Intrinsic aerobic capacity modulates Alzheimer's disease pathological hallmarks, brain mitochondrial function and proteome during aging

Abstract

Low aerobic capacity is strongly associated with all-cause mortality and risk for Alzheimer's disease (AD). Individuals with early dementia and AD have lower aerobic capacity compared to age-matched controls. The mechanism by which aerobic capacity influences AD risk is unknown but is likely mediated by sexual dimorphism and tissue-level differences in mitochondrial energetics. Here, we used rats selectively bred for large differences in intrinsic aerobic exercise capacity. Brain tissue from 18-month and 24-month-old female and male low-capacity runner (LCR) and high-capacity runner (HCR) rats were analyzed for markers of mitochondrial function and AD-associated pathologies. LCR rats, irrespective of sex, exhibited a greater increase in brain amyloid beta (Aβ₄₂) and tau hyperphosphorylation (pTau^thr181/total tau) with aging. In female LCR rats, brain mitochondrial respiration at states 3, 4, and FCCP-induced uncoupling, when stimulated with pyruvate/malate, was reduced at 18 and 24 months, leading to lower ATP-linked mitochondrial respiration compared to mitochondria from HCR rats. Male LCR rats also showed reduced complex II-stimulated mitochondrial respiration (succinate + rotenone) at 24 months compared to HCR rats. Differences in mitochondrial respiration were associated with tau hyperphosphorylation and Aβ42 alterations in both HCR and LCR strains. Proteomic analysis unveiled a distinct difference in the mitochondrial proteome, wherein female LCR rats displayed diminished mitochondrial translation and oxidative phosphorylation (OXPHOS) proteins at 18 months compared to female HCR rats. Conversely, male LCR rats exhibited increased OXPHOS protein abundance but reduced tricarboxylic acid (TCA) cycle proteins compared to male HCR rats. These findings underscore a robust association between intrinsic aerobic exercise capacity, brain mitochondrial function, and AD pathologies during aging.

Keywords: Aerobic capacity; Aging; Alzheimer’s disease; Amyloid beta; Bioenergetics; Mitochondria; Tau.

Fig. 1

Pathological hallmarks of AD. A Body weight (g). B Brain amyloid beta (Aβ₄₂). C Brain phosphorylation of tau ^thr181 and total tau ratio. Data are presents as means ± SEM. n = 4–8/group. * p < 0.05 aging effect, # p < 0.05 strain effect, ^p < 0.05 vs. indicated group

Fig. 2

Brain mitochondrial respiration. A ADP-stimulated state 3 O₂ flux. B Oligomycin-stimulated state 4 O₂ flux. C FCCP-stimulated uncoupled O₂ flux. D ATP-linked O₂ flux. E Complex II supported O₂ flux. F Respiratory control ratio (RCR). Data are presents as means ± SEM. n = 4–8/group. *p < 0.05 aging effect, #p < 0.05 strain effect, ^p < 0.05 vs. indicated group

Fig. 3

Brain mitochondrial respiration associated with pathological hallmarks of AD. A Correlation between brain state 3 O₂ flux and Aβ₄₂. B Correlation between brain state 3 O₂ flux and phosphorylation of tau ^thr181 and total tau ratio. Data are presents as means ± SEM. n = 4–8/group

Fig. 4

Female brain mitochondrial proteomics at 18 months of age. A Mitochondrial protein abundance. B Top 10 upregulated and downregulated mitochondrial proteins based on Log2-fold change. C IPA analysis of the mitochondrial proteome. D String pathway of proteins in upregulated pathways. E String pathway of proteins in downregulated pathways. F OXPHOS top 10 upregulated and downregulated mitochondrial proteins based on Log2. n = 4–5. *p < 0.05 vs. LCR

Fig. 5

Male brain mitochondrial proteomics at 18 months of age. A Mitochondrial protein abundance. B Top 10 upregulated and downregulated mitochondrial proteins based on Log2-fold change. C IPA analysis of the mitochondrial proteome. D String pathway of proteins in upregulated pathways. E String pathway of proteins in downregulated pathways. F TCA cycle top 10 upregulated and downregulated mitochondrial proteins based on Log2. G OXPHOS top 10 upregulated and downregulated mitochondrial proteins based on Log2. n = 3–4. *p < 0.05 vs. LCR