Tirzepatide activates the GIP receptor. Maridebart cafraglutide (AMG 133) blocks it. Both drive double-digit weight loss. Here is a receptor-first look at how agonism and antagonism at the same receptor land in the same place — and the four mechanistic hypotheses researchers use to reconcile it.
Two of the most effective investigational obesity compounds do opposite things at the glucose-dependent insulinotropic polypeptide (GIP) receptor. Tirzepatide is a GIP receptor agonist — it switches the receptor on [2]. Maridebart cafraglutide (AMG 133 / MariTide) is a GIP receptor antagonist — a blocking antibody conjugated to GLP-1 agonist peptides, so it switches the same receptor off [4,5]. Yet both produce large, durable weight loss: tirzepatide −20.9% at 72 weeks in SURMOUNT-1 [3]; maridebart cafraglutide −12.3% to −16.2% at 52 weeks in its Phase 2 trial [6].
The leading reconciliation is functional antagonism: sustained GIP receptor agonism desensitizes and downregulates the receptor, so chronic agonism can end up mimicking antagonism at the tissue level [7]. Layered on top: both molecules are strong GLP-1 receptor agonists (the GLP-1 arm carries much of the loss), and human genetics support that perturbing GIP signalling in either direction can move body weight [1,8]. This page is a receptor-first explainer for researchers, framed around the 2026 Annual Review of Nutrition synthesis by Holst, Rosenkilde and colleagues [1]. For research-use context only — see the companion GLP-1, GIP & glucagon mechanism of action explainer.
GIP receptor · ON
Agonist — activates GIPR (and GLP-1R) as a single dual-agonist peptide [2]
Peptide, once weekly (Mounjaro / Zepbound, Eli Lilly)
−20.9% at 72 weeks, 15 mg (SURMOUNT-1) [3]
GIP receptor · OFF
Antagonist — a GIPR-blocking antibody conjugated to two GLP-1R agonist peptides [4,5]
Antibody–peptide conjugate, once monthly (AMG 133 / MariTide, Amgen)
−12.3% to −16.2% at 52 weeks (Phase 2) [6]
The GIP receptor is one of the hottest targets in metabolic medicine, and the field cannot agree on which way to push it. Tirzepatide — the most effective approved obesity drug — activates the GIP receptor [2,3]. Maridebart cafraglutide — one of the most watched investigational obesity drugs — blocks it [4,5,6]. Both work. A 2026 Annual Review of Nutrition synthesis from Jens Juul Holst, Mette Rosenkilde and colleagues put the contradiction directly in its title: “The Paradox and Future of GLP-1/GIP Combination Therapies” [1].
This is not a rounding error between two similar molecules. Agonism and antagonism are pharmacological opposites, and yet the clinical readouts converge on double-digit weight loss. Resolving that is not just trivia — it tells researchers what the GIP receptor actually contributes to energy balance, and whether “switch it on” and “switch it off” are as opposite as they look.
Tirzepatide (development code LY3298176) is a single 39‑amino‑acid peptide engineered to activate two receptors at once: the GIP receptor and the GLP-1 receptor. Its discovery paper describes it explicitly as a dual GIP and GLP-1 receptor agonist, biased somewhat toward GIP relative to native peptide [2]. Functionally, it turns the GIP receptor on and keeps it on across the once-weekly dosing interval.
The efficacy is not in question. In SURMOUNT-1, tirzepatide 15 mg produced a mean body-weight reduction of −20.9% at 72 weeks, versus −3.1% on placebo, with the 5 mg and 10 mg doses at −15.0% and −19.5% respectively [3]. That result reset expectations for what incretin pharmacology could achieve and made GIP agonism look like a clear part of the winning formula — which is exactly what makes the antagonist result so jarring.
