Dulaglutide Research: The GLP-1 Fc Fusion Protein, Not a Peptide

What is dulaglutide?

What is dulaglutide? Its FDA prescribing information describes the molecule precisely: a fusion protein consisting of two identical, disulfide-linked chains, each containing an N-terminal GLP-1 analogue sequence covalently linked to the Fc portion of a modified human immunoglobulin G4 (IgG4) heavy chain by a small peptide linker, produced using mammalian cell (Chinese hamster ovary) culture. The GLP-1 analogue portion is stated to be 90% homologous to native human GLP-1(7-37), with structural modifications introduced at the site responsible for interaction with the enzyme DPP-4.

Unpack that and there are three distinct pieces of engineering in one molecule. The GLP-1 portion carries substitutions that resist DPP-4 cleavage — the same problem every molecule in this class has to solve. The IgG4 Fc portion is itself modified: the label describes changes in an area with a potential T-cell epitope, and in the regions of the Fc responsible for binding high-affinity Fc receptors and for half-antibody formation. And a small peptide linker joins the two. It is a designed protein, not a modified hormone.

The practical consequence is worth stating early, because it is what distinguishes this compound from every other entry in the GLP-1 group. At roughly 63 kDa, dulaglutide is about fifteen times the molecular weight of semaglutide (≈ 4114 Da) and cannot be produced by solid-phase peptide synthesis. It is a recombinant biologic that has to be expressed in a living cell line and purified as a protein.

Discovery and origin: solving the half-life problem a different way

Glaesner et al. (2010, Diabetes Metab Res Rev) published the engineering and characterisation of the molecule under its development code LY2189265, and the paper states the problem it was built to solve. Native GLP-1 has a half-life of minutes, secondary to rapid inactivation by DPP-4 and to excretion; the prior art at the time was exenatide, injected twice daily. The stated goal was a long-acting, efficacious GLP-1 analogue, and the approach taken was a GLP-1 immunoglobulin G4 Fc fusion protein.

This is where the compound diverges from its class. Liraglutide and semaglutide extend residence time by attaching a fatty-acid chain that binds reversibly to circulating albumin — a chemical trick applied to a synthetic peptide. Dulaglutide instead borrows the mechanism antibodies themselves use to persist in circulation, by building an antibody Fc region into the molecule. Two different answers to one question, and the reason dulaglutide is grouped with peptides that it does not otherwise resemble.

Glaesner et al. characterised the molecule in rodent and primate systems, reporting that it retained full receptor activity in vitro, that its half-life in rats and cynomolgus monkeys was 1.5–2 days with serum immunoreactivity representing active compound persisting beyond six days, and that in monkeys a single dose increased glucose-dependent insulin secretion for up to a week. Those are preclinical characterisation results reported by the developing company, described here as such.

Reference data

Sourced from the FDA prescribing information and the engineering literature. Dulaglutide has no PubChem compound record — small-molecule chemical registries do not cover a 63 kDa recombinant protein — so fields that would normally come from that record are absent rather than estimated. That absence is itself a useful signal about what kind of molecule this is.

Two rows deserve a second look. Polysorbate 80 is a surfactant included to suppress aggregation and surface adsorption — a formulation requirement of large proteins, not of small peptides. And the fact that the approved product ships as a solution rather than a freeze-dried powder reflects that its formulation was developed and validated around that presentation; lyophilisation of a multi-domain protein is a separate formulation problem with its own stability considerations.

Mechanism — one receptor, an unusual delivery vehicle

The receptor pharmacology is the least unusual part of this compound. Dulaglutide is a GLP-1 receptor agonist: a single receptor, the same one semaglutide and liraglutide engage. It is not a dual or triple agonist, and it does not engage the GIP or glucagon receptors that tirzepatide and retatrutide add. On the receptor axis it is the simplest molecule in the group.

