Voluntary wheel running decreases amyloidogenic pathway and rescues cognition and mitochondrial energy metabolism in middle-aged female 3xTg-AD mouse model of Alzheimer's disease

Abstract

Background

Evidence supports the neuroprotective effects of physical activity, either in experimental animal models or humans. However, the biological mechanisms by which physical exercise modulates dementia and Alzheimer's disease (AD) progression are still unclear.

Objective

This study investigated whether long-term (6 months) of voluntary wheel running induces neuroprotective effects in the pathogenesis of AD in middle-aged (8 months) female mice, focusing on energy metabolism.

Methods

A genetic mice model of AD (3xTg-AD) that performed wheel running presented changes in body metabolism and muscle oxidative profile, as well as restored discriminative and non-associative retention memories, evaluated by novel object recognition and open field tasks, respectively.

Results

In the hippocampus, these mice exhibited reduced levels of amyloidogenic AβPPβ fragment, phospho-Tau protein and phospho-Akt (activated form), without changes in phospho-AMPK (activated form). In addition, hippocampal mitochondria presented a restored respiratory function, characterized by lower coupling degree and weak contribution from complex I found in 3xTg-AD mice.

Conclusions

The results demonstrated that voluntary exercise improves cognitive parameters and biochemical hallmarks of AD, modulates Akt activation and enhances mitochondrial energy metabolism in hippocampus of middle-aged 3xTg-AD female mice, thereby reinforcing the neuroprotective role of physical exercise and the involvement of mitochondria in the etiology of the AD.

Keywords

Alzheimer's disease AβPP mitochondrial energy metabolism voluntary wheel running

Introduction

Alzheimer's disease (AD) is currently a global challenge, accounting for 60%–80% of dementia cases, with more than 55 million people affected worldwide.¹ Characterized by neurodegenerative disorder associated with learning, memory, and communication deficits, AD remains a disease without a treatment capable of halting its progression.²

The primary pathological marker of AD is the production and deposition of extracellular amyloid-β peptides (Aβ) and the intracellular aggregates of hyperphosphorylated tau proteins.^3,4 The production of Aβ occurs through the sequential cleavage of amyloid-β protein precursor (AβPP) by β-secretase (BACE) and γ-secretase enzymes.⁵ However, the complete understanding of the triggers that increase Aβ production and deposition, as well as tau aggregation, remains debatable. Metabolic dysfunctions are probable candidates, given their identification as risk factors for AD, including diabetes, obesity, and physical inactivity.^6,7 Mitochondrial metabolic dysfunction is also implicated in the etiology of AD,^8–13 reinforced by growing evidence demonstrating co-localization of AβPP (wild-type or mutant AβPPSwed) and Aβ with mitochondria in postmortem AD brains, AD animal and cell models.¹⁴ However, the detailed mechanism of this relationship is still elusive.^15,16

Physical exercise has emerged as a potential therapy that can reduce the prevalence of AD by indirectly modifying risk factors¹⁷ and directly reducing Aβ production and deposition and tau aggregation.^18,19 In a meta-analysis study conducted by Hamer and Chida,²⁰ it was suggested that physical exercise in both genders and across different training regimes was associated with a 28% reduction in dementia and a 45% reduction in AD. Therefore, this study aimed was to investigate whether long-term (6 months) voluntary wheel-running induces neuroprotective effects in the pathogenesis of AD in middle-aged (8 months) 3xTg-AD female mice, focusing on energy metabolism. This genetic mice model of AD present a set of neurodegenerative biomarkers, such as early long-term memory impairment (4 months), age-related Aβ plaques and tau phosphorylation species in hippocampal regions.²¹ Here we analyzed cognitive parameters (discriminative memory and habitual memory), the levels of pathological AD hallmarks (AβPP-C99 amyloidogenic fragments and phospho-Tau protein), body and muscle metabolisms, the activation levels of key metabolic sensors enzymes (Akt and AMPK) and mitochondrial energy metabolism in the hippocampus. Our results demonstrated that voluntary exercise improves cognitive and biochemical parameters of AD and modulate Akt activation and mitochondrial energy metabolism in the hippocampus of middle-aged 3xTg-AD female mice.

Methods

Animals

All experiments were approved by the Animal Care and Use Committee of the School of Pharmaceutical Sciences of Ribeirão Preto, USP (Protocols no. 22.1.292.60.9). Female 3xTg-AD mice (B6; 129-Psen^1tm1Mpm Tg (AβPPSwe, tauP301L) 1L fa/MmJax; 24 weeks old) and background strain B6129SF2/J (wide-type control; WT) were provided by the Center of animals from the School of Pharmaceutical Sciences of Ribeirão Preto, USP. The animals were maintained at the vivarium of the School of Pharmaceutical Sciences of Ribeirão Preto, housed in groups of 4–6 per cage (size 18 × 33.5 × 41 cm, H × B × L) under 20 ± 2° C, in a 12-h light/dark cycle with free access to standard chow and drink water.

Voluntary wheel running and experimental design

After acclimatization, 8-month-old mice (both wild-type [WT] and 3xTg-AD) were housed either without access to a running wheel (sedentary, SED) or with free access to a running wheel (voluntary wheel running, VWR) measuring 20 cm in diameter and 8 cm in width (Bragança Wired Exerciser, São Paulo, SP). Together, the group of mouse per cage (4–6 animals) moved the wheel an average of 12–18 km/day, without differences between WT and 3xTg-AD. After 4 months, the animals underwent an Incremental Load Test and Indirect Calorimetry (see Figure 1A) to verify the exercise adaption of each animal. Over the subsequent 2 months, mice were subjected to cognitive tests in the following sequence: the Open field (OF), Novel object recognition (NOR), and Active Avoidance (AA) tests. At 14 months of age, mice were euthanized, and the brain, heart, and soleus muscle were excised. Half of the hippocampus was promptly utilized for respirometry assays, while the remaining tissues were stored at −80°C.

Figure 1.

Six months of VWR improves the oxidative metabolism of middle-aged 3xTg-AD female mice. (A) Experimental timeline of 8-months-old WT and 3xTg-AD sedentary (SED) or voluntary wheel running (VWR) housing for 6 months, followed by behavioral tests: Open Field (OF), Novel Object Recognition (NOR), and Active Avoidance (AA) tasks. (B) Incremental load test. (C) Citrate synthase (CS) activity in heart muscle. (D) CS activity in soleus muscle. Data are expressed as the mean ± SEM; a = p < 0.05 versus Wt_SED (one-way ANOVA, Tukey's post-hoc test); N = 8 (WT_SED), 7 (WT_VWR), 6 (3xTg-AD_SED), and 6 (3xTg-AD_VWR).

