Exercise and Mitochondrial Function

How Training Builds a Better Mitochondrial Network

Exercise is the most reliable stimulus known for improving mitochondrial health in humans. Each bout of muscle contraction creates an energetic and redox challenge that the cell senses through calcium, AMP and reactive oxygen species (ROS). These signals converge on the transcriptional coactivator PGC-1α and set off a coordinated response: new mitochondrial proteins are made, the network is remodeled through fusion and fission, and damaged organelles are removed by mitophagy¹.

Repeated over weeks, this turnover produces more mitochondria that work better, with roughly 23–27% gains in mitochondrial content across endurance, high-intensity and sprint training. Inactivity and ageing push result in the opposite effect - toward fragmented, ROS-emitting networks¹,². For practitioners, the central message is that mitochondrial trainability is preserved across age, sex and disease³. The main limitation is that most mechanistic detail comes from rodent and cell models, and human tissue-level data are largely restricted to skeletal muscle.

The Importance of Exercise for Mitochondrial Health

Why does a brisk walk, a set of squats or a few hard intervals change how a cell makes energy for days afterwards? The question matters because physical inactivity is now the norm, not the exception. San-Millán makes the point sharply: movement is the condition our physiology evolved under, and it is sedentary living that has become the intervention, with mitochondrial dysfunction among its collateral effects².

Mitochondrial dysfunction and disrupted cellular bioenergetics are consistently reported in type 2 diabetes, cardiovascular disease, metabolic syndrome and neurodegenerative disease². Skeletal muscle sits at the centre of this picture. It’s the body’s largest metabolic organ, the main site of insulin-stimulated glucose disposal, and the tissue in which mitochondrial plasticity is most visible. The heart depends on mitochondria even more absolutely, deriving around 90% of its ATP from oxidative metabolism⁴.

The Energy Cost of Exercise

During exercise, oxygen consumption per gram of skeletal muscle can rise more than tenfold1, and ATP must be resupplied as fast as myosin and ion pumps consume it. The heart has an even narrower margin - it holds almost no ATP reserve, and without continuous resynthesis its stores would be exhausted within 2–10 seconds4. Meeting this demand has a cost - faster electron flow through the respiratory chain raises ROS production, calcium cycles rapidly between the sarcoplasmic reticulum and cytosol, and individual mitochondria are pushed close to their functional limits. A cell that responded to every bout of exercise only by working harder would accumulate damaged organelles, oxidised lipids and mitochondrial DNA (mtDNA) lesions over time. 

The cell treats each exercise bout as information. Rather than simply tolerating the stress, it reads the signals that stress generates and uses them to rebuild the mitochondrial pool. Two opposing processes run in parallel: biogenesis, which adds new mitochondrial protein and membrane to the existing network, and mitophagy, which removes segments that have lost membrane potential or are emitting excess ROS1. Fusion and fission connect the two, allowing healthy components to be shared and damaged ones to be isolated. The result is accelerated turnover. Hood and colleagues describe this as the fundamental life cycle of the organelle, and exercise is its most powerful physiological accelerator1. 

Successful Cellular Adaptation 

When the stimulus is repeated over months, muscle carries more mitochondria with a higher density of respiratory chains. Holloszy’s landmark 1967 study showed that training roughly doubled muscle respiratory capacity while mitochondrial content rose by 60–80%, indicating that each mitochondrion also worked better¹,⁶. Trained muscle relies more on fatty acid oxidation, spares glycogen and becomes more sensitive to ADP, because a larger mitochondrial pool shares the load¹.

Adaptation is also qualitative. Trained muscle emits less ROS during acute exercise, reflecting tighter control of the electron transport chain¹. Strength athletes show a distinct remodelling: similar mitochondrial volume to untrained controls, but smaller mitochondria with greater cristae density and surface-to-volume ratio⁷. Exercise therefore shapes mitochondria to the demand placed on them, not to a single template.

What Exercise Does to Mitochondria

Four processes explain most of what exercise does to mitochondria. They run concurrently and feed into one another, so the separation below is for clarity rather than a strict sequence. 

Exercise signals both build and clear mitochondria An exercise bout raises calcium, ROS and the AMP:ATP ratio, activating CaMK, p38 MAPK and AMPK. These signals converge on PGC-1 alpha to build and connect new mitochondria, while AMPK acting via ULK1 drives fission, mitophagy and lysosomal clearance. Together they produce an adapted mitochondrial network with more mitochondria, higher respiration and lower ROS emission. Exercise signals both build and clear mitochondria Exercise bout contraction raises energy and redox demand Calcium release activates CaMK ROS rise activates p38 MAPK AMP:ATP rises activates AMPK via ULK1 converge on PGC-1α Build and connect PGC-1α coactivates NRF-1/2 and Tfam New proteins imported via TOM/TIM Fusion via Mfn1/2 and OPA1 Isolate and clear Fission via Drp1 isolates damage Mitophagy via PINK1/Parkin, BNIP3/NIX Lysosomal capacity rises via TFEB Adapted mitochondrial network More mitochondria, higher respiration, lower ROS emission
Figure 1. Calcium, ROS and AMP signals converge on PGC-1α to build and connect new mitochondria, while AMPK–ULK1 signalling and fission route damaged segments to mitophagy. Adapted from Hood et al.1
Dysregulated process Conditions investigated Why they may be connected
Reduced biogenesis and mitochondrial content with inactivity Type 2 diabetes, insulin resistance, metabolic syndrome Lower oxidative capacity is associated with impaired substrate handling; mitochondrial change is one of several contributors, not a sole cause2
Fragmentation, impaired mitophagy and ROS emission in muscle Age-related muscle loss (sarcopenia), frailty, disuse atrophy Mitochondrial changes coincide with, and may precede, loss of muscle mass in animal models; human causality is not established1
Cardiac mitochondrial dysfunction, calcium mishandling, inflammasome activation Heart failure, coronary artery disease, hypertension Strong mechanistic data from animal models; human evidence is mainly associative4
Vascular mitochondrial ROS and endothelial dysfunction Peripheral artery disease, chronic kidney disease, vascular ageing Mitochondrial ROS impair nitric oxide signalling; supported by targeted-antioxidant trials (Section 10)21,22
Circulating mitochondrial DAMPs (cell-free mtDNA) Chronic low-grade inflammation mtDNA can activate innate immunity; exercise-related changes are poorly characterised5

