Urolithin A is a small molecule that gut bacteria make from ellagitannins and ellagic acid, polyphenols found in pomegranates, nuts and berries. Not everyone’s gut bacteria can make it. A synthetic version, sold under the brand name Mitopure by the Swiss company Amazentis, is marketed as a “mitophagy activator” for muscle and healthy aging.
Mitophagy is the process by which cells break down and recycle damaged mitochondria (the parts of the cell that produce energy). Urolithin A triggers mitophagy in cell and animal studies. In people, several small randomized trials have been run. The main result to know: in the three best-known human trials, the primary outcome was either safety (the first trial) or a functional measure that did not improve significantly versus placebo (the next two). The positive findings come from secondary outcomes and biomarkers. Most of the human research was funded by the manufacturer.
This page summarizes what the clinical trials, safety studies and the FDA GRAS record report. It is for education only and is not medical advice.
| Property | Detail |
|---|---|
| Class | Gut-microbial metabolite of ellagitannins and ellagic acid; sold as a “postbiotic” |
| Chemistry | 3,8-dihydroxybenzo[c]chromen-6-one; C₁₃H₈O₄, 228.20 g/mol |
| Main product | Mitopure (Amazentis SA), made by chemical synthesis |
| Route | Oral (softgels, powders, foods) |
| Doses in human trials | Single doses of 250–2,000 mg; 250–1,000 mg/day for 4 weeks to 4 months |
| Half-life | About 1 day in blood (first-in-human study); peak about 6 h after a dose |
| Key mechanism fact | Induces mitophagy in cells, worms and rodents; human evidence is limited to biomarkers |
| US status | Not an approved drug. GRAS notice GRN 791 for food use: FDA had “no questions” (December 20, 2018) |
| WADA | Not named on the 2026 Prohibited List |
What it is
Urolithin A is a dibenzopyranone, a three-ring compound with two hydroxyl groups (PubChem). The commercial ingredient is chemically identical to the molecule the gut produces. Amazentis makes it by synthesis from resorcinol and a brominated benzoic acid, to a specification of at least 97% purity (FDA GRN 791 response letter).
A product of gut bacteria
Gut bacteria convert dietary ellagitannins and ellagic acid into urolithin A in several steps (Pidgeon et al., Nat Commun 2025). The same 2025 study identified a gene cluster in Enterocloster gut bacteria that performs one of the last steps, the conversion of urolithin C to urolithin A. In samples of human stool tested in the lab, only those in which this gene cluster was active produced urolithin A, which may explain why people differ (Pidgeon et al., Nat Commun 2025).
People fall into three “urolithin metabotypes” (metabolic types) after eating ellagitannin-rich foods (Tomás-Barberán et al., J Agric Food Chem 2014):
| Metabotype | What they produce | Share of volunteers across trials |
|---|---|---|
| A | Urolithin A only | 25–80% |
| B | Urolithin A plus isourolithin A and/or urolithin B | 10–50% |
| 0 | No detectable urolithins | 5–25% |
These types were seen regardless of age, sex, BMI or food source. Metabotype B was more common in people with metabolic syndrome or colorectal cancer.
A manufacturer-sponsored trial in 100 healthy adults measured this more directly (Singh et al., Eur J Clin Nutr 2022):
- only 12% had detectable urolithin A in blood at baseline;
- after one glass of pomegranate juice, about 40% converted significant amounts of the precursors into urolithin A;
- 500 mg of urolithin A taken directly gave more than 6 times the exposure of the pomegranate juice.
Producers had higher gut-microbiome diversity than non-producers. Varying producer status is the main argument made for taking urolithin A directly rather than relying on diet.
How it works: mitophagy
The mitophagy findings come almost entirely from cell and animal work:
- In worms (C. elegans). Urolithin A induced mitophagy, prevented the build-up of dysfunctional mitochondria with age and extended lifespan (Ryu et al., Nat Med 2016).
- In rodents. The same study found better exercise capacity in two mouse models of age-related muscle decline and in young rats.
- In mice, immune cells. Urolithin A expanded T memory stem cells and naive T cells, according to the authors of the later human immune trial (Denk et al., Nat Aging 2025).
- Other pathways. In laboratory studies, urolithin A also triggers aryl hydrocarbon receptor signaling, which affects immune function and the gut barrier (Pidgeon et al., Nat Commun 2025).