Incremental load test

The test was conducted on a treadmill (AVS Special projects^®), initially set at 9 m/min velocity with a 20% incline and with increments of 2 m/min every 2 min until the mice reached voluntary exhaustion. Exhaustion was defined as the point at which the animals touched the end of the treadmill five times within a 1-min interval.²²

Citrate synthase activity

Soleus and heart muscles were homogenized in 250 mM Tris-HCl buffer (pH 7.4), 2 mM EGTA, and 250 mM sucrose. The Bradford method²³ (Bio-Rad, CA, USA) was used to determine protein concentration. The citrate synthase activity was measured following the increase in 5,5′-dithiobis (2-nitrobenzoic acid) absorbance at 412 nm in the presence of acetyl-CoA, according to previous protocol.²⁴ Reactions were adjusted to fit in a 96-well microplate platform, using 10 µg of protein/well.

Indirect calorimetry

Mice were placed individually in a hermetic chamber connected to an indirect calorimetry system (Oxylet Physiocage™, PanLab, Barcelona, Spain) with free access to chow and drink water. After 2 h of habituation, oxygen consumption (VO₂) and CO₂ release (VCO₂) were recorded for 24 h. Respiratory quotient (RQ) was calculated as VCO₂/VO₂ and used to estimate macronutrient oxidation.²⁵ Energy expenditure was calculated using the following formula: EE = [3.85 + (1.232 × RQ)] × VO₂ × 1.44.

Open-field (OF) task

This task evaluates motor performance in the training section and non-associative memory in the retention test session. Habituation to the open field was carried out in an acrylic arena (46 × 46 × 46 cm) surrounded by infrared sensors, responsible for detecting the animal's position throughout the monitoring period. Thus, it was possible to calculate the average speed rate and the total distance traveled throughout the experiment using dedicated software (Insight, Ribeirão Preto, Brazil). The animals were placed in the center of the arena to explore it for 5 min (training session) and, after 24 h, they were again subjected to a similar open field session (test session). The total distance covered and rearing in both sessions were counted. The decrease in the total distance covered and the number of rearings between the two sessions were used as measures of the retention of habituation.²⁶ The arena was cleaned with 30% ethanol and water between each session.

Novel object recognition (NOR) task

The NOR task evaluated discriminative short-term memory.²⁷ The mice were habituated to the empty arena (without any objects) for 10 min per 3 consecutive days. Twenty-four hours after the habituation days, during the training phase, the mice were subjected to two identical objects (shape and size) at two adjacent corners of the arena, each 10 cm away from the side of the walls. Each time, the mice were placed in the middle of the box, facing away from the objects, and they were allowed to explore the objects for 5 min. After a delay of 15 min, the test phase was conducted for 5 min. One previously introduced object was replaced by an object of the same size but different shape and color in the same places. Exploration time was analyzed for the arrangement of identical objects from the training phase (exploration time) and different objects from the test phase (discrimination time). A recognition index was calculated and reported as the ratio TB/(TA + TB) (TA = time exploring familiar object A, and TB = time spent exploring new object B). Sniffing or touching the object with the nose or front paws was defined as exploration.^28,29

Immunoblotting technique

The hippocampus and cortex were homogenized in RIPA buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1% Triton, 0,01% SDS, 1.2 mM EDTA, 0,5% deoxycholic acid) containing a 1% protease inhibitor cocktail (Sigma-Aldrich, P2714), and protein concentration was determined by the Bradford method.²³ In next step, the samples were centrifuged at 20.000 × g for 90 min at 4°C. Twenty micrograms (20 μg) of total protein were electrophoretically separated in 10% and 14% SDS-polyacrylamide gel (100 V, 90 min) and then transferred onto PVDF membrane (0.45 μm pore size; GE Healthcare Life Science) in a tank blotting system (100 V, 120 min). Membranes were blocked with a 5% casein solution for 1 h and incubated overnight (4°C) under agitation with the antibodies: AβPP-C99/AβPP-C83 (1:1000, Merck, A8717), α-actin (1:3000, DSHB, 224-236-1), phospho-Tau (1:1000, Millipore, AB8672), Tom20 (1:1000, Cell Signaling, D8T4N), anti-mouse secondary (1:5000, Cell Signaling, 7076S), anti-rabbit secondary (1:5000, Cell Signaling, 7074S), tTau (1:1000, Sigma, SAB4501821), Akt (pan) (1:1000, Cell Signaling, C67E7), phosphor-Akt (1:1000, Cell Signaling, D9E); AMPK (1:1000, Cell Signaling, D63G4), phospho-AMPK (1:1000, Santa Cruz, 33524), and Oxphos (1:1000, Abcam, Ab 110413). After three washes with TBS-T, the membranes were incubated with appropriate HRP-conjugated secondary antibodies (1:5000 in 1% BSA) for 1 h at room temperature. For detection, we used a chemiluminescence reaction kit. Immunoreactive protein bands were visualized in the ChemiDoc MP System (BioRad) and analyzed by the system's software (ImageLab 5.2.1). The image result for each band was normalized against the α-actin level (internal control) of the respective sample.

Respirometry

Fresh biopsies of the bilateral hippocampus (4–7 mg wet weight) were hand dissected in cubes (∼2 × 2 × 2 mm, 4–7 mg w.w.) and permeabilized in 1 mL ice-cold BIOPS solution (50 mM K⁺ -MES, 20 mM taurine, 0.5 mM dithiothreitol, 6.56 mM MgCl₂, 5.77 mM ATP, 15 mM phosphocreatine, 20 mM imidazole, and 10 mM Ca-EGTA buffer, pH 7.1) containing saponin (0.01%) for 10 min. Then, after 12–15 min after mice euthanasia, biopsies were placed in the Oroboros chamber (Oxygraph-2k, Oroboros, Innsbruck, Austria) containing 2.1 mL air-saturated MiR05 (0.5 mM EGTA, 3 mM MgCl₂, 60 mM K-lactobionate, 20 mM taurine, 10 mM KH₂PO₄, 20 mM HEPES, 110 mM sucrose, 1 g/L albumin, pH 7.1) at 37°C, with 300 rpm stirring, and the O₂ consumption rates were monitored. Mitochondrial respiratory States were determined as follow: non-phosphorylation (State L), in the presence of 10 mM glutamate (NADH-linked substrate, N); phosphorylation supported by N (State P-N), after addition of 1 mM ADP, or supported by N and succinate (State P-NS), after addition of 10 mM succinate; noncoupled or at maximal respiratory capacity supported by NS (State E-NS), after 2 additions of 1 µM carbonyl cyanide m-chlorophenyl hydrazone (CCCP, uncoupler), or supported by S (State E-S), after addition of 1 µM rotenone (complex I inhibitor). Lastly, 5 mM malonate and 2 mM NaCN, inhibited complexes II and IV respectively, and residual oxygen consumption (ROx) was determined. Rox rates were subtracted from every respiratory state.³⁰

Statistics

The results are expressed as mean ± SEM and were analyzed by two-way analysis of variance followed by Tukey's post-hoc test for multiple comparisons. Values of p ≤ 0.05 were considered statistically significant in all cases. The software used was Graph Pad Prism version 8.0 (GraphPad Software, Inc., San Diego, CA, USA, 2007).

