Amylin: The Other Hormone Beta Cells Release

Pancreatic beta cells are known for one product. They make two. Packed into the same secretory granules as insulin, and released with it after a meal, is a 37-amino-acid peptide called amylin — a hormone that was not identified until 1987, and then only because it had been found clogging the islets of people with type 2 diabetes.

Amylin matters to anyone following metabolic peptide research for a specific reason: it is the pathway that the newest generation of weight-management compounds targets alongside GLP-1. This guide covers what amylin is, what it does, why the human version is chemically awkward, and how far the evidence actually reaches.

Found in the wrong place first

Most hormones are discovered because something is being secreted. Amylin was discovered because something was stuck.

Pathologists had known for decades that the islets of people with type 2 diabetes often contain amyloid deposits — the same general class of insoluble protein aggregate seen in other amyloid diseases. The deposits were too insoluble, and present in too little tissue, to identify. In 1987, Cooper and colleagues at Oxford solved it: working from amyloid-rich pancreatic extracts from three patients with type 2 diabetes, they isolated a peptide of 37 amino acids and about 3,905 daltons. They named it diabetes-associated peptide, because they could not find it in six normal pancreases, and noted it was roughly 46% identical to calcitonin gene-related peptide — placing it in a recognisable hormone family rather than leaving it as a curiosity.

In the same year, Westermark and colleagues approached it from the other direction, showing that the amyloid fibrils in a human insulinoma and in the islets of a diabetic cat were derived from a neuropeptide-like protein that is also present in normal islet cells. That second finding is the important one conceptually: the material forming the deposits was not a disease-specific aberration but a normal islet product. The peptide acquired two names that are still used interchangeably — amylin, for its function, and islet amyloid polypeptide, for how it was found.

What amylin does

The clearest summary of amylin’s actions comes from a regulatory document rather than a review: the US prescribing information for pramlintide, the synthetic analogue. It states that amylin “is colocated with insulin in secretory granules and cosecreted with insulin by pancreatic beta cells in response to food intake”, that it “slows gastric emptying” without altering the overall absorption of nutrients, that it “suppresses glucagon secretion”, and that it “also regulates food intake due to centrally-mediated modulation of appetite”.

Where that central action happens has been mapped in detail. In a 2012 review in Cellular and Molecular Life Sciences, Thomas Lutz identifies the area postrema in the brainstem as the presumed primary site for most of amylin’s effects, and states plainly that “the AP is necessary for amylin’s effect on eating”. The area postrema is one of the few brain regions without a functional blood-brain barrier, so a circulating peptide can act on its neurons without needing a transport mechanism — a detail that distinguishes amylin from peptides whose central effects would require crossing that barrier. From the area postrema, signals pass to the nucleus of the solitary tract and the lateral parabrachial nucleus, and onward to hypothalamic regions.

Functionally, Lutz describes amylin as a satiation signal rather than an appetite suppressant in the loose sense: it “is believed to be a physiological control of meal size”. Plasma amylin rises within minutes of eating, low near-physiological amounts reduce eating, and blocking amylin increases meal size. Notably, the review records that the effect occurs without signs of conditioned taste aversion or visceral illness — relevant because an animal eating less can simply mean an animal feeling unwell, and distinguishing the two is what makes a satiation claim credible.

The receptor is a partnership

Amylin does not have a receptor of its own in the way most peptides do. It acts through the calcitonin receptor — a G protein-coupled receptor, from the family our guide to peptide receptors describes — but only once that receptor has acquired a partner protein.

Those partners are the receptor activity-modifying proteins, a family of three: RAMP1, RAMP2 and RAMP3. Pairing the calcitonin receptor with a RAMP changes what the receptor responds to. As Lutz puts it, “RAMPs confer specificity of the CTR to amylin”, with RAMP1 or RAMP3 the typical partners. The resulting complexes are conventionally labelled AMY1, AMY2 and AMY3 in the pharmacology literature.

This has a practical consequence for anyone designing a receptor assay: what a cell responds to depends on which proteins it happens to express, not on the receptor gene alone. It also leaves a real gap in the physiology. Lutz is explicit that “it still remains to be shown that the CTR, RAMP1, or RAMP3 all co-localize in amylin-sensitive AP neurons” — so the exact receptor composition at amylin’s own primary site of action was, at the time of that review, still inferred rather than demonstrated. A 2015 review in Pharmacological Reviews by Hay and colleagues surveys this pharmacology at length for readers who want the detail.

