Urolithin A

A reference work, not a recommendation to use: this profile describes mechanism and evidence and does not replace medical advice. The claims come from marketing and forums and are tested here, not endorsed.

Claims checked

• Activates mitophagy and increases muscle strength and endurance

  Mixed results

  Muscle & performance · healthy (middle-aged to older) · Human, randomised

  Both large studies missed their primary endpoint: in 66 older adults walking distance and maximal ATP production ^[7], in middle-aged people peak power ^[3]. Positive findings concern secondary endpoints (muscle endurance after two months ^[7]; strength and walk test ^[3]). The meta-analysis (5 RCTs, walking distance pooled from 2) found +17.03 m without significance (p = 0.135) for the six-minute walk test with low certainty ^[9]. Mitophagy itself has been shown directly only in worm and rodent ^[5].

• Better athletic performance and recovery in trained people

  No benefit

  Muscle & performance · healthy (highly trained runners) · Human, randomised small studiesEvery study behind this level and result had fewer than 100 participants. Small studies report larger effects on average than large ones, so the result is less certain.manufacturer-fundedAccording to their funding or competing-interest statements, all studies behind this result were paid for, wholly or in part, by the maker of the product or compound, or list employees of the maker among their authors. Manufacturer-funded studies on average come out more favourably for the product.

  In a study with 42 trained runners over four weeks, the 3000-meter time did not improve; the time trial, however, took place only in a subgroup of 22 runners. Maximal oxygen uptake rose without a significant group difference. Lower perceived exertion and less creatine kinase after exertion are secondary findings; the study was funded by the manufacturer ^[8].

• Rejuvenates the immune system and lowers silent inflammation

  Single unconfirmed finding

  Immune system & infections · healthy (middle-aged) · Human, randomised small studiesEvery study behind this level and result had fewer than 100 participants. Small studies report larger effects on average than large ones, so the result is less certain.manufacturer-fundedAccording to their funding or competing-interest statements, all studies behind this result were paid for, wholly or in part, by the maker of the product or compound, or list employees of the maker among their authors. Manufacturer-funded studies on average come out more favourably for the product.

  A manufacturer-funded study with 50 healthy people found small shifts in CD8 T cell subsets after four weeks (difference 0.50 percentage points; p = 0.0437) and a higher fatty acid oxidation capacity (p = 0.0061); further findings were secondary or exploratory ^[10]. C-reactive protein was lower in a second study with 66 participants; there, founders, board members and employees of Amazentis were among the authors ^[7]. Clinical endpoints such as frequency of infections were not studied.

• Protects the blood vessels

  No benefit

  Heart & blood vessels · Patients: obesity in middle age · Human, randomised small studiesEvery study behind this level and result had fewer than 100 participants. Small studies report larger effects on average than large ones, so the result is less certain.

  In an independent RCT with 50 participants, the primary endpoint, flow-mediated vasodilation, remained unchanged after four weeks, as did the other peripheral vascular measures ^[11]. For context, different population: in ten people with heart failure, ultrasound values and BNP did not change in a small crossover study ^[12]. For healthy people without obesity, there are no data.

• Promotes cerebral blood flow and protects the brain

  Single unconfirmed finding

  Cognition & focus · Patients: obesity in middle age · Human, randomised small studiesEvery study behind this level and result had fewer than 100 participants. Small studies report larger effects on average than large ones, so the result is less certain.

  In the same RCT with 50 participants, task-related cerebral blood flow responses (secondary endpoint) were stronger than with placebo after four weeks, but task performance remained unchanged ^[11]. This is a single surrogate finding without a clinical endpoint and without confirmation; entry of urolithin A into the brain is thus not shown.

Mechanism of action

Urolithin A does not arise in food but in the colon: gut bacteria break down ellagitannins and ellagic acid from pomegranate, walnuts and certain berries in several steps to urolithins. Whether urolithin A is formed depends on one’s own microbiome. In a study with 100 healthy people, only 12% had measurable urolithin A levels at all, and after pomegranate juice, about 40% of participants formed the compound in appreciable amounts; those who could had a more diverse gut flora. Direct intake of urolithin A bypassed this dependence and led to more than six-fold availability compared with the juice ^[1]. Blood lipid and risk markers also differ by these metabolic types (metabotypes A, B and 0), but only in cross-sectional comparisons ^[2].

Although urolithin A is marketed as a “postbiotic” and is called that even in manufacturer publications ^[3], it is not one: the ISAPP consensus explicitly excludes purified microbial metabolites without cellular components ^[4]. In this library, the compound is therefore filed under metabolism/mitochondria and in the family of microbial metabolites.

