pe
pep-10689 v1 CC-BY-SA-4.0

GIP: natural gut hormone that boosts insulin after meals

A hormone released by the small intestine after eating that signals the pancreas to produce insulin; also a key target of the diabetes and weight-loss drug tirzepatide (Mounjaro/Zepbound). Used as a lab research tool.

statussynthesized targetGIPR length42 aa refs10
status 4 / 5
prediction metrics boltz-2 1.0
ipTM0.739
pTM0.671
avg pLDDT58.7
ranking score0.617
STRUCTURE · PEP-10689 × GIPR
ranking0.617
target interface 4.5Å peptide drag rotate · ctrl+scroll zoom · right-click pan
boltz-2 1.0 · mmCIF ↓ download
sequence42 aa
151015202530354042
YAEGTFISDYSIAM DKIHQQDFVNWLLA QKGKKNDWKHNITQ
in the news 33 articles
overview readme

What this is

GIP (human) is the 42-amino-acid form of gastric inhibitory polypeptide — a hormone released by the small intestine after a meal that tells the pancreas to release more insulin. It is one of two "incretin" hormones in humans, alongside GLP-1 (Seino 2010). The full-length human hormone is 42 residues (Pederson 2016) and is the parent molecule for shorter naturally occurring forms — including the 39-residue variant tracked as /card/pep-10691 and the amidated 30-residue truncation GIP(1-30)NH₂. GIP matters today because its receptor, GIPR, is one of the two targets hit by tirzepatide and by newer investigational dual-incretin drugs being developed for type 2 diabetes and obesity (Véniant 2024, Bailey 2024).

History

The hormone that became GIP grew out of decades of work on "enterogastrones" — gut factors thought to inhibit stomach acid secretion. Pederson (2016) gives a first-person account of that arc: animal studies aimed at identifying acid-inhibitory factors led to the isolation of a 42-amino-acid polypeptide, which inhibited acid secretion in animal models, but whose role in human gastric physiology turned out to be unclear. Marks (2020) reviews the period 1969–2000, during which GIP's defining role was reframed from acid inhibition to glucose-dependent stimulation of insulin release. The GIP receptor was first characterized as a class B (secretin-VIP) family GPCR widely distributed in peripheral organs and the brain (Usdin 1993), and its ligand binding and signaling were mapped in transfected systems (Wheeler 1995).

What it does

GIP is released from K-cells of the upper small intestine in response to nutrients and acts on pancreatic β-cells to amplify glucose-stimulated insulin secretion — the core "incretin effect" (Seino 2010, Bailey 2024). Its action is glucose-dependent: GIP raises insulin output only when blood glucose is elevated, so it does not, on its own, drive hypoglycemia. GIPR is not restricted to the pancreas; it is also expressed in adipose tissue, bone, and the central nervous system (Usdin 1993), and GIP signaling at those sites has been studied in lipid handling and bone turnover (Bailey 2024). Together GIP and GLP-1 — both signaling through related class B GPCRs that raise intracellular cAMP in β-cells — account for the incretin component of post-meal insulin release (Seino 2010).

Mechanism

GIPR is a class B G-protein-coupled receptor that couples primarily to Gαs, raising intracellular cAMP in β-cells and other GIPR-expressing tissues; this potentiates glucose-triggered insulin secretion (Seino 2010, Wheeler 1995). The pharmacology of GIP at its receptor is exquisitely sensitive to terminal truncation: Hansen and colleagues (2016) showed that N- and C-terminally shortened forms of the naturally occurring amidated truncation GIP(1-30)NH₂ are high-affinity competitive antagonists rather than agonists at the human GIP receptor — a few residues at the N-terminus determine whether the ligand activates or blocks the receptor. Species differences are also large: Sparre-Ulrich and colleagues (2016) demonstrated that (Pro3)GIP, long used as a "GIP receptor antagonist" in rodent work, is in fact a full agonist at the human GIP receptor while behaving as a partial agonist/competitive antagonist at rat and mouse receptors. This is a meaningful caveat when reading any rodent GIP-antagonism literature: a tool compound called an antagonist in mice may be an agonist in humans. The stored sequence here is the canonical 42-residue human form; the active circulating pool in vivo includes both full-length GIP and the naturally truncated GIP(1-30)NH₂ form, which is itself C-terminally amidated (Hansen 2016) — an amidation that is not represented in the raw 42-letter sequence.

