A drug in search of a hormone
The usual order is hormone first, drug second. Someone finds a signal the body makes, works out what it does, and chemists then build something that imitates or blocks it. Insulin, GLP-1 and oxytocin all went that way. Ghrelin went the other way round, and the detour took fifteen years.
It started with a small synthetic peptide. Working from fragments of the opioid peptide enkephalin, Cyril Bowers and colleagues arrived at a six-residue molecule that made the pituitary release growth hormone, and did so specifically, without touching the other pituitary hormones (Bowers et al., 1984). It was not a copy of anything known. The body’s own growth-hormone-releasing hormone had been isolated only two years earlier, and this peptide did not work through its receptor. Whatever it was pressing, nobody had a name for it.
Drug companies followed the pharmacology anyway and built a family of growth hormone secretagogues, including the oral compound MK-677. In 1996 a team at Merck used one of them as bait and cloned the receptor it bound: a G-protein-coupled receptor in the pituitary and hypothalamus with no known natural activator (Howard et al., Science, 1996). An orphan. The field now had drugs and a receptor, and a hormone-shaped hole between them.
Found in the stomach
Everyone expected the missing hormone to come from the brain, where the receptor was. Masayasu Kojima and Kenji Kangawa found it in the stomach. Screening tissue extracts for anything that would activate the orphan receptor, they purified a 28-amino-acid peptide from rat stomach and named it ghrelin, from an ancient root meaning “to grow” (Kojima et al., Nature, 1999).
A growth hormone signal made by the gut was strange enough. The peptide’s structure was stranger. The purified hormone carried an eight-carbon fatty acid, octanoate, bolted onto its third amino acid, and without that fatty acid it did nothing at the receptor. A synthetic copy of the bare sequence was inert. Ghrelin remains the only hormone known to need this particular modification, and that oddity is why the search had been so hard: you cannot find a peptide by its sequence alone when the sequence alone does not work.
The signal that says eat
Within a year the story changed subject. Given to rodents, ghrelin made them fatter, and not through growth hormone: it increased food intake and shifted the body toward storing fat (Tschöp et al., Nature, 2000). In people the effect was immediate. Nine healthy volunteers given an intravenous infusion of ghrelin ate 28% more from a buffet than they did on saline, and every one of them ate more (Wren et al., 2001). The authors called it the first circulating hormone shown to stimulate eating in humans.
Its natural rhythm fit the role. Ghrelin in the blood climbs before each meal and drops soon after (Cummings et al., 2001). Almost every other signal in the gut’s report to the brain says the same thing in different words: stop, there is food here. GLP-1, PYY and CCK are all brakes. Ghrelin is the one clear voice on the other side, rising when the stomach is empty.
It also explained something dieters already knew. In a study of people who lost 17% of their body weight by dieting, the day-long ghrelin profile rose by 24%. In people who had lost even more weight after gastric bypass surgery, it was 77% lower than in normal-weight controls, and the mealtime peaks had vanished (Cummings et al., NEJM, 2002). Dieting turned the hunger signal up. Surgery, somehow, had turned it off.
A hormone that makes people eat, that rises when they diet, and that falls after the most effective obesity treatment then known. Block it, the reasoning went, and you have a weight-loss drug.
Why blocking it never worked
The reasoning had two problems, and both were visible early.
The first is that common obesity is not a high-ghrelin state. Fasting ghrelin is lower in people with obesity than in lean people (Tschöp et al., 2001). The body had already turned the signal down, and people were hungry anyway. The striking exception is Prader-Willi syndrome, a genetic condition defined by relentless hunger, in which ghrelin runs several-fold high (Cummings et al., 2002). That is a real lead for one rare disease and no help for the common one.
The second problem is that animals do not need ghrelin to eat. When Roy Smith’s group deleted the ghrelin gene in mice, the animals were, in the paper’s words, not anorexic dwarfs. They grew normally, ate normally and got fat on a high-fat diet just like their littermates. The authors drew the conclusion in their abstract: ghrelin antagonists were unlikely to be broadly useful against obesity (Sun et al., 2003). Appetite is held up by many overlapping signals. Take one away and the others carry on.
Two decades later that prediction has held. No ghrelin blocker has been approved for obesity, while drugs that amplify the brake signals have reshaped the field. This is the pattern laid out in mechanism is not proof: a signal can do exactly what the experiments say when you add it, and still not be the thing holding the system up when you remove it.
