Aerobic exercise during exposure to 40Hz light icker protects against early cognitive impairments in Alzheimer’s disease of 3xTg mice

Background: Alzheimer’s disease (AD) is a progressive degenerative brain disease and the primary cause of dementia. At an early stage, AD is generally characterized by memory impairment involving recent experiences owing to dysfunctions of the cortex and hippocampus. The lesion gradually spreads to the association cortex. Early amyloid-β (Aβ) deposition and tau protein expression result in a loss of synaptic function, mitochondrial damage, and increased cell death via microglia and astrocyte activation, which ultimately lead to cognitive decline. Exercise has been identied as a powerful tool for preventing AD-related neuroinammation and cognitive decline, and light ickering at 40 Hz light icker is known to stabilize gamma oscillations and reduce Aβ. Therefore, we investigated whether exercise under 40-Hz light ickering protects against cognitive decline based on analyses of neuroinammation, mitochondrial function, and neuroplasticity in the hippocampus in a 3xTg AD mouse model. Methods: Using a 3xTg-AD model, 5-month-old mice were subjected to 12 weeks of exercise treatment and 40-Hz light ickering independently and in combination. Various factors, including spatial learning and memory, long-term memory, hippocampal Aβ, tau, neuroinammation, pro-inammatory cytokine expression, mitochondrial function, and neuroplasticity, were analyzed. Results: Aβ and tau proteins levels were signicantly reduced in the early stage of AD, resulting in protection against cognitive decline by reduced neuroinammation and pro-inammatory cytokines, improved mitochondrial function, reduced apoptosis, and increased synapse-related protein expression. In particular, exercise under 40-Hz light ickering was more effective than exercise or 40-Hz light ickering alone, resulting in improvements in parameter values to levels in the non-transgenic aged-match control group. Conclusions: In this study, exercise under a special environment, such as 40-Hz light ickering, may exert a protective effect against cognitive decline. We detected synergistic effects of exercise and 40-Hz light ickering on pathological changes in the hippocampus in the early cognitive impairment of AD.

causing Aβ production and aggregation and the pathological accumulation of Aβ deposits beginning several years before the onset of symptoms. In addition to changes in Aβ, loss of synapses in early AD causes cognitive decline and tracks disease progression [1,5]. The activation of microglia and astrocytes around amyloid plaques contributes to AD-related in ammation [6]. Neuroin ammation generally includes the release of reactive oxygen species (ROS), nitric oxide, pro-in ammatory chemokines, and cytokines and potentially includes the release of neurotoxin molecules from activated glia [7]. The overexpression of these mediators induces nerve damage in AD and other neurodegenerative disorders by various mechanisms [8]. Furthermore, the early stage of AD is associated with decreased mitochondrial function and oxidative damage [9]. Mitochondria are essential organs for the normal functioning of neurons, including synapse activation, and mitochondrial dysfunction is related to neurodegenerative diseases [10][11][12]. Given the important role of mitochondrial dynamics in neurons and the fact that morphological abnormalities of mitochondria are common pathological changes observed in AD, mitochondrial abnormalities are likely to be a common pathway leading to critical neurological dysfunction in the development of AD [12][13][14][15]. As such, excessive production of Aβ or plaque deposition causes cognitive impairment owing to nerve damage and loss in various areas of the brain.
Although various AD treatments are currently under development, there are still challenges. Many studies have shown that exercise has a positive effect on brain function. In various age groups, exercise is associated with a reduced risk of cognitive impairment in later life [16] and has a strong effect on neuroin ammation and cognitive decline in AD [17]. Exercise elicits a whole-body response involving metabolism and immunity [18]. Despite these ndings, the bene cial effects of exercise are limited to either a delay or reduction in cognitive decline, rather than a complete improvement in cognitive function in AD. Accordingly, additional treatments are needed to complement the effects of exercise. It has recently been reported that various non-invasive gamma stimuli, such as 40-Hz light icker and auditory stimuli, improve AD-related pathological factors and cognitive function [19][20][21]. Therefore, using a 6-month-old 3xTg-AD mouse model, we investigated the combined effects of aerobic exercise under 40-Hz light icker on cognitive function and related pathological factors in the early stage of AD.