Evidence

  • Human: GIP has been studied in human physiology for decades as one of the two principal incretin hormones (Seino 2010, Pederson 2016, Marks 2020). The GIP receptor — not GIP-the-peptide — is the clinical target: it is engaged by approved dual-incretin drugs (tirzepatide) and by investigational GIPR-antagonist conjugates such as AMG 133 (maridebart cafraglutide), which has progressed through phase 1 with weight-loss signals (Véniant 2024).
  • Animal: GIPR pharmacology has been characterized across rat, mouse, and human receptor systems, with species differences large enough to flip an agonist into an antagonist (Sparre-Ulrich 2016). GIPR distribution beyond the pancreas — including adipose, bone, and brain — was mapped in early studies (Usdin 1993).
  • In vitro: Ligand binding, cAMP coupling, and the consequences of N- and C-terminal truncation have been mapped in transfected cell systems (Wheeler 1995, Hansen 2016, Sparre-Ulrich 2016).

Known effects

  • Glucose-dependent insulin secretion — Established physiological role; foundation of the incretin concept (Seino 2010).
  • Adipose and bone signaling — GIPR is expressed outside the pancreas; GIP signaling has documented roles in lipid handling and bone turnover (Usdin 1993, Bailey 2024).
  • Drug-target validation for obesity and T2D — GIPR is one of the two receptors engaged by tirzepatide and the converse target (antagonism) of the bispecific molecule AMG 133 (Véniant 2024, Bailey 2024).

Regulatory status

GIP (human) as the native 42-residue peptide is not a marketed drug. Its receptor, GIPR, is the clinical target: tirzepatide engages GIPR as one of its two receptor targets (Véniant 2024, Bailey 2024), and AMG 133 (maridebart cafraglutide) is an investigational bispecific molecule that antagonizes GIPR while agonizing GLP-1R (Véniant 2024). Regulatory approval status applies to those engineered molecules, not to GIP itself.

Related peptides

  • GIP (1-39) (/card/pep-10691) — the 39-residue natural variant of GIP; same biology, shorter form.
  • Glucagon (/card/pep-04430) — a related class B GPCR ligand; useful contrast for incretin/anti-incretin signaling.
  • GLP-1 receptor agonists including semaglutide (/card/pep-00016), liraglutide (/card/pep-10868), and exenatide (/card/pep-04439) — engage only the GLP-1 arm of the incretin system, in contrast to GIP, which engages GIPR. Tirzepatide engages both.
Hypotheses6 directions▾ collapse

Research directions for this peptide, selected from the current sources — hypotheses you can explore and model. None of it is proven yet; tap any one to see the full thinking.

openupdated 2026-06-05

Could the very beginning of the GIP hormone decide whether the cell releases insulin or starts to resist the drug?

If true, scientists could redesign the start of GIP to favor the signals that lower blood sugar while avoiding those that make the drug stop working over time, helping patients stay on therapy longer.

The hypothesis
The N-terminal Tyr1-Ala2-Glu3 motif of GIP is not merely a binding anchor but a dynamic trigger that allosterically propagates through the receptor transmembrane core, and its precise side-chain chemistry dictates signaling pathway selection.
Why it’s plausible
Class B GPCR peptide ligands typically use the N-terminus as the 'signal' domain and the C-terminus as the 'affinity' domain. GIP(1-42) has a free N-terminal tyrosine, whereas the shorter amidated GIP(1-30)NH2 truncates the C-helix but keeps the same N-terminus. Differences in potency and bias between these natural forms may originate from how the N-terminal triplet engages the receptor juxtamembrane region, not just affinity.
Why it matters
Signaling bias (Gαs vs beta-arrestin) is increasingly linked to therapeutic efficacy versus tachyphylaxis for incretin drugs. Understanding whether the N-terminal chemistry itself drives bias would open rational engineering of biased GIP analogues.
Plausibility.70
Novelty.35
Impact.70
Basis · grounding2 computed/notes
[1]
sequenceN-terminal sequence YAEGTF...; Tyr1 is unmodified and highly conserved across mammalian GIP sequences.
[2]
noteNatural truncations GIP(1-39) and GIP(1-30)NH2 exist; their differential activity implies the N-terminus carries signal information beyond simple binding.
openupdated 2026-06-05

Could tying two specific parts of GIP together with a chemical bridge make it work better and wear out the receptor less?

If true, this design trick could lead to more effective diabetes and obesity drugs that keep working longer and cause fewer side effects for patients.