Maridebart cafraglutide (AMG 133, brand-development name MariTide) is a fundamentally different kind of molecule. It is a monoclonal antibody that antagonizes the GIP receptor, chemically conjugated to two GLP-1 receptor agonist peptides [4]. One molecule therefore does two opposite-direction things: it blocks GIP signalling while activating GLP-1 signalling. The medicinal-chemistry discovery and engineering of that antibody–drug conjugate were reported in the Journal of Medicinal Chemistry in 2026 [5]; the preclinical and Phase 1 pharmacology were published in Nature Metabolism in 2024, confirming GIPR-antagonist plus GLP-1R-agonist activity in cell systems and dose-dependent weight loss in animals and humans, with effect sustained months after the last dose [4].
The Phase 2 obesity trial, published in the New England Journal of Medicine in 2025, reported mean weight change of −12.3% to −16.2% at 52 weeks across dose arms, versus −2.5% on placebo — and the curves had not clearly plateaued by the end of the study [6]. The antibody backbone also enables a once-monthly injection, a differentiator from the weekly incretin peptides. So a GIP receptor blocker delivers weight loss in the same double-digit range as a GIP receptor activator.
No single mechanism is settled, but the literature converges on four non-exclusive explanations. The 2026 Annual Review of Nutrition paper treats them as complementary rather than competing [1].
This is the central bridge. A 2020 Nature Communications study showed that chronic, high-level GIP receptor agonism desensitizes and downregulates the receptor in adipocytes, producing a state that mimics functional GIP receptor antagonism [7]. In other words, if you agonize the GIP receptor hard enough for long enough, the receptor stops responding — so the tissue-level endpoint of chronic agonism can converge on the tissue-level endpoint of antagonism. On this view, tirzepatide and maridebart cafraglutide may be arriving at a similar downstream GIP state by opposite routes. Notably, the group that published this desensitization work is the same group that later developed the AMG 133 antagonist.
Both molecules are potent GLP-1 receptor agonists. GLP-1 receptor agonism alone drives large weight loss through central appetite suppression and delayed gastric emptying, as documented across the semaglutide programme and detailed in our receptor mechanism-of-action explainer. If GLP-1 agonism is the dominant weight-loss engine in both compounds, then the GIP component is a modulator — tuning tolerability, insulin sensitivity, and adipose handling — rather than the primary driver. That reframes the paradox: the two drugs agree on GLP-1 (the big lever) and differ on GIP (the tuning dial) [1,4,6].
The GIP receptor is expressed in the brain, pancreatic islets, adipose tissue, and bone, and its role is not identical across those compartments. Agonism and antagonism may act on different GIP-receptor populations — for example central versus peripheral pools — so “blocking” and “activating” the receptor are not clean whole-body opposites. The Annual Review of Nutrition synthesis emphasizes that the net metabolic effect depends on where the receptor is engaged, not only on the direction of engagement [1].
A 2026 European Heart Journal Mendelian-randomization study used naturally occurring genetic variants in GIP signalling as a lifelong “natural experiment,” and linked perturbed GIP signalling to body-weight reduction and altered cardiovascular risk [8]. Genetic evidence like this is agnostic to a specific drug: it supports the broader idea that disturbing the GIP axis — in either direction — can influence body weight, which is exactly what the agonist-versus-antagonist convergence would predict.
| Tirzepatide | Maridebart cafraglutide (AMG 133) | |
|---|---|---|
| GIP receptor action | Agonist (on) | Antagonist (off) |
| GLP-1 receptor action | Agonist | Agonist |
| Molecular format | Single dual-agonist peptide | Anti-GIPR antibody + 2 GLP-1 agonist peptides |
| Dosing interval | Once weekly | Once monthly |
| Headline weight loss | −20.9% (72 wk, 15 mg) [3] | −12.3% to −16.2% (52 wk) [6] |
| Stage | Approved (Mounjaro / Zepbound) | Investigational (Phase 3 initiated) |
| Developer | Eli Lilly | Amgen |
Trial figures come from separate studies with different designs and populations and are not a head-to-head comparison. Weight-loss ranges for maridebart cafraglutide span the reported dose arms of its Phase 2 trial [6].
