Calorie Restriction and Muscle mtDNA: What CALERIE Found

Moderate calorie restriction can change gene activity in human muscle, but the best human trial has not shown that it builds more mitochondrial DNA. That distinction matters if you are considering calorie restriction as a mitochondrial "upgrade." The biology appears to respond upstream; the commonly cited marker of mitochondrial genome abundance does not.

The evidence comes from CALERIE, a two-year randomized trial of calorie restriction in non-obese adults. An exploratory muscle-biopsy analysis measured three different aspects of mitochondrial DNA (mtDNA) in quadriceps: copy number, deletion frequency, and mitochondrial gene expression. They did not move in lockstep. Calorie restriction produced no measurable increase in mtDNA copy number versus an ad-libitum control diet, while deletion frequency and mitochondrial gene expression showed directionally favorable signals. The result is more useful than a blanket claim that calorie restriction either "fixes" or "fails" mitochondria.

What CALERIE actually tested

CALERIE randomly assigned non-obese adults to calorie restriction or an ad-libitum (eat-as-usual) control condition for two years. The protocol aimed for 25% calorie restriction, but participants achieved about 12% on average. That is a meaningful dietary change, but it is also less than half the intended intervention dose.

In the mtDNA substudy, researchers analyzed remnant DNA from quadriceps biopsies collected at baseline, 12 months, and 24 months. The sample fell from 93 participants at baseline to 44 at 12 months and 31 at 24 months. Using droplet digital PCR, the team quantified nuclear DNA, mtDNA, and mtDNA deletion mutations, then related those measures to existing muscle gene-expression data. The authors describe the analysis as exploratory in their 2026 CALERIE mtDNA report.

That label is not decorative. With roughly 31 paired samples at the final visit, the substudy could reliably detect only relatively large changes. The copy-number null result could mean that 12% calorie restriction truly has no effect on mtDNA copy number in this population. It could also mean that a smaller effect escaped detection. The same caution applies in reverse to the favorable directional findings: they generate hypotheses, but they do not confirm protection.

Three mitochondrial endpoints, three answers

1. mtDNA copy number did not increase

MtDNA copy number is a proxy for the amount of mitochondrial genome present in muscle. If moderate calorie restriction caused substantial net mitochondrial biogenesis, researchers might expect this measure to rise after two years. It did not show a statistically meaningful amplification in the calorie-restriction group relative to controls.

This is the cleanest challenge to the popular shorthand that eating less automatically means "more mitochondria." In healthy, non-obese adults under this specific two-year intervention, the trial did not detect that outcome. It does not establish that calorie restriction can never affect mitochondrial content, nor does it answer whether a larger achieved restriction would do so. CALERIE achieved about 12%, not its 25% target.

2. Deletion frequency moved in a favorable direction

The same analysis found a directionally favorable signal for mtDNA deletion frequency. Deletions are damaged mitochondrial genomes missing stretches of DNA; over time, damaged copies can expand within muscle fibers. This endpoint reflects genome integrity rather than mitochondrial quantity.

The distinction has practical relevance. In a 2021 observational study of older men and women, both skeletal-muscle mtDNA copy number and deletion mutation frequency independently predicted physical-performance decline. That finding does not prove that changing either biomarker through diet will preserve function. It does show why researchers care about both measures rather than treating copy number as the whole story.

Still, CALERIE did not measure exercise performance, electron-transport-chain activity, or fiber-level function as outcomes of this biopsy analysis. A favorable biomarker trend is not evidence that participants gained stronger or more durable muscle.

3. Muscle gene expression suggested biological activity

Calorie restriction also showed a positive directional signal for mitochondrial gene expression. That fits a companion CALERIE analysis: Das et al. reported in 2023 that about 12% achieved calorie restriction over two years altered human skeletal-muscle transcription of genes related to stress response and longevity pathways.

Gene expression is an early step, not a receipt for a completed mitochondrial renovation. RNA must still be translated into proteins, assembled into functioning organelles, and maintained. CALERIE's transcriptomic response alongside unchanged mtDNA copy number shows why those stages should not be collapsed into one claim. Calorie restriction engaged molecular programs in muscle; it did not produce a measurable increase in mitochondrial genome abundance.

Why the deletion result is hard to interpret

Age shapes the signal available to detect. Herbst and colleagues found in 2021 that mtDNA deletion mutations accumulate exponentially with age in human skeletal muscle. CALERIE enrolled adults roughly spanning their third through fifth decades, when absolute deletion burdens are generally low compared with later life.

That creates two plausible explanations for a small favorable trend. Calorie restriction might slow damage before deletion burden accelerates with age. Or measurement variability at low deletion frequencies might produce an apparent difference in a small sample. The trial cannot distinguish those explanations.

Biology also gives no license to infer decades of benefit from two years of data. The age-related deletion curve makes early prevention an interesting possibility, but CALERIE did not follow participants long enough to test whether calorie restriction changes that curve or prevents later disability.

How does exercise compare?

Exercise has a stronger mechanistic case for driving mitochondrial biogenesis because it combines energy demand, muscle contraction, and repeated metabolic stress. In untrained middle-aged men, a 2022 training study found that endurance exercise increased muscle PGC-1α messenger RNA and improved peak power; the study did not establish a direct exercise effect on mtDNA copy number or deletion frequency as primary endpoints.

Functional evidence for training is also easier to find than functional evidence for calorie restriction in this setting. A randomized HIIT study in obese older adults reported measurable changes in physical performance and skeletal-muscle composition. Its participants, intervention, and endpoints differ from CALERIE's, so it cannot provide a head-to-head mtDNA comparison. It does show that structured exercise can produce measurable muscle adaptation in a population with more room for functional improvement.

Other clinical studies reinforce that baseline biology matters. A 2025 Parkinson's disease study found skeletal-muscle complex I deficiency in a subgroup of patients, while a multiple-sclerosis study reported impaired muscle oxidative capacity that did not fully normalize after exercise training. Neither study tested calorie restriction. They illustrate why mitochondrial defects may be easier to identify in disease states than in healthy middle-aged adults with low baseline damage.

The practical verdict

CALERIE gives the first randomized, multi-year human biopsy evidence on calorie restriction and skeletal-muscle mtDNA, but it does not validate moderate calorie restriction as a mitochondrial-biogenesis tool. At about 12% achieved restriction over two years, participants showed transcriptional engagement and a favorable directional integrity signal, yet no measurable rise in mtDNA copy number.

For healthy, non-obese adults, the honest claim is narrow: moderate calorie restriction may alter muscle mitochondrial biology, but current randomized evidence does not show that it increases muscle mtDNA abundance or improves physical function.

If your goal is muscle capacity, exercise has the better direct functional evidence and the more complete biological rationale. The deletion and gene-expression signals from CALERIE deserve replication in a larger, adequately powered study that measures mitochondrial function and physical performance alongside mtDNA. Until then, "calorie restriction protects your mitochondrial DNA" is a headline looking for data.