Glycine

What glycine is

Glycine is the simplest proteinogenic amino acid and the only one without a side chain. Exactly this makes it irreplaceable in collagen: in the triple helix, a glycine sits at every third position, because no larger residue fits there. This makes it the most abundant amino acid in collagen by quantity.

Glycine is nevertheless not a bottleneck substance. The body makes around 45 grams itself every day, and food adds 3 to 5 grams. A serving of 3 grams roughly doubles the intake from food and remains a small addition compared with the body's own production.

Three sleep studies, and what they measured

In 2006, Inagawa and colleagues gave 19 women who were dissatisfied with their sleep 3 grams of glycine or placebo for 4 days, 1 hour before bedtime, in a randomized crossover. Measurement was by two questionnaires: less fatigue, more freshness, a clearer head after waking up — all subjective.

In 2007 came the only study to date with polysomnography, in 2012 one on the day after too little sleep; both are described in detail further below. That is all. A systematic review that went through 47 articles with 50 studies across 11 organ systems classifies all three as works with small samples and a high risk of bias.

The temperature trick and the wrong receptor

The explanation that is almost always supplied: glycine lowers core body temperature, and because sleep onset is physiologically accompanied by a drop in temperature, you fall asleep faster. The mechanism behind this was cleanly elucidated in 2015 — in rats.

There, oral administration shortened the time to non-REM sleep and lowered core temperature, while blood flow in the skin of the sole of the foot rose in a dose-dependent manner and dissipated heat. What is decisive is what switching off individual components showed: NMDA receptor antagonists blocked the effect, the glycine receptor antagonist strychnine did not; after destruction of the suprachiasmatic nucleus it had disappeared entirely. The pathway therefore runs through the excitatory co-agonist site on the NMDA receptor, not through the inhibitory glycine receptor effect. The common formula that glycine calms like GABA and therefore makes you sleepy describes a real property of the amino acid — just not the one measured here.

In humans, the drop in temperature has not been shown so far. The only study that measured core temperature after a glycine-rich dose in the evening used 15 grams of collagen peptides in 13 men over 7 nights. It found fewer awakening events in polysomnography (21.3 versus 29.3; p = 0.028) — and no difference in core temperature.

Glutathione, GlyNAC and the largest study

The second argument for glycine is: it is one of the three building blocks of glutathione. This gave rise to the combination GlyNAC, glycine plus N-acetylcysteine. One research group has presented a series of small studies on it, most recently a randomized one in 24 older adults, 12 versus 12, over 16 weeks, with broad improvements. In 2022, an independent group designed a larger test — and missed the relevant primary endpoint, see below.

This fits with a study in 19 people with severe obesity: 100 milligrams per kilogram of body weight over 2 weeks clearly raised the glycine level in the blood and measurably increased the excretion of glycine-conjugated metabolites. So the detoxification pathway worked. Glutathione remained unchanged. The conjugation pathway can evidently be fed, the glutathione level not simply like that.

Collagen, lifespan and the rest

For collagen, the chain ends early. That glycine is the main building block is established. That more glycine leads to more collagen rests mainly on a cell culture experiment on bovine chondrocytes: from 1.0 millimoles per liter, type II collagen synthesis rose by 60 to 75 percent, but far above the concentrations usual in the body. There is no human study on this.

From aging research comes an animal finding that is taken seriously: in the Interventions Testing Program, a feed with 8 percent glycine extended the lifespan of genetically heterogeneous mice by 4 to 6 percent, in males (p = 0.002) as well as in females (p < 0.001). A solid mouse result, not a human finding.

Glycine has its best data where nobody sells it as a biohack: as an add-on to antipsychotics, in entirely different amounts. A meta-analysis of 40 placebo-controlled studies with 4,937 patients on NMDA receptor modulators, including glycine, D-serine and sarcosine, found improvements there mainly in negative symptoms.

What the studies show

The only polysomnography on glycine

Randomized crossover study in volunteers with persistently unsatisfactory sleep: 11 participants each for glycine and placebo, mean age 40.5 years, range 30 to 57, predominantly women, 2 consecutive nights, 3 grams 1 hour before bedtime versus a taste-matched placebo. Result: shortened time to fall asleep, shortened latency to deep sleep, higher sleep efficiency, unchanged sleep architecture, plus better subjective sleep quality. Exact effect sizes are not in the freely accessible part of the paper.