Biomarkers and Functional Readouts 

Because no single test captures mitochondrial health, the most useful approach pairs a whole-person functional measure with, where available, a marker of oxidative stress or vascular function, and interprets changes over time rather than single values. 

Measure What it may indicate Key limitation
VO2max or VO2peak (cardiopulmonary exercise test) Integrated aerobic capacity; percentage gains in mitochondrial content run about 2–3 times larger than VO2max gains3 Reflects cardiac output and oxygen delivery as well as mitochondria; effort-dependent
Muscle oxygen recovery kinetics (near-infrared spectroscopy) Non-invasive estimate of muscle oxidative capacity Equipment and protocol vary; mainly a research tool
Citrate synthase activity, cardiolipin, mitochondrial volume density (muscle biopsy) Mitochondrial content; citrate synthase and cardiolipin correlate best with content1 Invasive; content does not equal function of individual mitochondria
High-resolution respirometry (biopsy) Mitochondrial respiration and H2O2 emission Specialist laboratories only; ex vivo conditions
Plasma F2-isoprostanes, lipid hydroperoxides Systemic lipid peroxidation, including after exercise Not tissue-specific; timing relative to exercise matters
Flow-mediated dilation Endothelial function, partly influenced by vascular mitochondrial ROS Operator-dependent; affected by meals, caffeine and menstrual phase
Cell-free mtDNA Mitochondrial damage-associated signalling Rises transiently with strenuous exercise; no clinical reference ranges5
Patient-reported fatigue, energy and vitality Clinically meaningful whole-person outcome Subjective; not specific to mitochondria10

Supporting and Disrupting Factors 

The strongest human evidence in this field concerns exercise itself, and the most practical findings come from Mølmen and colleagues' meta-regression of 5,973 participants3. 

Factor Potential role Evidence and cautions
Training frequency More sessions per week produce larger gains in mitochondrial content and VO2max (6 > 4 > 2 sessions/week) Strong human meta-regression3; build frequency gradually in deconditioned patients
Training intensity Per hour of exercise, sprint interval training is about 3.9 times, and high-intensity training about 1.7 times, as efficient as continuous endurance training for mitochondrial content3 Total content gains are similar across formats once load is matched; intensity suits time-poor, medically cleared patients
Continuous endurance training Larger gains in capillary density; steady improvement over more weeks Strong human evidence3; well tolerated and suitable for most patients
Resistance training Remodels mitochondrial structure (cristae density, surface area) and preserves muscle mass Cross-sectional human data7; effects on mitochondrial content are less consistent5
Baseline fitness Lowest-fitness patients show the largest percentage gains Trainability is preserved regardless of age, sex, menopause or disease3
Prolonged sedentary time and disuse Rapid loss of mitochondrial content and rise in ROS Mainly animal mechanism with consistent human observation1,2
High-dose vitamin C and E around training May blunt redox signals needed for adaptation Human RCT and animal data8,9; does not apply to dietary antioxidant intake
Exhaustive or poorly recovered training Shifts dynamics towards fission, impairs respiration Mostly animal data4; relevant to overreaching and low-reserve patients

How to discuss supplements 

The key question for any compound is not only whether it lowers an oxidative stress marker, but whether it preserves or blunts training adaptation. That distinction is the main reason mitochondria-targeted antioxidants are studied separately from broad-spectrum antioxidants (Section 10). Any discussion should state the studied population, dose, duration and outcome, and should acknowledge that a biomarker change does not guarantee a functional or clinical benefit. 

The evidence supports exercise as a first-line mitochondrial intervention for almost every patient. Mixing continuous and interval formats covers both capillary and mitochondrial adaptation, and age or chronic disease is not a reason to expect a poor response3. It is reasonable to avoid routine high-dose antioxidant stacks around training sessions. It's not appropriate to infer a patient's mitochondrial status from symptoms alone, or to treat a single oxidative stress result as diagnostic. Patients with cardiovascular, renal or metabolic disease, or suspected primary mitochondrial disease, warrant medical clearance and, formal exercise testing before intensive training where relevant. 