Results

Six months of voluntary wheel running (VWR) improves aerobic capacity and the molecular marker of oxidative metabolism in soleus and cardiac muscles in middle-aged 3xTg-AD female mice

The animals were submitted to the incremental load test to verify if 6 months VWR improves in vivo aerobic capacity in middle-aged female mice. Figure 1B shows that both mice groups (WT and 3xTg-AD) with wheel access (exercised, VWR) presented higher exhaustion velocity compared to their respective pairs without wheel access (control condition, considered “sedentary” for comparative means, SED), in which the 3xTg-AD exercised mice (VWR) demonstrated the highest significant level. In addition, muscle oxidative capacity in these mice was checked by measuring citrate synthase activity, a classical biomarker of mitochondrial content.³¹ Hearts (Figure 1C) and Soleus (Figure 1D) muscles of 3xTg-AD exercised mice presented higher CS activity levels than 3xTg-AD under sedentary conditions.

Six months of VWR reduces the basal but not active energy expenditure in middle-aged female mice

These animals were submitted to the indirect calorimetry to verify if 6 months VWR changes body metabolic parameters in middle-aged female mice. Figure 2 shows that during the active period of the day (12 h dark) no differences in energy expenditure (EE) (Figure 2A) were observed between mice groups (WT and 3xTg-AD) with or without wheel access. However, when the mice were resting (12 h light), 3xTg-AD_SED mice presented elevated EE compared to both WT groups (Figure 2B) and VWR reduced EE in both WT and 3xTg-AD mice. Despite this, the groups of mice had similar body weights (g): WT_SED = 27.94 ± 0.87; WT_VWR = 27.83 ± 1.46; 3xTg-AD_SED = 28.5 ± 0.97 and 3xTg-AD_VWR = 29.3 ± 0.54 (p > 0.05). Respiratory quotient (RQ) was similar between all mice groups under resting or active periods (Figures 2C, D), indicating no changes in energy substrate oxidation.

Figure 2.

Six months VWR reduces body metabolic rates in the resting period of the day in middle-aged female mice. (A) Energy expenditure (EE) during the 12 h dark. (B) EE during the 12 h light; a = p < 0.05 versus WT_VWR; b = p < 0.05 versus 3xTg-AD_VWR; c = p < 0.05 versus WT_SED (Two-way ANOVA, Tukey's post hoc). C) Respiratory quotient (RQ) index during the 12 h light. (D) RQ during the 12 h dark. Data are expressed as the mean ± SEM; N = 9 (WT_SED), 9 (WT_VWR), 7 (3xTg-AD_SED), and 6 (3xTg-AD_VWR).

Six months of VWR recovers discriminative short-term and non-associative long-term memory in middle-aged 3xTg-AD female mice

After finding evidence that VWR has improved parameters of aerobic capacity and metabolism in 3xTg-AD middle-aged female mice, we tested if 6 months of VWR improved discriminative memory and performance related to retention of habituation, submitting the mouse groups to new object recognition (NOR, Figure 3A-C) and open field (OF, Figures 3D, E) tasks, respectively. Figure 3A presents the similar exploratory indexes (%) in training session of all mice groups. As expected, 3xTg-AD mice submitted to control condition (3xTg-AD_SED) presented lower exploratory index (Figure 3B) and time spent (Figure 3C) on NOR compared to control WT animals, demonstrating an expressive impairment of short-term discriminative memory. WT animals (SED and VWR) also showed a common characteristic of the normal adaptive learning process of habituation in the OF arena, in which the number of rearing (Figure 3D) and total distance traveled (Figure 3E) during the Test session is smaller than Training session. However, 3xTg-AD_SED presented an opposite profile, demonstrating a significant deficit of habituation memory. All these memories and learning parameters appeared fully recovered in 3xTg-AD mice with wheel access. Despite this, the VWR did not reverse the decreased total distance exploration observed in 3xTg-AD, a behavior associated with increased innate anxiety.³²

Figure 3.

Six months of VWR improves short-term memory and learning habituation of middle-aged 3xTg-AD female mice. (A) Percentual of exploratory behavior in the initial familiarization session in the novel object recognition (NOR). (B) Percentual of exploratory behavior of the new object in the test session in NOR; a = p < 0.05 versus object 1 (familiar) of the same group (unpaired Student's t-test). (C) Time expenditure in NOR; a = p < 0.05 versus WT_SED, WT_VWR, and 3xTg-AD_VWR (Two-way ANOVA, Tukey's post hoc). (D) Number of rearing in open field (OF) arena; a = p < 0.05 versus training session of the same group (unpaired Student's t-test). (E) Total distance traveled in OF arena; a = p < 0.05 versus training session of the same group (unpaired Student's t-test). Data are expressed as the mean ± SEM; N = 8 (WT_SED), 7 (WT_VWR), 6 (3xTg-AD_SED), and 6 (3xTg-AD_VWR).

Six months of VWR reverts tau phosphorylation in hippocampus in middle-aged 3xTg-AD female mice

After finding evidence that VWR has improved parameters of metabolism and memory in middle-aged 3xTg-AD female mice, next we verify in these animals if 6 months of VWR alters the levels of Tau phosphorylation (p-Tau), a widely considered pathological hallmark in AD.^33,34 The analyses of Tau by western blotting (Figure 4) showed that VWR reduced the elevated levels of p-Tau relative to total Tau (t-Tau) (Figure 4B) or α-actin (Figure 4C) in hippocampus of 3xTg-AD mice, to the levels of WT mice (SED or VWR). Notably, the cortex of middle-aged 3xTg-AD female mice did not present significant p-Tau levels compared to WT, and wheel access exerts no effect (Figure 4D, E).

Figure 4.

Six months VWR reduces the levels of tau phosphorylation in hippocampus of middle-aged 3xTg-AD female mice. (A) Western blotting representative image showing anti-Tau phospho (pTau), anti-Tau (total Tau, tTau) or anti-actin stainings in hippocampus of WT and 3xTg-AD sedentary (SED) or voluntary wheel running (VWR). (B) pTau/tTau band ratio in hippocampus; a = p < 0.05 versus WT_SED, WT_VWR, 3xTg-AD_VWR; (C) Band intensities of pTau relative to α-actin in hippocampus; a = p < 0.05 versus 3xTg-AD_SED, 3xTg-AD_VWR; (D) Western blotting representative image showing anti-pTau, anti-tTau or anti-actin stainings in cortex. (E) pTau/tTau band ratio in cortex; (F) Band intensities of pTau relative to α-actin in cortex. Two-way ANOVA, Tukey's post hoc. Data are expressed as the mean ± SEM; N = 8 (WT_SED), 7 (WT_VWR), 6 (3xTg-AD_SED), and 6 (3xTg-AD_VWR).