In people, the evidence consists of biomarkers measured in muscle biopsies. In the first human trial, 4 weeks of 500 or 1,000 mg/day changed mitochondrial gene expression in muscle and lowered plasma acylcarnitines (fat-burning byproducts that rise when mitochondria work less efficiently) (Andreux et al., Nat Metab 2019). In the ATLAS trial, 500 mg/day raised phospho-Parkin, a marker of active PINK1/Parkin mitophagy, in muscle biopsies from 6 participants. The authors noted that the gene-expression and protein pathways changed by urolithin A did not overlap, and that some changes appeared at only one of the two doses (Singh et al., Cell Rep Med 2022). These are markers of a mechanism, not proof that mitophagy causes any functional benefit in humans.
Urolithin A is often discussed alongside NAD+ precursors, which target mitochondria by a different route. See NAD+, NMN and NR (nicotinamide riboside).
Pharmacokinetics
| Parameter | Finding | Source |
|---|---|---|
| Bioavailability in people | Detected in plasma at every dose tested (250–2,000 mg) | Andreux 2019 |
| Absorption in rats | Under 2.2% of a radiolabeled dose absorbed | FDA GRN 791 letter |
| Main circulating forms | Urolithin A glucuronide, then urolithin A sulfate and the parent compound | Heilman 2017; Singh 2022, EJCN |
| Time to peak | About 6 h | Singh 2022, EJCN, citing the first-in-human study |
| Half-life | About 24 h; cleared 3–4 days after a single dose | Same |
| Levels after 4 months of 1,000 mg/day | Parent urolithin A 1,038 pg/mL; urolithin A glucuronide 1,014 ng/mL | Liu 2022 |
Most of what circulates is the conjugated (glucuronide or sulfate) form: in the Liu trial, the glucuronide was measured in nanograms per mL and parent urolithin A in picograms per mL.
Regulatory status
In June 2018, Amazentis submitted GRAS notice GRN 791 to the FDA. GRAS (“generally recognized as safe”) is a route by which a company concludes a food ingredient is safe and can notify the FDA. The notice covered urolithin A in the following foods (FDA GRN 791 inventory entry; FDA response letter):
- up to 500 mg per serving in protein shakes, meal replacement drinks, instant oatmeal, protein and nutrition bars, and yogurt drinks;
- up to 1,000 mg per serving in Greek yogurts, high-protein yogurts and milk-based protein shakes.
On December 20, 2018, FDA replied that it had “no questions at this time” about the company’s conclusion. The letter states that it is not an FDA affirmation of GRAS status. It also did not evaluate health claims. Amazentis estimated that consumers 12 and older would take in a mean of 1,183 mg/day, and 2,421 mg/day at the 90th percentile, from these foods. GRN 791 is the only urolithin A notice in the FDA GRAS inventory.
Urolithin A is not approved as a drug for any condition. It is not named on the WADA 2026 Prohibited List.
What the research shows
Who funded the research
Most human trials of urolithin A were sponsored by Amazentis, which makes Mitopure:
- First-in-human trial. Amazentis was the sponsor (ClinicalTrials.gov NCT02655393), and most authors worked for the company (Andreux 2019).
- Liu 2022 (older adults). The study was supported by Amazentis. The paper states the funder had a role in its design, data analysis and interpretation, and the decision to publish. Two authors were company employees, and the CEO and the board chair were co-authors (Liu 2022).
- Singh 2022 (ATLAS). Amazentis sponsored the study, and 6 of the 9 authors were company employees (Singh 2022).
- Later trials and safety work. Amazentis also funded the immune trial (Denk 2025), the runners trial (Whitfield 2025), and the producer-status trial (Singh 2022, EJCN). Amazentis scientists co-authored the rat safety study (Heilman 2017) and filed the FDA GRAS notice.
Industry funding does not make results wrong. But the independent trials so far are small, and their results are mixed (see “Other trials” below).
First-in-human trial (Andreux 2019)
This randomized, double-blind, placebo-controlled trial enrolled 60 healthy, sedentary elderly adults (ClinicalTrials.gov NCT02655393; Andreux et al., Nat Metab 2019). The design had two parts:
- Single doses of 250, 500, 1,000 or 2,000 mg, including a test with food.
- Daily doses of 250, 500 or 1,000 mg for 28 days.