Supplements Studied Alongside Exercise 

A range of supplements have been studied for their effects on mitochondria, exercise performance or recovery. The table below summarises those with the most relevant human evidence. Several are commonly used in clinical practice, but few have been tested in combination with a structured training programme, and fewer still with direct measures of mitochondrial function. Mitoquinol is discussed separately in Section 10. 

Supplement Proposed mitochondrial rationale Human evidence alongside exercise Practitioner interpretation
Creatine monohydrate Expands the phosphocreatine pool that buffers ATP and shuttles energy between mitochondria and myofibrils With resistance training in older adults (357 participants), creatine produced greater gains in fat-free mass, chest and leg press strength and chair-stand performance than training alone11. A low dose (0.03 g/kg/day, about 2.3 g) for 6 weeks improved fatigue resistance without weight gain12 The strongest human evidence in this group. Supports training quality and muscle mass rather than mitochondrial content directly
Dietary nitrate (e.g. beetroot) Increases nitric oxide availability and may improve the efficiency of oxidative phosphorylation In a double-blind crossover trial, inorganic nitrate improved the efficiency of muscle mitochondria (P/O ratio), and the improvement correlated with a lower oxygen cost of exercise13 Plausible, food-based option targeting efficiency rather than mitochondrial content; long-term training studies with mitochondrial outcomes are limited
Coenzyme Q10 Electron carrier in the respiratory chain and lipid-soluble antioxidant 200 mg/day for 6 weeks in middle-aged men mildly lowered muscle mitochondrial H2O2 without changing respiration or mitochondrial content29. Bioavailability and tissue uptake remain unresolved questions14 Biologically plausible, but little evidence that it enhances training adaptation in healthy people
PQQ (pyrroloquinoline quinone) Proposed cofactor that supports mitochondrial biogenesis signalling 20 mg/day with 6 weeks of endurance training in untrained men raised muscle PGC-1α protein more than placebo, but VO2peak and performance gains were no greater15. 21.5 mg/day for 12 weeks improved leg strength and walking tests in healthy adults around 54 years, without a training programme16 Early evidence. Marker changes have not yet translated into greater training gains
NAD+ precursors (e.g. NMN) Restore NAD+ for oxidative metabolism and sirtuin activity In 48 amateur runners training 5–6 times a week, 600–1,200 mg/day NMN for 6 weeks improved ventilatory-threshold measures more than training alone; VO2max was unchanged17 A single small trial; promising but needs replication and mitochondrial outcome measures
Urolithin A Gut-derived metabolite that induces mitophagy 1,000 mg/day for 4 months in adults aged 65–90 did not meet its primary endpoints (6-minute walk, muscle ATP production). Muscle endurance improved at 2 months, and acylcarnitines, ceramides and CRP fell18 Mechanistically aligned with mitochondrial quality control; not yet tested alongside a training programme
Resveratrol Activates AMPK and SIRT1, mimicking some exercise signals 150 mg/day for 30 days improved muscle mitochondrial respiration on fatty acid substrates in obese men19. However, 250 mg/day during 8 weeks of training in older men blunted training-induced improvements in blood pressure and VO2max20 Caution when combined with training; an example of a supplement that may interfere with adaptation
Omega-3 fatty acids (EPA/DHA) Incorporate into mitochondrial and cell membranes; modulate inflammation 600 mg EPA plus 260 mg DHA daily for 4 weeks partially reduced markers of muscle damage (range-of-motion loss, creatine kinase) after eccentric exercise21 Relevant to recovery and muscle damage rather than mitochondrial adaptation
High-dose vitamins C and E Broad-spectrum antioxidants that scavenge ROS 1,000 mg vitamin C plus 400 IU vitamin E daily blunted training-induced gains in insulin sensitivity and PGC-1α induction8; vitamin C reduced mitochondrial biogenesis in rats9 Avoid routine high doses around training; dietary antioxidant intake is not implicated

A pattern emerges from this evidence. Supplements that support energy buffering or delivery, such as creatine and dietary nitrate, have the most consistent functional data. Those aimed at mitochondrial signalling or biogenesis, such as PQQ, NAD+ precursors and urolithin A, change biomarkers but have not reliably improved outcomes beyond training alone. Broad antioxidants and resveratrol show that a supplement can work against training adaptation, which is why the effect on adaptation, not just on a biomarker, is the key question for any compound used alongside exercise. 

Practitioner Interpretation 

The evidence supports exercise as a first-line mitochondrial intervention for almost every patient. Frequency is a reliable lever, mixing continuous and interval formats covers both capillary and mitochondrial adaptation, and age or chronic disease is not a reason to expect a poor response³. It is reasonable to avoid routine high-dose antioxidant stacks around training sessions. It is not reasonable to infer a patient’s mitochondrial status from symptoms alone, or to treat a single oxidative stress result as diagnostic. Patients with cardiovascular, renal or metabolic disease, or suspected primary mitochondrial disease, warrant medical clearance and, where relevant, formal exercise testing before intensive training.

Exercise Protocols for Mitochondrial Health 

No single protocol is best for every patient. Continuous, interval and sprint training all raise mitochondrial content to a similar extent once the training load is matched3. The protocols below have been tested in human trials with mitochondrial or metabolic outcomes and can be adapted in clinic. 

