"Single", "dual" and "triple agonist" are counts, not rankings. They say how many of three related receptors a molecule activates — GLP-1, GIP and glucagon — and nothing about how strongly, how selectively, or to what end. The three receptors belong to one family, respond to three different hormones, and sit on overlapping but distinct tissues. Understanding what separates them is what makes the labels legible.
Three receptors, one family
All three are class B1 G-protein-coupled receptors: a large extracellular domain that captures the C-terminal end of a long peptide hormone, then a seven-transmembrane bundle that the hormone's N-terminus engages to trigger activation. Canonically all three couple to Gαs, raising intracellular cAMP. That shared architecture is why one peptide backbone can be engineered to reach more than one of them, and why selectivity is difficult rather than automatic.
| Receptor | Native ligand | Principal tissues | Canonical signal |
|---|---|---|---|
| GLP-1R | Glucagon-like peptide-1 (from proglucagon) | Pancreatic β-cells, brainstem, hypothalamus, vagal afferents | Gαs → cAMP; glucose-dependent insulin secretion |
| GIPR | Glucose-dependent insulinotropic polypeptide | Pancreatic islets, adipose tissue, bone, CNS | Gαs → cAMP |
| GCGR | Glucagon (from the same precursor as GLP-1) | Hepatocytes, adipose tissue, kidney | Gαs → cAMP; hepatic glucose output, lipolysis |
GLP-1 and glucagon are both cut from proglucagon, which is why their sequences are similar enough that a single engineered peptide can address both, and why the engineering problem is one of tuning rather than fusion.
What does "dual agonist" mean on a label?
That the molecule has measurable agonist activity at two of the three receptors. It does not say the two activities are balanced — most multi-agonists are deliberately unbalanced, with substantially different potency at each target, and the ratio is a design parameter rather than an accident. A compound described as a dual agonist with a hundred-fold difference between its two targets behaves very differently from one with a three-fold difference. The label is a count; the ratio is the pharmacology.
Why these targets, and why they were hard
The incretin effect is the observation that glucose taken by mouth provokes a larger insulin response than the same amount of glucose delivered intravenously. The difference is hormonal: GLP-1 and GIP are released from the gut in response to nutrients and potentiate insulin secretion. Both are also cleared within minutes by dipeptidyl peptidase-4, which cleaves after the second residue, and that short half-life is the reason native incretins were never practical research tools. Almost every design decision in this class follows from it.
Why is glucagon receptor agonism included at all?
It reads as contradictory, since glucagon raises hepatic glucose output while GLP-1 promotes insulin secretion. The rationale is that glucagon receptor activity also increases energy expenditure and hepatic lipid handling, and that concurrent GLP-1 activity constrains the glycaemic consequence that unopposed glucagon agonism would produce. Whether the combination behaves as intended is an empirical question that depends on the ratio between the activities, not something that can be reasoned from the receptor list. It is a genuinely different design premise from adding a second incretin.
The engineering that makes them usable
Three modifications recur across the class, and they are largely independent of which receptors a molecule targets.
DPP-4 resistance. Substituting the second residue — commonly alanine to 2-aminoisobutyric acid — removes the cleavage site. This is the single change that converts a peptide with a few minutes of stability into one with hours.
Acylation. A fatty acid chain is attached through a linker, usually to a lysine side chain. The chain binds reversibly to serum albumin, and the bound fraction acts as a circulating reservoir protected from renal filtration. Half-life extends from hours to days. The structural consequence is that these molecules are amphipathic, which matters for how they behave in a tube.
Backbone substitutions for stability and selectivity. Residue changes that resist oxidation, discourage aggregation, or shift the balance of activity between receptors.
Why are these peptides acylated?
To exploit albumin. An unmodified peptide of this size is filtered by the kidney quickly; bound to albumin it is not. The binding is non-covalent and in equilibrium, so a small free fraction remains available while the bulk is held in reserve. The practical consequence for handling is that an acylated peptide is a surfactant-like molecule — it concentrates at air–liquid interfaces, foams readily if shaken, and binds to hydrophobic plastics, which makes adsorption losses a live concern at working concentrations rather than a theoretical one.
The catalogue, by receptor
| Compound | GLP-1R | GIPR | GCGR | Other |
|---|---|---|---|---|
| Semaglutide | Yes | — | — | — |
| Tirzepatide | Yes | Yes | — | — |
| Retatrutide | Yes | Yes | Yes | — |
| Retatrutide + cagrilintide | Yes | Yes | Yes | Amylin/calcitonin receptors |
The three single compounds are compared molecule by molecule — backbone, acyl chain, mass and what each isolates — in semaglutide vs tirzepatide vs retatrutide.
Is cagrilintide an incretin?
No, and that is the reason it appears in combination rather than as a variant. Cagrilintide is an amylin analogue acting at the calcitonin receptor and the amylin receptor complexes — a separate receptor family with a separate native hormone. Pairing it with an incretin-family compound is combining two mechanisms, not intensifying one. Analytically that also makes the combination a blend rather than a single compound, with the additional certificate requirements that implies.
Why can potency figures not be compared across papers?
Because an EC50 is a property of an assay as much as of a molecule. The number depends on the cell line, the receptor expression level, whether the readout is cAMP accumulation or β-arrestin recruitment, the incubation time, and whether albumin is present in the buffer — and for an acylated peptide that last one alone can move the figure by orders of magnitude, since albumin sequesters the compound. Two laboratories can report honest EC50 values for the same molecule that differ substantially. Comparisons are meaningful within a single paper, where every compound saw the same assay, and unreliable between papers.
Does adding a receptor make a compound stronger?
It makes it broader, which is not the same property. A triple agonist engages more pathways than a dual agonist; whether that produces a larger effect depends entirely on the balance between the activities and on the system being measured. In an assay reading only GLP-1 receptor cAMP, a well-designed triple agonist may look weaker than a selective GLP-1 agonist, because its activity is distributed. Reading the receptor count as a potency ranking is the most common error in this area.
United Peptides supplies these compounds for laboratory and research use only. Nothing here describes clinical use, efficacy or safety in humans or animals. These products are not drugs, foods, supplements or cosmetics, and are not for human or veterinary use.
