The incretin–glucagon receptor axis

Scope and definitions

This page maps human receptor pharmacology, not treatment. GLP-1 and GIP are incretin hormones. Glucagon belongs to the same class B1 receptor family but is not an incretin. The map separates a molecule's exact mature form, its receptor relation, the main curated transduction route, and the source that supports each edge.

Target nomenclature and curated receptor relations are attributed to the IUPHAR/BPS Guide to Pharmacology (GtoPdb) web services, accessed 22 August 2026; mature peptide and receptor identities are cross-checked against UniProt and the cited primary studies.

Ligand identity and precursor genes

Human GLP-1 occurs here as two mature products: the 30-residue C-terminally amidated GLP-1(7-36)amide and the 31-residue GLP-1(7-37). Both are processed from the GCG precursor. Human glucagon(1-29) is also processed from GCG. There is no separate human GLP1 gene for these peptides; sharing a precursor does not mean that the products are processed or released identically in every tissue.

Human GIP(1-42) is processed from the separate GIP precursor. “Gastric inhibitory polypeptide” is the UniProt/GtoPdb name; “glucose-dependent insulinotropic polypeptide” is an established alternative name.

Shared signalling spine

GLP1R, GIPR and GCGR are human class B1 G-protein-coupled receptors. Their principal curated transduction route is Gs → adenylyl cyclase → increased intracellular cAMP. Structural work also resolves the three cognate ligand–receptor–Gs complexes.

That shared spine does not make the receptors interchangeable. It does not establish identical effects across pancreas, liver, brain, adipose tissue or the gastrointestinal tract. It is also not an “only Gs” claim: structural evidence shows that human GCGR can recognise both Gs and Gi.

Canonical relations and cross-pharmacology

The canonical endogenous relations are GLP-1(7-36)amide and GLP-1(7-37) → GLP1R, GIP(1-42) → GIPR, and glucagon(1-29) → GCGR. GtoPdb also records glucagon as an agonist at human GLP1R. In one primary human-receptor assay, glucagon behaved as a full agonist but was about 200-fold less potent than GLP-1(7-36)amide. That is an assay-specific cross-pharmacology result, not a redefinition of glucagon's canonical physiological pathway.

Absence of another relation from a curated registry does not prove biological impossibility. Potency, efficacy, receptor reserve, bias, internalisation and β-arrestin outcomes must remain tied to the ligand, species and assay system in which they were measured.

Engineered agonist layer

The separate engineered layer does not infer a drug relation from the endogenous map. Every edge has an agent-specific source: semaglutide → GLP1R; tirzepatide → GIPR and GLP1R; retatrutide → GIPR, GLP1R and GCGR; and survodutide and pemvidutide → GLP1R and GCGR.

“Approved medicine” means that a current official product label supports both medicine status and the stated receptor pharmacology for at least one product in the cited jurisdiction. It is not a blanket approval claim for every formulation, indication or country. “Investigational compound” is a dated development-status classification supported by the cited trial registry; registration or publication is not approval.

The edges show target activity only. They do not rank potency, balance, bias or clinical performance between agents, and they do not assign a share of an observed clinical outcome to any one receptor.

What this map cannot establish

A receptor edge cannot by itself establish insulin secretion, appetite change, gastric emptying, lipolysis, energy expenditure, weight change, safety or clinical benefit. Those claims require their own tissue, animal, human or regulatory evidence. The engineered layer therefore links to the full monographs instead of allowing receptor activity to inherit their clinical claims.