Glucagon-like peptide-2 (native)
GLP-2, GLP-2 (1-33), human GLP-2, proglucagon-derived peptide
GLP-2 is a 33-amino-acid hormone released from L-cells in the ileum and colon after eating, cut from the same proglucagon precursor that yields GLP-1. Its defining action is intestinotrophic: it makes intestinal mucosa grow, deepening crypts and lengthening villi, while also slowing gastric emptying and reducing acid secretion. The native peptide is destroyed within minutes by DPP-4, which is why it has never become a medicine in its own right. Its analogues have.
Mechanism
GLP-2 is generated when prohormone convertase 1/3 processes proglucagon in intestinal L-cells, the same cells and the same precursor that yield GLP-1, oxyntomodulin and glicentin. All are co-secreted in response to luminal nutrients, particularly fat and carbohydrate, with a biphasic pattern: a rapid neurally and hormonally mediated early phase followed by a sustained phase as nutrients reach the distal gut. GLP-2 acts through the GLP-2 receptor, a class B G-protein-coupled receptor closely related to the GLP-1 and glucagon receptors.
The receptor's distribution is the key to understanding the peptide. GLP-2 receptors are absent from enterocytes; they sit on subepithelial myofibroblasts, enteric neurons and a subset of enteroendocrine cells. Every effect on the epithelium is therefore indirect. Receptor activation raises cyclic AMP and drives release of insulin-like growth factor-1 from myofibroblasts (the principal mediator, acting through beta-catenin signalling in crypt stem cells), along with keratinocyte growth factor, ErbB ligands, vasoactive intestinal peptide and nitric oxide from enteric neurons. The consequences are increased crypt-cell proliferation, reduced enterocyte apoptosis, greater mucosal surface area, increased mesenteric blood flow, enhanced barrier function and reduced gastric motility and acid output. Native GLP-2 is cleaved at the Ala2-Asp3 bond by dipeptidyl peptidase-4 to the largely inactive GLP-2(3-33), giving a circulating half-life of roughly 7 minutes.
What the research shows
The foundational preclinical observation came from Drucker and colleagues in 1996, who showed that GLP-2 administration to mice produced marked small bowel epithelial proliferation, the first demonstration that a proglucagon-derived peptide was intestinotrophic. The pivotal human study is Jeppesen's 2001 report in Gastroenterology: eight patients with short bowel syndrome and no colon received native GLP-2 400 micrograms subcutaneously twice daily for 35 days, with formal balance studies before and after. Intestinal energy and wet weight absorption improved, body weight and lean body mass rose, and jejunal biopsies showed increased crypt depth and villus height. That is a genuine and mechanistically coherent result, but it was open-label, uncontrolled and involved eight people.
Development then moved entirely to protease-resistant analogues, because twice-daily injection of a peptide with a 7-minute half-life is not a viable product. Teduglutide, the [Gly2] analogue, is licensed. Glepaglutide, a long-acting analogue, showed dose-dependent reductions in faecal output and increased absorption in a randomised phase 2 trial reported in Lancet Gastroenterology and Hepatology in 2019, and subsequently reduced parenteral support in a phase 3 randomised controlled trial published in Gastroenterology in 2025. Apraglutide, a further long-acting analogue permitting weekly administration, has also progressed to phase 3. The interesting scientific tension in this field is that the very property that makes GLP-2 useful (driving mucosal growth) is also the source of the neoplasia concern that governs the monitoring requirements on every GLP-2 analogue label. Native GLP-2 itself remains a physiological tool rather than a therapy.
Evidence assessment
Limited evidence
Human data on native GLP-2 as a therapeutic consist of a small number of open-label, uncontrolled studies, most notably Jeppesen's 2001 balance study in eight short bowel patients. There are no randomised controlled trials of the native peptide itself. The randomised evidence in this record belongs to engineered analogues (teduglutide and glepaglutide), not to native GLP-2. The underlying physiology is extremely well characterised, but the therapeutic evidence for the native hormone is thin because development moved wholesale to DPP-4-resistant analogues.