The practical lesson for anyone comparing incretin-class compounds is that receptor direction on a datasheet does not fully predict the physiological outcome. A compound labelled “GIP agonist” and a compound labelled “GIP antagonist” can land in overlapping weight-loss territory because receptor desensitization, GLP-1 co-agonism, tissue distribution, and dosing kinetics all sit between the receptor label and the scale reading. For triple agonists such as retatrutide, which add glucagon agonism on top of GIP and GLP-1, the same caution applies: the receptor stack is a starting point, not a verdict.
It also means the GIP arm is still being actively decoded. Two large, well-funded programmes are running the natural experiment in opposite directions, and Phase 3 read-outs on the antagonist will be one of the most informative events in the field. Until then, the honest position is the one the Annual Review of Nutrition authors take: the paradox is real, several mechanisms plausibly contribute, and no one has cleanly separated them yet [1].
This article documents receptor pharmacology and trial-measured effects of GIP receptor agonism and antagonism. Remy Peptides supplies research peptides for in-vitro laboratory use only. Nothing on this page is medical advice, dosing guidance, a treatment recommendation, or guidance on personal supplementation or monitoring. Tirzepatide and maridebart cafraglutide are named here for scientific accuracy; they are not products sold by Remy Peptides. Researchers should consult primary literature and, where human health is involved, a qualified clinician.
Tirzepatide (LY3298176) is a GIP receptor agonist. It is a single peptide that activates both the GIP receptor and the GLP-1 receptor, which is why it is described as a dual GIP/GLP-1 receptor agonist. It switches the GIP receptor on. In SURMOUNT-1 the 15 mg dose produced a mean 20.9 percent body-weight reduction at 72 weeks.
Maridebart cafraglutide (AMG 133, also called MariTide) is a GIP receptor antagonist. It is a monoclonal antibody that blocks the GIP receptor, chemically conjugated to two GLP-1 receptor agonist peptides. So the same molecule blocks the GIP receptor while activating the GLP-1 receptor. In its phase 2 obesity trial it produced mean weight loss of 12.3 to 16.2 percent at 52 weeks.
This is the GIP paradox. The leading explanation is functional antagonism: chronic, sustained agonism of the GIP receptor desensitizes and downregulates the receptor, so long-term agonism can end up mimicking antagonism at the tissue level. Both molecules are also strong GLP-1 receptor agonists, and the GLP-1 arm carries much of the weight-loss effect while the GIP manipulation tunes tolerability and metabolic handling. Receptor direction on paper does not fully predict the net physiological outcome.
Functional GIP receptor antagonism is the observation that continuous, high-level agonism of the GIP receptor drives receptor desensitization and internalization, so the sustained-agonism state resembles the blocked-receptor state. A 2020 study showed that chronic GIPR agonism desensitizes adipocyte GIPR activity in a way that mimics functional GIPR antagonism, which is the central bridge used to reconcile why both a GIP agonist and a GIP antagonist can lower body weight.
In SURMOUNT-1, tirzepatide 15 mg produced a mean 20.9 percent body-weight reduction at 72 weeks versus 3.1 percent on placebo. In the phase 2 maridebart cafraglutide obesity trial, mean weight change at 52 weeks ranged from 12.3 to 16.2 percent versus 2.5 percent on placebo, without a plateau at the end of the study. The trials used different designs and populations, so these figures describe each compound rather than a head-to-head comparison.
A 2026 European Heart Journal Mendelian-randomization study used genetic variants in GIP signalling as a natural proxy for perturbing the pathway and linked them to body-weight reduction and cardiovascular risk. Genetic evidence like this supports the idea that modulating GIP signalling in either direction can influence body weight, which is consistent with the pharmacology seen for both the agonist and the antagonist.