Urolithin A is a compound formed when gut bacteria convert ellagitannins and ellagic acid, found in foods such as pomegranates, walnuts and blackberries. It isn't present in free form in the food itself; it is made not by you but by specific bacteria in your gut. In laboratory animals it has been shown to trigger the clearance of damaged mitochondria (mitophagy). In humans, what we have is a handful of small, short randomized trials: some muscle and immune measures improved, while several of the outcomes the trials were actually designed around showed no difference. There are no data showing that it extends human lifespan.
The reason the topic draws so much interest is that mitochondrial function declines with age and few interventions target it. Urolithin A is being put forward as a candidate to fill that gap. To see how solid that candidacy is, let's go through the evidence in order.
Two people who eat the same pomegranate can end up with very different amounts of urolithin A in their blood. Researchers explain the difference with the concept of the "urolithin metabotype"; the literature describes three groups, known as UM-A, UM-B and UM-0. A study by the CEBAS-CSIC team in Spain of 839 people aged 5–90 (Cortés-Martín et al., 2018) reported that age was the main factor determining which metabotype a person falls into.
The clearest figure for the share of producers comes from a study published in 2021 in the European Journal of Clinical Nutrition. Among 100 healthy adults in the US, only 12% had measurable urolithin A in their blood under everyday conditions. Giving participants a glass of pomegranate juice didn't change the picture much: about 40% were able to convert the precursor compounds into urolithin A at a meaningful level. Producers had a more diverse gut microbiome. This study was run by researchers at Amazentis, the company that developed the supplement; the person-to-person difference in production, however, had already been shown independently by the Spanish team. We cover the link between gut bacteria and health more broadly in our gut microbiome article.
The study that put urolithin A on the aging-research agenda appeared in Nature Medicine in 2016 (Ryu et al.). The team showed that the compound triggers mitophagy when taken orally. Mitophagy is the mitochondria-specific form of autophagy: the cell dismantles mitochondria that can no longer do their job and reuses the parts.
In the roundworm C. elegans, urolithin A prevented the age-related buildup of dysfunctional mitochondria and extended lifespan. In two different mouse models of age-related muscle decline and in young rats, exercise capacity increased. These are animal findings. A longer life for a one-millimeter worm generates a hypothesis for humans, not evidence. Amazentis employees are among the authors of this study as well.
Middle-aged adults (Singh et al., Cell Reports Medicine 2022). A total of 88 healthy, overweight middle-aged adults took 500 mg of urolithin A, 1,000 mg of urolithin A or placebo daily for 4 months. The trial's prespecified primary endpoint was peak power output, and neither dose separated significantly from placebo. The authors themselves name this as the study's main limitation. Among secondary measures, hamstring muscle strength was 12% higher than placebo in the 500 mg group and 9.8% higher in the 1,000 mg group. Part of that difference comes from a decline in the placebo group: in those taking placebo, average torque in the same muscle fell by 9.8% over the course of the trial. In the 6-minute walk test and in peak oxygen consumption, the high-dose group improved relative to its own baseline, but the comparison with placebo fell short of statistical significance (p = 0.098 and p = 0.058, respectively).
Older adults (Liu et al., JAMA Network Open 2022). In Seattle, 66 people aged 65–90 took 1,000 mg of urolithin A or placebo daily for 4 months. There were two primary endpoints: 6-minute walk distance and maximal ATP production in a hand muscle. Neither showed a significant difference. Walk distance rose by an average of 60.8 meters in the urolithin A group and 42.5 meters in the placebo group; in other words, those on placebo improved markedly too. Among secondary measures, though, the number of contractions the hand and leg muscles could perform before fatiguing increased significantly in favor of urolithin A at month 2, and C-reactive protein and some acylcarnitine levels fell at month 4.
On the safety side, no problems emerged: in the older-adult trial the frequency of adverse events did not differ between groups, and in the middle-aged trial the group reporting the most adverse events was the placebo group (17 people, versus 13 in the 500 mg group and 15 in the 1,000 mg group). The product used in both trials was Amazentis's Mitopure, and company employees appear on the author lists. For the method with far stronger evidence for preserving muscle strength, see our strength training article.
Immune aging (Denk et al., Nature Aging 2025). A total of 50 healthy adults aged 45–70 took 1,000 mg of urolithin A or placebo daily for 4 weeks. In this trial the primary endpoints were significant: the share of "naive-like", less exhausted CD8 T cells in the blood rose (treatment difference 0.50 percentage points), and these cells' capacity for fatty acid oxidation increased (14.72 percentage points). The findings are at the cellular level. The trial's endpoints did not include clinical outcomes such as infection frequency or vaccine response. Amazentis employees are again among the authors.