The mechanism of action is considered to be mitophagy, i.e., the targeted breakdown of damaged mitochondria. Urolithin A was described as the first natural compound that induces mitophagy: in roundworms, it prevented the accumulation of damaged mitochondria, extended lifespan and preserved mobility, and in two mouse models of age-related muscle decline and in young rats, it improved endurance (animal experiments, level 2) ^[5]. In humans, this mechanism is documented only indirectly: in the first human study with older, low-activity people, safety was the primary endpoint; after four weeks, acylcarnitines in plasma and the expression of mitochondrial genes in muscle changed, both secondary endpoints ^[6]. In later studies, acylcarnitines, ceramides and C-reactive protein fell ^[7], and in muscle samples of trained men, an upregulation of mitochondria-related signaling pathways was found without a measurable change in mitochondrial function ^[8].

The two main studies on muscle performance missed their primary endpoints. In 66 older adults aged 65 to 90 years, neither the distance in the six-minute walk test nor the maximal ATP production in the hand muscles changed significantly after four months; the muscle endurance of two muscles was better after two months (secondary endpoint) ^[7]. In middle-aged people, the study missed the primary endpoint peak power; the manufacturer’s author group reports about 12% better muscle strength as well as “clinically meaningful” improvements in endurance and the six-minute walk test ^[3]. A meta-analysis of five randomized studies with a total of 236 participants could pool only the six-minute walk test from two studies: +17.03 m (95% CI −5.33 to 39.40; p = 0.135) with low certainty according to GRADE; it rates all other endpoints as exploratory ^[9]. In 42 well-trained runners at an altitude training camp, the 3000-meter time did not improve, though it was measured only in a subgroup of 22; perceived exertion and a muscle damage marker were lower ^[8].

Outside muscle, there are isolated findings. In a study with 50 healthy middle-aged people, after four weeks, proportions of T cell subsets and the fatty acid oxidation capacity of CD8 cells shifted; the differences were small and the study was funded by the manufacturer ^[10]. In an independently conducted study with 50 middle-aged people with obesity, the primary endpoint, flow-mediated vasodilation, remained unchanged; task-related cerebral blood flow responses differed, which the author group describes as a reason for further research ^[11]. In ten people with heart failure, urolithin A did not change cardiac ultrasound values or BNP ^[12]. On safety, animal data are available: urolithin A was unremarkable in genotoxicity tests, the main metabolites are glucuronides and sulfates, and in a 90-day study in rats, the highest tested feed concentration remained without discernible harm; the paper comes from the manufacturer ^[13].

Targets

Effect profile

Which signaling systems the compound affects according to current data. ↑ enhanced, ↓ dampened, ≈ modulated, ○ no effect; dots = evidence level.

• ·Dopamineno data
• ·Noradrenalineno data
• ·Serotoninno data
• ·Acetylcholineno data
• ·Glutamate · AMPAno data
• ·Glutamate · NMDAno data
• ·GABAno data
• ·Histamineno data
• ·Orexinno data
• ·Adenosineno data
• ·Opioidsno data
• ·Melatoninno data
• ·Neurotrophinsno data
• ○Blood flowno effect · Flow-mediated vasodilation unchanged in obesity; cerebral blood flow responses altered as a secondary finding^[11]
• ·Nitric oxideno data
• ≈Energy metabolismmodulated · Mitophagy in worm and rodent (level 2); in humans only gene expression in muscle and acylcarnitines in plasma, no better ATP production^[5][6][7]
• ·Insulin · glucoseno data
• ↓Inflammation · oxidative↓ · C-reactive protein lower than with placebo; immune cell findings small and manufacturer-funded^[7][10]
• ·Stress axisno data
• ·Testosterone axisno data
• ·Endocannabinoidsno data
• ·Gut barrierno data
• ≈Microbiomemodulated · Formation depends on one’s own microbiome; only a minority forms urolithin A from ellagitannins^[1][2]
• ·Short-chain fatty acidsno data
• ·Bile acidsno data
• ·Gut hormonesno data
• ·Gut–brain axisno data
• ·Autophagy · senescenceno data
• ·Immune systemno data
• ·Lipid metabolismno data
• ·Coagulation · fibrinolysisno data
• ·Methylationno data

Routes of administration and bioavailability

How much of a compound reaches the blood or target tissue depends on the route into the body and on the dosage form. All routes of administration explained

• Oral (swallowed) · synthetic urolithin A (pure compound)

  more than six-fold availability compared with pomegranate juice; in blood predominantly as glucuronide and sulfate

  Direct intake makes availability independent of the microbiome and of diet ^[1]; in the first human study, urolithin A was measurable in plasma at all tested doses ^[6]. Animal data show glucuronides and sulfates as main metabolites ^[13].

