GIP and GIP receptor agonists: how they work

GIP raises insulin when blood glucose is high and glucagon when it is low. In mice, both activating and blocking its receptor lower weight. What is known, and what is still open.

Illustration after the structure PDB 7DTY (Zhao et al., eLife 2021): GIP bound to the GIP receptor. The G protein in the structure is not shown.

GIP was the first incretin hormone to be discovered and remains the least understood. It raises insulin release when blood glucose is high and raises glucagon release when it is low. It acts on fat tissue, though in people how it does so is poorly measured. And on appetite the findings point in opposite directions: in mice, both activating and blocking the GIP receptor lower body weight. That is not a contradiction in the data but an open question about how this receptor works, and it is precisely why GIP has ended up in combination drugs.

The body's own hormone

GIP stands for glucose-dependent insulinotropic polypeptide. Its older name, gastric inhibitory polypeptide, gives away what it was isolated for in the early 1970s: a substance from the gut that suppresses gastric acid. The insulin effect was shown in people shortly afterwards, when Dupre and Brown demonstrated that a GIP infusion during a glucose load amplified insulin release (Dupre 1973). Historically, GIP is therefore the first incretin; GLP-1 followed only in the 1980s.

The hormone comes from the K-cells in the upper small intestine, and fat is the strongest trigger. Where GLP-1 responds mainly to carbohydrate, GIP rises most after a fatty meal. Like GLP-1, it is cut by the enzyme DPP-4 within minutes.

What GIP does at the pancreas depends on the glucose level, and it runs in two directions. Christensen and colleagues in Copenhagen infused ten healthy young men with a physiological dose of GIP or saline under three conditions: an insulin-induced hypoglycaemia, normal glucose and hyperglycaemia. In hyperglycaemia, GIP more than doubled insulin release and did nothing to glucagon. In hypoglycaemia and at normal glucose the reverse happened: GIP raised glucagon and had little effect on insulin. In the first thirty minutes of hypoglycaemia, glucagon rose by 76 pmol/l per 30 minutes against 28 with saline (Christensen 2011). The investigators call GIP a bifunctional glucose stabiliser: it pushes blood glucose towards the middle, from both sides.

That glucagon effect is the major difference from GLP-1, which suppresses glucagon. It also explains an old puzzle in diabetology. Nauck compared nine patients with mild type 2 diabetes with nine healthy controls in 1993, under a hyperglycaemic clamp at 8.75 mmol/l. Both hormones raised insulin release, but the maximal effect of GIP was 54 percent lower in the patients than in the controls, while GLP-1 retained 71 percent of its action. GIP also failed to lower glucagon; GLP-1 did (Nauck 1993). In type 2 diabetes, then, GIP has largely stopped working as an incretin. For decades that was the reason to leave GIP alone as a drug target.

Fat tissue: much animal data, little human

Fat cells carry the GIP receptor, and that is the basis for the idea that GIP directs surplus energy after a fatty meal into storage. The best-known evidence comes from Kyoto. Miyawaki and colleagues bred mice without the GIP receptor and fed them a high-fat diet. Ordinary mice developed heavy fat deposition and insulin resistance; the mice without the receptor were clearly protected against both. They preferentially burned fat, with a lower respiratory quotient, and stayed lean. Crossed with genetically obese ob/ob mice, the receptor-deficient animals also gained less weight (Miyawaki 2002). The authors' conclusion was that GIP links overnutrition directly to obesity.

In people this is far harder to measure. The human infusion studies on what GIP does to blood flow and fat storage in subcutaneous fat tissue come from a small number of groups; one of those studies, on the recruitment of capillaries in fat tissue, was withdrawn by the journal and does not count in this overview. What remains is the conclusion of the review by Hammoud and Drucker (2023): the effects of GIP on fat tissue are consistent in animals and insufficiently supported in people to build a clinical statement on.

The mouse data do throw new light on insulin sensitivity. Samms and colleagues at Eli Lilly asked why tirzepatide, which activates both incretin receptors, improves insulin sensitivity in obese mice more than a GLP-1 agonist. In mice without a GLP-1 receptor, where the weight loss through GLP-1 dropped away, tirzepatide still improved insulin sensitivity, through higher glucose uptake in white fat tissue. A long-acting selective GIP agonist did the same (Samms 2021). GIP would then do two things to fat tissue: make it store, and make it more sensitive to insulin. Whether this holds in people has not been measured.

Appetite: the agonist-antagonist paradox

Here GIP becomes interesting and awkward at the same time. Two lines of animal research draw opposite conclusions, and both are reproducible.

