GLP-1 Research Peptides: The Complete Guide (Semaglutide, Tirzepatide, Retatrutide and more)

GLP-1 research peptides are synthetic analogues of the incretin hormones glucagon-like peptide-1 and glucose-dependent insulinotropic polypeptide, supplied as lyophilised powder for laboratory investigation of receptor pharmacology, signalling bias and metabolic physiology. This guide explains the incretin axis, how mono-, dual- and triple-agonist molecules differ at the receptor level, what the published literature reports, and how these compounds are handled, characterised and verified in a research setting.

Updated 10 September 2026~14 min readReviewed by the GenoPept technical team

What are GLP-1 research peptides?

GLP-1 research peptides are laboratory-grade synthetic analogues of glucagon-like peptide-1 and its sister incretin GIP, chemically modified so they resist enzymatic degradation and persist long enough to be studied in receptor assays, cell culture and preclinical models. They are supplied as lyophilised (freeze-dried) powder in sealed vials and reconstituted by the researcher immediately before use.

The category is defined by pharmacology rather than by any single sequence. What unites semaglutide, tirzepatide, retatrutide, liraglutide, mazdutide, survodutide and their relatives is that each engages one or more class B G-protein-coupled receptors in the secretin receptor family, driving cyclic AMP accumulation through Gs coupling. What separates them is which receptors they engage, with what relative potency, and how long the modified backbone survives in a biological matrix.

Researchers in this area need three things from a supplier: an unambiguous identity, a purity figure generated by an appropriate method, and a physical form that survives shipping and storage.

The incretin axis: where GLP-1 comes from and what it does

The incretin effect describes the observation that nutrients delivered into the gut provoke a larger insulin response than the same glucose load delivered intravenously. Two hormones account for most of that difference: GLP-1, released from intestinal L-cells, and GIP, released from K-cells in the proximal small intestine.

GLP-1 is a product of post-translational processing of proglucagon. The biologically active forms are GLP-1(7-37) and the amidated GLP-1(7-36)NH₂, both 30–31 residues long. Once released, they engage the GLP-1 receptor, a class B GPCR expressed on pancreatic beta cells, on neurons in the hindbrain and hypothalamus, on gastric smooth muscle and in several other tissues.

Drucker’s 2018 review in Cell Metabolism sets out the breadth of that receptor distribution and the downstream physiology, and remains the standard orientation text for anyone entering this field.¹ The key mechanistic points are that beta-cell signalling is glucose-dependent, that central receptor populations are implicated in the regulation of food intake in animal models, and that gastric motility is modulated in a manner that shows tachyphylaxis for some analogues but not others.

GIP acts on a distinct receptor, GIPR, with its own tissue distribution that includes adipose tissue and central nervous system sites. Because GIPR and GLP-1R are separate proteins with overlapping but non-identical expression maps, a molecule that engages both has a pharmacological profile that is not simply the sum of two mono-agonists — a point demonstrated directly in receptor-level work on tirzepatide.⁴

Why the native hormones cannot be used as research tools

Dipeptidyl peptidase-4 cleaves the N-terminal dipeptide from both GLP-1 and GIP, producing metabolites with sharply reduced receptor potency. The circulating half-life of native GLP-1 is on the order of two minutes. Renal clearance removes what DPP-4 leaves behind.

For a laboratory tool this is fatal outside the shortest of acute assays. Every compound in the GLP-1 research category exists because chemists solved that problem, and the way each one solves it explains most of its behaviour.

How GLP-1 analogues are engineered for stability

Two modifications dominate this class: replacing the DPP-4 substrate residue with a non-natural amino acid, and attaching a long-chain fatty diacid through a hydrophilic linker so the peptide binds reversibly to serum albumin. Together these extend circulating persistence from minutes to days.

The first modification is usually α-aminoisobutyric acid (Aib), a symmetric, doubly methylated alanine. Placed at the position that DPP-4 recognises, it makes the peptide a poor substrate without materially reducing receptor affinity. Lau and colleagues described this design logic explicitly in the medicinal-chemistry paper that reported semaglutide’s discovery.²

The second modification is acylation. A C16, C18 or C20 fatty acid — usually a diacid, so one carboxyl remains free — is coupled to a lysine side chain through a spacer built from γ-glutamic acid and one or two units of 8-amino-3,6-dioxaoctanoic acid (AEEA, also written ADO). The fatty tail docks into albumin’s hydrophobic binding sites. Because the interaction is reversible, the peptide is released slowly, and the albumin reservoir also shields it from renal filtration.

A third, quieter modification is substitution of residues that are chemically vulnerable. Replacing a lysine elsewhere in the sequence with arginine, for example, prevents the acylation chemistry from attaching a fatty acid at the wrong site during synthesis.

Receptor coverage: mono, dual and triple agonists

Receptor coverage is the most useful way to organise this category. A mono-agonist engages GLP-1R alone. A dual agonist adds either GIPR or the glucagon receptor (GCGR). A triple agonist engages all three. Amylin analogues sit outside the incretin family entirely, acting at calcitonin-receptor complexes.