  Human, randomised

• Oral (swallowed) · Ellagitannins from pomegranate, walnuts or berries

  strongly dependent on the individual; only about 40% form urolithin A in appreciable amounts

  The conversion takes place only in the colon by bacteria; people are distinguished by metabotype A, B or 0 ^[1][2].^[1][2]

  Human, randomised

Human studies

• First human study (safety and mitochondrial markers)

  RCT, placebo-controlled, single and multiple doses · healthy (older, low-activity people) · single dose or 4 weeks

  Measured: Safety (primary); plasma levels, acylcarnitines, mitochondrial gene expression in muscle (secondary)

  Result: Favorable safety profile, measurable in plasma; acylcarnitines and mitochondrial gene expression altered.^[6]

  Human, randomised

• Muscle endurance in older adults

  RCT, double-blind, placebo-controlled · healthy (65 to 90 years) · n = 66 · 4 months

  Measured: Six-minute walking distance and maximal ATP production in the hand muscles (primary); muscle endurance of two muscles (secondary)

  Result: Primary endpoints without significant difference; muscle endurance better after two months; acylcarnitines, ceramides and C-reactive protein lower after four months. No differences in adverse events.^[7]

  Human, randomised

• Strength and performance in middle age

  RCT, placebo-controlled, two dose levels · healthy (middle-aged) · 4 months

  Measured: Peak power (primary); muscle strength, endurance, walk test, plasma markers, muscle proteins

  Result: Primary endpoint missed; muscle strength about 12% better according to the report, endurance and walk test “clinically meaningful”; acylcarnitines and C-reactive protein lower. The manufacturer is the sponsor.^[3]

  Human, randomised

• Meta-analysis muscle health

  systematic review with meta-analysis (5 RCTs) · mixed (healthy people and patients) · n = 236

  Measured: Six-minute walk test (poolable, 2 studies); strength, endurance, biomarkers (narrative)

  Result: +17.03 m (95% CI −5.33 to 39.40; p = 0.135; I² = 0%), certainty low (GRADE); other endpoints heterogeneous and reported only exploratorily.^[9]

  Human, randomised

• Highly trained runners at an altitude camp

  RCT, double-blind, parallel · healthy (highly trained male runners) · n = 42 · 4 weeks

  Measured: 3000-meter time (subgroup of 22 only), maximal oxygen uptake, running economy, muscle damage markers, muscle proteome (subgroup of 20)

  Result: Running time unchanged; perceived exertion lower (p = 0.02), creatine kinase area smaller (p < 0.0001); mitochondria-related protein patterns upregulated, mitochondrial function unchanged. Manufacturer-funded.^[8]

  Human, randomised

• Immune aging

  RCT, double-blind, placebo-controlled · healthy (middle-aged) · n = 50 · 4 weeks

  Measured: Phenotype of CD3 T cell subsets and immune metabolism (primary)

  Result: More naive-like CD8 cells (difference 0.50 percentage points; 95% CI 0.16–0.83; p = 0.0437), higher fatty acid oxidation capacity (14.72 percentage points; 6.46–22.99; p = 0.0061); further findings secondary to exploratory. Funded by the manufacturer.^[10]

  Human, randomised

• Vascular and cerebral blood flow in obesity

  RCT, double-blind, placebo-controlled · Patients: obesity, 40 to 64 years · n = 50 · 4 weeks

  Measured: Flow-mediated dilation of the brachial artery (primary); further vascular and cerebral blood flow measures

  Result: Primary endpoint without group difference, as were the other vascular measures; task-related cerebral blood flow responses stronger, task performance unchanged. 7 participants lost (14%). No conflicts of interest declared.^[11]

  Human, randomised

• Heart failure with reduced pumping function

  RCT, double-blind, placebo-controlled, crossover · Patients: heart failure with reduced ejection fraction · n = 10 · 2 × 4 weeks with washout

  Measured: Echocardiography (ejection fraction, among others), pro-BNP, blood lipids

  Result: No change in ultrasound values, pro-BNP, glucose and C-reactive protein; HDL cholesterol slightly higher (p = 0.026).^[12]

  Human, randomised

• Metabotypes and direct intake

  randomized comparison study (pomegranate juice vs. urolithin A product) · healthy · n = 100 · single and repeated measurements