Animals
Mice were kept under the following conditions: arbitrary adjustment of food and water, 25 ± 1 °C, and light from 7 am to 7 pm. Mice were randomly divided into a wild-type (CON) group, 3xTG-AD (AD) group, 3xTg-AD and 40-Hz light icker (AD + 40) group, 3xTg-AD and exercise (AD + EX) group, and 3xTg-AD and exercise with 40-Hz light icker (AD + 40 + EX) group (n = 10 per group). All test animals were 5 months of age. The genotype of each animal was con rmed by a PCR analysis of DNA from tail biopsies. BrdU (Sigma, St. Louis, MO, USA) was administered intraperitoneally (i.p.) at 100 mg/kg/day for 7 days for 4 weeks prior to killing the animal to observe neurogenesis.
Exercise protocol and exposure to 40-Hz light ickering Exercise sessions were initiated in 5-month-old 3xTg mice. Animals in the exercise groups exercised on a treadmill made for animal use once daily in the dark, 6 days per week for 12 consecutive weeks. For acclimation, mice were subjected to 5 min of warm up at a 0° incline at 3 m/min, 30 min of the main exercise at 10 m/min, and 5 min of cool down at 3 m/min for the rst 3 weeks. Subsequently, mice were subjected to 40 min of the main exercise at 11 m/min from weeks 4 to 6, 50 min of the main exercise at 12 m/min from weeks 7 to 9, and 50 min of the main exercise at 13 m/min from weeks 10 to 12. During treadmill running, electrical stimulation was removed to minimize stress. The exposure time to 40-Hz light ickering was the same as the exercise time.

Preparation of tissue samples
Mice were euthanized immediately after the behavior test. To prepare brain slices, the animals were fully anesthetized with ethyl ether, perfused transcardially with 50 mM phosphate-buffered saline (PBS), and then xed with a freshly prepared solution of 4% paraformaldehyde in 100 mM phosphate buffer (pH 7.4). The brains were then removed, post-xed in the same xative overnight, and transferred to a 30% sucrose solution for cryoprotection. Coronal sections with 40-µm thickness were created using a freezing microtome (Leica, Nussloch, Germany). From each group of 10 animals, 5 were used for immunohistochemistry and 5 for used for western blotting and analyses of mitochondrial function. The hippocampal for western blot analyses were immediately stored at -70 °C until use. For immunohistochemistry, two sections from each group were analyzed, resulting in a total of 10 slices.

Morris water maze
The Morris water maze test was conducted to measure spatial learning and memory ability. The test animals were acclimated by swimming freely for 60 s in a pool without a platform one day before the start of training. The educational training was performed 3 times a day for 5 days with a platform. If the animal was unable to nd the location of the platform within 60 s, the experimenter guided the animal to the platform. Then, the animal remained on the platform for 30 s. A probe trial was conducted 24 h after the training session; free swimming for 60 s without a platform was automatically evaluated by video tracking to determine whether the memory of the previous platform was retained or not.
Step-through avoidance test A step-through avoidance test was conducted to measure long-term memory. On the rst day of training, the test animal was placed on a platform illuminated by a halogen light bulb, and the box door was opened. When the test animal entered the dark box, the door was closed, and the animal was allowed to stay for 20 s. This process was performed twice. In the third or nal trial, the door was closed, and a foot shock at 0.2 mA current and 2 s in duration was administered once. After 24 h, the test animal was placed on a platform illuminated with a halogen light bulb. When the box door was opened, the latency to enter the dark box was measured. All times exceeding 300 s were recorded as 300 s.