The distinction matters because each receptor arm contributes different physiology in the published preclinical and clinical literature. GLP-1R activation is associated with glucose-dependent insulin secretion and central appetite signalling. GIPR pharmacology is more contested; work on tirzepatide showed that the molecule is an imbalanced agonist, behaving essentially as a full agonist at GIPR while being weaker and biased at GLP-1R.⁴ GCGR engagement is generally discussed in terms of hepatic substrate handling and energy expenditure.

Structural work has begun to explain how one backbone can satisfy two receptors. Sun and colleagues reported structural determinants of tirzepatide’s dual agonism, showing how a GIP-derived scaffold retains enough of the GLP-1 recognition motif to activate GLP-1R while remaining optimised for GIPR.⁵

The compound families in the GLP-1 research category

Seven incretin analogues and two blends make up the GLP-1 research category at GenoPept. The table below summarises the identity and pharmacological class of each. All figures are physicochemical properties of the compound, not administration guidance.

What the published research programmes report

The clinical literature for this class is unusually large for research peptides, because several of these molecules progressed through licensed pharmaceutical development. Reading it is the fastest way to understand what each receptor arm contributes, and it is the primary reason researchers select one analogue over another.

Semaglutide’s obesity programme was reported by Wilding and colleagues in the New England Journal of Medicine in 2021.⁶ Tirzepatide’s equivalent programme was reported by Jastreboff and colleagues in 2022.⁷ A direct head-to-head comparison of tirzepatide and semaglutide in type 2 diabetes was published by Frías and colleagues in 2021.⁸

Retatrutide’s phase 2 obesity trial was published by Jastreboff and colleagues in 2023 and is the reference point for triple-agonist pharmacology in humans.⁹ The molecule’s discovery and receptor characterisation had been reported the previous year by Coskun and colleagues in Cell Metabolism.¹⁰

For research purposes the useful content of these papers is not the headline outcome but the pharmacokinetic and pharmacodynamic detail: exposure–response relationships, receptor occupancy inferences, gastric emptying data, and the tolerability signals that constrain how quickly exposure can be increased in any model system.

Reading the literature without over-reading it

Two cautions apply. Trial results describe the licensed pharmaceutical form under controlled conditions and do not transfer to a research-grade compound. And receptor pharmacology measured in transfected cell lines does not always predict tissue-level behaviour, which is why the structural and signalling-bias papers matter alongside the outcome trials.^4,5

Handling, solution preparation and storage

Acylated incretin analogues are surface-active, which shapes almost every handling rule. The fatty tail that binds albumin also binds plastic, glass and air–liquid interfaces. Foaming, vortexing and repeated freeze–thaw are the three most common causes of unexplained potency loss in this class.

1. Equilibrate before opening. Allow the sealed vial to reach ambient temperature before removing the flip cap, so atmospheric moisture does not condense onto the lyophilised cake.
2. Introduce diluent down the vial wall. Direct the stream against the glass rather than onto the cake. This avoids driving the powder into suspension as a foam.
3. Dissolve by slow inversion. Rotate the vial gently until the cake clears. Do not shake, and do not vortex.
4. Inspect the solution. A correctly reconstituted incretin analogue is clear and colourless with no visible particulates or opalescence.
5. Aliquot for repeat use. Divide the stock into single-use volumes so that the working stock is never subjected to repeated temperature transitions.
6. Label and record. Note the compound, batch number, concentration in mg/mL, diluent identity and preparation date on every aliquot.

The table below is a concentration reference for solution preparation. It converts vial mass and diluent volume into a solution concentration, which is the only quantity a laboratory protocol should express.

Purity, identity and certificate of analysis

A purity figure without a method is meaningless. For acylated incretin analogues, identity and purity should be established by reversed-phase HPLC — which separates the target peptide from deletion sequences, truncations and mis-acylated species — combined with mass spectrometry, which confirms the deconvolved molecular mass against the theoretical value.

Mass spectrometry is especially important here. Two of the commonest synthesis defects in this class are the absence of the fatty acid entirely and the attachment of an incomplete linker. Both change the mass by a predictable, detectable amount while barely shifting HPLC retention. A COA that reports only an HPLC area percentage cannot exclude either.

A further distinction that catches researchers out is purity versus net peptide content. HPLC purity describes what fraction of the peptide-related material is the target. Net peptide content describes what fraction of the total vial mass is peptide at all, the remainder being counter-ions (typically trifluoroacetate or acetate) and residual water. A 99% pure peptide can still be only 80% peptide by mass.

GenoPept publishes a per-batch third-party certificate of analysis covering HPLC purity and mass-spectrometric identity for every product; batch documents are available at the COA certificates page. Our guide to peptide purity testing explains how to read those chromatograms.

Choosing between GLP-1 peptides for a research plan

Receptor question first, chemistry second. The compound selection follows from what the experiment is designed to isolate: a single receptor arm, an interaction between two arms, or a comparison across the whole family.

Practical constraints matter too. Blends save a reconstitution step but remove the ability to vary the ratio between the two peptides, so they suit confirmatory work rather than exploratory mapping.