Insulin's forgotten twin

Ask anyone to name what the pancreas secretes to control blood sugar and the answer is insulin. It is one of the most famous molecules in medicine — a hundred-year-old first miracle of peptide therapy. What almost no one names is the hormone packed into the very same storage granules, released in the same pulse, from the same beta cells: amylin.

They are not neighbors who happen to fire together. Insulin and amylin are co-packaged and co-secreted — a built-in partnership, roughly a single hormonal system with two chemically distinct signals. Yet amylin wasn’t even isolated until 1987, and it was found not in a hunt for a new hormone but in the wrong place entirely: buried in the amyloid — the pathological protein clumps — that scar the islets of a diabetic pancreas. Its formal name still carries that origin: islet amyloid polypeptide, IAPP. The twin was discovered as debris.

Here is the first genuinely elegant thing about amylin: it has no receptor of its own. There is no gene that codes for a dedicated “amylin receptor.” Instead, amylin hijacks a receptor that already exists for another hormone — the calcitonin receptor — and changes what it listens for by adding a single accessory protein.

That accessory is a RAMP — a receptor activity-modifying protein. On its own, the calcitonin receptor responds to calcitonin. Clamp a RAMP onto it and the same core receptor is re-tuned: now it prefers amylin. Swap which RAMP (there are three) and you get the family of amylin receptors. The body didn’t evolve a whole new receptor for its second beta-cell hormone; it took one it already had and gave it a different hat. It is molecular economy — reuse the machine, change the adapter — and it is why amylin’s pharmacology is knitted into the same calcitonin family that governs bone and calcium.

If insulin’s job is to put nutrients away — pulling glucose into cells after a meal — amylin’s job is to govern how fast those nutrients arrive and when the body decides it has had enough. It pulls three distinct brakes:

• It slows gastric emptying. food leaves the stomach more gradually, so glucose enters the blood as a gentle rise rather than a spike insulin then has to chase
• It suppresses inappropriate glucagon. it silences the counter-regulatory signal that would otherwise tell the liver to add its own glucose on top of the meal
• It promotes satiety centrally. it acts on the area postrema, a hindbrain outpost that sits outside the blood-brain barrier and reads circulating signals, to produce the sense of fullness

Read those together and amylin is the rate-and-enough hormone to insulin’s storage hormone. They aren’t redundant — they cover different halves of the same post-meal problem, which is exactly why evolution ships them in the same granule.

Now the twist that hid amylin for so long. Human amylin is chemically sticky. Its native sequence is prone to misfolding and aggregating into amyloid fibrils — the same broad class of pathological protein clumping seen in other amyloid diseases. In the type 2 diabetic pancreas, that aggregated amylin deposits in the islets and is toxic to the very beta cells that made it, part of the slow attrition of insulin-producing capacity. The hormone meant to partner insulin can, misfolded, help kill the cells that produce both.

That same stickiness made native human amylin nearly undruggable — you cannot bottle a peptide that clumps in the vial. The fix was engineering. The approved analog pramlintide borrows a trick from rodent amylin, whose sequence resists aggregation, substituting a few proline residues to break the misfolding-prone stretch while keeping the biology. Newer long-acting analogs like cagrilintide extend that idea for once-weekly dosing. The “amylin drugs” are, in a real sense, amylin with its self-destruct sequence edited out.

So why is a hormone discovered in 1987 suddenly one of the hottest bets in metabolic medicine? Because it does its work through a different door than the incretins. GLP-1 drives satiety through the GLP-1 receptor; amylin drives it through amylin receptors in the area postrema. Two separate pathways converging on the same goal means their effects stack rather than overlap — the appetite suppression adds up instead of hitting one receptor’s ceiling twice.

That is the logic behind the current wave: cagrilintide paired with semaglutide (an amylin analog plus a GLP-1 analog in one regimen), and amycretin, a single molecule engineered to be both a GLP-1 and an amylin agonist at once. It is the same move the triple agonist made with glucagon — recruit another arm of the body’s own machinery into the chord — except the arm here isn’t a third incretin. It’s insulin’s original partner, brought back into the fold. When we ask what comes after the triple agonist, amylin is the clearest answer: not a bigger number, but a hormone the body was already using all along.

All of this remains an active story — cagrilintide and amycretin are investigational, and the durable head-to-head evidence is still being written. But the biology underneath is a century old and hiding in plain sight. The most modern idea in the field turns out to be the pancreas’s oldest one: never released insulin without its twin.