Foods and Supplements for Mitochondrial Health: What the Evidence Says
Hundreds of products promise to "boost your mitochondria." Few have human evidence. We graded the leading foods and supplements against what they actually show in human studies — dose ranges, safety notes, and sources included.
What does "supporting mitochondrial health" actually mean?
A claim "supports mitochondrial health" is only meaningful if it changes something measurable: mitochondrial respiration, ATP output, mitochondrial protein content, mitophagy (recycling of damaged mitochondria), or a clinical outcome tied to energy metabolism like muscle strength or exercise capacity. Most products on the market have never been tested against any of these endpoints. The sections below apply one rule throughout: mechanistic plausibility is not evidence, animal results are not human results, and a blood marker is not mitochondrial function. Each item is graded Established, Preliminary, or Marketing ahead of evidence — and every recommendation starts with the cheapest, best-supported intervention of all: correcting genuine nutrient deficiency.
Does the Mediterranean diet help mitochondria?
The Mediterranean dietary pattern — vegetables, fruits, olive oil, legumes, whole grains, fish — carries the most respectable dietary evidence in this space, though it remains preliminary for mitochondrial endpoints specifically. A 2023 review (PMID 37019345) concluded that components of the pattern can reduce mitochondrial reactive oxygen species production, preserve membrane potential and respiration, and modulate mitochondrial enzymes — but acknowledged that most of the mechanistic data are preclinical and human confirmation is still needed. A 2026 USC study added a striking human signal: older adults with closer Mediterranean-diet adherence had higher blood levels of the mitochondria-encoded microproteins humanin and SHMOOSE — though this was observational, so it shows correlation, not causation. The honest summary: the pattern is excellent general health advice with plausible mitochondrial mechanisms and emerging human signals, but "eating Mediterranean improves your mitochondrial function" is not yet an established claim.
What about berries, olive oil, cocoa, and red wine — the polyphenols?
Polyphenol-rich foods — berries, olives and olive oil, cocoa, red grapes, green tea — show genuinely interesting mitochondrial biology in cell models. Delphinidin (found in berries and red wine) restored mitochondrial respiration, mtDNA content, and complex IV activity in cellular models; resveratrol and lycopene modulated mitochondrial enzymes (PMID 37019345 review). A 2024 Nature Metabolism study identified oleuropein, an olive polyphenol, as an activator of mitochondrial calcium uptake via the MICU1 regulator in aged human muscle cells and mouse muscle, improving bioenergetics — preclinical and human-cell work, not yet clinical trials. But the polyphenol with the most marketing behind it deserves a warning: resveratrol. A human study of a single 300 mg dose found no activation of the AMPK/SIRT1 axis in muscle or fat, and resveratrol actually inhibited mitochondrial respiration in ex vivo muscle. Human efficacy for resveratrol's anti-aging and mitochondrial claims remains unproven — sold on animal data, contradicted by the human data we have. That is Marketing ahead of evidence.
Is urolithin A the best-studied mitophagy supplement?
Yes — urolithin A carries the most promising human signal for engaging mitophagy, the recycling of damaged mitochondria. It is produced in the gut from ellagitannins in pomegranate, walnuts, and berries, but only in people whose gut bacteria perform the conversion — a large fraction of people produce little, which is the rationale for supplementing directly rather than relying on foods. The pivotal trial (Singh et al., Cell Reports Medicine 2022; n=88 middle-aged adults, 4 months, 500 or 1,000 mg/day versus placebo) reported roughly 10–12% greater leg-muscle strength gains, improved VO2 peak and six-minute walk distance at the 1,000 mg dose, lower plasma acylcarnitines and C-reactive protein, and increased mitophagy-protein expression in muscle biopsies. The caveats are real and must be stated: the prespecified primary endpoint (peak power) did not reach statistical significance, and the trials were sponsored by the product's manufacturer (Amazentis/Timeline). Grade: Preliminary but a genuine human signal — not a proven therapy, and food-dose effects should not be equated with supplement-trial doses. Doses studied: 500–1,000 mg/day; safe in trials up to 2,000 mg/day.
Does CoQ10 help if you're not deficient?
Coenzyme Q10 sits at the center of this category's paradox: impeccable biochemistry, disappointing clinical data in healthy people. CoQ10 is an integral component of the electron transport chain — that is Established — and genetic primary CoQ10 deficiency is genuinely treatable with CoQ10 supplementation Established. But in healthy, non-deficient adults the record is null: oral CoQ10 raises plasma levels but generally does not raise muscle CoQ10 or improve VO2max and exercise economy (small muscle-biopsy trial). Ubiquinol (the reduced form) at 300 mg/day for six weeks improved oxidative-phosphorylation coupling efficiency in muscle without changing exercise tolerance or VO2 kinetics in healthy active men; 200 mg/day for one month in young athletes reduced exercise-induced CoQ depletion and oxidative markers without improving performance or muscle-damage markers; one small trial found enhanced peak power in trained athletes at 300 mg/day, without measuring plasma CoQ10. The pattern: benefits appear in deficiency or disease states, not in replete healthy adults. Dose range studied: 100–300 mg/day; generally well tolerated. Statin users note: statins can lower CoQ10, and replenishment has mechanistic rationale — but proven outcome benefit is lacking.
What about PQQ, L-carnitine, creatine, and alpha-lipoic acid?