The day after too little sleep

Randomized, single-blind crossover study: 10 healthy men aged 30 to 61 enrolled, 7 evaluated. Over 3 nights, time in bed was cut by 25 percent, from an average of 7.3 to 5.5 hours, plus 3 grams of glycine or placebo 30 minutes before bedtime. On day 1, fatigue was significantly lower (p = 0.022), sleepiness was not (p = 0.098); in the psychomotor vigilance test, reaction time was shorter on day 1 and day 3, and on day 3 the difference in fatigue had disappeared. All participants worked for the manufacturer that conducted the study.

The glutathione test that went wrong

114 healthy older adults with a mean age of 65 received glycine plus N-acetylcysteine in a 1:1 ratio for 2 weeks in daily doses of 2.4, 4.8 or 7.2 grams, or placebo; 20 young participants provided reference values. The primary endpoint, total glutathione and redox ratio, was missed: 903.5 versus 959.6 milligrams per liter (p = 0.278) and 12.49 versus 12.65 (p = 0.739). That the older adults had more oxidative stress was clearly confirmed by the study. Only in a subgroup formed post hoc did glutathione rise (p = 0.016).

Where the data stop

On sleep there are exactly three human studies, together fewer than 50 evaluated people, over 2 to 4 nights. The systematic review notes that of all 50 included studies only two had a low risk of bias, both in sick people, not a single one in healthy people. A scoping review that searched up to July 9, 2026 finds the same 3 papers and calls for larger studies with uniform sleep endpoints. So the state of things is not that the effect has been refuted — it has never been properly tested.

Status, approval and legal

In the EU, glycine is authorized as food additive E 640, under the name glycine and its sodium salt. There is no authorized health claim: glycine does not appear in the consolidated text of Regulation (EU) No 432/2012, neither for sleep nor for collagen nor for detoxification. For amino acids in food supplements, Germany has no general regulation on maximum amounts. A general administrative order of the German Federal Office of Consumer Protection and Food Safety (BVL) of March 28, 2003 on the import of such products lists a maximum amount for glycine of 51 milligrams per recommended daily serving — not a national limit, but an indication of the order of magnitude at the time. This page makes no statement on doping relevance, because the prohibited list could not be accessed for this research.

Safety

At the study doses, glycine was well tolerated; the study with 114 participants explicitly calls the administration safe and well tolerated. There is no official upper limit from EFSA or BfR. A review of tolerable amino acid amounts gives a NOAEL for humans of 129.0 milligrams per kilogram of body weight per day with a usual total intake of 3.21 grams daily; at a factor of 2.8, the margin is narrower than in the rat, where it is 9.9. Controlled data in healthy people extend only to 14 days; over months, glycine has been given in controlled settings only to sick people, for example 15 grams daily over 3 months in metabolic syndrome. Two limitations are documented: a single dose of 0.8 grams per kilogram of body weight impaired sensorimotor gating in healthy men 45 minutes after administration. And the added benefit in psychiatry was seen only in patients not taking clozapine — anyone taking clozapine should not use larger amounts of glycine without consulting a physician.

Sources

• Soh et al., GeroScience 2024 (systematic review, 50 studies)
• Fan, Nelson and Wallace, Journal of Dietary Supplements 2026 (scoping review)
• Yamadera et al., Sleep and Biological Rhythms 2007 (polysomnography)
• Inagawa et al., Sleep and Biological Rhythms 2006
• Bannai et al., Frontiers in Neurology 2012 (sleep restriction)
• Kawai et al., Neuropsychopharmacology 2015 (mechanism, rat)
• Lizzo et al., Frontiers in Aging 2022 (114 participants, GlyNAC)
• Tan et al., Scientific Reports 2025 (glycine in severe obesity)
• Blachier et al., Amino Acids 2021 (tolerable amounts, NOAEL)
• Regulation (EU) No 432/2012, consolidated text (authorized claims)