Six months of VWR reduces the accumulation of amyloidogenic aβpp-c99 fragments in hippocampus in middle-aged 3xTg-AD female mice

In line with Tau phosphorylation, next we verify if 6 months VWR alters the amyloid cascade of AD in these animals. The starting point is the AβPP cleavage by BACE1, producing a 99-residue C-terminal peptide (AβPP-C99), that is subsequently cleaved to Aβ protein by γ-secretase.³⁵ As expected, the analyses of AβPP-C99 fragments by western blotting demonstrated that the hippocampus (Figure 5A, B) and cortex (Figure 5E, F) of middle-aged 3xTg-AD female mice have ∼ 4- and 2-fold higher AβPP-C99 content than WT mice under control conditions (SED), respectively. Despite the non-amyloidogenic AβPP-C83 fragments, generated from AβPP cleavage by α-secretase, also elevated in 3xTg-AD animals (Figures 5A, C, E, G), the higher AβPP-C99/C83 ratios (Figures 5D, H) indicates a more intense amyloidogenic pathway in hippocampus and cortex of these animals. Surprisingly, middle-aged 3xTg-AD female mice with access to a running wheel presented a reduction in AβPP-C99 and AβPP-C83 content, and a significant reduction in AβPP-C99/C83 ratio only in hippocampus; in cortex, VWR reduced AβPP-C83 fragments but not the AβPP-C99/C83 ratio.

Figure 5.

Six months of VWR reduces the levels of APP-C99 and APP-C83 fragments in hippocampus and cortex of middle-aged 3xTg-AD female mice. (A) Western blot representative image showing anti-APP (APP-C99 or APP-C83 fragments) and anti-actin stainings in hippocampus of WT and 3xTg-AD sedentary (SED) or voluntary wheel running (VWR). (B) Band intensities of APP-C99 band relative to α-actin in hippocampus; a = p < 0.05 versus WT_SED, WT_VWR, 3xTg-AD_VWR; b = p < 0.05 versus WT_SED, WT_VWR. (C) Band intensities of APP-C83 relative to α-actin in hippocampus; a = p < 0.05 versus WT_SED, WT_VWR, 3xTg-AD_VWR; b = p < 0.05 versus WT_SED, WT_VWR; (D) APP-C99/APP-C83 band ratio in hippocampus; a = p < 0.05 versus WT_SED, WT_VWR, 3xTg-AD_VWR; b = p < 0.05 versus WT_SED, WT_VWR. (E) Western blot representative image showing APP-C99, APP-C83 and α-actin expression levels in cortex. (F) Band intensities of APP-C99 relative to α-actin in cortex; a = p < 0.05 versus 3xTg-AD_SED, 3xTg-AD_VWR; b = p < 0.05 versus 3xTg-AD_SED. (G) Band intensities of APP-C83 relative to α-actin in cortex; a = p < 0.05 versus WT_SED, WT_VWR, 3xTg-AD_VWR; (H) APP-C99/APP-C83 band ratio in cortex; a = p < 0.05 versus 3xTg-AD_SED, 3xTg-AD_VWR; Two-way ANOVA, Tukey's post hoc. Data are expressed as the mean ± SEM; N = 8 (WT_SED), 7 (WT_VWR), 6 (3xTg-AD_SED), and 6 (3xTg-AD_VWR).

VWR reduces the activation of insulin signaling intermediary Akt in hippocampus of middle-aged 3xTg-AD female mice

After finding evidence that VWR improved parameters of metabolism, memory, and markers of AD in middle-aged 3xTg-AD female mice, we next investigated whether 6 months of VWR changes the activation levels of metabolic sensors AMP-activated protein kinase (AMPK) and protein kinase B (Akt) in hippocampus. Both kinases play several vital roles in neuronal metabolism and function: AMPK is a sensor of low ATP concentration and regulates essential metabolic pathways, such as glycolysis and mitochondrial biogenesis,^36,37 and cellular processes of autophagy, ion flow and neuronal excitability;³⁸ while Akt regulates glucose uptake and utilization, as well as protein synthesis, neurotransmitter signaling and synaptic plasticity.³⁹ The levels of phosphorylated-AMPK (pAMPK, Figure 6A, B) and phosphorylated-Akt (pAkt Ser⁴⁷³, Figure 6C, D) in hippocampus of middle-aged 3xTg-AD female mice were similar compared to WT, indicating the similar levels of activations at sedentary conditions. However, in mice with wheel access (VWR) the levels of p-Akt were decreased while pAMPK were not changed, compared to animals without wheel access (SED). These results indicate that the reduction in AKT activation in the hippocampus induced by VWR was unrelated to intracellular energetic changes.

Figure 6.

Six months of VWR reduces the activation of insulin signaling intermediary Akt in hippocampus of middle-aged female mice. (A) Western blotting representative image showing anti-AMPK phospho (p-AMPK) or anti-actin stainings in hippocampus of WT and 3xTg-AD sedentary (SED) or voluntary wheel running (VWR). (B) Band intensities of p-AMPK relative to α-actin in hippocampus; (C) Western blotting representative image showing anti-Akt phospho (p-Akt Ser⁴⁷³) or anti-actin staining in cortex. (D) Band intensities of p-Akt relative to α-actin in cortex; a = p < 0.05 versus WT_SED, 3xTg_AD_SED; Data are expressed as the mean ± SEM; Two-way ANOVA, Tukey's post hoc. N = 8 (WT_SED), 7 (WT_VWR), 6 (3xTg-AD_SED), and 6 (3xTg-AD_VWR).

