Glycine: The 3-Gram Sleep Candidate
Glycine is a sweet-tasting amino acid your body makes by the gram daily, and it opens this audit as the tier's one candidate with human sleep trials behind it — small, from one company's laboratories, but placebo-controlled, in one case monitored overnight with electrodes. Here is what they measured, what they missed, and who should think twice before spooning a sweet powder into bedtime.
What the evidence supports
- Three grams before bed shortened the latency to sleep onset (p = 0.01) and to slow-wave sleep (p = 0.019) on overnight polysomnography in a randomized crossover trial (Yamadera et al., 2007).
- Both randomized crossover trials found better mornings: less fatigue, more liveliness, clearer-headedness (Inagawa et al., 2006).
- Daytime sleepiness suggested improvement, though significance was limited to one measure and time point; memory-recognition performance improved on the days after glycine nights (Yamadera et al., 2007).
What remains uncertain
- Every sleep trial came from Ajinomoto-affiliated teams — the company manufactures glycine — and no independent group has replicated the polysomnography findings.
- Eleven volunteers, all Japanese, with unsatisfactory sleep rather than diagnosed insomnia; generalizability is unknown.
- Sleep architecture was untouched, and no trial tested long-term use or any health outcome beyond the morning after.
Evidence last reviewed: September 16, 2026. Conclusions may change as new research is published.
the sweetest amino acid, audited
Product picks are generic categories, not brands. We may earn a commission on Amazon or iHerb purchases at no cost to you — this never changes our evidence conclusions. Full disclosure
Glycine powder
The sleep trials used 3 g of glycine before bed; plain powder in water is the closest retail equivalent.
⚠️ Sweet taste; GI upset is possible at high doses. A sleep aid for some, not a treatment for insomnia.
Check price on Amazon →Where the 3-Gram Story Came From
Glycine is the smallest of the twenty amino acids your body builds proteins from, and it wears three hats. It is the most abundant amino acid in collagen — the structural protein of skin, tendons, and bone. It is also a signaling molecule: an inhibitory neurotransmitter at its own receptors and a co-agonist at NMDA-type glutamate receptors. Your liver makes grams of it daily; diet adds more. Nothing about it screams "sleep aid" until you notice the inhibitory signaling and ask the obvious question.
The question was asked inside Japan's industrial amino-acid programs. Glycine is an Ajinomoto product — the company grew from MSG into a broad amino-acid manufacturer — and its Research Institute for Health Fundamentals ran the two sleep trials this page rests on: a subjective study (2006) and a polysomnography study (2007). A third small Ajinomoto-led study (2012) examined next-day function during partial sleep restriction and reported similar small effects. The Longevity Supplement Candidates audit carries the summary; this page goes deeper on the tier's one objective sleep claim.
What the Trials Actually Showed
The 2007 trial (Yamadera and colleagues, Sleep and Biological Rhythms) matters most because it measured sleep objectively. Eleven volunteers — eight women, three men, aged 30 to 57 — all with chronically unsatisfactory sleep (mean Pittsburgh Sleep Quality Index above 8) slept two monitored nights in a hospital lab, twice, a week or more apart, taking either 3 g of glycine or a flavor-matched placebo within an hour of lights-out. Randomized crossover, but single-blinded — worth knowing when you weigh the subjective scores.
On the monitored nights, glycine shortened the time to fall asleep — latency to stage-2 sleep dropped (p = 0.01), latency to slow-wave sleep too (p = 0.019). What did not change matters as much: REM latency held steady and the night's architecture — the share of each sleep stage — was untouched. Glycine moved the doorway, not the building. The next mornings, the St Mary's Hospital questionnaire showed higher satisfaction (p = 0.046), less difficulty falling asleep (p = 0.008), shorter subjective time to fall asleep (p = 0.002), and better sleep efficacy (p = 0.018). Through the day, sleepiness showed a suggested improvement — for that measure and time point only — and a memory-recognition task improved.
The 2006 trial (Inagawa and colleagues) skipped the electrodes and measured only the morning after: a randomized, double-blind crossover in volunteers with sleep complaints, 3 g or placebo, scored with the St Mary's questionnaire and a fatigue checklist. Fatigue fell; liveliness and clear-headedness rose. Two trials, same dose, same direction — that consistency is the honest core of glycine's case.
The Two Trials, Side by Side
Side by side — including the columns marketing summaries drop — the shape of this evidence base shows at a glance.
| Trial | Design & people | Dose | Objective (PSG) | Subjective mornings | Funding | Verdict |
|---|---|---|---|---|---|---|
| 🌙 Yamadera 2007 | Randomized single-blind crossover; 11 volunteers (8 women, 3 men, aged 30–57) with unsatisfactory sleep | 3 g within 1 h of bedtime | Sleep-onset and SWS latencies shortened (p = 0.01; p = 0.019); REM latency and architecture unchanged | Satisfaction, ease of onset, sleep efficacy up; daytime sleepiness suggested improvement (one measure, one time point) | Ajinomoto co-authors; company ethics board co-approved | Positive |
| 📋 Inagawa 2006 | Randomized double-blind crossover; volunteers with sleep complaints | 3 g or placebo before bed | None — no overnight monitoring | Fatigue, liveliness, clear-headedness improved | Ajinomoto-affiliated team | Positive |
| 🕳️ The gap | No independent polysomnography replication since | — | — | — | Independent labs: none | Missing |
The Mechanism: A Cooling Story
Why would a sweet amino acid hurry sleep along? The proposed pathway runs through body temperature — presented here as mechanism, a plausible story consistent with the findings, not a demonstrated cause. Sleep onset normally follows a small decline in core temperature, driven by widened blood vessels in hands and feet radiating heat from the core. That thermal doorway is part of the physiology in the Science of Sleep pillar.