Protocol Example session Weekly dose in trials What the evidence shows Best suited to
Moderate continuous endurance training 30–60 min of cycling, brisk walking or jogging at a conversational pace (below the second ventilatory threshold, roughly under 87% of maximum heart rate3); 40–60% of VO2peak in clinical trials34 3 or more sessions/week; benefits keep accruing over months3 Mitochondrial content up about 23% and the largest gains in capillary density3; improved oxidative, inflammatory and cardiac markers in hypertensive men after 6 weeks34 Deconditioned, older or cardiometabolic patients; the foundation of most programmes
Low-volume HIIT (1-minute intervals) 10 × 60 s at about 90% of maximum heart rate, with 60–75 s easy recovery; about 20–25 min including warm-up23,28 3 sessions/week In type 2 diabetes, 6 sessions over 2 weeks raised citrate synthase by about 20% and Mfn2 by about 71%, and lowered average 24 h glucose from 7.6 to 6.6 mmol/L23 Time-poor patients, including those with type 2 diabetes, once medically cleared
Longer-interval HIIT (4 × 4) 4 × 4 min at 90–95% of maximum heart rate (or above 90% of VO2peak), with 3 min active recovery24,26 3 sessions/week, often with 1–2 moderate sessions added24 Over 12 weeks, HIIT increased muscle mitochondrial respiration in both young and older adults and reversed many age-related differences in mitochondrial proteins24 Healthy adults, including older adults, aiming for maximal aerobic and mitochondrial gains
Sprint interval training 4–6 × 30 s all-out efforts, with 4 min easy recovery22 3 sessions/week with a rest day between sessions22 6 sessions over 2 weeks raised citrate synthase by 38% and doubled cycling endurance22; the most time-efficient format, but gains plateau after about 2 weeks unless volume increases3 Fit, motivated, medically cleared patients; not a starting point for frail or high-risk patients
Resistance training Progressive multi-joint resistance exercise for the major muscle groups 2–4 sessions/week Preserves muscle mass and insulin sensitivity and remodels mitochondrial structure7,24; on its own it did not increase mitochondrial respiration, whereas combined training did24 Everyone, especially older adults; combine with aerobic work for mitochondrial gains

Programming Principles 

  • Frequency is the most reliable lever: Mitochondrial content and VO₂max rose more with 6 than with 4, and with 4 than with 2, sessions per week³. For patients starting from a low base, adding sessions matters more than making each one harder.

  • Use intensity for fast gains and volume for continued gains: High-intensity and sprint training produce the fastest early increases, but further gains over months depend on accumulating more training time³. A practical progression is to begin with continuous training, add one interval session per week, then increase total weekly time.

  • Expect early, measurable change: Mitochondrial content can rise by 13–22% and VO₂max by 4–8% within 2 weeks, depending on the format³. Capillary gains occur mostly within the first 4 weeks³. Early wins can support adherence.

  • Prescribe intensity by feel as well as by numbers: Heart-rate and VO₂ thresholds were defined mainly for trained people, so untrained patients may experience “moderate” percentages as hard³. Rating of perceived exertion and the talk test are practical complements.

  • Build in recovery: Effective sprint protocols used rest days between sessions²². Exhaustive or poorly recovered training impairs mitochondrial function in animal models⁴, and hard intervals transiently increase mtDNA damage in humans²⁶.

  • Keep the redox signal intact: Avoid routine high-dose vitamin C and E around training sessions⁸,⁹.

An example week consistent with this evidence is 2–3 moderate continuous sessions, 1 interval session (low-volume HIIT or 4 × 4) and 2 resistance sessions. Patients with cardiovascular, renal or metabolic disease should have medical clearance before interval or sprint training.

Current Evidence 

  • Exercise training increases skeletal muscle mitochondrial content, respiratory capacity and capillarisation in humans, and these changes track with improvements in VO₂max¹,³. The magnitude of adaptation depends mainly on baseline fitness and training load (intensity multiplied by volume), and relative trainability is maintained across the lifespan³. Inactivity rapidly reverses these gains¹. Exercise also improves patient-reported energy and fatigue¹⁰.

  • The relative importance of intensity and volume remains debated; Mølmen and colleagues’ data suggest both matter and that one can partly compensate for the other³. Researchers are moving from static protein markers towards measures of mitophagy flux, and from biopsy-based content markers towards non-invasive functional readouts¹. The roles of the mitochondrial unfolded protein response, lysosomal biogenesis and extracellular mitochondria in exercise adaptation are active questions¹,⁵,⁷.

  • Most mechanistic understanding of dynamics, mitophagy and cardiac adaptation comes from rodents, and human adaptations often differ in magnitude and timing⁴. Human cardiac mitochondria are rarely sampled. The threshold at which exercise-induced ROS shift from useful signal to harmful excess is not defined for any individual, and the way supplements interact with that threshold is only beginning to be studied in controlled trials.

Why Mitoquinol Is Relevant 

If exercise adaptation depends on ROS signals, the obvious question is whether any antioxidant can reduce mitochondrial oxidative damage without silencing those signals. Mitoquinol mesylate (MitoQ) offers a way to test this. It is a ubiquinone attached to a lipophilic triphenylphosphonium cation, which drives accumulation inside mitochondria in response to membrane potential. There the ubiquinone is reduced to its active ubiquinol form by the respiratory chain and recycled after neutralising lipid peroxyl radicals²⁵. Its action is therefore concentrated at the inner mitochondrial membrane rather than spread across the whole cell, unlike vitamins C and E.

