Retatrutide vs Tirzepatide: Receptor Targets Compared
Tirzepatide and retatrutide look almost identical on paper. Both are 39-residue synthetic peptides built on a modified glucose-dependent insulinotropic polypeptide backbone, both carry a fatty diacid chain that binds albumin and stretches their circulating half-life to roughly a week, and both came out of the same laboratories at Eli Lilly within a few years of each other. Their molecular weights differ by about eighty daltons. Side by side on a chromatogram they are close cousins.
The difference between them is not chemical bulk but receptor arithmetic. Tirzepatide engages two receptors; retatrutide engages three. That single addition — the glucagon receptor — changes the pharmacology enough that the two compounds are studied as distinct mechanistic classes rather than as one molecule and its successor. This article sets out what the published literature reports about each, where the receptor profiles diverge, and what that divergence does and does not explain. Everything described below refers to work in cell systems, rodent models or registered clinical trials.
Chemistry and structure
Tirzepatide has the molecular formula C225H348N48O68, a molecular weight of approximately 4,813 Da, and is catalogued under CAS 2023788-19-2. Its sequence derives from native GIP rather than from GLP-1, which is the first non-obvious design choice: the molecule is a GIP analogue that has been engineered to also recognise the GLP-1 receptor, not the other way round. Two positions carry α-aminoisobutyric acid, a non-proteinogenic residue that resists cleavage by dipeptidyl peptidase-4, the enzyme that inactivates native incretins within minutes. A C20 fatty diacid is conjugated through a linker to a lysine side chain in the mid-sequence, and it is this lipid that drives reversible albumin binding and slows renal clearance.
Retatrutide, development code LY3437943, has the formula C221H342N46O68, a molecular weight of approximately 4,731 Da, and CAS number 2381089-83-2. It is likewise a 39-residue GIP-based scaffold with an acylation site and protective non-standard residues. The reported terminal half-life is around six days, comparable to tirzepatide's five, and in both cases the pharmacokinetics have been described as dose-proportional.
For contrast, semaglutide (CAS 910463-68-2) sits in an earlier generation: a 31-residue GLP-1 analogue with a single α-aminoisobutyric acid substitution at position 8 and a C18 diacid on the lysine at position 26. Lau and colleagues, writing in the Journal of Medicinal Chemistry (2015), described the albumin-binding strategy that underlies its protraction. It engages one receptor. The progression from one target to two to three is the organising story of this compound class.
What each receptor contributes
The three receptors are all class B G-protein-coupled receptors, all signal predominantly through Gs and cyclic AMP, and all sit in different tissues doing different work.
The GLP-1 receptor is expressed on pancreatic beta cells and at several sites in the central nervous system, including the area postrema and hypothalamic nuclei involved in appetite regulation. Its activation potentiates glucose-dependent insulin secretion and, through the central sites, reduces food intake in animal models. This is the best-characterised of the three arms and the one that single-agonist compounds rely on entirely.
The GIP receptor is the more contested. GIP is the other major incretin, and receptor agonism enhances insulin secretion in a glucose-dependent manner much as GLP-1 does. But GIP receptors are also expressed in adipose tissue and in the central nervous system, and the consequences of activating those populations have been debated for two decades. The rationale for including the GIP arm in a dual agonist is that it adds insulinotropic capacity and appears, in preclinical work, to improve tolerability of the GLP-1 component.
The glucagon receptor is the one that makes retatrutide structurally different in kind. Glucagon is conventionally understood as the counter-regulatory hormone that raises blood glucose, which makes it an unlikely partner for two insulin-promoting signals. The argument for including it rests on the liver. Work published in Diabetes (2018) reported that hepatic glucagon receptor signalling increases energy expenditure and reduces liver fat content through induction of lipid oxidation, with farnesoid X receptor-mediated hepatic cycling and secretion of the hepatokine FGF21 implicated in the pathway. Notably, the effect on energy expenditure was reported in diet-induced obese mice but not in lean animals, and was traced to hepatic rather than adipose receptor populations. The design logic, then, is that GLP-1 and GIP agonism reduce energy intake while glucagon agonism raises energy output, and that the glycaemic penalty of the glucagon arm is offset by the two incretin arms.
Tirzepatide: imbalanced and biased
The most informative pharmacology paper on tirzepatide is Willard and colleagues in JCI Insight (2020), which characterised the molecule as both imbalanced and biased — two distinct properties that are easy to conflate.
Imbalanced refers to relative affinity. Tirzepatide binds the GIP receptor with affinity equivalent to that of native GIP, but binds the GLP-1 receptor roughly fivefold more weakly than native GLP-1. It is therefore not an even dual agonist; it leans toward GIP.
Biased refers to which downstream pathway a receptor engages once bound. At the GLP-1 receptor, tirzepatide was reported to favour cyclic AMP generation over β-arrestin recruitment, with a correspondingly weaker capacity to drive receptor internalisation than native GLP-1 produces. Because β-arrestin recruitment normally desensitises and internalises the receptor, an agonist that triggers less of it leaves more receptor at the cell surface. The same paper reported that in primary islets β-arrestin1 limits the insulin response to GLP-1 but not to GIP or to tirzepatide — an observation consistent with the bias being functionally meaningful rather than a signalling curiosity. At the GIP receptor, by contrast, the compound was reported to behave much like the native ligand.