Immunohistochemistry
To visualize Aβ, neuroin ammation, and cell differentiation, immunohistochemistry was performed to detect glial brillary acidic protein (GFAP) and ionized calcium-binding adaptor protein-1 (Iba-1) with antigen retrieval in CA1 and the dentate gyrus (DG), doublecortin (DCX) in the DG, and Aβ in the CA1 region of the hippocampus. The sections were incubated in PBS for 10 min and then washed three times Laboratories). Antibody-biotin-avidin-peroxidase complexes were visualized using the 3,3, Diaminobenzidine (DAB) Substrate Kit (Vector Laboratories). The slides were air-dried overnight at room temperature, and the coverslips were mounted using Permount.

Immuno uorescence
NeuN/BrdU-positive cells in the DG were evaluated by immuno uorescence. In brief, the brain sections were permeabilized by incubation in 0.5% Triton X-100 in PBS for 20 min, incubated in 50% formamide-2 × standard saline citrate at 65 °C for 2 h, denatured in 2 N HCl at 37 °C for 30 min, and then rinsed twice in 100 mM sodium borate (pH 8.5). The sections were incubated overnight with rat anti-BrdU antibody antibodies were anti-mouse IgG Alexa Fluor-488 and anti-rat IgG Alexa Fluor-560. Images were captured using an FV3000 confocal microscope for a subsequent Z-section (1 µm) analysis (Olympus, Tokyo, Japan).

TUNEL staining
To visualize DNA fragmentation, TUNEL staining was performed using an In Situ Cell Death Detection Kit (Roche Diagnostics, Risch-Rotkreuz, Switzerland) according to the manufacturer's protocol. Sections were post-xed in ethanol-acetic acid (2:1), rinsed, incubated with proteinase K (100 mg/mL), and then rinsed again. Next, they were incubated in 3% H 2 O 2 , permeabilized with 0.5% Triton X-100, rinsed, and incubated in the TUNEL reaction mixture. The sections were rinsed and visualized using Converter-POD with 0.03% DAB, counterstained with Cresyl violet, and mounted onto gelatin-coated slides. The slides were air-dried overnight at room temperature. A cover slip was added using Permount mounting medium.

Western blotting
Hippocampal tissues were homogenized on ice and lysed in lysis buffer containing 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 0.5% deoxycholic acid, 1% Nonidet P40, 0.1% sodium dodecyl sulfate, 1 mM PMSF, and leupeptin 100 mg/mL. The protein content was measured using a Colorimetric Protein Assay Kit (Bio-Rad, Hercules, CA, USA). Thirty micrograms of protein were separated on sodium dodecyl sulfatepolyacrylamide gels and transferred onto a nitrocellulose membrane, which was incubated with mouse β-

Isolation of hippocampal mitochondria
Mitochondria were isolated from the mouse brain using a Mitochondria Isolation Kit for Tissue (Thermo #89801; Thermo Fisher Scienti c) according to the manufacturer's instructions. Brie y, homogeneous suspensions of hippocampal tissue were prepared using a pre-chilled homogenizer (T 10 Basic Ultra-Turrax; Ika, Staufen im Breisgau, Germany). The hippocampal suspensions were spun at 1,000 × g for 5 min at 4 °C. The pellet was suspended in 800 µL of bovine serum albumin/reagent A solution and incubated for 2 min before adding 10 µL of isolation reagent B. After incubation for 5 min with intermittent vortexing, 800 µL of reagent C was added. The resulting cell lysate was centrifuged at 700 × g for 10 min at 4 °C, after which the supernatant was transferred to a new tube and spun at 3,000 × g for 15 min at 4 °C. The supernatant (cytosolic fraction) was collected for analysis. After washing with mitochondrial isolation reagent C, the mitochondrial pellet was lysed with 2% CHAPS in Tris-buffered saline containing protease inhibitors. The samples were stored at -80 °C until use.

Mitochondrial Ca 2+ retention capacity
The mitochondrial calcium retention capacity was tested to assess the susceptibility of the permeability transition pore (PTP) to opening. Brie y, after grinding the hippocampal tissue, overlaid traces of changes in uorescence induced by Calcium Green-5 N were measured continuously (ΔF/min) at 37 °C during state 4 respiration using a Spex FluoroMax 4 spectro uorometer (Horiba Scienti c, Edison, NJ, USA).

