Melatonin discovery — how a frog skin extract became the world's most-taken sleep aid

What followed was one of the great feats of chemical patience in twentieth-century biology. Lerner and his team extracted and processed approximately 250,000 bovine pineal glands — cattle pineal tissue, sourced from slaughterhouses, processed in bulk — looking for a substance present in such vanishingly small quantities that it took the glands of a quarter million cows to yield enough to characterize. In 1958, they announced they had found it. They named it melatonin, from melanocyte (the pigment cell) and tonin (from serotonin, its biochemical precursor). The name described what they thought it did: contract melanocytes, lighten skin.

That turned out to be a minor footnote.

The compound Lerner isolated did indeed lighten frog skin. In amphibians, melatonin causes melanosomes — the pigment-containing organelles inside melanocytes — to aggregate toward the cell center, producing a lightening effect. It was a real finding. But in humans, it turns out melatonin has essentially no effect on skin pigmentation. The whole premise of the search, the thing that had motivated the extraction of a quarter million pineal glands, was a biological dead end in the species Lerner cared about. What he had found instead, though he wouldn't know it for years, was one of the body's central timekeeping molecules.

The realization came slowly, through the work of researchers who kept asking what melatonin was actually doing if it wasn't lightening human skin. The answer began to emerge in the 1960s and 1970s, as studies in mammals — first in rodents, then in larger animals — revealed a striking pattern. Melatonin levels in the blood were not constant. They rose sharply at night and fell during the day, in a pattern that tracked precisely with darkness and light. This was not a trivial observation. It meant the pineal gland was doing something that no other endocrine organ had been found to do: it was measuring time. It was converting the light-dark cycle into a hormonal signal that the rest of the body could read.

The mechanism, worked out over subsequent decades, is elegant in its directness. Light entering the retina travels along a dedicated neural pathway — the retinohypothalamic tract — to the suprachiasmatic nucleus (SCN), a small paired structure in the hypothalamus that functions as the body's master circadian clock. The SCN projects, via a multi-synaptic pathway, to the pineal gland. In darkness, the SCN releases its inhibition of the pineal, and the pineal synthesizes melatonin from serotonin via two enzymatic steps, releasing it into the bloodstream. Light suppresses the pathway and shuts melatonin production off. The result is that melatonin blood levels are high at night — typically peaking somewhere between 2 and 4 a.m. in most adults — and near-undetectable during daylight hours. The body doesn't just experience darkness. It chemically encodes it.

This made melatonin fundamentally different from other hormones. Cortisol fluctuates. Insulin responds to meals. Growth hormone pulses during sleep. But melatonin's rhythm is uniquely tied to the light environment, not to behavior or metabolic state. It is, researchers came to understand, a signal of darkness rather than a signal of sleep. The two often coincide — humans evolved sleeping in the dark — but they are not the same thing. This distinction turns out to be critical for understanding both what melatonin does and where the consumer market for it has gone wrong.

By the 1980s, the circadian role of melatonin was reasonably well characterized in animal models. The question was what it did in humans. A series of studies established that blind individuals — particularly those with no light perception — often had disrupted circadian rhythms, and that exogenous melatonin at appropriately timed doses could help entrain their sleep-wake cycles. People living in extreme northern latitudes, where winter darkness was prolonged, showed melatonin profiles that differed from those of people in temperate zones. Shift workers and jet-lag subjects showed the kind of melatonin phase-misalignment that predicted their subjective dysfunction. The picture that assembled itself was of melatonin as a pacemaker signal — not a sedative, not a sleep-inducing drug, but a timing molecule whose presence told the body when it was night.

The consumer story diverges sharply from the science at this point, and the divergence traces largely to a single publication.

In 1995, Russel Reiter — a prolific researcher at the University of Texas Health Science Center who had spent years studying melatonin's antioxidant properties — published a popular book arguing that melatonin was not merely a circadian hormone but a master anti-aging molecule. The antioxidant research was real: melatonin does have hydroxyl radical-scavenging properties, and in cell culture studies it had shown protective effects against oxidative damage. Reiter's claims extended this into assertions about aging, cancer prevention, and longevity that substantially outran the evidence available at the time. The book became a bestseller. Melatonin, which had been available in the United States as a dietary supplement since the mid-1990s, flew off the shelves.

The timing was also convenient for market reasons that had little to do with science. In 1994, the Dietary Supplement Health and Education Act had significantly loosened the regulatory framework for supplements in the United States, allowing melatonin to be sold without FDA approval as long as no direct disease claims were made. The combination of a charismatic longevity narrative, a permissive regulatory environment, and genuine scientific interest in circadian biology produced a market explosion. Melatonin became, and remains, one of the most widely sold supplements in the world. In the United States alone, consumer spending on melatonin supplements now runs into the hundreds of millions of dollars annually. It is taken by roughly one in five Americans who report any supplement use.