The hypothesis
N-terminal cyclization of GIP via a lactam bridge between Glu3 and Lys30 introduces a conformational constraint that selectively enhances Gαs coupling over beta-arrestin recruitment, producing a biased agonist with improved metabolic efficacy and reduced receptor desensitization.
Why it’s plausible
The GIP sequence contains Glu at position 3 and Lys at position 30 (YAEGTFISDYSIAMDKIHQQDFVNWLLAQKGKKNDWKHNITQ), spaced roughly one helical turn apart in a predicted alpha-helical C-terminal region. Lactam bridges between i and i+27 positions have been successfully used in other peptide hormones to stabilize bioactive conformations. If the N-terminal region needs flexibility for signaling while the C-helix provides the scaffold, a strategically placed lactam could lock the receptor-activating geometry without disrupting binding.
Why it matters
Biased GPCR agonism is a major goal in incretin drug development. A conformationally constrained GIP analogue with proven bias would be a direct competitor to existing dual agonists and could be developed with reduced side effects.
Plausibility.50
Novelty.55
Impact.60
Basis · grounding1 paper · 1 computed/note
[1]
sequenceGlu3 and Lys30 are present in the native sequence; in a predicted C-terminal helix these residues would be separated by approximately one helical turn, suitable for lactam cyclization.
[2]
paper
GIPR signals through Gαs and other intracellular messengers; engineering selective Gαs bias is a recognized therapeutic strategy for class B GPCRs.
doi: 10.1111/bph.13384
openupdated 2026-06-05

Could the GIP receptor have a hidden shape that current computer models do not see?

If true, drug designers could build better GIP-based medicines by targeting the receptor's true moving shape, potentially improving how well they work for people with diabetes or obesity.

The hypothesis
The modest pLDDT (58.7) of the GIP:GIPR complex prediction reflects genuine receptor flexibility rather than model error, and a functionally relevant GIPR conformational state remains un-captured.
Why it’s plausible
Structure predictions with ipTM ~0.74 but pLDDT below 60 often indicate a dynamic or partially disordered interface. GIPR is a class B GPCR whose extracellular domain undergoes large movements upon peptide binding. The current model may lock the receptor in a single state, missing a physiologically relevant conformation that alters affinity or signaling bias.
Why it matters
If true, optimizing GIP analogues against a single static receptor model could miss the conformation that drives therapeutic efficacy or side-effect profiles, explaining why some GIPR-directed drugs show unexpected in vivo behavior.
Plausibility.60
Novelty.40
Impact.55
Basis · grounding1 paper · 1 computed/note
[1]
structureboltz-2/complex ipTM=0.738, pLDDT=58.7 for GIP(1-42) bound to GIPR; high interface confidence paired with low per-residue confidence suggests a dynamic interface rather than a false-positive prediction.
[2]
paper
GIPR is a class B1 GPCR; this family is known for large extracellular domain rearrangements and multiple signaling pathways beyond canonical Gαs.
doi: 10.1111/bph.13384
openupdated 2026-06-05

Could this gut hormone enter the brain and switch brain immune cells from attack mode to repair mode?

If true, GIP-based drugs might one day slow diseases like Parkinson's or Alzheimer's, giving patients and families more healthy years beyond just treating diabetes.

The hypothesis
GIP crosses the blood-brain barrier in pharmacologically relevant amounts and activates GIPR-expressing microglia to shift their phenotype from pro-inflammatory M1-like toward homeostatic M0, conferring disease-modifying potential in neurodegenerative disorders beyond metabolic disease.
Why it’s plausible
The readme states GIPR is distributed in the brain, yet nearly all clinical development focuses on peripheral metabolic effects. Several class B GPCR ligands (GLP-1 analogues) have shown neuroprotective effects in Parkinson and Alzheimer trials. GIP is smaller and has a different receptor distribution; if it accesses the brain and modulates microglial activation, it could represent an underexplored neuroimmune mechanism with implications for diseases currently lacking disease-modifying therapies.
Why it matters
This would expand the therapeutic scope of GIP from diabetes and obesity into neurodegeneration, potentially offering a new class of brain-penetrant immunomodulatory peptides with a better safety profile than biologic anti-inflammatory agents.
Plausibility.40
Novelty.55
Impact.70
Basis · grounding2 computed/notes
[1]
noteGIPR is widely distributed in peripheral organs and the brain, yet clinical development is overwhelmingly focused on metabolic indications.
[2]
sourceGIPR belongs to the B1 family of 7TM receptors with multiple intracellular messengers, suggesting signaling flexibility that could mediate diverse cellular responses including immune modulation.
openupdated 2026-06-05

Could taking GIP and GLP-1 as two separate injections control weight and blood sugar better than one dual drug?

If true, doctors could adjust each hormone dose to fit an individual patient, potentially getting better results with fewer side effects for people with obesity or type 2 diabetes.