Six months of VWR reversed the higher mitochondrial coupling degree in hippocampus in middle-aged 3xTg-AD female mice

As mitochondrial dysfunction is a common finding in several in vivo and in vitro models of AD,⁴⁰ next we investigated if VWR modulated energetic mitochondrial function in middle-aged 3xTg-AD female mice. Respiratory rates were monitored by HRR in permeabilized fresh biopsies of hippocampus, a technique that maintain the native cellular environment and interactions (mix of cell types and extracellular matrix), providing a more physiologically relevant assessment of mitochondrial function, reflecting the in vivo situation more accurately. Respiratory rates in the states of non-phosphorylation (State L, without ADP), phosphorylation (State P, after ADP addition) or at maximal respiratory capacity (State E, after CCCP uncoupler addition), in the presence of NADH-linked (or complex I) substrate (N, glutamate) alone or with complex II substrate (S, succinate) (Figure 7A). Respiratory rates at State P were similar between all mice groups, independently of the respiratory substrate (Figures 7C, D). However, State L respiration was lower in 3xTg-AD_SED (Figure 7B), leading to an increased P/L ratio (with substrates N, for complex I) (Figure 7H) compared to WT_SED, which indicates a higher coupling degree in these mitochondria. Respiratory rates at State E presented no differences between all mice groups when sustained by substrates for both complexes (NS, Figure 7E), but after complex I inhibition by rotenone, State E sustained by succinate (S) appeared elevated in 3xTg-AD_SED (Figure 7F). Still, in these animals the lower rates at NS-S pathway control efficiency (State E: NS - S), Figure 7G) demonstrate a low contribution of complex I relative to complex II in respiration, indicating a relative complex I inhibition in these mitochondria. The limitation of OXPHOS capacity by the phosphorylation system, expressed by the P/E control ratio, was similar between all mice groups (Figure 7I). The wheel access (VWR) reversed all mitochondrial changes found in hippocampus of 3xTg-AD animals, and in WT animals reduced State L respiration, without changes in coupling degree (P/L ratio).

Figure 7.

Six months of VWR reversed the altered mitochondrial respiration in hippocampus of middle-aged 3xTg-AD female mice. (A) Representative figure of respiratory monitoring by HRR. (B) Respiratory rates at State L (Leak) supported by complex I substrates glutamate 10 mM (N); a = p < 0.05 versus WT_VWR, 3xTg-AD_SED; (C) Respiratory rates at State P (oxidative phosphorylation) supported by NADH-linked substrates (N). (D) Respiratory rates at State P supported by N and 10 mM succinate (NS). (E) Respiratory rates at State E (maximal electron transfer) supported by N and 10 mM succinate (NS). (F) Respiratory rates at State E supported by succinate in the presence of 2 μM rotenone (S); a = p < 0.05 versus WT_VWR, 3xTg-AD_VWR; (G) NS-S pathway control efficiency (State E: NS - S); a = p < 0.05 versus WT_SED; WT_VWR, 3xTg-AD_SED; (H) P/L control ratio. a = p < 0.05 versus WT_VWR, 3xTg-AD_VWR; (I) P/E control ratio. Two-way ANOVA, Tukey's post hoc. Data are expressed as the mean ± SEM; N = 8 (WT_SED), 7 (WT_VWR), 6 (3xTg-AD_SED) and 6 (3xTg-AD_VWR).

In addition, the content of mitochondrial proteins was analyzed by western blotting. Hippocampus of middle-aged 3xTg-AD female mice did not present significant differences in the contents of Tom20, a marker of mitochondrial mass, or RC components (from Complex I to IV) compared to WT under sedentary conditions. The wheel access seemed to exert antagonistic effects on mitochondrial content in each strain, as seen by the significantly high Tom20 in WT_VWR compared to 3xTg-AD_VWR mice (Figure 8).

Figure 8.

Six months of VWR exerts antagonistic effects on mitochondrial content in hippocampus of middle age wt and 3xTg-AD female mice. Quantification by western blotting of mitochondrial proteins of respiratory chain complexes (CI, CII, CIII and CIV) and Tom20 normalized by α-actin. WT and 3xTg-AD sedentary (SED) or voluntary wheel running (VWR). a = p < 0.05 versus 3xTg-AD_VWR. Data are expressed as the mean ± SEM; Two-way ANOVA, Tukey's post hoc. N = 8 (WT_SED), 7 (WT_VWR), 6 (3xTg-AD_SED), and 6 (3xTg-AD_VWR).

Discussion

This study aimed to investigate whether 6 months of voluntary exercise induces neuroprotective effects in the pathogenesis of AD in 8-month-old female mice, focusing on energy metabolism. Together, the results demonstrate that VWR improved the parameters of body and muscle metabolisms, memory and learning, biochemical markers of AD (p-Tau and AβPP-C99), modulated Akt activation and mitochondrial coupling degree in hippocampus of middle-aged 3xTg-AD female mice. These beneficial effects of VWR were observed during a stage of significant cognitive decline in the 3xTg-AD female mice, which is first apparent at 6 months and worsens significantly with aging.⁴¹

Our results shown that the expressive impairments in discriminative and the non-associative retention memories, evaluated by NOR and OF tasks, respectively, were fully recovered in middle-aged 3xTg-AD mice that performed 6 months of wheel running. The protective effects of exercise against cognitive decline have been demonstrated in various training protocols and AD models.^18,19,42 For example, in young (7-week-old) AβPP/PS1 transgenic mice submitted to 9 weeks of mandatory treadmill running (5 days/week, 20–60 min/day at 10 m.min⁻¹, with increments of 10 min per day), the exercise increased the learning and memory performances associated with amygdala and hippocampus, assessed by Pavlovian fear conditioning task.⁴³ Long-term voluntary running (4 months) in different periods of daily access to a running wheel (0, 1, 3, and 12-h), beginning in 14-month-old Tg2576 mice, was able to reduce behavioral dysfunction, evaluated by OF and Y-maze, with a better response with 3-h daily exercise.¹⁸ Nevertheless, in male AβPP/PS1 mice 4 months of voluntary exercise improved cognitive function in young (9-weeks old) but not in middle-aged (6-month old) animals, analyzed by Morris water maze test.⁴⁴ Possibly, this different result may be related to sex differences in underlying mechanisms, with females responding better to exercise.⁴⁵ In addition to aerobic exercise, 9-month-old 3xTg mice underwent a resistance training protocol and showed improved cognitive performance assessed by the NOR and Y-maze tests.⁴⁶ Other studies, including resistance exercise, have shown neuroprotective effects related to a decrease in amyloid load and plaques, an increase in neurotrophic factors, and an improvement in synaptic transmission.^47–49 Thereby, the protective effects of exercise are still up for debate, but seem to be dependent on the exercise program, including regime (voluntary versus mandatory), type, start time, volume and intensity, type of AD model, and molecular condition of the disease, as well as the cognitive task used to evaluate.