The primary outcome was safety, and the authors reported a favorable safety profile. Secondary outcomes showed that urolithin A reached the blood at all doses, and that 500 and 1,000 mg/day changed plasma acylcarnitines and mitochondrial gene expression in muscle.
Older adults: Liu 2022 (JAMA Network Open)
This double-blind randomized trial in Seattle enrolled 66 adults aged 65–90 (mean 71.7, 76% women). They took 1,000 mg/day of urolithin A or placebo for 4 months (Liu et al., JAMA Netw Open 2022). Both primary endpoints were null:
| Primary endpoint | Urolithin A | Placebo | Result |
|---|---|---|---|
| 6-minute walk distance, change | +60.8 m (SD 67.2) | +42.5 m (SD 73.3) | Not significant |
| Maximal ATP production, hand muscle (first dorsal interosseus) | +0.07 mM/s (SD 0.23) | +0.06 mM/s (SD 0.20) | Not significant |
| Maximal ATP production, leg muscle (tibialis anterior) | −0.03 mM/s (SD 0.10) | +0.03 mM/s (SD 0.10) | Not significant |
Both groups improved their walking distance by more than the 30 m usually considered clinically meaningful, so the placebo response was large.
Secondary results were mixed:
- Muscle endurance. The number of contractions until fatigue rose more with urolithin A at 2 months (hand: +95.3 vs +11.6; leg: +41.4 vs +5.7). At 4 months, the placebo group had caught up and the difference was no longer significant.
- Biomarkers. Several acylcarnitines, ceramides and C-reactive protein (CRP, a marker of inflammation) fell with urolithin A compared with placebo.
Middle-aged adults: Singh 2022 (Cell Reports Medicine, ATLAS)
This single-center, double-blind trial in Canada enrolled 88 untrained, overweight adults aged 40–64 with low aerobic fitness. They took placebo, 500 mg/day or 1,000 mg/day for 4 months (Singh et al., Cell Rep Med 2022). The primary endpoint, peak power output on a cycle test, did not differ significantly from placebo. It rose by about 4% in both urolithin A groups and by 0.7% with placebo.
Secondary outcomes (changes from baseline after 4 months):
| Outcome | Placebo | 500 mg/day | 1,000 mg/day |
|---|---|---|---|
| Hamstring average peak torque | −9.8% | +12% (p = 0.027 vs placebo) | +9.8% (p = 0.029 vs placebo) |
| Knee flexion maximum torque | −9.3% | +10.6% (p = 0.017) | +10.5% (p = 0.022) |
| Quadriceps average peak torque | −2.5% | +2.3% | +4.7% (not significant) |
| Hand-grip strength | +2.4% | +4.6% | +5.1% (not significant) |
| Peak VO₂ (aerobic capacity) | −1.1% | +1.6% | +10.2% (p = 0.058 vs placebo) |
| 6-minute walk distance | −0.1% | −0.2% | +7% (+33.4 m; p = 0.098 vs placebo) |
| Lean mass (DXA scan) | −0.7% | +0.1% | −1.0% (unchanged) |
Points to note:
- The significant strength results were in the hamstrings only. The quadriceps and grip did not differ.
- Part of the hamstring difference came from a significant decline of about 10% in the placebo group.
- The walk and VO₂ gains at 1,000 mg were significant only compared with baseline, not compared with placebo.
- CRP fell significantly at 1,000 mg. Acylcarnitines fell at 500 mg but not at 1,000 mg.
- Urolithin A did not change lean mass or fat mass. Despite the “muscle” marketing, it did not build muscle in this trial.
The authors call the trial a proof-of-concept study and list the null primary endpoint as its main limitation.