Three findings are most informative for practitioners.

  1. Training adaptations appear to be preserved. In untrained middle-aged men, 20 mg/day MitoQ augmented the acute exercise-induced rise in muscle PGC-1α mRNA and increased training-induced gains in peak power, while gains in mitochondrial content, VO₂peak and 20 km time trial performance were similar to placebo²⁸. Earlier work also found no impairment of VO₂max or oxidative capacity adaptations to endurance training³⁰. This contrasts with the blunting reported for high-dose vitamins C and E⁸.

  2. Exercise-induced oxidative damage is reduced. Twenty-one days of MitoQ attenuated high-intensity exercise-induced mtDNA and nuclear DNA damage in muscle and lymphocytes; a single dose one hour before exercise did not²⁶. In trained middle-aged cyclists, 28 days of MitoQ lowered post-exercise F₂-isoprostanes and produced a 1.3% faster 8 km time trial, with 4.4% higher mean power²⁷. A 2024 meta-analysis of eight studies (188 participants) confirmed a reduction in exercise-induced oxidative damage but no overall improvement in endurance performance³¹.

  3. Signals of benefit in vascular-limited populations. In peripheral artery disease, a single dose improved endothelial function, maximal walking time and time to claudication³². In chronic kidney disease, four weeks improved flow-mediated dilation but not exercise capacity³³. In hypertensive men, MitoQ and endurance training each improved oxidative, inflammatory and echocardiographic markers, with the largest changes when combined³⁴.

Study Population MitoQ protocol Main exercise-related finding
Broome et al. 202127 19 trained male cyclists, 44 ± 4 years (crossover) 20 mg/day, 28 days 8 km time trial 1.3% faster, mean power 4.4% higher; post-exercise F2-isoprostanes lower
Broome et al. 202228 23 untrained men, 44 ± 7 years 20 mg/day, 10 days then 3 weeks HIIT Augmented PGC-1α mRNA after exercise; greater peak power gain; mitochondrial content gains unchanged
Williamson et al. 202026 24 healthy men 20 mg acute and 21 days Chronic dosing attenuated exercise-induced mtDNA damage; acute dosing ineffective
Pham et al. 202029 20 men, about 50 years 20 mg/day, 6 weeks (vs CoQ10 200 mg) Mild reduction in muscle mitochondrial H2O2; respiration and content unchanged
Bispham et al. 201736 18 adults, 60–79 years (crossover; conference abstract) 20 mg/day, 6 weeks Leg-extension power 11% higher; VO2max unchanged
Masoumi-Ardakani et al. 202234 52 hypertensive men, 40–55 years 20 mg/day, 6 weeks, ± endurance training Combination produced the largest improvements in TAC, MDA, IL-6 and cardiac indices
Park et al. 202032 11 adults with peripheral artery disease (crossover) Single acute dose Maximal walking time +73.8 s; time to claudication +44.2 s; FMD improved
Kirkman et al. 202333 18 adults with chronic kidney disease 20 mg/day, 4 weeks FMD improved; exercise capacity unchanged
Hughes et al. 202335 9 physically inactive women (crossover) 80 mg single dose VO2max lower than placebo (21.0 vs 23.5 mL/kg/min)

Research Opportunities and Gaps 

Most trials are small, short and predominantly male, and only a handful combined MitoQ with a structured training programme²⁸,³⁰,³⁴. The meta-analysis authors call for 4–16 week training studies and dose comparisons³¹. The acute 80 mg finding in inactive women³⁵ is a reminder that dose, timing and training status may change the outcome. MitoQ is also a useful research tool: because it acts on mitochondrial rather than whole-cell redox state, it can help separate the contributions of mitochondrial ROS to adaptation, fatigue and vascular function. Whether these effects translate into long-term functional or clinical outcomes remains unknown.

Key Takeaways 

  • Exercise is a signalling event: calcium, AMP and ROS converge on PGC-1α to coordinate mitochondrial biogenesis, remodelling and mitophagy. 

  • Training produces more and better mitochondria, with content gains of about 23–27% across endurance, high-intensity and sprint formats; frequency and training load are the most reliable levers. 

  • Inactivity and ageing fragment the network and blunt its signalling, yet relative trainability is preserved across age, sex and chronic disease. 

  • ROS are part of the adaptive signal, so high-dose broad-spectrum antioxidants around training may be counterproductive. 

  • Mitoquinol reduces exercise-induced oxidative damage without evidence of blunting adaptation, but performance effects are inconsistent and long-term outcomes are untested. 