Cognitive improvements in the context of AD are frequently associated to reductions of pathological hallmarks, such as Tau phosphorylation.^50–53 Such effects induced by exercise in 3xTg-AD mice were previously found in adult (4-months old) and aged (24-months old) mice, submitted to 3 months of mandatory treadmill running (5 days/week, 5 min at 5 m.min⁻¹, 20 min at 10 m.min⁻¹, 5 min at 5 m.min⁻¹).¹⁹ Here, in addition to diminished Tau phosphorylation, we show for the first-time reduced levels of amyloidogenic AβPP-C99 fragments in hippocampus of middle-aged 3xTg-AD female mice that performed voluntary wheel running. A known feature of this AD model is the overproduction of AβPP-C99, driven by expression of the mutant human AβPP (AβPPSwe - Swedish double mutation; K670N, M671L),⁵⁴ a mutation occurring at the amino terminal of Aβ that strongly increases BACE1 affinity for AβPP.⁵⁵ In addition to AβPP-C99, non-amyloidogenic AβPP-C83 fragments were diminished in 3xTg-AD VWR, indicating that BACE1 and α-secretase cleavages were reduced. However, the diminished AβPP-C99/C83 ratio in 3xTg-AD mice engaging in voluntary wheel running suggests that the amyloidogenic pathway undergoes a more pronounced decrease than the non-amyloidogenic pathway. This shift promoted by VWR was not found in the cortex of these animals, highlighting the potential protective impact of physical exercise specific in the hippocampus, at the initial pathogenesis of AD. A reduction in the AβPP-C99/C83 ratio was also found in the streptozotocin (STZ)-induced AD rat model after 4 weeks of treadmill exercise (5 days/week, 30 min/day).⁵⁶

AβPP processing products can enter the cell and, even as pTau, accumulate in mitochondria via a mitochondrial import mechanism, triggering mitochondrial dysfunction; in turn, mitochondrial function and bioenergetic intermediates, such as reactive oxygen species (ROS), and NAD+/NADH, can modulate AβPP processing and trafficking pathways (for review).^15,57 Our results showed that VWR could hamper this feedback loop, by reversing the mitochondrial complex I inhibition and higher coupling degree found in the hippocampus in middle-aged 3xTg-AD female mice, and reducing AD pathological hallmarks. Substantial data support a role for pTau, AβPP, and Aβ in modulating mitochondrial function in several AD models,^58–62 inducing respiratory impairment by directly influencing the activities of RC complexes, such as complex I and III,⁶³ also leading to increasing ROS generation⁶⁰ specifically in hippocampus of 3xTg-AD mice, Aβ accumulation, and mitochondrial respiratory dysfunction start early (3 months old) and increase in an age-dependent way⁶¹ Aβ and pTau interacts with VDAC1, as found in the brains from AD patients and AD models, in which is observed mitochondrial dysfunction.^58,63 In the context of reduced complex I activity, the higher coupling degree found in hippocampus of middle aged 3xTg-AD female mice can be suggested as an adaptive mechanism to improve efficiency of oxidative phosphorylation, since the metabolic sensor AMPK did not show any changes in cell energy levels (AMP/ATP ratio). However, a high coupling degree linked to slow leak respiration is associated with greater ROS formation, since under these conditions the mitochondrial membrane potential is high and electron flow rate is low, increasing the probability of an electron reducing oxygen to ROS before it reaches complex IV.⁶⁴

AMPK signaling is an essential modulator of mitochondrial biogenesis and metabolism,³⁷ and its role in AD is intriguing.⁶⁵ AMPK activation by pharmacological or dietary interventions has reduced^66–68 or increased.^69,70,13,71 AD hallmarks, mainly BACE1 mRNA and Aβ levels. However, the improvements in AD markers in middle aged 3xTg-AD mice promoted by VWR demonstrated here were not associated with activating this kinase (pAMPK) in the hippocampus. On the other hand, the high energy metabolic sensor Akt⁷² was decreased in the hippocampus of mice with wheel access, characterizing a specific effect of VWR. In AD, reductions in Akt activation have been shown to promote beneficial effects, such as protecting against Aβ-induced toxicity in mutant PDK1^K465E/K465E neurons⁷³ and increasing α-secretase AβPP cleavage.⁷⁴ Taken together, physical exercise's metabolic regulations involved in neuroprotection is still debatable, but it appears to involve a finely-tuned balance between catabolic and anabolic pathways, with a pronounced tendency towards a low-energy state condition.

In summary, the results presented here demonstrated that long-term voluntary wheel running improved phenotypic characteristics of AD (response in memory tasks), reduced amyloidogenic pathway (AβPP-C99) and molecular marker of AD (pTau), and restored mitochondrial metabolism in hippocampus of middle-aged female mice, genetic model of AD. These effects could be mediated by reductions in Akt activation (pAkt) unrelated to changes in the intracellular energy levels (pAMPK). Then, the present work reinforces the neuroprotective role of physical exercise and mitochondrial involvement in the etiology of AD.

Footnotes

Acknowledgments

The authors are deeply thankful to Prof. Dr Germán G. Sgro for their insightful advice and assistance throughout the course of Immunoblotting Technique, and Alcides Pereira for technical assistance.

ORCID iD

Luciane Carla Alberici

Author contributions

Jonathas Rodrigo dos Santos (Conceptualization; Data curation; Investigation; Methodology; Writing – original draft); Carlos Rocha Catalão (Formal analysis; Methodology; Writing – review & editing); Anderson Vulczak (Methodology); Ana Elisa Azzolini (Data curation; Methodology; Supervision); Luciane Carla Alberici (Conceptualization; Data curation; Funding acquisition; Writing – original draft; Writing – review & editing).

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work received financial support from the São Paulo Research Foundation (FAPESP—2018/10089-2 to L.C.A) and Coordination for the Improvement of Higher Education Personnel - Brazil (CAPES) - Finance Code 001. LCA has a National Council for Scientific and Technological Development (CNPq) Research Fellowship.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability

The data supporting the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

References

World Health Organization. Dementia:Key facts, https://www.ncbi.nlm.nih.gov/pubmed/ (2024, accessed 21 March 2024).

Teng

. Novel development and prospects in pathogenesis, diagnosis, and therapy of Alzheimer's disease. J Alzheimers Dis Rep 2024; 8: 345–354.

Hardy

Selkoe

. The amyloid hypothesis of Alzheimer's disease: progress and problems on the road to therapeutics. Science 2002; 297: 353–356.

Hardy

. The amyloid hypothesis for Alzheimer's disease: a critical reappraisal. J Neurochem 2009; 110: 1129–1134.

Hampel

Vassar

De Strooper

, et al. The β-secretase BACE1 in Alzheimer's disease. Biol Psychiatry 2021; 89: 745–756.

Liang

Mintun

Fagan

, et al. Exercise and Alzheimer's disease biomarkers in cognitively normal older adults. Ann Neurol 2010; 68: 311–318.

Edwards

Gamez

Escobedo

Jr , et al. Modifiable risk factors for Alzheimer's disease. Front Aging Neurosci 2019; 11: 146.