The hypothesis
Co-administration of a GIP-selective peptide with a GLP-1-selective peptide at a fixed molar ratio produces synergistic, not merely additive, weight loss and glycemic control by engaging distinct neuronal populations in the brainstem and hypothalamus.
Why it’s plausible
Tirzepatide combines GIPR and GLP-1R agonism, but it is a single molecule with fixed pharmacokinetics and receptor affinities. The readme notes GIPR is widely distributed in peripheral organs and the brain. If GIP and GLP-1 act on partially non-overlapping neuronal circuits, a rationally tuned combination of two separate selective peptides could outperform a single dual agonist by allowing independent dose optimization and tissue-specific targeting.
Why it matters
This would challenge the current industry assumption that dual agonists must be single molecules, and could enable personalized incretin therapy where GIP and GLP-1 doses are titrated independently.
Plausibility.45
Novelty.30
Impact.50
Basis · grounding2 computed/notes
[1]
noteGIPR is widely distributed in peripheral organs and the brain; tirzepatide and newer dual-incretin drugs target both GIPR and GLP-1R.
[2]
sourceGIPR signals through Gαs and other intracellular messengers, suggesting signaling complexity that could interact differently with GLP-1R pathways in neurons.
openupdated 2026-06-05

Could the back end of the GIP hormone grab onto the cell surface to make its insulin signal last longer?

If true, drug makers could design GIP drugs with a stronger or weaker tail grip to control how long the medicine works, possibly creating longer-acting diabetes treatments without bigger injections.

The hypothesis
The C-terminal amphipathic helix of GIP (residues ~30-42) makes transient contact with the extracellular leaflet of the plasma membrane, and this membrane interaction stabilizes the receptor-bound conformation and prolongs signaling duration.
Why it’s plausible
The C-terminal region of GIP is predicted to be helical in many class B peptide hormones. The sequence contains a cluster of hydrophobic residues (LLAQ) followed by charged residues (KGKKNDWKHNITQ). Such amphipathic C-termini in other peptide hormones have been shown to embed partially in the membrane, effectively increasing local concentration and slowing dissociation. If GIP uses the same mechanism, truncation to GIP(1-30)NH2 would remove this membrane anchor and shorten signaling, explaining observed potency differences.
Why it matters
Engineering the C-terminal helix for stronger or weaker membrane interaction could tune residence time at the receptor, offering a new pharmacokinetic knob independent of plasma half-life extension via fatty acylation or Fc fusion.
Plausibility.35
Novelty.50
Impact.45
Basis · grounding2 computed/notes
[1]
sequenceC-terminal residues 30-42: LLAQKGKKNDWKHNITQ; contains hydrophobic patch LLAQ followed by basic residues KGKK, consistent with amphipathic helix and membrane interaction motifs seen in other peptide hormones.
[2]
noteNatural truncation GIP(1-30)NH2 removes the C-terminal 12 residues, which correlates with reduced activity compared to full-length GIP(1-42).
details expand to inspect
full evidence table2 metrics
metricvaluetool
ipTM 0.7386026382446289 boltz-2
ranking score 0.6171028017997742 boltz-2
structural qualityopenfold3
metricvaluenote
gpde1.353global PDE — lower = better
disorderNaNfraction disordered
3-letter notation
Tyr-Ala-Glu-Gly-Thr-Phe-Ile-Ser-Asp-Tyr-Ser-Ile-Ala-Met-Asp-Lys-Ile-His-Gln-Gln-Asp-Phe-Val-Asn-Trp-Leu-Leu-Ala-Gln-Lys-Gly-Lys-Lys-Asn-Asp-Trp-Lys-His-Asn-Ile-Thr-Gln
recipeboltz-2 1.0
parametervalue
modelboltz-2 1.0
weights
hardwarenvidia_nim_api
mlx version
python
random seed
msa strategynone
diffusion samples1
runtime
predicted bymlx@peptide
predicted at2026-04-24
citationbibtex
peptidemodel (2026). GIP: natural gut hormone that boosts insulin after meals (pep-10689, v1). PeptideModel. https://peptidemodel.com/card/pep-10689
@peptide{pep10689,
  sequence = {YAEGTFISDYSIAMDKIHQQDFVNWLLAQKGKKNDWKHNITQ},
  target   = {gipr},
  author   = {peptidemodel},
  year     = {2026},
  status   = {synthesized}
}
related peptides 5 by signal overlap
clinical trials 571 on ct.gov · 24 on EUCTR · checked 2026-05-22
ct.gov trials 571
with results 36
EUCTR 24
by phase
1phase 22phase 38no phase
by status
5completed1recruiting1not yet recruiting2unknown
references 10 papers
[4] supporting
[6] supporting
discussion no comments
sign in to comment
peptidemodel.com CC-BY-SA-4.0 research only · not for human use