Statistical analyses
Cell counting and optical density quanti cation were performed using Image-Pro Plus (Media Cyberbetics Inc., Rockville, MD, USA) attached to a light microscope (Olympus). The data were analyzed using oneway analysis of variance, followed by Tukey post-hoc tests. All values are expressed as means ± standard error of the mean, and p values < 0.05 were considered signi cant.

Results
Effect of exercise and 40-Hz light icker on spatial working learning, memory, and long-term memory in the early 3xTg-AD As summarized in Fig. 1, Morris water maze and step-through avoidance tests were performed to evaluate spatial learning, memory, and long-term memory. Spatial learning was evaluated as the time until the animal visited the platform. In all groups except the AD group (p < 0.001), the time to nd the platform was shorter shortened from day 3, and there were no differences among treatment groups. With respect to the spatial learning ability, the time to nd the platform for each group was as follows: Day 3; CON group (p = 0.022), AD + 40 (p = 0.008), AD + EX (p = 0.001), AD + 40 + EX (p < 0.001), Day 4; CON group (p = 0.001), AD + 40 (p = 0.008), AD + EX (p = 0.001), AD + 40 + EX (p < 0.001), Day 5 (p < 0.001). For the spatial memory test at 24 h after 5 days of training, spatial memory was lower in the AD group than in the CON group (p < 0.001) and increased in the 40-Hz light icker group (p < 0.001), exercise group (p < 0.001), and combination treatment group (p < 0.001). The effect of the combination of the two treatments on spatial memory was greater than that of single treatments (p < 0.001) and spatial memory in the combined treatment group was equal to or better than that in the non-transgenic CON group (p = 0.049). Even long-term memory, as evaluated by the step-through avoidance test, was lower (p < 0.001) in the AD group and higher in the 40-Hz light icker group (p = 0.001), exercise group (p = 0.001), and combination treatment group (p < 0.001) than in the CON group. The combination treatment showed better results than those for the single treatment groups, i.e., the 40-Hz light icker group (p = 0.004) and exercise group (p = 0.005), with an increment to the level observed in the non-transgenic CON group.
The combination treatment showed better results than those for the single treatment with 40-Hz light icker (t-tau/p-tau ratio, each APP p < 0.001) or exercise (t-tau/p-tau ratio, APP each p < 0.001). To identify Aβ-positive cells, CA1 of the hippocampus was analyzed by immunohistochemistry. In the CON group, Aβpositive cells were not found in CA1 of the hippocampus. Compared with the AD group, hippocampal CA1 Aβ-positive cells (p < 0.001) were decreased in the 40-Hz light icker group, exercise group, and combination treatment group, and the effects for the combination treatment were greater than those for the single treatment of 40-Hz light icker (p = 0.001) and exercise (p < 0.001). In particular, Akt/GSK3β/tau/APP protein expression levels improved to the level observed in the non-transgenic CON group.
Effect of exercise exposure with 40-Hz light icker on neuroin ammation and pro-in ammatory cytokines in the hippocampus in early 3xTg-AD To evaluate neuroin ammation in the hippocampus, GFAP-positive astrocytes and Iba-1-positive microglial cells were analyzed by immunohistochemistry (Fig. 3). Compared with the CON group, neuroin ammation was increased in the hippocampal CA1 and DG in the AD group (GFAP and Iba-1: p < 0.001). Compared with the AD group, GFAP and Iba-1 levels were lower in the 40-Hz light icker group, exercise group, and combination treatment group (GFAP (p = 0.002) and Iba-1 (p = 0.001) from CA1, and GFAP (p < 0.001) and Iba-1 (p < 0.001) from DG). Additionally, the combined treatment with 40-Hz light icker and exercise decreased these markers in the CA1 and DG of the hippocampus more substantially than the decreases in response to single treatments (GFAP and Iba-1: p < 0.001). Levels of proin ammatory cytokines in the hippocampus, i.e., TNF-α and IL-6, were analyzed by western blotting. Levels