Dual-channel receptor atlas · human tissue only

One molecule, two channels, about a dozen cell types that matter.

Tirzepatide does nothing anywhere without a receptor. This maps every human cell type known to carry GLP-1R or GIPR, what happens when the drug lands there, and what you can actually do about it. Rodent-only findings are flagged, because most of the popular story rests on them.

This page describes mechanism. It is not medical advice, it is not a treatment plan, and reading it creates no clinical relationship. Dosing, safety, and treatment decisions belong with your own clinician, who knows things about you that this page cannot.

showing all evidence tiers

GLP-1R

GIPR

both - the channels overlap and read white

Dot size = receptor density (0-3)

Tirzepatide was built by engineering GLP-1 activity into the GIP backbone. It is not a balanced dual agonist, and it does not signal at GLP-1R the way GLP-1 does. Both facts are load-bearing.

Signalling profile, relative to each native hormone

At GIPR it behaves like the real hormone. At GLP-1R it drives cAMP while barely recruiting β-arrestin - so the receptor internalises less and stays available on the cell surface. In primary islets, β-arrestin1 limits the insulin response to GLP-1 but not to tirzepatide.

Imbalanced toward GIPR

Tirzepatide binds GIPR with affinity matching native GIP, but binds GLP-1R roughly 5× weaker than native GLP-1. Receptor-occupancy modelling at clinical doses confirms greater GIPR engagement.

Why this is practical: GLP-1R dose escalation is limited by nausea and vomiting. GIPR engagement isn't. Leaning the molecule toward GIPR is how you get more total incretin effect per unit of gastrointestinal misery.

The rodent trap

Tirzepatide is far less potent at mouse GIPR than at human GIPR. Any mouse study of this drug systematically underweights the GIP arm - the half that most distinguishes it from a GLP-1-only agonist.

How to use this: when you meet a mechanistic claim about tirzepatide, check the species first. Mouse work here is a weak proxy in a specific, predictable direction.

What adapts, and what doesn't

Delayed gastric emptying fades with repeated dosing, but less completely than it is usually described. In the four-week phase 1 study the effect was pronounced after the first dose and had largely gone by the end in healthy participants - but in participants with type 2 diabetes on an escalating schedule a residual delay was still measurable after multiple doses, which is the population actually taking the drug. Appetite suppression, glucose lowering and weight loss do not fade the same way.

Consequence: nausea and reflux are worst early and after each escalation, and genuinely improve. Constipation, driven by enteric neurons along the whole gut, is more persistent, and unlike the nausea it does not reliably settle on its own.

Evidence tiers - human protein / human function: validated antibody, radioligand binding, or measured response in people · human transcript: single-cell or single-nucleus RNA sequencing of human tissue · rodent only: not confirmed in human tissue.

Sources

Pyke C, Heller RS, Kirk RK, Orskov C, Reedtz-Runge S, Kaastrup P, Hvelplund A, Bardram L, Calatayud D, Knudsen LB. 2014. GLP-1 receptor localization in monkey and human tissue: novel distribution revealed with extensively validated monoclonal antibody. Endocrinology 155(4):1280-1290. doi:10.1210/en.2013-1934 - the anchor for the GLP-1R map. Its whole point is that an extensively validated antibody finds the receptor in far fewer places than earlier work claimed, which is why this atlas is shorter than the popular story.
Willard FS, Douros JD, Gabe MB, et al. 2020. Tirzepatide is an imbalanced and biased dual GIP and GLP-1 receptor agonist. JCI Insight 5(17):e140532. doi:10.1172/jci.insight.140532 - the anchor for the pharmacology panel: GIPR affinity matching native GIP, weaker GLP-1R binding, cAMP over beta-arrestin bias, and the species difference at mouse GIPR.
Urva S, Coskun T, Loghin C, et al. 2020. The novel dual GIP and GLP-1 receptor agonist tirzepatide transiently delays gastric emptying similarly to selective long-acting GLP-1 receptor agonists. Diabetes, Obesity and Metabolism 22(10):1886-1891. doi:10.1111/dom.14110 - the four-week study behind the tachyphylaxis claim, including the residual delay in participants with type 2 diabetes that the simple version of the story leaves out.
Hansen MS, Soe K, Christensen LL, et al. 2023. GIP reduces osteoclast activity and improves osteoblast survival in primary human bone cells. European Journal of Endocrinology 188(1):144-157. doi:10.1093/ejendo/lvac004 - GIPR in primary human osteoclasts and osteoblasts, and the reason the bone entries are tiered as human rather than rodent.
Emont MP, Jacobs C, Essene AL, et al. 2022. A single-cell atlas of human and mouse white adipose tissue. Nature 603:926-933. doi:10.1038/s41586-022-04518-2 - the human white adipose single-nucleus reference behind the adipocyte entries, which are transcript-level evidence rather than measured function.
Sun B, Willard FS, Feng D, et al. 2022. Structural determinants of dual incretin receptor agonism by tirzepatide. PNAS 119(13):e2116506119. doi:10.1073/pnas.2116506119 - how one peptide engages two receptors, and the structural basis of the imbalance the pharmacology panel describes.

Scope: this atlas covers human cell types with published GLP-1R or GIPR evidence, tiered by what that evidence is. The tiers are the substance of the page: human protein or function claims are checkable against the sources above, transcript-level entries say only that the message is present, and rodent-only entries are marked because they are where the popular account of this drug most often outruns the human data. Effects listed under "where it doesn't bind" are real but indirect, and this page does not attribute them to a receptor. Nothing here is medical advice, and dosing and safety decisions belong with your clinician.