Why the human peptide sticks to itself

Human amylin has a property that no drug developer wants: it forms amyloid fibrils. That is how it was discovered, and it is the reason the hormone itself has never been usable as a medicine.

The chemistry is reasonably well localised. A 2026 paper in Endocrines by Hassan and colleagues describes the peptide as having an N-terminal domain spanning residues 1 to 19, an “amyloidogenic core (20–29)”, and a C-terminal region from 30 to 37. Species differences map onto that core: rat amylin, which the same authors used as a negative control and which “is reported to not form fibrils”, carries proline residues at positions 25, 28 and 29. Proline is a beta-sheet breaker — it lacks the backbone amide hydrogen that beta-sheet hydrogen bonding requires — and the authors state directly that “the proline substitutions in rIAPP and pramlintide disrupt β-sheet formation”.

Which brings the story to the drugs.

The pramlintide substitutions come straight off its US label, which describes it as “a synthetic analog of human amylin” that “differs in amino acid sequence from human amylin by replacement with proline at positions 25 (alanine), 28 (serine), and 29 (serine)”. Three residues, borrowed from a rodent, turned an unusable hormone into a drug.

From pramlintide to the current wave

Pramlintide solved aggregation but not convenience. The Hassan paper notes that it is unstable at physiological pH, must be held at pH 4.0, and cannot be co-formulated with insulin — which, for a hormone whose whole physiology is co-secretion with insulin, is an awkward limitation.

The current generation attacks the half-life problem as well. Cagrilintide is cited in that review as an amylin agonist with an extended half-life, studied most prominently in combination with semaglutide; our cagrilintide research guide works through its trial record, and the GLP-1 research map places it among the incretin-pathway compounds it is usually discussed beside. Two further analogues appear in the same review: petrelintide, described as a potent, stable, long-acting human amylin analogue from a 2025 development paper in the Journal of Medicinal Chemistry, and eloralintide, a selective amylin receptor agonist with a 48-week phase 2 trial published in The Lancet in 2025.

Two cautions are worth stating plainly. First, none of these long-acting analogues is an approved medicine; phase 2 and phase 3 results are not approvals, a distinction our guide to clinical trial phases sets out. Second, amylin is not an incretin and cagrilintide is not a GLP-1 agonist, however often the two classes are discussed in one breath — amylin signals through the calcitonin receptor family, while GLP-1 and GIP have their own receptors. Combination programmes pair two pathways; they do not merge them.

What this means at the bench

The amylin family is unusually demanding to work with, for reasons that follow directly from the biology above.

• Aggregation is the dominant failure mode. A peptide whose native form builds fibrils will not forgive careless handling. Our guide to peptide stability and degradation covers the mechanisms, and what lyophilized means covers why the dry state is the stable one.
• pH is not a detail. Pramlintide’s requirement for an acidic formulation is a reminder that solubility and stability for this family can be narrow and specific.
• Chromatography can show aggregation. Broadened or shifted peaks and new late-eluting species are the visible signature; see how to read an HPLC chromatogram.
• Receptor assays depend on what the cell expresses. Without the relevant RAMP, a calcitonin receptor is not an amylin receptor.
• Mind the nomenclature. Amylin, IAPP and diabetes-associated peptide are the same molecule; AMY1, AMY2 and AMY3 are receptor complexes, not peptides. Our amino acid codes guide helps with reading substitution notation such as S28P.

What is established and what is not

Sourcing amylin-pathway reagents in Canada

Native human amylin is a research material, not a product we stock; what the catalogue carries are the engineered analogues and the metabolic compounds they are studied alongside. For any of them:

• Confirm identity against the right mass, remembering that lipidated analogues are heavier than the native 37-residue chain. See how to read a mass spectrum.
• Match the batch number on the certificate to your vial, and keep it with your records. Our COA guide covers what to look for.
• Store cold, dry and dark, and treat aggregation-prone material carefully once reconstituted. See the storage guide.
• Do not confuse a reagent with a medicine. None of these compounds is an approved drug in Canada.

CagrilintideThe long-acting amylin analogueView product RetatrutideIncretin pathway, studied alongsideView product MazdutideGLP-1 and glucagon receptorsView product SurvodutideGLP-1 and glucagon receptorsView product

Frequently asked questions