Mitochondrial Aging and Physical Decline: Insights From Three Generations of Women

Abstract

Decline in mitochondrial DNA (mtDNA) copy number, function, and accumulation of mutations and deletions have been proposed to contribute to age-related physical decline, based on cross sectional studies in genetically unrelated individuals. There is wide variability of mtDNA and functional measurements in many population studies and therefore we assessed mitochondrial function and physical function in 18 families of grandmothers, mothers, and daughters who share the same maternally inherited mtDNA sequence. A significant age-related decline in mtDNA copy number, mitochondrial protein expression, citrate synthase activity, cytochrome c oxidase content, and VO2 peak were observed. Also, a lower abundance of SIRT3, accompanied by an increase in acetylated skeletal muscle proteins, was observed in grandmothers. Muscle tissue-based full sequencing of mtDNA showed greater than 5% change in minor allele frequency over a lifetime in two locations, position 189 and 408 in the noncoding D-loop region but no changes were noted in blood cells mtDNA. The decline in oxidative capacity and muscle function with age in three generations of women who share the same mtDNA sequence are associated with a decline in muscle mtDNA copy number and reduced protein deacetylase activity of SIRT3.

Keywords: Acetylated protein; Aging; Mitochondrial DNA; Oxidative capacity; SIRT3; Skeletal muscle.

Figure 1.

Changes with age in whole-body VO₂ peak (A), skeletal muscle citrate synthase activity (B), mtDNA copy number in skeletal muscle (C) and blood (D), protein levels of total oxidative phosphorylation muscle proteins (E) PGC-1α expression (F), cytochrome c oxidase protein content (G), and activity (H). These data show significant correlation of age to oxidative capacity measured at the level of the whole-body (beta = −3.225; p < 0.001) and skeletal muscle (beta = 0.536; p < 0.001). There was significant correlation of age with muscle mtDNA (beta = −0.095; p = 0.009), but not with blood mtDNA (beta = 0.088; p = 0.54). Skeletal muscle expression levels of oxidative phosphorylation proteins were significantly correlated to age (beta = −1.153; p = 0.013) but not that of PGC-1α protein levels and age (beta = −0.063; p = 0.39). Beta represents the estimated change in each measurement per 10-year increase in age and is estimated from a hierarchal linear model with a random intercept for each family unit.

Figure 2.

Changes in minor allele frequency (MAF) by position number. Changes are based on a position-by-position analysis of using a logistic regression model with a quasi-binomial variance structure and family-specific intercept. The change in MAF was estimated from the fitted model for a 50-year difference (80 years vs 30 years of age). Changes >5% in the MAF over this 50-year period were considered clinical relevant. Only two locations (positions 189 and 408) reached this criterion, both in the noncoding D-loop region.

Figure 3.

Associations of total accelerations units (AU) with citrate synthase (CS) activity (Panel A) and VO₂ peak (Panel B). Greater total accelerations were associated with increased CS activity (beta = 306.3; p = 0.010) and VO₂ peak (beta = 109.1; p < 0.001). Beta represents the estimated change in AU for each unit increase in CS activity and VO₂ peak estimated from a hierarchal linear model with a random intercept for each family unit.

Figure 4.

The abundance of acetylated lysine residues was significantly increased in skeletal muscle proteins of grandmothers compared to daughters. The expression of mitochondrial-localized SIRT3 was decreased in grandmothers compared to daughters. Data represent means ± SEM. * denotes significant (p < 0.05) difference in a pairwise comparison of grandmothers and daughters.