  Measured: Frequency of urolithin A producers, microbiome differences, plasma levels

  Result: Only 12% with measurable urolithin A at baseline; about 40% formed it after juice; producers with higher microbiome diversity; direct intake more than six-fold availability (p < 0.0001).^[1]

  Human, randomised

Risks and warnings

• Short-term safety unremarkable, long-term data lacking

  The studies ran four weeks to four months without notable differences in adverse events ^[6][7][10]. The basis of the safety assessment is animal experiments by the manufacturer: no genotoxicity, no discernible harm in a 90-day study in rats up to the highest tested feed concentration ^[13]. Data over years, in pregnant and lactating women and in children are lacking.

  Human, randomised

• Body of evidence almost entirely from the manufacturer

  The studies ^{[1][3][6][7][8][10]} were funded by Amazentis or co-authored by employees, board members and advisory board members, partly with patents on the application ^[3][7]. Independent so far are the vascular study ^[11] and the heart failure study ^[12], both without a positive primary finding.

  Human, randomised

• Correction to the immune study

  An author correction was published for the immune aging study; its content was not read here and is noted under “pruefung.nicht_verifiziert” ^[10].

  Human, randomised

Interactions

• Antibiotics and gut flora: Endogenous formation from ellagitannins depends on gut bacteria; antibiotics can impair it. For the ingested pure compound, this is irrelevant; data on this are lacking ^[1].
• Substances with phase II metabolism (glucuronidation, sulfation): Urolithin A is predominantly glucuronidated and sulfated ^[13]; interactions at this level are plausible but not studied.

Fact check: common claims

• refuted

  “Urolithin A is a postbiotic”

  The ISAPP consensus excludes largely purified microbial metabolites without cellular components from the definition ^[4]. Even so, a manufacturer publication calls the compound a “postbiotic compound” ^[3]; this is marketing, not a classification according to the consensus.

• not supported

  “Urolithin A demonstrably increases muscle strength and endurance”

  Both large studies missed their primary endpoints ^[3][7], there was no performance benefit in trained people ^[8], and the meta-analysis found no significant effect on walking distance at low certainty ^[9].

• partly supported

  “Because most people do not form urolithin A themselves, they need it as a supplement”

  That only a minority forms the compound in appreciable amounts and that direct intake produces higher levels is documented ^[1]. Whether higher levels bring a clinical benefit is open: the studies with clinical primary endpoints missed them ^[3][7][11].

• not supported

  “Urolithin A extends life (marketed as a “postbiotic” for longevity)”

  Life extension has been shown only in roundworms, plus better endurance in mouse and rat; these are animal findings (level 2) ^[5]. There are no data on lifespan in humans; the human studies measure walking distance, strength, blood markers and immune cells over weeks to months ^[3][7][10]. For the designation “postbiotic,” see the first claim ^[4].

• not supported

  “Pomegranate or pomegranate juice has the same effect”

  After juice, only about 40% of participants formed urolithin A, and levels were markedly lower than after direct intake ^[1]. Whether the lower levels are sufficient was not tested clinically; differences between metabotypes are so far only cross-sectional findings ^[2].

Pharmacokinetics

Absorption

Arises in humans only in the colon from ellagitannins and is thus microbiome-dependent ^[1]; taken as a pure compound, reliably measurable in plasma ^[6].

Half-life

Not verified.

Metabolism

In animal experiments predominantly glucuronides and sulfates after oral and intravenous administration ^[13]; in humans, phase II conjugates also circulate ^[1].

Entry into the brain

Not documented. Altered cerebral blood flow responses in one study are not evidence of entry into the brain ^[11].

Legal status

Germany

Not verified.

EU

No entry for “urolithin” in the consolidated Union list of novel foods, as of August 10, 2026 (German and English versions searched, 0 hits) ^[14]; no EU legal act with “urolithin” in the title found via Cellar and no EFSA novel food opinion found in PubMed. The Commission’s Novel Food Catalogue was not checked; no statement on marketability can be derived from this.

USA

Not verified.

Related compounds

Open questions

• Is there an independent study with a clinical primary endpoint that shows a benefit?
• Do people who do not form urolithin A themselves (metabotype 0) benefit specifically? The studies did not stratify for this.
• Can mitophagy be measured directly in humans, instead of via acylcarnitines and gene expression?
• What is the significance of the altered cerebral blood flow responses with unchanged task performance?
• How does the compound behave over years, especially with permanent stimulation of mitophagy?