Other trials
| Trial | Population, dose | Primary or main outcome | Funding or disclosures |
|---|---|---|---|
| Denk et al., Nat Aging 2025 | 50 healthy middle-aged adults; 1,000 mg/day for 4 weeks | Naive-like CD8+ T cells +0.50 percentage points (95% CI 0.16–0.83); CD8+ fatty-acid oxidation capacity +14.72 points (95% CI 6.46–22.99) | Amazentis |
| Whitfield et al., Sports Med 2025 | 42 highly trained male runners at an altitude camp; 1,000 mg/day for 4 weeks | No improvement in 3,000 m time trial; no significant VO₂max interaction (p = 0.138); lower perceived exertion and creatine kinase (a marker of muscle damage) after the trial | Amazentis |
| Zhao et al., JISSN 2024 | 20 resistance-trained men; 1 g/day for 8 weeks | Bench and squat 1RM not significantly different from placebo; isometric strength and reps to failure improved | Jiangsu provincial education grant |
| da Costa Pinaffi-Langley et al., JAMA Netw Open 2026 | 50 adults aged 40–64 with obesity; 1,000 mg/day for 4 weeks | Primary outcome null: brachial flow-mediated dilation (a measure of blood-vessel function) +1.3% vs +0.3%; secondary blood-vessel outcomes also null; larger task-evoked brain blood-flow responses, task performance unchanged | No conflicts of interest reported |
Systematic reviews
- Dao 2026. A meta-analysis of 5 randomized trials with 236 participants could pool only one outcome, the 6-minute walk test, from 2 trials (Dao et al., Front Nutr 2026). The pooled difference was +17.03 m (95% CI −5.33 to 39.40), not significant, with low certainty of evidence (GRADE). The authors describe the strength, endurance and biomarker findings as “exploratory” signals, not reproducible effects.
- Kuerec 2024. An earlier review of 5 studies in 250 healthy people found anti-inflammatory and mitochondrial-gene effects, and gains in strength and endurance. It found no effect on maximal ATP production, body measurements, cardiovascular outcomes or physical function (Kuerec et al., Ageing Res Rev 2024).
Side effects and risks
Human safety data
The trials have been short (4 weeks to 4 months) and have reported no safety signal:
- Liu 2022. Treatment-emergent adverse events occurred in 16 people on urolithin A and 15 on placebo, with no serious events. Vital signs, blood chemistry, hematology and urinalysis showed no differences (Liu 2022).
- ATLAS. There were 42 adverse events in 17 people on placebo, 24 in 13 people on 500 mg, and 36 in 15 people on 1,000 mg. Many were related to the muscle biopsies. Safety lab tests did not differ between groups (Singh 2022).
- Denk 2025. Nine adverse events were recorded across both groups in 28 days. Kidney function and liver enzymes did not change (Denk 2025).
Long-term safety beyond 4 months has not been studied in controlled trials (Dao 2026).
Animal toxicology
In rats, the manufacturer’s safety program found the following (Heilman et al., Food Chem Toxicol 2017):
- urolithin A was not genotoxic in a standard battery of tests;
- 28-day and 90-day feeding studies showed no target organs and no changes in clinical, blood-chemistry or hematology measures;
- the no-observed-adverse-effect level (NOAEL) was the highest dose tested: 3,451 mg/kg/day in males and 3,826 mg/kg/day in females over 90 days.
These are animal results and set no human dose.
Unknowns
- Interactions. No drug-interaction studies were found in the trials reviewed here.
- Populations. The published trials enrolled adults, mostly healthy. No safety data were found for pregnancy, breastfeeding, children or people with serious illness.
- Product equivalence. The GRAS notice and the trials used Amazentis’s synthetic urolithin A with its own purity specification. Results may not transfer to other products sold as “urolithin A,” or to pomegranate or ellagitannin extracts, which depend on whether a person can make urolithin A.
Key takeaways
- A gut-bacteria metabolite, not a vitamin. Urolithin A comes from ellagitannins in pomegranates, nuts and berries, and only some people make meaningful amounts. One trial found detectable levels at baseline in 12% of adults.
- Mitophagy is shown mainly in cells and animals. It extended lifespan in worms and improved exercise capacity in rodents. Human data are limited to biomarkers in blood and muscle biopsies.
- Primary endpoints were safety or null. The first-in-human trial’s primary outcome was safety. Liu 2022 found no effect on walk distance or muscle ATP production, and ATLAS found no effect on peak power. The positive strength and endurance results are secondary outcomes.
- Most of the research is industry-funded. Amazentis, the maker of Mitopure, sponsored nearly all human trials and filed the GRAS notice. The independent trials so far were small, with null or mixed results.
- Short-term safety looks good. Trials of up to 1,000 mg/day for up to 4 months reported adverse events similar to placebo, and rat studies found no toxicity at very high doses. Long-term human data are lacking.
- GRAS is not approval. In 2018, FDA had “no questions” about Amazentis’s conclusion that urolithin A is GRAS for use in foods. That letter is not an FDA affirmation of safety and does not cover any health claim.