References

  1. Hood DA, Memme JM, Oliveira AN, Triolo M. Maintenance of skeletal muscle mitochondria in health, exercise, and aging. Annu Rev Physiol. 2019;81:19–41. doi:10.1146/annurev-physiol-020518-114310 

  2. San-Millán I. The key role of mitochondrial function in health and disease. Antioxidants. 2023;12(4):782. doi:10.3390/antiox12040782 

  3. Mølmen KS, Almquist NW, Skattebo Ø. Effects of exercise training on mitochondrial and capillary growth in human skeletal muscle: a systematic review and meta-regression. Sports Med. 2025;55:115–144. doi:10.1007/s40279-024-02120-2 

  4. Cai J, Wang T, Li S. Mitochondria: the central hub linking exercise to enhanced cardiac function. Front Physiol. 2026;17:1747133. doi:10.3389/fphys.2026.1747133 

  5. Pedrosa MB, Santos LL, Ferreira R, Magalhães J. Extracellular mitochondria: a potential player involved in exercise health benefits. Biochimie. 2026;242:97–107. doi.org/10.1016/j.biochi.2025.12.011

  6. Holloszy JO. Biochemical adaptations in muscle: effects of exercise on mitochondrial oxygen uptake and respiratory enzyme activity in skeletal muscle. J Biol Chem. 1967;242(9):2278–2282. doi.org/10.1016/S0021-9258(18)96046-1

  7. Botella J, Schytz CT, Pehrson TF, et al. Increased mitochondrial surface area and cristae density in the skeletal muscle of strength athletes. J Physiol. 2023;601(14):2899–2915. doi:10.1113/JP284394 

  8. Ristow M, Zarse K, Oberbach A, et al. Antioxidants prevent health-promoting effects of physical exercise in humans. Proc Natl Acad Sci USA. 2009;106(21):8665–8670. doi:10.1073/pnas.0903485106

  9. Gomez-Cabrera MC, Domenech E, Romagnoli M, et al. Oral administration of vitamin C decreases muscle mitochondrial biogenesis and hampers training-induced adaptations in endurance performance. Am J Clin Nutr. 2008;87(1):142–149. doi:10.1093/ajcn/87.1.142

  10. Wender CLA, Manninen M, O'Connor PJ. The effect of chronic exercise on energy and fatigue states: a systematic review and meta-analysis of randomized trials. Front Psychol. 2022;13:907637. doi:10.3389/fpsyg.2022.907637 

  11. Devries MC, Phillips SM. Creatine supplementation during resistance training in older adults: a meta-analysis. Med Sci Sports Exerc. 2014;46(6):1194–1203. doi:10.1249/MSS.0000000000000220 

  12. Rawson ES, Stec MJ, Frederickson SJ, Miles MP. Low-dose creatine supplementation enhances fatigue resistance in the absence of weight gain. Nutrition. 2011;27(4):451–455. doi:10.1016/j.nut.2010.04.001 

  13. Larsen FJ, Schiffer TA, Borniquel S, et al. Dietary inorganic nitrate improves mitochondrial efficiency in humans. Cell Metab. 2011;13(2):149–159. doi:10.1016/j.cmet.2011.01.004

  14. Mantle D, Lopez-Lluch G, Hargreaves IP. Coenzyme Q10 metabolism: a review of unresolved issues. Int J Mol Sci. 2023;24(3):2585. doi:10.3390/ijms24032585 

  15. Hwang PS, Machek SB, Cardaci TD, et al. Effects of pyrroloquinoline quinone (PQQ) supplementation on aerobic exercise performance and indices of mitochondrial biogenesis in untrained men. J Am Coll Nutr. 2020;39(6):547–556. doi:10.1080/07315724.2019.1705203 

  16. Shiojima Y, Takahashi M, Takahashi R, et al. Efficacy and safety of a novel dietary pyrroloquinoline quinone disodium salt on muscle strength and physical function in healthy volunteers: a randomized, double-blind, placebo-controlled study. J Funct Foods. 2024;113:106012. doi:10.1016/j.jff.2024.106012 

  17. Liao B, Zhao Y, Wang D, Zhang X, Hao X, Hu M. Nicotinamide mononucleotide supplementation enhances aerobic capacity in amateur runners: a randomized, double-blind study. J Int Soc Sports Nutr. 2021;18(1):54. doi:10.1186/s12970-021-00442-4 

  18. 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. doi:10.1001/jamanetworkopen.2021.44279 

  19. Timmers S, Konings E, Bilet L, et al. Calorie restriction-like effects of 30 days of resveratrol supplementation on energy metabolism and metabolic profile in obese humans. Cell Metab. 2011;14(5):612–622. doi:10.1016/j.cmet.2011.10.002 

  20. Gliemann L, Schmidt JF, Olesen J, et al. Resveratrol blunts the positive effects of exercise training on cardiovascular health in aged men. J Physiol. 2013;591(20):5047–5059. doi:10.1113/jphysiol.2013.258061 

  21. Tsuchiya Y, Ueda H, Yanagimoto K, Kato A, Ochi E. 4-week eicosapentaenoic acid-rich fish oil supplementation partially protects muscular damage following eccentric contractions. J Int Soc Sports Nutr. 2021;18(1):18. doi:10.1186/s12970-021-00411-x 

  22. Burgomaster KA, Hughes SC, Heigenhauser GJF, Bradwell SN, Gibala MJ. Six sessions of sprint interval training increases muscle oxidative potential and cycle endurance capacity in humans. J Appl Physiol. 2005;98(6):1985–1990. doi:10.1152/japplphysiol.01095.2004 

  23. Little JP, Gillen JB, Percival ME, et al. Low-volume high-intensity interval training reduces hyperglycemia and increases muscle mitochondrial capacity in patients with type 2 diabetes. J Appl Physiol. 2011;111(6):1554–1560. doi:10.1152/japplphysiol.00921.2011 