The fat on the hormone
The fatty acid took another nine years to explain. In 2008 two groups independently identified the enzyme that attaches it, a previously uncharacterized member of a family of membrane-bound acyltransferases, and named it ghrelin O-acyltransferase, or GOAT (Yang et al., Cell, 2008; Gutierrez et al., PNAS, 2008). As far as anyone has found, ghrelin is the only thing it works on: one enzyme, one substrate.
The raw material turned out to come partly from dinner. When mice were fed medium-chain fats, the fatty acids on their newly made ghrelin matched the chain length of what they had eaten, down to an unnatural seven-carbon version when they were fed a seven-carbon fat (Nishi et al., 2005). The hunger hormone is assembled, in part, from the food it asks for.
The enzyme also gave the clearest answer yet to what ghrelin is for. Mice lacking GOAT, and so lacking all active ghrelin, looked normal on ordinary or high-fat food. Put on severe calorie restriction, they could not hold their blood sugar: by day seven it had collapsed and the animals were near death, while normal mice on the same rations stayed active. Infusing either ghrelin or growth hormone rescued them (Zhao et al., PNAS, 2010). On that reading the two halves of ghrelin’s history are one job. It raises growth hormone and drives hunger for the same reason: to keep an animal alive through famine.
Ghrelin looks less like a hunger switch than a starvation defense. That is why removing it changes little when food is plentiful, and why it made a poor target for a disease of plenty.
The body's own blocker, and the drugs that did arrive
The receptor had one more surprise. In 2018 a liver- and gut-derived peptide called LEAP2, filed away years earlier as a minor antimicrobial, was shown to be a natural antagonist of the ghrelin receptor. It rises with feeding and falls with fasting, the mirror image of ghrelin (Ge et al., Cell Metab, 2018). Infused into healthy men, it lowered food intake at a free meal and blunted the rise in blood sugar afterwards (Hagemann et al., 2022). So the body runs the receptor with two hormones, an accelerator from the stomach and a brake from the liver, and what matters may be their ratio. Whether that can be turned into a therapy is an open, early question.
The drugs that did reach patients push the receptor the original way. Macimorelin, an oral ghrelin-receptor agonist, is approved as a diagnostic: a single dose provokes a growth hormone surge, and a weak response identifies adult growth hormone deficiency about as well as the older, riskier insulin test (Garcia et al., 2018; FDA label). Anamorelin went after appetite. In two phase 3 trials in people with lung cancer and cachexia it increased lean body mass over twelve weeks but did not improve handgrip strength, the other primary endpoint (Temel et al., Lancet Oncol, 2016). It has been approved in Japan. More mass without more strength is a result this catalog has met before.
And the secretagogues that started it all are still here. Ipamorelin and its relatives are descendants of that 1984 hexapeptide, pressing the same receptor for its growth hormone effect. How they compare with the GHRH route is the subject of two ways to ask for growth. It is worth remembering, reading about them, that the receptor they act on is the hunger receptor too.
What the detour teaches
Ghrelin’s history is a run of reversed expectations. The drug came before the hormone. The growth signal came from the stomach. The peptide needed a fat. The hunger hormone was low in obesity. And the obvious drug, a blocker, was the one that never worked.
- Discovered backwards. Synthetic growth hormone secretagogues (1984) led to an orphan receptor (1996), which led to the hormone itself (1999). The pharmacology was fifteen years ahead of the physiology.
- A peptide that needs a fatty acid. Ghrelin is inactive until the enzyme GOAT attaches an eight-carbon fatty acid, some of it taken straight from dietary fat. No other hormone is known to require it.
- The one clear hunger signal. Ghrelin rises before meals, rises with dieting, and makes people eat more when infused. Most other gut hormones tell the brain to stop.
- Not the cause of obesity. Ghrelin is low in common obesity, and mice without it eat and grow normally. Its essential role appears to be protecting blood sugar during starvation.
- The useful drugs activate it. A ghrelin-receptor agonist is an approved diagnostic for growth hormone deficiency, and another adds lean mass in cancer cachexia. Blocking the receptor has produced no approved drug.
Keep going
- Ghrelin reference (receptor, source, mechanism)
- Two ways to ask for growth — the secretagogues as growth hormone drugs
- The gut's full report — the satiety signals on the other side
- The hormone discovered twice — leptin, the long-term counterpart
- The gut & appetite family
- Ask the research agent about ghrelin
Educational reference on mechanism, summarized from public scientific literature and clinical-trial disclosures and simplified in places. Not individualized medical advice or a prescription. Where a dose or regimen appears it is reported from its source as research literacy — so you can read the evidence yourself — not an instruction to dose. Specific compounds and trials are named to explain the science; verify any claim against the linked primary sources.