Sources
- Ryu D, Mouchiroud L, Andreux PA, et al. Urolithin A induces mitophagy and prolongs lifespan in C. elegans and increases muscle function in rodents. Nat Med. 2016;22(8):879–888. PubMed.
- Andreux PA, Blanco-Bose W, Ryu D, et al. The mitophagy activator urolithin A is safe and induces a molecular signature of improved mitochondrial and cellular health in humans. Nat Metab. 2019;1(6):595–603. PubMed.
- Amazentis SA. A single and multiple dose study of AMAZ-02 to evaluate safety and pharmacokinetics in elderly subjects. NCT02655393. ClinicalTrials.gov.
- Liu S, D’Amico D, Shankland E, et al. Effect of urolithin A supplementation on muscle endurance and mitochondrial health in older adults: a randomized clinical trial. JAMA Netw Open. 2022;5(1):e2144279. PubMed.
- Singh A, D’Amico D, Andreux PA, et al. Urolithin A improves muscle strength, exercise performance, and biomarkers of mitochondrial health in a randomized trial in middle-aged adults. Cell Rep Med. 2022;3(5):100633. PubMed.
- Singh A, D’Amico D, Andreux PA, et al. Direct supplementation with urolithin A overcomes limitations of dietary exposure and gut microbiome variability in healthy adults to achieve consistent levels across the population. Eur J Clin Nutr. 2022;76(2):297–308. PubMed.
- Tomás-Barberán FA, García-Villalba R, González-Sarrías A, et al. Ellagic acid metabolism by human gut microbiota: consistent observation of three urolithin phenotypes in intervention trials, independent of food source, age, and health status. J Agric Food Chem. 2014;62(28):6535–6538. PubMed.
- Pidgeon R, Mitchell S, Shamash M, et al. Diet-derived urolithin A is produced by a dehydroxylase encoded by human gut Enterocloster species. Nat Commun. 2025;16(1):999. PubMed.
- Heilman J, Andreux P, Tran N, et al. Safety assessment of urolithin A, a metabolite produced by the human gut microbiota upon dietary intake of plant derived ellagitannins and ellagic acid. Food Chem Toxicol. 2017;108(Pt A):289–297. PubMed.
- Denk D, Singh A, Kasler HG, et al. Effect of the mitophagy inducer urolithin A on age-related immune decline: a randomized, placebo-controlled trial. Nat Aging. 2025;5(11):2309–2322. PubMed.
- Whitfield J, McKay AKA, Tee N, et al. Evaluating the impact of urolithin A supplementation on running performance, recovery, and mitochondrial biomarkers in highly trained male distance runners. Sports Med. 2025;55(12):3183–3200. PubMed.
- Zhao H, Zhu H, Yun H, et al. Assessment of urolithin A effects on muscle endurance, strength, inflammation, oxidative stress, and protein metabolism in male athletes with resistance training: an 8-week randomized, double-blind, placebo-controlled study. J Int Soc Sports Nutr. 2024;21(1):2419388. PubMed.
- da Costa Pinaffi-Langley AC, Gulej R, Kaposzta Z, et al. Urolithin A and endothelial and cerebrovascular function in middle-aged adults with obesity: a randomized clinical trial. JAMA Netw Open. 2026;9(10):e2638047. PubMed.
- Dao T, Nguyen TT, Gariani K, et al. Effects of urolithin A supplementation on muscle health outcomes in humans from randomized controlled trials. Front Nutr. 2026;13:1834344. PubMed.
- Kuerec AH, Lim XK, Khoo AL, et al. Targeting aging with urolithin A in humans: a systematic review. Ageing Res Rev. 2024;100:102406. PubMed.
- Food and Drug Administration. GRN No. 791: Urolithin A (notifier Amazentis SA; filed June 15, 2018; closed December 20, 2018). FDA GRAS Notice Inventory.
- Keefe DM, Office of Food Additive Safety, FDA. Response letter to GRAS Notice No. GRN 000791 (urolithin A). December 20, 2018. FDA.
- National Center for Biotechnology Information. PubChem Compound Summary for CID 5488186, Urolithin A. PubChem.
- World Anti-Doping Agency. World Anti-Doping Code International Standard: Prohibited List 2026. WADA 2026 Prohibited List.