What is distinctive is everything around the receptor interaction:

• Fc-mediated persistence. The IgG4 Fc region is the half-life mechanism, built into the molecule rather than added as a side chain. This also means the molecule’s pharmacokinetics are governed in part by antibody-recycling biology rather than by albumin binding alone.
• Deliberate Fc de-tuning. The label records modifications in the Fc regions responsible for binding high-affinity Fc receptors — that is, the developers kept the persistence properties of an antibody Fc while altering the immune-effector engagement an unmodified IgG would bring. Half-antibody formation, an IgG4-specific instability, was likewise engineered against.
• Immunogenicity as a measured endpoint. Because a large engineered protein can provoke an antibody response in a way a small peptide generally does not, anti-drug antibodies were studied directly. Milicevic et al. (2016, Diabetes Obes Metab) reported on their incidence across the phase 2 and phase 3 programme. That such a study exists at all is a consequence of the molecule’s class.

What the trial programme investigated

Reported as trial designs and programme structure, attributed to the investigators and to Eli Lilly, which funded and ran the work. No outcome figures are restated here beyond design descriptions.

AWARD — the phase 3 diabetes programme

A numbered series of randomised phase 3 trials in type 2 diabetes, each defined by its active comparator rather than by placebo alone. AWARD-1 compared dulaglutide added onto pioglitazone and metformin against exenatide; AWARD-5 (Weinstock et al., 2015) was a two-year randomised phase 3 comparison against sitagliptin in metformin-treated patients. Regional extensions followed, including an AWARD-CHN2 subgroup analysis against insulin glargine in a Chinese population. Comparator-controlled design is the defining feature of the programme.

REWIND — the cardiovascular-outcomes trial

Gerstein et al. (2019, Lancet) reported REWIND, a double-blind, randomised, placebo-controlled cardiovascular-outcomes trial in type 2 diabetes. A companion paper in the same issue reported an exploratory analysis of renal outcomes from the same trial. The distinction between a prespecified primary outcome and an exploratory analysis is preserved in the publications themselves and is preserved here.

Dulaglutide was approved by the FDA in 2014 and marketed as Trulicity. It is a prescription medicine in the jurisdictions where it is registered; supplied research material is not that medicine, was not made by its manufacturer, and inherits none of its clinical or regulatory record.

Evidence status and what is not established

What is established: a published engineering and characterisation paper describing the molecule’s design and preclinical pharmacology; a phase 3 programme of comparator-controlled randomised trials; a cardiovascular-outcomes trial with a registered protocol and peer-reviewed publication; a published immunogenicity analysis; and regulatory approval in multiple jurisdictions. As clinical records go this is a mature one.

What is not established, and is deliberately not claimed here: (1) anything about research-grade supplied material, which is not the approved product and for which no equivalence has been demonstrated — a point that carries more weight for a recombinant protein than for a synthetic peptide, because the manufacturing process is a larger determinant of what the molecule actually is; (2) any property of a lyophilized preparation of this molecule, since the characterised product is a solution; (3) any outcome in any reader, since every result cited belongs to a controlled trial of a manufactured medicine given under medical supervision; and (4) any comparison with other GLP-1 molecules beyond the comparator arms the AWARD trials actually ran.

This page reports no efficacy or benefit claim in its own voice, gives no dosing or administration guidance, and describes material supplied for laboratory research use only — not for human or veterinary use.

How to verify this compound yourself

This is the section where dulaglutide differs most sharply from everything else in the catalogue, and the difference is not a matter of degree. Standard peptide quality assurance — reversed-phase HPLC purity plus a mass-spec identity confirmation — was designed for molecules two orders of magnitude smaller, and it characterises a 63 kDa homodimer far less completely.

• Purity by RP-HPLC alone is not enough. For proteins of this size the standard purity question is answered by size-exclusion chromatography, which detects aggregates and fragments that a reversed-phase method may not resolve, usually alongside an electrophoretic method run under reducing and non-reducing conditions to confirm the disulfide-linked dimer. A single RP-HPLC percentage does not address aggregation at all.
• Intact mass is a coarse identity check here. Confirming roughly 63 kDa says little: at that size, a deamidation, an oxidised methionine or a scrambled disulfide is a small fractional mass change or none at all. Protein identity is normally established by peptide mapping — enzymatic digestion followed by LC-MS of the resulting fragments — not by a single intact-mass number.
• Higher-order structure has no mass signature. A protein that has lost its fold can have exactly the correct molecular weight and exactly the correct sequence and still not be the same material. Nothing on a typical peptide COA addresses this.
• Manufacturing route is a legitimate question to ask. Because this molecule cannot be made by peptide synthesis, any supplier offering it is offering a recombinant protein produced in cell culture and purified. Host-cell protein and residual DNA are quality attributes specific to that route and are not part of peptide testing.
• Endotoxin — where tested, an independent attribute, and one with particular relevance to a cell-culture-derived product.