In the same laboratories' follow-up work — in rats, not humans — oral glycine produced exactly that peripheral pattern: widened vessels at the heat-dissipation surfaces and a slight drop in core temperature — a pre-bed bolus, the interpretation goes, nudges the heat-dumping your body starts naturally on the way to sleep. A tidy, sensible account, with three caveats: the sleep trials did not measure temperature, the thermoregulation experiments are rodent studies, and whether this mechanism explains the human sleep findings is unproven. The inhibitory-neurotransmitter role may contribute too. A mechanism this plausible is exactly why independent replication matters — good stories are easy to tell in both directions.
Side Effects and Watch-Items
- 🍬 Sweet by nature — glycine is genuinely sweet; both trials masked it with a flavor-matched malt-sugar placebo. A 3 g scoop in water is a sweet pre-bed drink — short trials at this dose reported no serious adverse events, which is tolerability, not proof of long-term safety.
- 🚽 GI upset at higher doses — mild stomach complaints are the most common issue at larger servings. No serious adverse events were reported in these small, short trials at 3 g; the same group's high-dose tolerance testing also reported none — with the same funding asterisk. Long-term safety at 3 g nightly is not established.
- 💊 A clozapine interaction flag — glycine acts at inhibitory glycine receptors and NMDA receptors, and clozapine's pharmacology overlaps that glycinergic territory, and high-dose glycine has been studied clinically alongside antipsychotics; high-dose glycine trials alongside antipsychotics produced conflicting findings, and what that means at a 3 g bedtime dose is unresolved. Anyone on clozapine or other psychiatric medication should involve the prescriber before adding glycine.
- 💰 Funding transparency, stated plainly — both trials were authored by teams including Ajinomoto researchers, and the 2007 protocol was reviewed by the company's own ethics board alongside the hospital's. Ajinomoto sells glycine. None of this falsifies the data, but the entire human file originates from the party that profits from a positive answer — and no independent group has repeated the overnight measurements since.
- 🩺 Who asks a clinician first — people on psychiatric medication, especially clozapine; anyone with a treated sleep disorder, since a supplement that feels effective can displace real evaluation; and pregnancy or breastfeeding, where the data do not exist.
⚠️ When this becomes clinician territory
If you take clozapine or any medication that acts on the nervous system, adding a compound that modulates the same receptors is a decision for the person who prescribed it — and so is stopping anything. This page describes evidence; it does not advise starting, stopping, or changing any treatment.
Where This Leaves Glycine
Ranked within its tier, the glycine evidence remains limited: two small placebo-controlled crossover trials, one small company-led study of next-day function, and one objective overnight readout, all from the same research ecosystem. The asterisk is equally real: eleven volunteers, one research ecosystem, a 2007 paper, effects that shortened the doorway without rebuilding the night. Faster sleep onset is not a health outcome — nothing here shows glycine adds healthy years; no trial has tried.
The pragmatic reading: glycine is cheap, sweet, and well tolerated in these short trials — a reasonable low-friction candidate after the fundamentals that own most sleep variance. Light, caffeine timing, and alcohol matter more than any powder; the Sleep protocol and Caffeine, Alcohol & Blue Light pages own that evidence. What would raise the rating: independent replication, a larger sample, a trial in diagnosed insomnia. Until then, three stars is the ceiling.
Questions, Answered Briefly
- 🥄 How much, and when? Every trial used 3 g within the hour before bed — the dose with evidence behind it, stated as what researchers tested, not a recommendation.
- 🌙 Does it treat insomnia? Not tested — volunteers had unsatisfactory sleep, not a diagnosed disorder. That evaluation is medical; When Sleep Won't Come covers the honest first steps.
- 🍲 Can I get it from food? Collagen-rich foods — gelatin, bone broth, tough cuts — deliver glycine, and your body synthesizes grams daily. No trial has tested food sources against sleep; that claim is an extrapolation.
- 🧩 Can I stack it with magnesium or melatonin? No trial has tested glycine combined with anything — combination claims are marketing inferences.
The Bottom Line
- A small objective sleep signal — two placebo-controlled crossover trials, one with overnight monitoring, pointing the same modest direction: faster onset, better mornings at 3 g before bed.
- Small, funded, and dated — eleven monitored volunteers, an author team intertwined with the manufacturer, papers from 2006–2007, no independent replication since.
- Know what it doesn't do — sleep architecture, REM latency, and total sleep structure were untouched; a shorter doorway into sleep is not deeper sleep, and neither is a health outcome.
- Low-risk in short trials, not risk-free — sweet taste, GI upset at high doses, and an unresolved clozapine interaction flag mean psychiatric medication users involve their prescriber first; everyone else gets an honest candidate, no promises.
Related Topics
- Yamadera W. et al., "Glycine ingestion improves subjective sleep quality in human volunteers, correlating with polysomnographic changes," Sleep and Biological Rhythms 5(2):126–131 (2007)
- Inagawa K. et al., "Subjective effects of glycine ingestion before bedtime on sleep quality," Sleep and Biological Rhythms 4(1):75–77 (2006)
- Bannai M., Kawai N., Ono K., Nakahara K., Murakami N., "The effects of glycine on subjective daytime performance in partially sleep restricted healthy volunteers," Frontiers in Neurology 3:61 (2012)