The Gila Monster and GLP-1 Drugs: What Came From Lizard Venom, What Did Not, and Why Ozempic Is Not Made From It

The GLP-1 class began with a lizard, but only one of its drugs is a copy of the lizard's peptide. In 1992, a researcher at a New York veterans' hospital isolated a peptide from the venom of the Gila monster that turned out to switch on the same receptor as the human gut hormone GLP-1. Thirteen years later a synthetic copy, exenatide, became the first GLP-1 drug approved in the United States. The medicines most people now mean by "GLP-1" — semaglutide in Ozempic and Wegovy, tirzepatide in Mounjaro and Zepbound — were built from human hormones instead.

This page sets out the documents behind that story: the papers that isolated the peptide, the receptor work, the approvals, and a sequence comparison glp1ledger ran itself. Records were read on 11 October 2026.

1990–1992: two lizards, two peptides

The Gila monster (Heloderma suspectum) and its relative the Mexican beaded lizard (Heloderma horridum) are venomous lizards, and their venom contains hormone-like peptides. In the early 1990s Dr John Eng's group at the Solomon A. Berson Research Laboratory of the Bronx VA Medical Center isolated two of them.

• 25 November 1990 — exendin-3, from beaded-lizard venom, described in its title as "a new pancreatic secretagogue" (Eng et al., Journal of Biological Chemistry).
• 15 April 1992 — exendin-4, from Gila monster venom: a 39-amino-acid peptide that differs from exendin-3 at just two positions (glycine and glutamate at positions 2 and 3, where exendin-3 has serine and aspartate). The paper records that the change gave it different activity in guinea-pig pancreatic cells, acting only on what was then called "the exendin receptor" (Eng et al., JBC 1992).

1993: the receptor turns out to be GLP-1's

The step that turned a curiosity into a drug came the following year. Bernard Thorens and colleagues cloned the human GLP-1 receptor from pancreatic islets and showed that exendin-4 activated it — an agonist — while the shortened fragment exendin-(9-39) blocked it (Thorens et al., Diabetes 1993). A peptide from lizard venom could therefore do what the human hormone does — in the words of the exenatide label, enhance "glucose-dependent insulin secretion", suppress "inappropriately elevated glucagon secretion", and slow gastric emptying.

The practical advantage was durability. The human-based drugs that came later needed engineering to resist the enzyme that breaks GLP-1 down — the Ozempic label, for example, describes a change at position 8 "to provide stabilization against degradation by the enzyme dipeptidyl-peptidase 4 (DPP-4)", and the Trulicity label describes modifications to the part of the molecule that interacts with the same enzyme. Exenatide's label gives a mean terminal half-life of 2.4 hours — short by today's standards, but long enough for a twice-daily injection.

How much of the lizard peptide matches the human hormone

The labels describe how closely each human-based drug resembles human GLP-1, but the exenatide label says only that its sequence "partially overlaps" it. glp1ledger ran the comparison. Aligning the exenatide sequence printed in its label against human GLP-1(7-36) from the UniProt protein record P01275, position by position from the start:

Two notes on the comparison. Our 53% is a simple count of identical residues over GLP-1's 30 positions, without gaps; exenatide also carries a nine-residue tail that human GLP-1 lacks. And the labels' word "homology" may count positions differently, so the figures in the table show scale rather than an exact like-for-like comparison. The scale is the point: the lizard peptide shares about half the human sequence, the later drugs nearly all of it. One of the differences sits at the second residue, where human GLP-1 has alanine and exenatide has glycine. That is the same position — numbered 8 in the GLP-1(7-37) convention — that the Ozempic label says semaglutide modifies against DPP-4, and tirzepatide's label places an unnatural amino acid (Aib) at its own position 2.

2005 onward: from Byetta to generic exenatide

The FDA's records, read through its openFDA interface:

The labels for both Byetta (2025 version) and the Amneal generic (2026 version) open their description section the same way: exenatide "is a synthetic peptide, GLP-1 receptor agonist, that was originally identified in the lizard Heloderma suspectum." The medicine is manufactured; no lizard is involved in making it.

Why Ozempic and Mounjaro are not lizard drugs

The popular phrase "Gila monster Ozempic" merges two different lineages:

• Semaglutide and liraglutide (Ozempic, Wegovy, Rybelsus, Victoza, Saxenda) are engineered versions of human GLP-1 — 94% and 97% homologous to it, by their labels. How semaglutide differs from liraglutide is on semaglutide vs liraglutide.
• Tirzepatide (Mounjaro, Zepbound) is, in its label's words, "based on the GIP sequence" — a second human gut hormone — with changes that let it act on the GLP-1 receptor as well. The difference between the two receptors is set out on GIP and GLP-1.

What the newer drugs inherited from the Gila monster is the proof of principle: exenatide was the first evidence, in an approved medicine, that switching on the GLP-1 receptor with a long-lasting molecule could lower blood sugar. Its trial record also contains one of the more unusual findings in the class — exenatide is being studied for raised pressure inside the skull, set out on GLP-1 and headache. How the class works in the body is on the mechanism explainer, and the full list of drugs in it on which drugs are GLP-1s.

What this page cannot tell you

• The 53% figure is glp1ledger's own count, an ungapped alignment; published alignments using other methods may report a slightly different percentage.
• The research history here is the published record, not the full story of exenatide's commercial development, which involved companies whose internal documents are not public.
• Shared positions are not the whole story of how a molecule behaves; the sequence comparison says nothing about potency, duration or side effects, which come from trials.

Sources

• Eng J, Kleinman WA, Singh L, Singh G, Raufman JP, "Isolation and characterization of exendin-4, an exendin-3 analogue, from Heloderma suspectum venom", Journal of Biological Chemistry 1992;267:7402–7405, PubMed 1313797.
• Eng J, et al., "Purification and structure of exendin-3, a new pancreatic secretagogue isolated from Heloderma horridum venom", Journal of Biological Chemistry 1990;265:20259–20262, PubMed 1700785.
• Thorens B, Porret A, Bühler L, et al., "Cloning and functional expression of the human islet GLP-1 receptor. Demonstration that exendin-4 is an agonist and exendin-(9-39) an antagonist of the receptor", Diabetes 1993;42:1678–1682, doi:10.2337/diab.42.11.1678.
• UniProt P01275, Pro-glucagon (human), chain "Glucagon-like peptide 1(7-36)", rest.uniprot.org, read 2026-10-11.
• Labels read on DailyMed, 2026-10-11: BYETTA (AstraZeneca, set ID 53d03c03-ebf7-418d-88a8-533eabd2ee4f, effective 2025-09-02), sections 11 and 12; exenatide injection (Amneal, e6cb5c8f-e97f-4a6a-95a4-939fd2393949, effective 2026-05-27), section 11; VICTOZA (5a9ef4ea-c76a-4d34-a604-27c5b505f5a4), OZEMPIC (adec4fd2-6858-4c99-91d4-531f5f2a2d79), TRULICITY (463050bd-2b1c-40f5-b3c3-0a04bb433309) and MOUNJARO (d2d7da5d-ad07-4228-955f-cf7e355c8cc0), sections 11 and 12.1. At dailymed.nlm.nih.gov.
• FDA Drugs@FDA records via openFDA, read 2026-10-11: NDA 021773, NDA 022200, NDA 209210, ANDA 206697.