Three Generations, Counted by Receptors
The simplest way to order the incretin analogues is by how many receptors each one activates. Semaglutide engages one. Tirzepatide engages two. Retatrutide engages three. Everything else — the acylation chemistry, the half-life engineering, the backbone each is built on — follows from that choice.
This article compares the three side by side: what each one binds, how each is held together, what the published data show, and which of those differences actually change how you design an experiment. It is written for laboratory work with reference material, not for anything resembling clinical use.
A Note on the Supplier Shorthand
Before the pharmacology, a correction that matters. Research-chemical suppliers frequently label these compounds GLP-1, GLP-2 TRZ and GLP-3 RT. Those are internal product codes for first-, second- and third-generation incretins. They are not receptor names, and two of the three are actively misleading.
There is no GLP-3 receptor. Tirzepatide has nothing to do with GLP-2, which is a real receptor with a real and entirely unrelated biology in the intestinal epithelium. If you are matching a certificate of analysis against a catalog entry, match on the CAS number — 910463-68-2, 2023788-19-2 and 2381089-83-2 respectively — rather than on the product name.
What Each One Actually Binds
Semaglutide is a selective GLP-1 receptor agonist. One receptor, one pathway, and the cleanest control of the three for any experiment where a second incretin signal would confound the readout.
Tirzepatide is a dual GIP and GLP-1 receptor agonist. In-vitro characterisation describes an imbalanced profile: GIP-receptor affinity comparable to native GIP, with weaker GLP-1-receptor affinity than native GLP-1. At the GLP-1 receptor it favours cAMP generation over β-arrestin recruitment, a bias associated with reduced receptor internalisation in cell-based assays.
Retatrutide adds the glucagon receptor to both of those. One published review describes full agonism at all three, with the highest potency at GIP. Reported EC50 figures differ between sources — one review gives 0.0037 nM at GIP where vendor data sheets report roughly 0.0643 nM — so treat any single value as provisional.
A fourth pattern exists and is worth knowing about: mazdutide pairs GLP-1 with glucagon and leaves GIP out entirely. It is covered separately in our note on the glucagon arm.
Three Different Backbones
The three are not variations on one scaffold. Semaglutide is built on GLP-1(7–37), with Aib8 and Arg34 substitutions and a C18 fatty diacid on Lys26 through a γGlu–2×OEG linker. Tirzepatide is built on the native GIP sequence — 39 residues, Aib at positions 2 and 13, a C20 diacid on Lys20, C-terminal amide. Retatrutide is a 39-residue chain with Aib at 2 and 20, an α-methyl-leucine at 13, and a C20 diacid on Lys17 through a γGlu–AEEA linker.
The common thread is the fatty diacid. All three bind albumin reversibly, which is what produces a multi-day half-life and what makes assay medium composition a variable rather than a detail.
Molecular weights: 4113.58, 4813.45 and 4731.33 g/mol. Note that retatrutide is lighter than tirzepatide despite targeting an additional receptor — a useful sanity check when reading a mass-spec identity result.
What the Published Data Show
Semaglutide and tirzepatide are the active ingredients of approved prescription medicines with large trial programmes. The head-to-head SURPASS-2 trial reported greater HbA1c reduction with tirzepatide than semaglutide in type 2 diabetes, and SURMOUNT-5 compared them directly in obesity.
Retatrutide's phase 2 results appeared in the New England Journal of Medicine in 2023, and Eli Lilly has since announced topline phase 3 data. Sources disagree on its current regulatory stage, and some widely cited figures come from vendor and telehealth pages rather than primary literature.
All of that describes approved or investigational pharmaceutical products under clinical supervision. None of it applies to research-grade material, which is not manufactured or released for human use. See our research use disclaimer.
What Changes in the Laboratory
Albumin in the medium is not neutral. All three are engineered to bind it. Serum or BSA in your assay medium will bind a fraction of the peptide and lower the free concentration, shifting apparent potency. Keep albumin content constant across comparisons, and state it in your methods.
Receptor expression determines what you can see. A cell line expressing only GLP-1R cannot distinguish tirzepatide from semaglutide on anything but potency, and cannot distinguish retatrutide from either. Separating a triple agonist's arms needs three single-receptor systems, not one.
Signalling bias needs more than one readout. Tirzepatide's cAMP-versus-β-arrestin bias is invisible in a cAMP-only assay. If bias is the question, measure both.
Reconstitution is the same for all three. Each dissolves in water and bacteriostatic water, with reduced solubility under acidic conditions. Our peptide reconstitution guide covers volumes and technique; bacteriostatic water is the usual diluent for multi-draw stock.
Which to Use as a Comparator
If your question is about GLP-1 signalling alone, semaglutide is the control, because nothing else is happening.
If it is about GIP's contribution, tirzepatide against semaglutide isolates one variable cleanly.
If it is about glucagon-receptor engagement, retatrutide against tirzepatide isolates that arm — or mazdutide against semaglutide, which adds glucagon without adding GIP.
Every lot we ship carries a batch certificate; look yours up in the COA verification hub.
Frequently Asked Questions
Is retatrutide a GLP-3 agonist?
No. There is no GLP-3 receptor. Retatrutide is a triple agonist at the GIP, GLP-1 and glucagon receptors. "GLP-3 RT" is a supplier product code for a third-generation incretin, not a pharmacological description.
Why is retatrutide lighter than tirzepatide if it hits more receptors?
Receptor count and molecular weight are unrelated. Retatrutide is 4731.33 g/mol against tirzepatide's 4813.45, because the two are built on different backbones with different substitutions. Triple agonism comes from how the sequence presents to three binding pockets, not from added mass.
Can one cell line compare all three?
Only on potency at whichever receptor it expresses. Separating a dual agonist from a triple agonist requires systems expressing GIP, GLP-1 and glucagon receptors individually; a single GLP-1R line will rank all three on one axis and tell you nothing about the others.
Which has the longest half-life?
All three use a fatty diacid for reversible albumin binding and are designed for once-weekly administration in their clinical forms. Published half-lives are broadly comparable and depend heavily on the system measured, so treat cross-source comparisons with care.
References & Further Reading
Coskun T et al., Cell Metab 2022; Jastreboff AM et al., N Engl J Med 2023;389:514-526; Doggrell SA, Expert Opin Investig Drugs 2023; Frías JP et al., N Engl J Med 2021 (SURPASS-2); structural analysis of retatrutide receptor binding, Nat Commun 2024.