Quadriceps mitochondrial DNA quantity, quality, and gene expression after 2 years of calorie restriction: exploratory results from the CALERIE trial

The Comprehensive Assessment of Long-term Effects of Reducing Intake of Energy (CALERIE)™ trial was a randomized, 2-year controlled trial of caloric restriction (CR) versus an ad libitum (AL) control condition in nonobese humans. We performed exploratory analyses of muscle mitochondrial DNA (mtDNA) integrity, one element of mitochondrial quality. Our aims were to assess the feasibility of this approach, explore associations to inform future hypotheses, estimate effect sizes and statistical power for subsequent studies, and contribute additional data to the CALERIE database. We used droplet digital PCR to quantitate the copy numbers of nuclear DNA, mtDNA, and mtDNA deletion mutations in remnant total DNA samples extracted from quadriceps biopsies at baseline (n = 93), 12 months (n = 44), and 24 months (n = 31). MtDNA copy number and mutation frequency were correlated with existing gene expression data from the same muscle biopsies. MtDNA copy number was lower in females (p = 0.0005) and declined over time (p = 0.0001), with no statistically significant differences observed for CR versus AL (p = 0.2898) or age across both groups (p = 0.4644). Baseline copy number correlated positively with baseline physiological measures including fat-free mass (r = 0.43, p = 2e-05), self-reported energy intake (r = 0.34, p = 0.00077), resting metabolic rate (r = 0.42, p = 0.00261), total energy expenditure (r = 0.31, p = 0.00261), and V̇O₂max (r = 0.41, p = 1e-04), and with gene expression related to mitochondrial function, while showing negative correlations with nuclear genome maintenance, RNA splicing, and ribosomes. MtDNA mutation frequency increased with age (p = 0.0041) and showed a weak negative correlation with V̇O₂max (r = -0.23, p = 3e-05). MtDNA mutation frequency was not statistically different with CR. Mutation frequency was positively correlated with gene expression changes in inflammation and negatively correlated with protein translation. We observed that, as expected, mtDNA copy number declined over time, was lower in females, and correlated with metabolic measures and mitochondrial gene expression, suggesting its potential utility as a marker of metabolic health. These findings contribute to understanding mitochondrial adaptations to CR and highlight potential benefits of CR interventions.