A team led by Russel Reiter, the researcher who discovered melatonin’s antioxidant properties in 1993, has assembled a 217-reference defense of two linked claims: that melatonin directly neutralizes free radicals without needing a cell-surface receptor, and that most of the body’s melatonin is manufactured inside mitochondria rather than delivered from the pineal gland. The review pulls together spin-resonance spectroscopy, pulse radiolysis, quantum-chemical calculations, and cell studies to argue that melatonin sits inside the few-angstrom “damage radius” where destructive radicals are born and die. It also advances two newer ideas: that melatonin controls the physical state of mitochondrial protein droplets that govern mitophagy, and that it stabilizes cholesterol-rich membrane microdomains. This is a synthesis and a rebuttal, not a discovery.
For three decades a quiet argument has run through melatonin research. Everyone agrees melatonin regulates sleep and seasonal rhythms. What splits the field is whether melatonin’s much-advertised antioxidant power is real biology or a laboratory artifact.
The skeptical position, argued forcefully by Jean Boutin and colleagues, runs like this. Melatonin circulates in blood at concentrations of a few hundred picograms per millilitre, roughly a thousandth of the amount needed to quench radicals in a test tube. Anything melatonin appears to do to oxidative stress must therefore be indirect, mediated by its membrane receptors switching on the cell’s own antioxidant enzymes.
Reiter’s team answers with a geographical argument. Blood concentration is the wrong measurement. Free radicals such as the hydroxyl radical survive for nanoseconds and travel only a few angstroms before hitting something. A scavenger that is not already standing at the point of impact is irrelevant no matter how much of it circulates. The relevant question is not how much melatonin is in blood but how much is at the electron transport chain, and the answer the group reports is that mitochondria hold roughly thirty times more melatonin than cytosol and that this pool does not rise and fall with the pineal rhythm. Remove the pineal gland surgically and mitochondrial melatonin is unchanged. Knock out the melatonin-making enzyme inside neuronal mitochondria and superoxide accumulates.
If that picture is right, melatonin is not a hormone that happens to be an antioxidant. It is an ancient bacterial molecule that mitochondria inherited when they were engulfed two billion years ago, retained for local damage control, and that vertebrates later repurposed for timekeeping.
The review adds a genuinely novel layer. Inside the mitochondrial matrix, proteins organize into liquid droplets that must stay fluid to recruit the machinery that clears damaged mitochondria. Oxidative stress thickens that fluid until the droplets solidify and quality control stalls. Melatonin, the authors argue, keeps the matrix in the fluid window.
The catch is that almost none of this has been measured in a living human.
Actionable Insights
The paper is chemistry and cell biology, not a trial, so nothing here proves a health benefit. Here is what the numbers actually say.
In head-to-head test-tube comparisons melatonin beat the standard antioxidants on a per-molecule basis. Against the ABTS radical, the concentration needed to neutralize half the radicals was 5 micromolar for melatonin versus 11 for glutathione and 15.5 for vitamin C and Trolox, a vitamin E analogue. That is roughly 2 to 3 times more potent per molecule. Against the hydroxyl radical the gap was wider: 21 micromolar for melatonin versus 123 for glutathione, about 6-fold. Melatonin also neutralized 2 to 4 radicals per molecule while the others managed one or fewer, because its breakdown products are themselves scavengers.
Dose-wise, the authors report that animal-derived data scale to a human-equivalent dose of 1.0 to 1.5 mg per kilogram, meaning 70 to 105 mg daily for a 70 kg adult. This is 25 to 100 times a typical sleep dose. They cite 1000 mg daily for 30 days without observed toxicity. Oral bioavailability is only 10 to 30 percent and varies widely with CYP1A2 activity.
Treat the high-dose antioxidant framing as an untested hypothesis, not a protocol.
Context and Source
- Open Access Paper: Melatonin Targets Mitochondrial Redox Homeostasis: Optimizing the Intracellular Microenvironment
- Article type: Narrative review (labelled “Review” by the journal)
- Lead institution: Department of Cell Systems and Anatomy, UT Health San Antonio Long School of Medicine, Texas, USA
- Collaborating institutions: Baptist University College of Osteopathic Medicine (USA); Sao Paulo State University UNESP and FAMERP (Brazil); Universidad Autonoma Metropolitana-Iztapalapa (Mexico); IMBECU-CONICET, Mendoza (Argentina)
- Journal: International Journal of Molecular Sciences, MDPI, Basel, Switzerland
- Journal Impact Evaluation The impact score of this journal is 5.6 (2025 Journal Citation Reports, released June 2026; CiteScore 10.0), evaluated against a typical high-end range of 0 to 60+ for top general science, therefore this is a Medium impact journal.