Structural work has since supported this picture. Sun and colleagues, in PNAS (2022), reported cryo-electron microscopy structures of tirzepatide bound to both receptors and identified the residues responsible for its dual recognition, showing how a single sequence accommodates two binding pockets that native ligands do not cross between.
Retatrutide: adding the third arm
Coskun and colleagues described retatrutide's receptor pharmacology in Cell Metabolism (2022). Reported half-maximal effective concentrations at the human receptors were approximately 0.064 nM at GIPR, 0.775 nM at GLP-1R and 5.79 nM at GCGR. Expressed relative to each native ligand, the molecule was characterised as roughly 8.9 times as potent as GIP at the GIP receptor, and about 0.4 and 0.3 times as potent as GLP-1 and glucagon at their respective receptors.
Those ratios are worth reading carefully, because the compound is frequently described as a "balanced" triple agonist and the numbers do not straightforwardly support that word. Absolute potency is highest at GIPR by an order of magnitude. What is balanced is the profile relative to endogenous signalling: the molecule is deliberately sub-native at the glucagon and GLP-1 receptors and supra-native at GIPR, which produces a composite signal that no single hormone generates. Whether that specific ratio is optimal, or merely the one that emerged from the medicinal chemistry, is not established.
What the clinical literature reports
Both molecules have been studied in registered trials as investigational pharmaceutical products. Jastreboff and colleagues reported a phase 2 trial of retatrutide in The New England Journal of Medicine (2023), in which 48 weeks of once-weekly administration produced least-squares mean body weight changes of −8.7 percent at 1 mg, −17.1 percent at 4 mg, −22.8 percent at 8 mg and −24.2 percent at 12 mg, against −2.1 percent with placebo. A separate phase 2a trial published in Nature Medicine (2024) examined the compound in metabolic dysfunction-associated steatotic liver disease and reported reductions in hepatic fat content, an endpoint of particular interest given the hepatic localisation of the glucagon arm. Reported adverse events in both trials were predominantly gastrointestinal and concentrated during dose escalation, a pattern common across the incretin class.
This body of work describes specific pharmaceutical products administered under clinical protocol to defined patient populations. It has no bearing on research-grade material, which is supplied for laboratory use only and is not a medicine.
Laboratory handling and analytical verification
Both compounds are supplied lyophilised. Their shared architecture creates a handling profile distinct from small peptides such as BPC-157 or GHK-Cu. The fatty acid chain makes both molecules amphiphilic, which promotes self-association and adsorption to container surfaces at low concentrations; the resulting losses are not always visible. Lyophilised material should be held frozen and protected from light and moisture, and reconstituted solutions aliquoted rather than repeatedly thawed. As with any acylated peptide, the lipid linkage is a potential site of hydrolysis under extremes of pH.
Analytically, the close structural relationship between the two is the point that deserves attention. They differ by roughly 82 Da on a mass of nearly 4,800, share a common backbone scaffold, and elute in a similar region on a reversed-phase gradient. Low-resolution mass spectrometry on a multiply charged envelope is not, on its own, a confident discriminator between them. Identity confirmation should rest on high-resolution accurate mass against the calculated monoisotopic value, ideally corroborated by retention time against a characterised reference standard. On the purity side, the characteristic impurities of a long synthetic sequence are deletion peptides missing one residue and incompletely acylated species — both of which can sit close to the main peak and require an adequately resolving gradient to separate. A purity figure is only as informative as the chromatogram it came from. Third-party analytical documentation for our catalogue is published on the Lab Results page.
What remains unresolved
The most interesting open question concerns the GIP arm, and it is a genuine one. Tirzepatide and retatrutide both activate the GIP receptor, and both are associated with substantial metabolic effects. Yet compounds taking the opposite approach — pairing GLP-1 receptor agonism with GIP receptor antagonism — have also been reported to reduce body weight in early-phase clinical work. Two mechanistically contradictory strategies producing directionally similar outcomes is not a resolved pharmacology. Proposed explanations include receptor desensitisation, in which sustained agonism functionally resembles blockade, and tissue-specific differences between central and adipose GIP receptor populations. None has been demonstrated decisively.
A second question is how much of retatrutide's observed effect is attributable to the glucagon arm as opposed to simply greater incretin engagement. The phase 2 trial compared doses of one molecule, not receptor combinations against one another, so it cannot separate the contributions. Answering that would require head-to-head work with matched dual and triple agonists, which has not been published.
For laboratories working with these compounds as reference materials, the comparison is instructive precisely because the two molecules are so nearly alike. Nearly the same mass, nearly the same backbone, nearly the same half-life — and a different receptor profile that the analytical data alone will not reveal. Readers working with the broader class may also find our guide to reconstituting lyophilised peptides relevant, since acylated peptides of this size have solubility behaviour that differs from shorter sequences.
Research use only. Not a medicine, supplement or therapeutic good. Not for human or veterinary consumption.