  24. Robinson MM, Dasari S, Konopka AR, et al. Enhanced protein translation underlies improved metabolic and physical adaptations to different exercise training modes in young and old humans. Cell Metab. 2017;25(3):581–592. doi:10.1016/j.cmet.2017.02.009 

  25. Kelso GF, Porteous CM, Coulter CV, et al. Selective targeting of a redox-active ubiquinone to mitochondria within cells: antioxidant and antiapoptotic properties. J Biol Chem. 2001;276(7):4588–4596. doi:10.1074/jbc.M009093200 

  26. Williamson J, Hughes CM, Cobley JN, Davison GW. The mitochondria-targeted antioxidant MitoQ attenuates exercise-induced mitochondrial DNA damage. Redox Biol. 2020;36:101673. doi:10.1016/j.redox.2020.101673 

  27. Broome SC, Braakhuis AJ, Mitchell CJ, Merry TL. Mitochondria-targeted antioxidant supplementation improves 8 km time trial performance in middle-aged trained male cyclists. J Int Soc Sports Nutr. 2021;18(1):58. doi:10.1186/s12970-021-00454-0 

  28. Broome SC, Pham T, Braakhuis AJ, et al. MitoQ supplementation augments acute exercise-induced increases in muscle PGC1α mRNA and improves training-induced increases in peak power independent of mitochondrial content and function in untrained middle-aged men. Redox Biol. 2022;53:102341. doi:10.1016/j.redox.2022.102341 

  29. Pham T, MacRae CL, Broome SC, et al. MitoQ and CoQ10 supplementation mildly suppresses skeletal muscle mitochondrial hydrogen peroxide levels without impacting mitochondrial function in middle-aged men. Eur J Appl Physiol. 2020;120(7):1657–1669. doi:10.1007/s00421-020-04396-4 

  30. Shill DD, Southern WM, Willingham TB, et al. Mitochondria-specific antioxidant supplementation does not influence endurance exercise training-induced adaptations in circulating angiogenic cells, skeletal muscle oxidative capacity or maximal oxygen uptake. J Physiol. 2016;594(23):7005–7014. doi.org/10.1113/JP272491

  31. Gonzalo-Skok O, Casuso RA. Effects of mitoquinone (MitoQ) supplementation on aerobic exercise performance and oxidative damage: a systematic review and meta-analysis. Sports Med Open. 2024;10:77. doi:10.1186/s40798-024-00741-5 

  32. Park SY, Pekas EJ, Headid RJ, et al. Acute mitochondrial antioxidant intake improves endothelial function, antioxidant enzyme activity, and exercise tolerance in patients with peripheral artery disease. Am J Physiol Heart Circ Physiol. 2020;319(2):H456–H467. doi:10.1152/ajpheart.00235.2020 

  33. Kirkman DL, Stock JM, Shenouda N, et al. Effects of a mitochondrial targeted ubiquinol (MitoQ) on vascular function and exercise capacity in chronic kidney disease: a randomized controlled pilot study. Am J Physiol Renal Physiol. 2023. doi:10.1152/ajprenal.00067.2023

  34. Masoumi-Ardakani Y, Najafipour H, Nasri HR, et al. Moderate endurance training and MitoQ improve cardiovascular function, oxidative stress, and inflammation in hypertensive individuals: the role of miR-21 and miR-222. Cell J. 2022;24(10):577–585. doi:10.22074/cellj.2022.8089 

  35. Hughes et al. Mitochondrial-targeted antioxidant ingestion acutely blunts VO2max in physically inactive females. Physiol Rep. 2023.

  36. Bispham NZ, Santos-Parker JR, Steward CAC, et al. MitoQ supplementation improves leg-extension power in healthy late middle-aged and older adults. FASEB J. 2017;31(S1):lb852 (conference abstract). doi:10.1096/fasebj.31.1_supplement.lb852 

Turning Contraction Into a Signal 

Three signals rise in proportion to the work being done:

  • Cytosolic calcium increases with every contraction and activates calcium-calmodulin kinase (CaMK).

  • ATP breakdown raises the AMP:ATP ratio, which activates AMP-activated protein kinase (AMPK).

  • Mitochondrial and cytosolic ROS rise and activate p38 mitogen-activated protein kinase (MAPK)¹.

All three converge on PGC-1α¹,⁵. This redundancy matters clinically, because exercise does not depend on any single pathway, which is one reason mitochondrial health can still be restored by training when individual regulators such as PGC-1α or p53 are deficient in experimental models¹.

Building capacity: coordinating two genomes 

A mitochondrion is assembled from roughly 1,200 proteins, yet mtDNA encodes only 13 of them, around 1% of the total¹. Biogenesis therefore requires tight coordination between the nuclear and mitochondrial genomes. PGC-1α coactivates nuclear respiratory factors 1 and 2 (NRF-1/2), increasing transcription of nuclear genes encoding mitochondrial proteins. Mitochondrial transcription factor A (Tfam), assisted by p53, drives mtDNA transcription and replication¹.

New proteins must then be imported across both membranes via the TOM and TIM translocase complexes. Chronic exercise increases the expression and efficiency of this import machinery, whereas disuse impairs it¹. Biogenesis is thus a supply chain, and exercise upgrades every step of it.

Quality control: remodelling and removing 

  • In muscle, mitochondria form an interconnected reticulum that distributes membrane potential rapidly across the fibre¹.

