From Gila monster venom to weekly injections: a short history of GLP-1 drugs

In 1987, researchers reported that the glucagon gene encodes a proglucagon precursor that contains glucagon plus additional glucagon-like peptides, and that a 31-residue form, GLP-I(7-37), stimulated insulin release in the perfused rat pancreas at very low concentrations, whereas the unprocessed GLP-I(1-37) form had no effect ¹. Later physiology work described GLP-1 as a 30-amino-acid hormone released by intestinal L-cells after meals, acting chiefly to stimulate insulin secretion, inhibit glucagon secretion and slow gastrointestinal motility ².

A hormone with a delivery problem

In 1993, a small study of ten people with poorly controlled type 2 diabetes reported that an intravenous infusion of GLP-1(7-36 amide) raised insulin, lowered glucagon and brought fasting plasma glucose into the normal range within four hours, while placebo did not ³. Dipeptidyl peptidase IV was shown to remove the N-terminal dipeptide of GLP-1(7-36)amide in human serum ⁴, and a 1995 study found that subcutaneously injected GLP-1 was rapidly converted to a truncated metabolite in both healthy and diabetic participants ⁵. A current liraglutide label puts the half-life of native GLP-1(7-37) at 1.5 to 2 minutes ⁶.

A clue from a lizard

In 1992, investigators isolated a 39-amino-acid peptide, exendin-4, from the venom of the Gila monster (Heloderma suspectum) and described its activity in guinea pig pancreatic acini ⁷. The synthetic version, exenatide, is described in its FDA label as a GLP-1 receptor agonist originally identified in that lizard ⁸. Drugs@FDA lists an original approval date of April 2005 for Byetta ⁹. In a 30-week placebo-controlled trial of 336 metformin-treated adults with type 2 diabetes, exenatide was associated with lower HbA1c and with weight loss relative to placebo ¹⁰. The label reports a mean terminal half-life of 2.4 hours ⁸.

Engineering a longer-lasting molecule

In 2000, Novo Nordisk chemists described GLP-1 derivatives attached to fatty acids to promote binding to serum albumin, and reported that derivatives carrying fatty acids of 12 carbons or more were markedly more protracted than GLP-1 in their assays ¹¹. Liraglutide followed, and its label describes a plasma half-life of 13 hours after subcutaneous administration together with stability against DPP-4 and neutral endopeptidases ⁶. A 2015 paper described the design of semaglutide, which carries two amino acid substitutions relative to human GLP-1 and a fatty-acid derivatisation at lysine 26, and which had a plasma half-life of 46.1 hours in mini-pigs ¹². In people, the Wegovy label gives an elimination half-life of approximately one week ¹³. Another weekly design, dulaglutide, fuses a GLP-1 analogue to a modified IgG4 Fc fragment and has an elimination half-life of about five days ¹⁴.

From glucose control to weight and cardiovascular outcomes

In SUSTAIN-6, 3,297 adults with type 2 diabetes were randomised to once-weekly semaglutide or placebo for 104 weeks; the primary composite cardiovascular outcome occurred in 6.6% versus 8.9% of patients (hazard ratio 0.74), while retinopathy complications were significantly more frequent in the semaglutide group ¹⁵. In STEP 1, adults with overweight or obesity but without diabetes had a mean body-weight change of -14.9% over 68 weeks with semaglutide versus -2.4% with placebo ¹⁶. Reviews describe GLP-1 actions on appetite, gastric emptying, islet function and the cardiovascular system, and emphasise that important questions remain unanswered ¹⁷.

Beyond a single receptor

In 2013, a team reported a unimolecular peptide with balanced co-agonism at the GLP-1 and GIP receptors that showed greater antihyperglycaemic and insulinotropic efficacy than selective GLP-1 agonists in rodents, monkeys and humans ¹⁸. Tirzepatide, a fatty-acid-modified agonist of both receptors, was later evaluated in the 72-week SURMOUNT-1 trial, in which mean weight change ranged from -15.0% to -20.9% across three doses versus -3.1% with placebo ^19,20. Reviews have also noted the development of orally bioavailable agonists, allosteric modulators and unimolecular multi-agonists, all targeting the GLP-1 receptor ¹⁷. Triple-receptor and glucagon-containing designs have since reached phase 3 testing ^21,22.

Key points

• GLP-1 was identified through the glucagon gene in the 1980s, and an intravenous infusion was reported to lower fasting glucose in a small 1993 study of people with type 2 diabetes ^1,3.
• Native GLP-1 is degraded within minutes, mainly through DPP-4, which drove the search for modified analogues ^4,6.
• Exendin-4 from Gila monster venom became exenatide, with Drugs@FDA listing an April 2005 approval date ^7,9.
• Fatty-acid derivatisation and albumin binding extended half-life from minutes to roughly a week for semaglutide ^11,13.