Tirzepatide's Dual GIP/GLP-1 Receptor Mechanism

Most published summaries of tirzepatide describe it in one line: a “dual GIP/GLP-1 receptor agonist.” That phrase is accurate but it flattens the part of the pharmacology that the primary literature actually spends its time on. Two receptors are engaged, but they are not engaged equally, and the signal each receptor sends downstream is not a simple copy of what its native hormone produces. This post walks through what the mechanistic literature — receptor binding assays, signalling panels, cryo-electron microscopy structures, and isolated islet work — has reported about how the molecule interacts with each receptor.

What are the GIP and GLP-1 receptors, and how do they differ?

Both receptors are class B G protein-coupled receptors, and both are part of the incretin system. The incretin effect describes the observation that oral glucose delivery produces a larger insulin response than an equivalent intravenous glucose load, an effect attributed primarily to two gut-derived peptide hormones: glucose-dependent insulinotropic polypeptide (GIP) and glucagon-like peptide-1 (GLP-1) (source 5).

The two receptors share roughly 42 percent amino-acid sequence homology and both are expressed across multiple tissues, including pancreatic islets (source 5). Despite that structural family resemblance, the reviewed literature reports distinct distribution patterns and distinct downstream consequences of activation, which is the reason a single molecule engaging both is regarded as pharmacologically interesting rather than redundant.

Incretin effect

The larger insulin response observed after oral versus intravenous glucose administration, attributed to GIP and GLP-1 release from the gut.

Imbalanced agonism

A dual agonist engaging its two target receptors to measurably different degrees, rather than with matched potency at both.

Biased agonism

A ligand that preferentially activates one downstream pathway at a receptor (for example, G protein–mediated cAMP generation) over another (for example, β-arrestin recruitment).

Receptor internalisation

The process by which an activated receptor is drawn from the cell surface into the cell, typically reducing how much further signalling that receptor can produce.

How was tirzepatide engineered to engage both receptors?

The molecule was described in a 2018 Molecular Metabolism paper by Coskun and colleagues under its development code LY3298176 (source 2). Its 39-residue sequence is built on a GIP backbone, with residues drawn from GIP, GLP-1 and semaglutide alongside a small number of unique substitutions (source 4). The peptide is amidated at the C-terminus, and a C20 fatty diacid moiety is conjugated through a spacer to the lysine at position 20.

That fatty diacid is the pharmacokinetic half of the design rather than the receptor-engagement half. It supports reversible albumin binding, which published population pharmacokinetic work associates with slowed renal filtration and enzymatic degradation and a reported half-life of approximately five days (source 4). Population pharmacokinetics were described by a two-compartment model with first-order absorption and elimination (source 4). Separating those two design goals — receptor engagement from circulating persistence — is useful when reading the structural papers, because the acylation site sits away from the receptor-binding interface and several structural studies were performed using a non-acylated form of the peptide.

What does “imbalanced agonism” mean in the tirzepatide literature?

The term comes directly from the title of a 2020 JCI Insight paper, “Tirzepatide is an imbalanced and biased dual GIP and GLP-1 receptor agonist” (source 1). That study reported a greater degree of engagement at the GIP receptor than at the GLP-1 receptor, and framed this asymmetry as a defining feature of the molecule rather than an artefact of assay conditions.

The comparison is not that tirzepatide is weak at GLP-1R in absolute terms. One analysis reported tirzepatide binding GLP-1R with somewhat higher affinity than native GLP-1 itself (pIC50 8.90 versus 8.62) (source 4). The imbalance is relative: engagement at GIPR is stronger still. Researchers reading this literature should note that potency values vary with the assay system, the receptor species, and whether the acylated or non-acylated peptide was used, so cross-paper numerical comparisons need those conditions checked before they are read as equivalent.

What is biased signalling, and what did studies observe at the GLP-1 receptor?

Class B GPCRs can couple to more than one downstream effector. The two most commonly measured in this literature are G protein–mediated cAMP generation and β-arrestin recruitment, the latter of which is closely tied to receptor internalisation and desensitisation.

The 2020 signalling work reported a split behaviour across the two receptors (source 1). At GIPR, tirzepatide broadly mimicked the actions of native GIP. At GLP-1R, it did not mimic native GLP-1: the study observed bias toward cAMP generation over β-arrestin recruitment, together with a weaker ability to drive GLP-1R internalisation than GLP-1 produced. In other words, one molecule behaved like the native ligand at one receptor and like a pathway-selective ligand at the other.

What did cryo-EM structures reveal about how tirzepatide binds each receptor?

A 2022 Nature Communications study determined five cryo-electron microscopy structures to examine the structural basis of multi-receptor engagement, including tirzepatide-bound GIPR and tirzepatide-bound GLP-1R complexes, alongside the triagonist peptide 20 bound at GIPR, GLP-1R and the glucagon receptor (source 3). Corresponding coordinates were deposited in the Protein Data Bank, including entry 7FIM for the tirzepatide-bound human GLP-1R–Gs complex and 7VAB for the non-acylated tirzepatide-bound human GIPR–Gs complex.

Structural work of this kind is what moves a mechanistic claim from a functional observation to a physical account: rather than only reporting that cAMP output and β-arrestin output diverge, the structures show how a single sequence is accommodated at two related but non-identical binding pockets. For a laboratory verifying peptide identity, the deposited structures also serve a practical purpose — they fix the exact sequence and modification state the published pharmacology refers to, which is the reference point any identity or purity work should be measured against.

What did islet studies report about β-arrestin and insulin secretion?

The functional consequence of the GLP-1R bias was examined in primary islet preparations. In those experiments, researchers observed that β-arrestin1 limited the insulin response to GLP-1, but did not limit the response to GIP or to tirzepatide (source 1). The authors interpreted this as evidence that reduced β-arrestin recruitment is not incidental — it removes a brake that native GLP-1 signalling is subject to.

The 2020 paper's own summary is that the imbalance toward the GIP receptor, combined with the distinct signalling properties observed at the GLP-1 receptor, together may account for the potency reported for the molecule in earlier work (source 1). A broader 2023 review of incretin-based dual and tri-agonist mechanisms in pancreatic islets places this within the wider literature on how multi-receptor peptides behave in islet tissue (source 5). Both should be read as mechanistic hypotheses supported by the reported data, not as settled causal accounts.

How does the dual-receptor mechanism compare with single- and triple-agonist peptides?

Placing tirzepatide next to a selective GLP-1R agonist and a triple agonist makes the design space clearer. The table below summarises receptor engagement as described across the cited sources.

The contrast with peptide 20 is the useful one. A triagonist described as balanced across three receptors is a different design philosophy from a dual agonist described as imbalanced across two. Neither is presented in the literature as inherently superior; they are different points in the same design space, and the published structures were determined partly to understand why.

What remains unresolved in the mechanistic literature?

Several questions are still open in the sources above. The degree to which the GIPR-favouring imbalance is required, rather than merely present, has not been isolated experimentally in a way that separates it from the GLP-1R bias. Assay-dependent variation in reported potency values complicates cross-study synthesis. And the islet findings on β-arrestin1 come from ex vivo preparations, which constrain how far they generalise.

For laboratories sourcing tirzepatide as a research material, the practical implication of this literature is that identity and modification state matter to reproducibility. Acylated and non-acylated forms behave differently in structural work, and sequence-level substitutions are what distinguish this molecule from the GIP backbone it was built on. Every Steadfast Research Group batch ships with a batch-matched Certificate of Analysis documenting confirmed identity and purity by the stated analytical method, which is the baseline any mechanistic comparison against published data depends on.