Swerdlow

. Mitochondria and mitochondrial cascades in Alzheimer's disease. J Alzheimers Dis 2018; 62: 1403–1416.

Weidling

Swerdlow

. Mitochondria in Alzheimer's disease and their potential role in Alzheimer's proteostasis. Exp Neurol 2020; 330: 113321.

10.

Terada

Obi

Bunai

, et al. In vivo mitochondrial and glycolytic impairments in patients with Alzheimer disease. Neurology 2020; 94: e1592–e1604.

11.

Flores-León

Pérez-Domínguez

González-Barrios

, et al. Palmitic acid-induced NAD+ depletion is associated with the reduced function of SIRT1 and increased expression of BACE1 in hippocampal neurons. Neurochem Res 2019; 44: 1745–1754.

12.

Ryu

W-I

Bormann

Shen

, et al. Brain cells derived from Alzheimer’s disease patients have multiple specific innate abnormalities in energy metabolism. Mol Psychiatry 2021; 26: 5702–5714.

13.

Camacho-Castillo

Phillips-Farfán

Rosas-Mendonza

, et al. Increased oxidative stress contributes to enhance brain amyloidogenesis and blunts energy metabolism in sucrose-fed rat: effect of AMPK activation. Sci Rep 2021; 11: 19547.

14.

Pinho

Teixeira

Glaser

. Mitochondrial import and degradation of amyloid-β peptide. Biochim Biophys Acta 2014; 1837: 1069–1074.

15.

Wilkins

Swerdlow

. Amyloid precursor protein processing and bioenergetics. Brain Res Bull 2017; 133: 71–79.

16.

Pedros

Patraca

Martinez

, et al. Molecular links between early energy metabolism alterations and Alzheimer's disease. Front Biosci (Landmark Ed) 2016; 21: 8–19.

17.

Barnes

Yaffe

. The projected effect of risk factor reduction on Alzheimer's disease prevalence. Lancet Neurol 2011; 10: 819–828.

18.

Francis

Robison

Popescu

, et al. Voluntary wheel running reduces amyloid-β42 and rescues behavior in aged Tg2576 mouse model of Alzheimer's disease. J Alzheimers Dis 2020; 73: 359–374.

19.

Cho

Shin

Kim

, et al. Treadmill running reverses cognitive declines due to Alzheimer disease. Med Sci Sports Exerc 2015; 47: 1814–1824.

20.

Hamer

Chida

. Physical activity and risk of neurodegenerative disease: a systematic review of prospective evidence. Psychol Med 2009; 39: 3–11.

21.

Javonillo

Tran

Phan

, et al. Systematic phenotyping and characterization of the 3xTg-AD mouse model of Alzheimer's disease. Front Neurosci 2022; 15: 785276.

22.

Wisloff

Helgerud

Kemi

, et al. Intensity-controlled treadmill running in rats: vO(2 max) and cardiac hypertrophy. Am J Physiol Heart Circ Physiol 2001; 280: H1301–H1310.

23.

Bradford

. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976; 72: 248–254.

24.

Spinazzi

Casarin

Pertegato

, et al. Assessment of mitochondrial respiratory chain enzymatic activities on tissues and cultured cells. Nat Protoc 2012; 7: 1235–1234.

25.

Vulczak

Souza

Ferrari

, et al. Moderate exercise modulates tumor metabolism of triple-negative breast cancer. Cells 2020; 9: 628.

26.

Vianna

Alonso

Viola

, et al. Role of hippocampal signaling pathways in long-term memory formation of a nonassociative learning task in the rat. Learn Mem 2000; 7: 333–340.

27.

Morice

Andreae

Cooke

, et al. Preservation of long-term memory and synaptic plasticity despite short-term impairments in the Tc1 mouse model of down syndrome. Learn Mem 2008; 15: 492–450.

28.

Lueptow

. Novel object recognition test for the investigation of learning and memory in mice. J Vis Exp 2017; 126: 55718.

29.

Bevins

Besheer

. Object recognition in rats and mice: a one-trial non-matching-to-sample learning task to study ‘recognition memory’. Nat Protoc 2006; 1: 1306–1311.

30.

Pesta

Gnaiger

. High-resolution respirometry: OXPHOS protocols for human cells and permeabilized fibers from small biopsies of human muscle. Methods Mol Biol 2012; 810: 25–58.

31.

Larsen

Nielsen

Hansen

, et al. Biomarkers of mitochondrial content in skeletal muscle of healthy young human subjects. J Physiology 2012; 590: 3349–3360.

32.

Škop

Guo

Liu

, et al. The metabolic cost of physical activity in mice using a physiology-based model of energy expenditure. Mol Metab 2023; 71: 101699.

33.

Yang

Wang

. Nature of tau-associated neurodegeneration and the molecular mechanisms. J Alzheimers Dis 2018; 62: 1305–1317.

34.

Congdon

Tetlow

, et al. Tau-targeting therapies for Alzheimer disease: current status and future directions. Nat Rev Neurol 2023; 19: 715–736.

35.

O'Brien

Wong

. Amyloid precursor protein processing and Alzheimer's disease. Annu Rev Neurosci 2011; 34: 185–204.

36.

Marin

Gongol

Zhang

, et al. AMPK Promotes mitochondrial biogenesis and function by phosphorylating the epigenetic factors DNMT1, RBBP7, and HAT1. Sci Signal 2017; 10: eaaf7478.

37.

Herzig

Shaw

. AMPK: guardian of metabolism and mitochondrial homeostasis. Nat Rev Mol Cell Biol 2018; 19: 121–135.

38.

Culmsee

Monning

Kemp

, et al. AMP-activated protein kinase is highly expressed in neurons in the developing rat brain and promotes neuronal survival following glucose deprivation. J Mol Neurosci 2001; 17: 45–58.

39.

Wang

Zhou

, et al. Brain development and Akt signaling: the crossroads of signaling pathway and neurodevelopmental diseases. J Mol Neurosci 2017; 61: 379–384.

40.

Flannery

Trushina

. Mitochondrial dysfunction in Alzheimer’s disease and progress in mitochondria-targeted therapeutics. Curr Behav Neurosci Rep 2019; 6: 88–102.

41.

Belfiore

Rodin

Ferreira

, et al. Temporal and regional progression of Alzheimer's disease-like pathology in 3xTg-AD mice. Aging Cell 2019; 18: e12873.

42.

Kim

Cho

Kang

. Protective effect of exercise training against the progression of Alzheimer's disease in 3xTg-AD mice. Behav Brain Res 2019; 374: 112105.

43.

Lin

Shih

Chen

, et al. Running exercise delays neurodegeneration in amygdala and hippocampus of Alzheimer's disease (APP/PS1) transgenic mice. Neurobiol Learn Mem 2015; 118: 189–197.

44.