  • Fusion, mediated by mitofusins 1 and 2 (Mfn1/2) and OPA1, allows mitochondria to share mtDNA, proteins and metabolites. Fission, mediated by Drp1 and its receptors, separates damaged segments.

  • Inter-mitochondrial junctions act as circuit breakers, electrically isolating failing organelles so the wider network is protected¹.

  • Isolated segments are cleared by mitophagy. When membrane potential collapses, PINK1 accumulates on the outer membrane and recruits Parkin, which tags the organelle for engulfment and lysosomal degradation; BNIP3 and NIX provide a parallel receptor-mediated route¹,⁵.

  • In mice, acute treadmill exercise increases mitophagy about 6 hours later through AMPK and ULK1, and exercise also activates the lysosomal regulator TFEB¹. Endurance training favours fusion overall, increasing OPA1 and Mfn2 and lowering Drp1¹.

Redox signalling: why some oxidative stress is necessary 

ROS are often framed as damage, but during exercise they are also messengers. ROS activate p38 MAPK and the transcription factors that regulate PGC-1α, and the outcome depends on dose: moderate ROS can elaborate the mitochondrial network, while excess ROS fragments it¹.

This hormetic logic has direct clinical consequences:

  • In a frequently cited human trial, daily vitamin C (1,000 mg) and vitamin E (400 IU) during four weeks of training prevented exercise-induced improvements in insulin sensitivity and the induction of PGC-1α⁸.

  • Vitamin C alone reduced training-induced mitochondrial biogenesis and endurance gains in rats, although VO₂max gains in the human arm of that study were not significantly affected⁹.

The heart and circulating mitochondria 

The same principles operate in the heart. Regular exercise increases mitochondrial respiration, cardiolipin content and fusion in cardiomyocytes, improves calcium handling, and dampens ROS-driven NLRP3 inflammasome activation, largely in animal models⁴. A newer line of research asks whether exercise also changes extracellular mitochondria: whole organelles, mitochondria-derived vesicles and cell-free mtDNA in the circulation, much of it platelet-derived⁵. Strenuous exercise acutely raises circulating mtDNA in healthy people, but whether exercise-released mitochondria carry signals between tissues remains largely hypothetical⁵.

Real-World Example: A Middle-Aged Patient Starts Interval Training 

Consider a sedentary 45-year-old who begins three sessions a week of high-intensity interval training on a bike, a design used in several of the trials discussed below¹⁷.

  • The first session: Ten one-minute efforts drive calcium cycling, ATP turnover and ROS production well above resting levels. AMPK, CaMK and p38 MAPK are activated, and within hours PGC-1α mRNA rises in the working muscle¹,¹⁷. Animal data suggest mitophagy peaks around 6 hours after exercise, clearing organelles that did not cope well with the load¹.

  • The first weeks: Repeated pulses of transcription and turnover begin to add up - capillary growth occurs mostly within the first four weeks of training in untrained to moderately trained people³. Sprint and high-intensity formats produce the fastest early gains in mitochondrial content and VO₂max, while continuous endurance training improves more slowly but keeps improving over more weeks³.

  • The longer term: Across 353 studies, mitochondrial content rose by about 23% with endurance training, 27% with high-intensity training and 27% with sprint interval training after adjusting for frequency, duration and baseline fitness³. Our patient would likely see a meaningful rise in VO₂max, better fat oxidation at a given workload and, often, a lift in perceived energy: a meta-analysis of 81 randomised trials found small-to-moderate improvements in fatigue, energy and vitality with exercise training¹⁰.

  • Where it stops being adaptive: The same stimulus becomes harmful when it is excessive or recovery is inadequate. In rodents, exhaustive exercise reduces activity of respiratory complexes I, II and IV in the heart and shifts dynamics towards fission and mitochondrial loss⁴. In humans, high-intensity intervals measurably increase mtDNA and nuclear DNA damage in muscle and lymphocytes¹⁵. In healthy people this damage is repaired, but it illustrates why load must be matched to capacity in patients with limited reserve.

From adaptation to dysfunction 

The mitochondrial network fails when the signal is withdrawn, blunted or overwhelmed. Disuse is the fastest route. In denervation and immobilisation models, PGC-1α falls early, mitochondrial import of Tfam declines within three days, mtDNA copy number drops, and ROS emission rises¹. Elevated ROS then activate apoptotic and ubiquitin-proteasome pathways that drive muscle atrophy¹. The heart shows a parallel picture: sedentary conditions in animal models reduce fusion protein expression, impair mitochondrial calcium uptake and promote ROS-driven NLRP3 inflammasome activation⁴.

Ageing reaches a similar endpoint by a different route. Aged muscle shows increased fission protein expression, a fragmented reticulum, and reduced activation of AMPK, CaMK and p38 in response to exercise, so the same session produces a smaller PGC-1α response¹. Basal mitophagy may be elevated, but the acute exercise-induced response is blunted and lysosomal breakdown appears inefficient, with lipofuscin accumulation suggesting incomplete clearance¹. Physical activity level confounds many ageing studies, and part of the decline is intrinsic to ageing itself rather than inactivity alone¹.

In each case, the result is a smaller pool of lower-quality mitochondria that produce more ROS and less ATP per unit of oxygen.

Conditions Associated With This Biology