Tiers are applied consistently across the library and re-checked when new trials read out. Read the grading method.
Key studies
Induction of intestinal epithelial proliferation by glucagon-like peptide 2 Preclinical only
Identified GLP-2 as the proglucagon-derived peptide responsible for intestinal epithelial proliferation, establishing the entire therapeutic rationale for the class.
Glucagon-like peptide 2 improves nutrient absorption and nutritional status in short-bowel patients with no colon Preclinical only
Improved intestinal energy and wet weight absorption, increased body weight and lean mass, and increased crypt depth and villus height on biopsy. The key human proof of concept, but uncontrolled and very small.
Glepaglutide, a novel long-acting glucagon-like peptide-2 analogue, for patients with short bowel syndrome: a randomised phase 2 trial Preclinical only
Dose-dependent reduction in faecal wet weight output and increased intestinal absorption, supporting progression of long-acting GLP-2 analogues to phase 3.
Glepaglutide, a long-acting glucagon-like peptide-2 analogue, reduces parenteral support in patients with short bowel syndrome: a phase 3 randomized controlled trial Preclinical only
Confirmed reduction in parenteral support requirement with a long-acting GLP-2 analogue, extending the randomised class evidence beyond teduglutide.
Safety
Native GLP-2 has not been characterised in any large safety dataset. In the small short bowel studies it was reported as well tolerated, with local injection reactions the main issue. The theoretical safety concerns are inherited by the whole class and are taken seriously by regulators: because GLP-2 stimulates crypt-cell proliferation, there is a plausible risk of promoting existing intestinal neoplasia, which is why licensed analogues require colonoscopy before starting and periodic surveillance thereafter, and are contraindicated in active gastrointestinal malignancy. Fluid overload is a class effect in patients whose parenteral fluids are not reduced as absorption improves, and pancreaticobiliary events require monitoring. Native GLP-2 sold as a research chemical carries the usual problems of that market (no pharmacopoeial identity or purity assurance, no sterility guarantee, and no clinical oversight), compounded by the fact that its 7-minute half-life makes meaningful effect from intermittent self-administration implausible in the first place.
Regulatory status
| Jurisdiction | Status |
|---|---|
| United Kingdom | Native GLP-2 has no UK marketing authorisation and is not a licensed medicine. Teduglutide, the [Gly2] analogue, is licensed. Native GLP-2 sold to consumers would be an unlicensed product; it is legitimately supplied only as a laboratory research chemical, and its extremely short half-life makes any consumer use pharmacologically pointless. |
| United States | Native GLP-2 is not an approved medicine in the United States and has no FDA-licensed product. Its DPP-4-resistant analogue teduglutide is approved. Native GLP-2 is available only as a research reagent. |
| WADA (sport) | Not listed on the WADA Prohibited List. GLP-2 receptor agonists are not covered by any current category. |
Questions
Because dipeptidyl peptidase-4 cleaves it at the second amino acid within minutes, giving a half-life of around 7 minutes. No practical dosing schedule can maintain useful concentrations. Every GLP-2 medicine in clinical use (teduglutide, and the investigational glepaglutide and apraglutide) is engineered specifically to resist that cleavage.
No, and this is one of the more elegant facts in gut endocrinology. GLP-2 receptors are not present on enterocytes at all. They sit on subepithelial myofibroblasts and enteric neurons, which respond by releasing IGF-1, keratinocyte growth factor and other mediators that then act on crypt cells. The trophic effect is entirely indirect.
Closely. Both are cut from the same proglucagon precursor by the same enzyme in the same intestinal L-cells, and both are released together after a meal. They act on different receptors and do entirely different things: GLP-1 on insulin, appetite and gastric emptying; GLP-2 on intestinal growth and absorption.
No. Native GLP-2 has alanine at position 2; teduglutide has glycine. That difference is what makes teduglutide durable enough to be a medicine. Native GLP-2 offered for sale is a research reagent with no licensed status, no purity guarantee, and a half-life short enough that intermittent injection would achieve little.