Here is where the gap between physiology and practice becomes striking. The physiological nighttime melatonin peak in healthy young adults is approximately 100 to 300 picograms per milliliter of blood — a range corresponding to roughly 0.1 to 0.3 milligrams of circulating hormone. Consumer melatonin supplements are almost universally sold in doses of 5, 10, or even 20 milligrams. Standard dosing in the supplement market delivers approximately thirty to one hundred times the body's normal nighttime concentration.

This is not a minor discrepancy. The research on what low-dose, precisely timed melatonin can do is actually reasonably supportive — specifically for jet lag (where 0.5 to 3 mg taken at the destination bedtime appears to accelerate circadian re-entrainment), for shift workers (where timed administration may support daytime sleep), and for individuals with non-24-hour sleep-wake disorder, the condition affecting most totally blind people in which the internal clock runs on a cycle slightly longer than 24 hours. For these applications, low-dose timed melatonin addresses a genuine circadian mismatch by providing the darkness signal the body is missing or misreading. In 2014, tasimelteon — a melatonin receptor agonist — received FDA approval specifically for non-24-hour sleep-wake disorder in the blind, a regulatory milestone that reflects the strength of evidence for this narrow indication.

The supraphysiological doses in most consumer products are a different matter. At 5 or 10 mg, melatonin does cause sedation — not through its circadian signaling function, but through off-target effects that essentially drug the system. The compound binds not only its dedicated MT1 and MT2 receptors in the SCN and elsewhere but begins acting on a broader range of receptor systems when present at extraordinary concentrations. The sedation is real. Whether it constitutes healthy sleep architecture is a separate question, and the evidence is not reassuring. Some research suggests that supraphysiological doses may actually suppress the body's own melatonin production over time, through the receptor downregulation that follows sustained receptor overstimulation. The person taking 10 mg of melatonin every night to sleep may be systematically blunting the signaling system they're trying to support.

There are also off-target effects at high doses that the consumer market rarely discusses. Melatonin influences reproductive hormone signaling — in some animal models it suppresses LH and FSH release — and there has been concern, though not definitive human evidence, that chronic high-dose use in children and adolescents might affect pubertal timing. Gastrointestinal effects, morning grogginess, and disrupted dreams are commonly reported at high doses. The pharmacokinetics of oral melatonin are notably variable between individuals: absorption can differ by a factor of ten or more, meaning the same 5 mg pill may produce blood levels in one person that are ten times what it produces in another.

The research picture for melatonin is thus genuinely split. There is good evidence, largely from controlled trials, that low-dose timed melatonin may help support circadian re-entrainment in specific contexts: jet lag, shift work adaptation, and circadian rhythm disorders in blind individuals. Tasimelteon, the FDA-approved synthetic melatonin receptor agonist, validates the receptor target for the non-24-hour sleep-wake indication. The evidence for using high-dose melatonin as a general sleep aid — which is what most consumer use amounts to — is considerably weaker, and the mechanistic case for it actually runs against the physiology.

What the melatonin story reveals is a pattern that repeats throughout supplement history. A compound with a real, specific, well-characterized function in the body gets identified. Researchers find that in targeted applications it does something useful. That finding reaches the public through a popular translation — sometimes accurate, sometimes not — that strips away the dose-dependence, the timing-specificity, and the context-dependence that made the original finding meaningful. The supplement industry supplies a product at whatever dose is inexpensive to manufacture and easy for consumers to accept. The resulting consumer behavior has almost nothing in common with the clinical research, which was done at doses orders of magnitude lower, with precise timing, in people with a specific physiological problem.

Lerner never intended to create a sleep aid. He was looking for a skin-lightening molecule and found a circadian timer in a quarter million cows' worth of pineal glands. The molecular biology turned out to be exquisite — a tiny peptide that converts darkness into hormonal signal with extraordinary precision. The consumer product that followed, decades later, uses that molecule in doses and patterns that bear no relationship to the biology Lerner uncovered. The gap between the discovery and the product is, in a sense, a measure of how far commerce can drift from the science it claims to be based on.

What melatonin actually teaches is not that supplementation is wrong but that dose and timing are the entire argument. A molecule that signals darkness at physiological concentrations, precisely timed, may help support the body's circadian system when that system has been disrupted. The same molecule at thirty times that concentration, taken casually at bedtime for general insomnia, is a different intervention with a different mechanism and a thinner evidence base than most people taking it realize. The frog skin research pointed at something real. Whether what ended up on the shelf has much to do with that something is a question worth asking.