For context on how often this material class fails basic checks: a 2026 preprint by Mendias and Awan analysed 6,441 samples across fourteen compounds from a large public independent-testing dataset and reported that between 41.6% and 71.1% failed basic quality criteria depending on which of two acceptance frameworks was applied, with measurable endotoxin in 15%. Dulaglutide was not among the fourteen compounds analysed; the figures are cited as a base rate for the market this material is sold in, not as a measurement of this compound. It is a preprint and has not completed peer review.

See how to read a COA, how to verify peptide purity and how to spot a fake COA — reading each with the caveat above, since all three describe peptide-scale methods. Check the exact batch on the self-serve verify tool.

Research-grade sourcing and verification

Dulaglutide is not held in Cellworks stock. It appears in the sourcing catalogue as available to order at the 5 mg and 10 mg quantities the supplier lists. Sourcing takes roughly two to three weeks, no price is quoted until availability is confirmed, and there is no cart button.

Anything sourced this way is a laboratory research reagent supplied with a per-batch Certificate of Analysis. Given everything in the section above, the honest statement for this compound is that a peptide-format certificate answers fewer of the relevant questions here than it does for a synthetic peptide, and that is worth knowing before ordering rather than after.

Frequently asked questions

Literature cited

1. Glaesner W, Vick AM, Millican R, et al. “Engineering and characterization of the long-acting glucagon-like peptide-1 analogue LY2189265, an Fc fusion protein.” Diabetes Metab Res Rev. 2010;26(4):287–296. PMID 20503261. pubmed.ncbi.nlm.nih.gov/20503261.
2. Gerstein HC, Colhoun HM, Dagenais GR, et al. “Dulaglutide and cardiovascular outcomes in type 2 diabetes (REWIND): a double-blind, randomised placebo-controlled trial.” Lancet. 2019;394(10193):121–130. PMID 31189511. pubmed.ncbi.nlm.nih.gov/31189511.
3. Gerstein HC, Colhoun HM, Dagenais GR, et al. “Dulaglutide and renal outcomes in type 2 diabetes: an exploratory analysis of the REWIND randomised, placebo-controlled trial.” Lancet. 2019;394(10193):131–138. PMID 31189509. pubmed.ncbi.nlm.nih.gov/31189509.
4. Weinstock RS, Guerci B, Umpierrez G, et al. “Safety and efficacy of once-weekly dulaglutide versus sitagliptin after 2 years in metformin-treated patients with type 2 diabetes (AWARD-5): a randomized, phase III study.” Diabetes Obes Metab. 2015;17(9):849–858. PMID 25912221. pubmed.ncbi.nlm.nih.gov/25912221.
5. Milicevic Z, Anglin G, Harper K, et al. “Low incidence of anti-drug antibodies in patients with type 2 diabetes treated with once-weekly glucagon-like peptide-1 receptor agonist dulaglutide.” Diabetes Obes Metab. 2016;18(5):533–536. PMID 26847401. pubmed.ncbi.nlm.nih.gov/26847401.
6. Li Y, Li L, De Peng Y, et al. “Efficacy and Safety of Dulaglutide Versus Insulin Glargine in Chinese T2DM Patients: A Subgroup Analysis of a Randomized Trial (AWARD-CHN2).” Diabetes Ther. 2019;10(4):1435–1452. PMID 31228090. pubmed.ncbi.nlm.nih.gov/31228090.
7. US Food and Drug Administration. “TRULICITY (dulaglutide) injection — prescribing information, Section 11 (Description).” Structure, ~63 kDa molecular weight, CHO-cell production and excipients quoted from the label via openFDA drug-label API.
8. Mendias CL, Awan TM. “Evaluation of Research Grade Peptides Marketed Directly to Consumers Reveals Extensive Variability in Purity and Measured Abundance.” Preprints (not peer reviewed). 2026. doi.org/10.20944/preprints202604.1748.v1. (Market base rate; dulaglutide not among the compounds analysed.)