PQQ (pyrroloquinoline quinone): 20 mg/day for six weeks combined with endurance training (n=23 men) produced no added aerobic-performance benefit versus training alone, but significantly raised PGC-1α, a mitochondrial-biogenesis marker — a marker-only result, and oral PQQ did not stimulate mitochondrial biogenesis in human skin cells in vitro. Marketing claims of "creating new mitochondria in humans" run ahead of this evidence. L-carnitine: its physiology is textbook — it transports long-chain fatty acids into mitochondria Established — but 4 g/day for three months in healthy men did not raise muscle carnitine content, mitochondrial enzyme activities, or performance, because healthy muscle appears saturated. Benefits concentrate in deficient or diseased populations (metabolic syndrome, heart failure) rather than healthy adults. Note: L-carnitine raises TMAO, a gut-microbiome metabolite of uncertain long-term cardiovascular significance. Creatine monohydrate: strong rationale via phosphocreatine energy buffering; in mitochondrial cytopathy patients, small studies showed improved muscle endurance and body composition while others showed nothing (Tarnopolsky review). A combination of L-carnitine + creatine + leucine improved lean mass and strength in healthy older adults in an eight-week RCT. Typical dose: 3–5 g/day; well tolerated. Alpha-lipoic acid (ALA): a mitochondrial antioxidant by rationale, but 300 mg/day plus acetyl-L-carnitine for four weeks in trained cyclists did not improve aerobic or anaerobic performance — inconclusive rather than disproven, given shorter duration and lower dose than some positive older studies. Dose range in exercise studies: 600–1,200 mg/day. Caution: ALA can lower blood sugar — relevant for anyone on diabetes medications.
Which nutrients actually matter — magnesium and B vitamins?
Start with what is settled. B vitamins (B1, B2, B3) are direct cofactors in the citric acid cycle and electron transport chain, and magnesium is required for ATP to be biologically usable — ATP must bind Mg²⁺ to drive cellular work. Roughly 48% of Americans fall below the estimated average requirement for magnesium (NIH Office of Dietary Supplements). Correcting deficiency is established good medicine, and leafy greens, legumes, nuts, and whole grains are the food-first route. What is not supported: "more is better for mitochondria" beyond sufficiency. Supplementation in already-replete people has no demonstrated mitochondrial benefit. This is the category's most common error — taking an essential nutrient and extrapolating deficiency data into super-dose marketing. Melatonin, sometimes sold for mitochondrial antioxidant effects, remains at the investigational stage: mostly animal and in vitro evidence, thin human mitochondrial-endpoint data.
What about NR and NMN?
Nicotinamide riboside and NMN — NAD+ precursors — are covered in depth in our dedicated article, NR: the human trial record, and we do not duplicate that analysis here. The one-paragraph version: oral NR reliably raises blood NAD+ in humans, but four muscle-biopsy trials found no improvement in measured mitochondrial respiration, content, or bioenergetics — human translation of the animal evidence has failed so far. Read the full graded record there.
What are the safety notes?
All compounds above were generally well tolerated at studied doses: urolithin A safe in trials to 2,000 mg/day; CoQ10 safe at 100–300 mg/day; creatine 3–5 g/day has one of the best safety records in sports nutrition. Specific cautions: alpha-lipoic acid can lower blood sugar — coordinate with diabetes medications. L-carnitine raises TMAO with uncertain long-term cardiovascular significance — disclose, don't panic. No supplement is FDA-approved to treat disease, product quality varies widely (choose third-party-tested brands), and "mitochondrial support" on a label is a structure/function claim, not an efficacy finding. Anyone with active cancer should note that NAD+ precursor biology in oncology is unstudied in humans and mixed in animals. And the standing rule: none of these treat mitochondrial disease — see what mitochondrial disease is.
The foods-and-supplements bottom line
A Mediterranean-style diet has the best-supported (though still preliminary) dietary evidence for mitochondrial endpoints Preliminary. Urolithin A (500–1,000 mg/day) has the strongest human mitophagy signal of any supplement, from one company-sponsored trial whose primary endpoint missed Preliminary. CoQ10 is biochemically central and genuinely treats genetic deficiency, but does not improve mitochondrial function in healthy non-deficient adults Established · null. PQQ, L-carnitine, creatine, and alpha-lipoic acid show marker-level or population-specific signals, not general benefits Preliminary · mixed. Magnesium and B vitamins are essential cofactors — correct deficiency; super-dosing replete people has no demonstrated benefit Established nutrient · unproven beyond sufficiency. Resveratrol's human data contradict its marketing Marketing ahead of evidence.
Frequently asked questions
What is the single best-supported supplement for mitochondrial health?
Urolithin A has the strongest human evidence of engaging mitophagy (Singh et al. 2022, 500–1,000 mg/day), but the evidence is still preliminary — one company-sponsored trial, primary endpoint missed. No supplement has proven clinical benefit for mitochondrial function in healthy people.
Should healthy people take CoQ10?
Human trials show oral CoQ10 raises blood levels but generally does not raise muscle CoQ10 or improve exercise capacity in healthy, non-deficient adults. It is genuinely indicated for genetic CoQ10 deficiency and has rationale (not proof) for statin users.
Can I get urolithin A from food?
Pomegranate, walnuts, and berries supply its precursors, but conversion depends on gut bacteria that many people lack. Food-dose effects should not be equated with the 500–1,000 mg/day doses used in trials.
Do any of these treat mitochondrial disease?
No. Supplements do not treat mitochondrial disease. Management is symptomatic and supportive, with one approved disease-specific therapy (KYGEVVI for TK2 deficiency, November 2025). See our mitochondrial disease guide.
This article is for informational purposes only and is not medical advice, diagnosis, or treatment. Talk to your physician before starting any supplement, especially if you take prescription medications or have a medical condition.