The antagonist line builds on the Kyoto mice. Killion and colleagues at Amgen developed antibodies that block the GIP receptor. In obese mice on a high-fat diet, such an antibody prevented weight gain, with less food intake; in obese monkeys the weight loss was more pronounced than in mice. Combined with a GLP-1 receptor agonist, weight loss increased further in both species (Killion 2018).

The agonist line is equally strong. Mroz and colleagues from DiMarchi's laboratory made a series of GIP analogues with improved stability and receptor selectivity. In obese mice the agonists lowered weight consistently and dose-dependently, through less food intake and not through more energy expenditure; the effect persisted in mice without a GLP-1 receptor and vanished in mice without a GIP receptor. A long-acting antagonist from the same series did not change weight, not even in combination with liraglutide (Mroz 2019). The authors concluded that agonism, not antagonism, lowers weight, with a modest effect compared with GLP-1 agonists.

Two later studies try to reconcile the lines. Adriaenssens and colleagues in Cambridge made mice in which cells carrying the GIP receptor are visible and found them in three hypothalamic nuclei: the arcuate, the dorsomedial and the paraventricular nucleus, confirmed in human brain tissue. Artificially activating those cells suppressed food intake (Adriaenssens 2019). Zhang and colleagues in Munich went a step further. Mice without a GIP receptor in the central nervous system gained less weight on a high-fat diet, which supports the antagonist line. But an acylated GIP agonist lowered weight and food intake in ordinary mice, and that effect disappeared in the mice lacking the receptor in the brain. In those same mice, the combination of GLP-1 plus GIP lost its advantage over GLP-1 alone (Zhang 2021). Killion added a possible explanation in 2020: chronic activation of the GIP receptor on fat cells desensitises that receptor, and functionally behaves like an antagonist (Killion 2020).

As for people, the picture is more modest. Bergmann and colleagues in Copenhagen gave seventeen men with overweight or obesity, on four separate days, a glucose infusion accompanied by saline, GIP, GLP-1 or both. GLP-1 lowered energy intake at a free meal from 4483 to 2715 kJ. GIP alone did nothing (4062 kJ), and GIP alongside GLP-1 did not amplify the effect; intake with the combination was in fact higher than with GLP-1 alone (Bergmann 2019). An acute infusion of the natural hormone does not dampen appetite in people, and the question is whether a long-acting agonist does something different.

What a receptor agonist or antagonist does differently

The GIP receptor, like the GLP-1 receptor, is a G-protein-coupled receptor that raises cAMP. A selective GIP agonist as a stand-alone drug has barely been studied in people; in Finan's 2013 study a selective GIP agonist produced negligible weight loss in obese rodents, while the dual agonist reduced fat mass synergistically (Finan 2013). The human evidence for GIP receptor activation therefore comes almost entirely from compounds that also activate the GLP-1 receptor, such as tirzepatide and retatrutide, and is discussed in parts 3 and 4 of this series.

The antagonist line now has a human study of its own. Maridebart cafraglutide is an antibody against the GIP receptor with a GLP-1 agonist attached, given once every four weeks. In a phase 2 trial with 592 participants, weight in the obesity cohort fell by 12.3 to 16.2 percent after 52 weeks, against 2.5 percent on placebo; in participants with obesity and type 2 diabetes it fell by 8.4 to 12.3 percent against 1.7 percent. Gastrointestinal complaints were common, less so with slower escalation (Jastreboff 2025). This trial shows that GIP blockade alongside GLP-1 activation leads to substantial weight loss in people. What it does not show is how much of that is due to the GIP blockade, because an arm with the GLP-1 component alone was missing.

Adverse events and limitations

About adverse events of GIP receptor activation on its own, little can be said in people, because the compound has not been developed clinically on its own. In the combination drugs, nausea, vomiting and diarrhoea are the most reported events, and the share the GIP component has in them cannot be established. In the infusion studies with the hormone itself, the investigators reported no relevant complaints, but those are short experiments in small groups of healthy men.

The largest limitation is the step from mouse to human. The mouse GIP receptor responds differently to human GIP analogues than the human receptor does, and Mroz's studies show how sensitive the outcome is to exactly that difference. On top of that, the hypothalamus studies, however elegant, describe artificial activation of cells and not the action of a drug in the blood.

What remains open is the central question: is GIP in people a hormone that helps lower weight, or a hormone whose blockade does so? Both lines now have a clinical compound that works. If both compounds indeed add something through the GIP receptor, then Killion's desensitisation hypothesis is the explanation that has to assume the least. Proven in people it is not. Part 3 of this series looks at what the combination of GLP-1 and GIP receptor activation has produced in clinical trials, and which explanations for it are still uncertain.