Wang

Zhou

Luo

, et al. Voluntary wheel running is capable of improving cognitive function only in the young but not the middle-aged male APPSwe/PS1De9 mice. Neurochem Int 2021; 145: 105010.

45.

Barha

Liu-Ambrose

. Sex differences in exercise efficacy: is midlife a critical window for promoting healthy cognitive aging? FASEB J 2020; 34: 11329–11336.

46.

Liu

Chu

JMT

Yan

, et al. Short-term resistance exercise inhibits neuroinflammation and attenuates neuropathological changes in 3xTg Alzheimer's disease mice. J Neuroinflammation 2020; 17: 4.

47.

Campos

Ribeiro

Hashiguchi

, et al. Neuroprotective effects of resistance physical exercise on the APP/PS1 mouse model of Alzheimer's disease. Front Neurosci 2023; 17: 1132825.

48.

Jafarzadeh

Shakerian

Farbood

, et al. Effects of eight weeks of resistance exercises on neurotrophins and Trk receptors in Alzheimer model male Wistar rats. Basic Clin Neurosci 2021; 12: 349–359.

49.

Azevedo

Hashiguchi

Campos

, et al. The effects of resistance exercise on cognitive function, amyloidogenesis, and neuroinflammation in Alzheimer's disease. Front Neurosci 2023; 17: 1131214.

50.

Jack

Jr Bennett

Blennow

, et al. NIA-AA Research framework: toward a biological definition of Alzheimer's disease. Alzheimers Dement 2018; 14: 535–562.

51.

Hanseeuw

Betensky

Jacobs

HIL

, et al. Association of amyloid and tau with cognition in preclinical Alzheimer disease: a longitudinal study. JAMA Neurol 2019; 76: 915–924.

52.

Jack

Wiste

Botha

, et al. The bivariate distribution of amyloid-β and tau: relationship with established neurocognitive clinical syndromes. Brain 2019; 142: 3230–3242.

53.

Jack

Jr Knopman

Jagust

, et al. Tracking pathophysiological processes in Alzheimer's disease: an updated hypothetical model of dynamic biomarkers. Lancet Neurol 2013; 12: 207–216.

54.

Oddo

Caccamo

Shepherd

, et al. Triple-transgenic model of Alzheimer's disease with plaques and tangles: intracellular abeta and synaptic dysfunction. Neuron 2003; 39: 409–421.

55.

Hass

Lamere

Capell

, et al. The Swedish mutation causes early-onset Alzheimer's disease by beta-secretase cleavage within the secretory pathway. Nat Med 1995; 1: 1291–1296.

56.

Dong

Tucker

, et al. Treadmill exercise exerts neuroprotection and regulates microglial polarization and oxidative stress in a streptozotocin-induced rat model of sporadic Alzheimer's disease. J Alzheimers Dis 2017; 56: 1469–1484.

57.

Atlante

Valenti

. Mitochondrial complex I and β-amyloid peptide interplay in Alzheimer's disease: a critical review of new and old little regarded findings. Int J Mol Sci 2023; 24: 15951.

58.

Manczak

Reddy

. Abnormal interaction of VDAC1 with amyloid beta and phosphorylated tau causes mitochondrial dysfunction in Alzheimer's disease. Hum Mol Genet 2012; 21: 5131–5146.

59.

Resende

Moreira

Proença

, et al. Brain oxidative stress in a triple-transgenic mouse model of Alzheimer disease. Free Radic Biol Med 2008; 44: 2051–2057.

60.

David

Hauptmann

Scherping

, et al. Proteomic and functional analyses reveal a mitochondrial dysfunction in P301L tau transgenic mice. J Biol Chem 2005; 280: 23802–23814.

61.

Yao

Irwin

Zhao

, et al. Mitochondrial bioenergetic deficit precedes Alzheimer's pathology in female mouse model of Alzheimer's disease. Proc Natl Acad Sci U S A 2009; 106: 14670–14675.

62.

Birnbaum

Wanner

Gietl

, et al. Oxidative stress and altered mitochondrial protein expression in the absence of amyloid-β and tau pathology in iPSC-derived neurons from sporadic Alzheimer's disease patients. Stem Cell Res 2018; 27: 121–130.

63.

Eckert

Schmitt

Gotz

. Mitochondrial dysfunction - the beginning of the end in Alzheimer's disease? Separate and synergistic modes of tau and amyloid-β toxicity. Alzheimers Res Ther 2011; 3: 15.

64.

Berry

Trewin

Amitrano

, et al. Use the protonmotive force: mitochondrial uncoupling and reactive oxygen species. J Mol Biol 2018; 430: 3873–3891.

65.

Salminen

Kaarniranta

Haapasalo

, et al. AMP-activated protein kinase: a potential player in Alzheimer's disease. J Neurochem 2011; 118: 460–474.

66.

Won

Kim

, et al. Involvement of AMP-activated-protein-kinase (AMPK) in neuronal amyloidogenesis. Biochem Biophys Res Commun 2010; 399: 487–491.

67.

Donmez

Wang

Cohen

, et al. SIRT1 Suppresses beta-amyloid production by activating the alpha-secretase gene ADAM10 [retracted in: cell 2014; 158: 959]. Cell 2010; 142: 320–332.

68.

Kuhn

Wang

Dislich

, et al. ADAM10 Is the physiologically relevant, constitutive alpha-secretase of the amyloid precursor protein in primary neurons. EMBO J 2010; 29: 3020–3032.

69.

Son

Shin

Byun

, et al. Metformin facilitates amyloid-β generation by β- and γ-secretases via autophagy activation. J Alzheimers Dis 2016; 51: 1197–1208.

70.

Chen

Zhou

Wang

, et al. Antidiabetic drug metformin (GlucophageR) increases biogenesis of Alzheimer's amyloid peptides via up-regulating BACE1 transcription. Proc Natl Acad Sci U S A 2009; 106: 3907–3912.

71.

Domise

Didier

Marinangeli

, et al. AMP-activated protein kinase modulates tau phosphorylation and tau pathology in vivo. Sci Rep 2016; 6: 26758.

72.

Giebultowicz

Kapahi

. Circadian clocks and metabolism: the nutrient-sensing AKT and TOR pathways make the link. Curr Biol 2010; 20: R608–R609.

73.

Yang

Pascual-Guiral

Ponce

, et al. Reducing the levels of Akt activation by PDK1 knock-in mutation protects neuronal cultures against synthetic amyloid-beta peptides. Fron Aging Neurosci 2018; 9: 435.

74.

Pietri

Dakowski

Hannaoui

, et al. PDK1 Decreases TACE-mediated α-secretase activity and promotes disease progression in prion and Alzheimer's diseases. Nat Med 2013; 19: 1124–1131.