Vascular function (da Costa Pinaffi-Langley et al., JAMA Network Open 2026). In this trial, conducted at the University of Oklahoma and published in October 2026, 50 adults aged 40–64 with obesity took 1,000 mg of urolithin A or placebo for 4 weeks. The primary endpoint was flow-mediated dilation of the brachial artery, and no significant difference was found between groups (placebo 0.3%, urolithin A 1.3%). Secondary vascular measures showed no effect either. During a working-memory task, the cerebral blood flow response was higher in the urolithin A group, but task performance did not change; the authors note this as a finding that warrants future research.
Athletes. In 42 male distance runners at an altitude camp, 4 weeks of 1,000 mg urolithin A did not improve 3,000-meter run time; perceived exertion and creatine kinase, a marker of muscle damage, were lower (Whitfield et al., Sports Medicine 2025). In 20 resistance-trained male athletes there was no significant difference in one-repetition maximum lifts after 8 weeks, whereas isometric contraction strength and repetitions to failure increased relative to placebo (Zhao et al., 2024). A single-blind pilot study in 20 academy soccer players reported a 239-meter difference in an intermittent running test (p = 0.048). And in a crossover study of 10 patients with heart failure, no effect at all was seen on echocardiography measures.
The 2026 meta-analysis. In July 2026, Dao and colleagues surveyed all randomized trials on muscle health in Frontiers in Nutrition. They found 5 trials and 236 participants in total. Quantitative pooling was possible only for the 6-minute walk test: mean difference +17.03 meters, 95% confidence interval −5.33 to 39.40, p = 0.135. The direction is positive, the result statistically uncertain, and the certainty of evidence low by GRADE criteria. The authors do not regard the strength and endurance findings as a "replicable effect"; they describe them as exploratory signals.
When you eat pomegranate, what you get is the raw material for urolithin A. How much of it becomes urolithin A depends on your gut bacteria; in some people the amount is close to zero. A supplement, by contrast, delivers the finished molecule and skips the bacterial step. In the 100-person study above, a product containing 500 mg of urolithin A produced more than 6 times the blood exposure to urolithin A of a glass of pomegranate juice.
That doesn't mean the supplement is "better" than pomegranate. They are different things. Pomegranates, walnuts and blackberries also contain fiber and other polyphenols; their benefits can't be reduced to whether or not they are converted into urolithin A. Equally, someone who eats pomegranate shouldn't assume they are reaching the blood levels seen in clinical trials. On the dietary side, the data we have are observational: in the 294-person DIRECT PLUS trial, among those following a polyphenol-rich Mediterranean diet that included 28 grams of walnuts a day, people with higher urinary urolithin A levels lost more visceral fat. That is an association; it doesn't show that the loss was caused by urolithin A.
Urolithin A isn't found ready-made in foods. Gut bacteria convert the ellagitannins and ellagic acid in foods such as pomegranates, walnuts and blackberries into urolithin A. In other words, food is the source and the gut microbiota is the producer. In people who lack the bacteria that carry out this conversion, the same food does not turn into urolithin A.
No. In a study of 100 healthy adults, urolithin A was detected at baseline in the blood of only 12% of participants; even after a glass of pomegranate juice, only about 40% were able to produce urolithin A at a meaningful level. Production capacity depends on the makeup of the gut bacteria and, according to a Spanish study of 839 people, changes with age.
Partly, and not conclusively. In a randomized trial of 88 middle-aged adults, hamstring strength increased relative to placebo, but there was no significant difference in peak power output, the trial's primary endpoint. In a trial of 66 older adults, the primary endpoints (6-minute walk distance and the muscle's maximal ATP production) were not significant either; improvements were seen in secondary measures such as muscle endurance. A meta-analysis published in 2026 rated the certainty of the evidence as low.
There are no data showing that it extends lifespan in humans. The lifespan-extension finding comes from the roundworm C. elegans. Human trials have lasted 4 months at most and looked at intermediate measures such as muscle, immune cells and vascular function; no trial has measured outcomes such as disease or death.
In the two muscle trials, doses of up to 1,000 mg a day were well tolerated for 4 months; in the trial in older adults, the frequency of adverse events was no different from placebo. There are no data on longer-term use. Most trials were conducted in healthy adults; people who have a chronic illness, take medication, or are pregnant or breastfeeding should not start without consulting a doctor.
"For urolithin A, the mechanism is intriguing and the animal data are consistent. The human data, however, rest on five small muscle trials and a few short studies, and most of the outcomes they were primarily aiming at showed no difference. Today's evidence is enough to say 'promising but unproven.'"
— Vivelong Longevity Science Team