Registered nurse with eleven years in psychiatry and founder of Peptalis. Writes the platform's knowledge layer: compound profiles, evidence reviews and the quality methodology. Works from primary literature (PubChem for chemistry, PubMed for studies) and states where evidence is absent.

Sources checked via PubMed · 14 references

• The body's own hormone
• Fat tissue: much animal data, little human
• Appetite: the agonist-antagonist paradox
• What a receptor agonist or antagonist does differently
• Adverse events and limitations

• Christensen et al. 2011 ↗
• Nauck et al. 1993 ↗
• Bergmann et al. 2019 ↗
• Miyawaki et al. 2002 ↗
• Jastreboff et al. 2025 ↗

• BASICSGLP-1: how a gut hormone steers appetite, the stomach and the pancreas
• DEEP DIVEThe GLP class: from one receptor to three

Compounds in this article

• Tirzepatide · Not in catalogue · Research profile

For laboratory research use only. Not for human use.

SCIENTIFIC REFERENCES

These references are provided for informational and research purposes only. They do not constitute medical advice.

1. 01Human studyStimulation of insulin secretion by gastric inhibitory polypeptide in man.Dupre J, et al. J Clin Endocrinol Metab. 1973.View on PubMed →DOI →
2. 02Human RCTGlucose-dependent insulinotropic polypeptide: a bifunctional glucose-dependent regulator of glucagon and insulin secretion in humans.Christensen M, et al. Diabetes. 2011.View on PubMed →DOI →
3. 03Human studyPreserved incretin activity of glucagon-like peptide 1 [7-36 amide] but not of synthetic human gastric inhibitory polypeptide in patients with type-2 diabetes mellitus.Nauck MA, et al. J Clin Invest. 1993.View on PubMed →DOI →
4. 04Preclinical · animalInhibition of gastric inhibitory polypeptide signaling prevents obesity.Miyawaki K, et al. Nat Med. 2002.View on PubMed →DOI →
5. 05ReviewBeyond the pancreas: contrasting cardiometabolic actions of GIP and GLP1.Hammoud R, Drucker DJ. Nat Rev Endocrinol. 2023.View on PubMed →DOI →
6. 06Preclinical · animalGIPR agonism mediates weight-independent insulin sensitization by tirzepatide in obese mice.Samms RJ, et al. J Clin Invest. 2021.View on PubMed →DOI →
7. 07Preclinical · animalAnti-obesity effects of GIPR antagonists alone and in combination with GLP-1R agonists in preclinical models.Killion EA, et al. Sci Transl Med. 2018.View on PubMed →DOI →
8. 08Preclinical · animalOptimized GIP analogs promote body weight lowering in mice through GIPR agonism not antagonism.Mroz PA, et al. Mol Metab. 2019.View on PubMed →DOI →
9. 09Preclinical · animalGlucose-Dependent Insulinotropic Polypeptide Receptor-Expressing Cells in the Hypothalamus Regulate Food Intake.Adriaenssens AE, et al. Cell Metab. 2019.View on PubMed →DOI →
10. 10Preclinical · animalThe glucose-dependent insulinotropic polypeptide (GIP) regulates body weight and food intake via CNS-GIPR signaling.Zhang Q, et al. Cell Metab. 2021.View on PubMed →DOI →
11. 11Preclinical · animalChronic glucose-dependent insulinotropic polypeptide receptor (GIPR) agonism desensitizes adipocyte GIPR activity mimicking functional GIPR antagonism.Killion EA, et al. Nat Commun. 2020.View on PubMed →DOI →
12. 12Human RCTEffects of combined GIP and GLP-1 infusion on energy intake, appetite and energy expenditure in overweight/obese individuals: a randomised, crossover study.Bergmann NC, et al. Diabetologia. 2019.View on PubMed →DOI →
13. 13Preclinical · animalUnimolecular dual incretins maximize metabolic benefits in rodents, monkeys, and humans.Finan B, et al. Sci Transl Med. 2013.View on PubMed →DOI →
14. 14Human RCTOnce-Monthly Maridebart Cafraglutide for the Treatment of Obesity - A Phase 2 Trial.Jastreboff AM, et al. N Engl J Med. 2025.View on PubMed →DOI →

Peptalis is a knowledge platform. All content on this page is provided for scientific and educational reference purposes only and does not constitute medical advice. Compounds discussed are research compounds and are not intended for human or veterinary use.