4 (Gi/o). |
| PTHR24246 | OLFACTORY RECEPTOR AND ADENOSINE RECEPTOR | A mixed family name; check what the adenosine receptors share with ORs. |
| PTHR26452, PTHR26453, PTHR26454 | OLFACTORY RECEPTOR | Deferred (see Decisions). |
| PTHR11394 | TASTE RECEPTOR TYPE 2 | Bitter-taste receptors; extra-oral roles. |
| PTHR24060 | METABOTROPIC GLUTAMATE RECEPTOR | Class C; glutamate vs GABA-B vs CaSR divergence. |
| PTHR24061 | CALCIUM-SENSING RECEPTOR-RELATED | CASR, GPRC6A, taste type 1. |
| PTHR12011 | ADHESION G-PROTEIN COUPLED RECEPTOR | Many members have no demonstrated G protein coupling; tethered-agonist mechanism. |
| PTHR11309 | FRIZZLED | Wnt receptors: is GO:0004930 justified, or should coupling be left to SMO-like evidence? |
| PTHR11334 | MAS-RELATED G-PROTEIN COUPLED RECEPTOR | Primate-expanded MRGPRs; itch/mast-cell ligands. |
| PTHR11792 | ARRESTIN | Visual vs beta-arrestins; scaffolding functions. |
| PTHR24355 | G PROTEIN-COUPLED RECEPTOR KINASE/RIBOSOMAL PROTEIN S6 KINASE | A family that mixes GRKs with S6 kinases; kinase substrate terms must not cross the split. |
Several PANTHER families are named only "G-PROTEIN COUPLED RECEPTOR"
(e.g. PTHR22750, PTHR24230, PTHR24232); their contents are to be identified
from the member index, not guessed from the name.
Deliverable per family: a family review (coherence call, term-scope calls for
GO:0004930, the coupling BP terms, and the ligand-specific MF terms), and
IBA findings feeding IBA_REVIEW.
Workstream 2 — Modules
Two GPCR modules exist as drafts: gpcr_camp_pka_signaling (Gs) and
gpcr_plc_calcium_pkc_signaling (Gq). Related existing modules:
chemokine_signaling, wnt_signaling, hedgehog_signaling.
Planned work, using the module-curation skill:
- Promote the two drafts: ground GNAS, ADCY5, PRKACA, CREB1 and the Gq
parts in PAINT nodes where the evidence is clear; otherwise leave PTN
claims out. - Gi/o inhibitory module: Gi/o alpha inhibition of adenylyl cyclase plus
G beta-gamma activation of GIRK channels and inhibition of voltage-gated
calcium channels (the opioid, D2, alpha2, M2 and GABA-B arm). - G12/13-RhoA module: Galpha12/13, RhoGEFs (ARHGEF1/11/12), RhoA-ROCK.
- Receptor desensitization and arrestin module: GRK phosphorylation,
arrestin recruitment, clathrin-dependent internalization, recycling vs
degradation. This is whereGO:1904022positive regulation of G
protein-coupled receptor internalization and the arrestin-scaffold terms
belong. - RGS GAP module: RGS-accelerated GTP hydrolysis terminating signaling
(RGS4 and the R7/R4 subfamilies). - Rod phototransduction as a well-characterized exemplar (rhodopsin,
transducin, PDE6, GRK1, visual arrestin), if no existing module already
covers it.
Each module should list the receptor gene reviews it depends on; a module
part with no reviewed gene is a curation gap recorded in the module notes.
Workstream 3 — Gene functions
Per-gene reviews with the review skill. Order by how much each review
unblocks: a receptor that anchors a dismech drug-mechanism module, a module
part, or a family review comes first.
Tranche 1 — drug-target receptors (each anchors a dismech module):
ADRB1, ADRB2 (refresh), ADRA1A, ADRA2A, DRD2, HTR2A, HTR1B, HTR1D, OPRM1,
HRH1, HRH2, AGTR1, GLP1R, GIPR, CASR (refresh), CHRM3, S1PR1, P2RY12, CNR1, MC4R,
CALCRL, TSHR.
Tranche 2 — transducers and regulators (module parts): GNAS (refresh), GNAI1,
GNAI2, GNAO1, GNAQ, GNA11, GNA12, GNA13, GNB1, GNG2, ARRB1, ARRB2, GRK2,
GRK3, GRK5, GRK6, RGS2, RGS4, ADCY5, PLCB1.
Tranche 3 — disease-gene receptors without a drug (feed dismech disease
modules): RHO, OPN1LW, MC2R, PTH1R, AVPR2, GNRHR, KISS1R, LHCGR, FSHR,
GPR143, ADGRG1, ADGRV1 (refresh), FZD4.
Recurring function questions to resolve while reviewing:
GO:0004930alone is uninformative when a ligand-specific child is
supported; useMODIFYtoward the specific term (e.g.GO:0004941).- Coupling BP terms: annotate the Galpha class with direct evidence, not the
one inferred from family membership. - Orphan receptors: keep
GO:0004930only with evidence of G protein
coupling; constitutive-activity evidence (GPR101, GPR161) counts. - Arrestin-biased and G protein-independent signaling: decide when an
arrestin-dependent BP is warranted (cf.GO:0160213
beta-arrestin-dependent dopamine receptor signaling pathway). - Frizzled and adhesion GPCRs: whether the GPCR MF terms reflect the
receptor's actual mechanism.
First pairing: mu-opioid receptor
Chosen as the first receptor (2026-10-15) because dismech already has many
opioid-treated entries waiting for a shared mechanism module. Gene side:
- OPRM1 gene review (then OPRD1, OPRK1, OPRL1, so one family call covers
the classical opioid receptors; their PANTHER family is to be identified
from the member index, since no PANTHER family name mentions them) - Gi/o module: adenylyl cyclase inhibition, G beta-gamma activation of GIRK
channels and inhibition of voltage-gated calcium channels - Desensitization/arrestin module: GRK phosphorylation, beta-arrestin
recruitment, internalization (the tolerance-relevant arm)
The dismech side builds the mu-opioid agonism and antagonism modules on top.
Plan
-
[x] OPRM1 gene review
-
[ ] Pull GOA term counts for all human GPCRs to size the over-annotation problem
- [ ] Family reviews for PTHR24248, PTHR24247, PTHR24249 (aminergic)
- [ ] Family reviews for PTHR12011 (adhesion) and PTHR11309 (Frizzled)
- [ ] Taste receptor family call (PTHR11394)
- [ ] Promote
gpcr_camp_pka_signalingandgpcr_plc_calcium_pkc_signaling - [ ] New modules: Gi/o, G12/13-RhoA, desensitization/arrestin, RGS
- [ ] Tranche 1 gene reviews (agree the list with the dismech side)
- [ ] Tranche 2 gene reviews
- [ ] Tranche 3 gene reviews
- [ ] Cross-link each reviewed receptor to the dismech module that uses it
Decisions
- Olfactory receptors are deferred (decided 2026-10-15): no family or
per-gene review of the ~400 human ORs for now.
Open questions
- Should a GPCR family index page (like the CASPL one) be generated, or do
family reviews live only underinterpro/panther/? - Do we use GPCRdb / IUPHAR-Guide to Pharmacology classifications as a
reference alongside PANTHER, and if so how are they cited?