in the CON group were set to 1.0 and used to determine relative values in each group. The expression levels of pro-in ammatory cytokines were higher in the AD group than in the CON group (TNF-α: p < 0.001; IL-6: p < 0.001). These levels were lower in the 40-Hz light icker group (TNF-α: p = 0.003; IL-6: p = 0.004), exercise group (TNF-α: p = 0.008; IL-6: p < 0.001), and combination treatment group (TNF-α, IL-6: each p < 0.001) than in the AD groups. Additionally, the combination treatment of 40-Hz light icker and exercise resulted in greater decreases than either single treatment (40-Hz light icker: TNF-α: p = 0.004, IL-6: p < 0.001; exercise: TNF-α, IL-6: both p = 0.001). Neuroin ammation and pro-in ammatory cytokine expression recovered to the levels in the non-transgenic CON group.
Effect of exercise exposure with 40-Hz light icker on mitochondrial function and PTP (mPTP) opening in the hippocampus of early 3xTg-AD To determine mitochondrial function and PTP in the hippocampus, mitochondrial Ca 2+ retention capacity, H 2 O 2 emission, and levels of ANT1/2, VDAC1, and Cyp-D were evaluated (Fig. 4). In the hippocampus, the mitochondrial Ca 2+ retention capacity was lower in the AD group than in the CON group (p < 0.001). It was higher in the 40-Hz light icker group (p = 0.015), exercise group (p = 0.006), and combined treatment group (p < 0.001) than in the AD group. In addition, the increase in the Ca 2+ retention capacity was greater for the combined treatment than for single treatments (40-Hz light icker: p = 0.006; exercise: p = 0.015).
To determine the mPTP opening sensitivity, the CON group of Ca 2+ retention in the CON group (set to 100%) was used as the baseline. Compared with the CON group, the mPTP opening sensitivity was lower in the AD group (p < 0.001) and higher in 40-Hz light icker (p = 0.040), exercise (p = 0.036), and combined treatment groups (p < 0.001). Additionally, the increase in sensitivity was greater for the combination treatment than for single treatments (40-Hz light icker: p = 0.041; exercise: p = 0.045). Furthermore, mPTP-related membrane proteins were analyzed in isolated mitochondria. Mitochondrial membrane proteins, including ANT1/2, VDAC1, and Cyp-D, were more highly expressed in the AD group than in the CON group (all p < 0.001). Expression was suppressed in the 40-Hz light icker group (ANT1/2: p < 0.001, VDAC1: p = 0.001, Cyp-D: p = 0.001), exercise group (all p < 0.001), and combination treatment group (all p < 0.001). Protein levels were lower for the combination treatment than for single treatments ( emission rate was higher in the AD group (p = 0.001) and was signi cantly lower in the combination treatment group (p = 0.005) but not in other treatment groups. Finally, for the lipid substrate, the H 2 O 2 emission rate was also higher in the AD group than in the CON group (p < 0.001). The emission rate decreased in the 40-Hz light icker group (p = 0.001), exercise group (p = 0.007), and combination treatment group (p < 0.001). The reduction in the emission rate was greater for the combination treatment than for each treatment group (40-Hz light icker: p = 0.031, exercise: p = 0.003). The combination treatment was more effective in improving mitochondrial function and permeability of the hippocampus, with recovery to the levels observed in the non-transgenic CON group.
Effect of exercise exposure with 40-Hz light icker on apoptosis and cell death in the hippocampus in early 3xTg-AD Changes in apoptosis-related proteins in the hippocampus, including Bax, Bcl-2, cytochrome c, and cleaved caspase-3, were analyzed, and TUNEL-positive cells were analyzed to determine cell death (Fig. 5). Values in the CON group value were set to 1.0 and used to obtain relative values for comparisons among groups. Compared with the CON group, Bax, cytochrome c, and cleaved caspase-3 levels were higher and Bcl-2 levels were lower in the AD group (all p < 0.001). The expression levels of Bax, and combined treatment group (p < 0.001) than in the CON group. Additionally, cell death was greater in the combination treatment group than in the single treatment groups (all p < 0.001). Therefore, simultaneous treatment was more effective in inhibiting hippocampal apoptosis and cell death, with recovery to levels observed in the non-transgenic CON group.
Effect of exercise exposure with 40-Hz light icker on BDNF, synaptophysin, and PSD95 in the hippocampus in early 3xTg-AD The expression levels of the synaptic proteins BDNF, PSD 95, and synaptophysin were investigated in the hippocampus (Fig. 6). Levels in the CON group were set to 1.0 to obtain relative values in each group. Compared with the CON group, protein expression levels were decreased in the AD group (all p < 0.001) but were higher in each treatment group, including the 40-Hz light icker group (BDNF: p = 0.032, synaptophysin: p = 0.005, PSD95: p = 0.046), exercise group (BDNF: p = 0.015, synaptophysin: p = 0.001, PSD95: p < 0.001), and combination treatment group (all p < 0.001). The increases in expression were higher in the combination treatment group than in single treatment groups (40-Hz light icker: all p < 0.001, exercise: BDNF and synaptophysin: p < 0.001, PSD95: p = 0.001). Accordingly, the combined treatment was more effective in increasing the expression of synapse-related proteins; in particular, BDNF levels were equal to or better than those in the non-transgenic CON group.
Effect of exercise exposure with 40-Hz light icker on cell differentiation and neurogenesis in the hippocampal DG of early 3xTg-AD DCX-positive cells and NeuN/BrdU-positive cells were evaluated to investigate cell differentiation and neurogenesis in the hippocampus (Fig. 7). The frequencies of hippocampal DCX-positive cells and NeuN/BrdU-positive cells were lower in the AD group than in the CON group (both p < 0.001) but were increased in the 40-Hz light icker group (DCX: p < 0.001, NeuN/BrdU: p = 0.004), exercise group, and combination treatment group (both p < 0.001). The increases in DCX-positive cells (p < 0.001) and NeuN/BrdU-positive cells (40-Hz light icker: p < 0.001, exercise: p = 0.001) were greater with the combined treatment than with single treatments. Therefore, the combination of the two treatments increased cell differentiation and neurogenesis in the hippocampus, with recovery to the levels observed in the non-transgenic CON group. Discussion AD, the primary cause of dementia, begins in the temporal lobe and encroaches on all areas of the brain. Memory impairment is a major symptom because the hippocampus, the central organ involved in memory, is located in the temporal lobe. Similar to many diseases, treatment at the initial stage of AD is crucial. Various studies have shown that 3xTg-AD mice begin to develop de cits in short-term and longterm meomory, spatial learning and memory at 4 months [22], 6 months, and 6.5 months of age [23,24].
Although these ndings provide insight into the starting point of various cognitive de cits, they develop continuously over time [23]. Consistent with previous studies, we detected the impairment of spatial learning, memory, and long-term memory in the AD group. These AD-related cognitive dysfunctions support the amyloid cascade hypothesis, in which disease onset is characterized by changes in Aβ, followed by a series of events, including the accumulation of toxic forms of tau, which induces apoptosis [1]. Amyloid plaque deposition begins in the neocortex several years before the onset of symptoms and gradually spreads to the hippocampus, diencephalon, striatum, brainstem, and nally the cerebellum [25]. In both the human brain and animal models, the expression of APP, presenilin mutations, and plaques, which are related to AD, are not only associated with synaptic loss but also with a de ciency of memory and synaptic plasticity [26][27][28][29]. Pathological forms of Aβ and tau as well as glia-mediated neuroin ammation play roles in synatptotoxicity [30].
Microglial cells and astrocytes exhibit changes in gene expression, morphology, and secretion in response to toxic stimuli in the brain, and these alterations affect other cells, including neurons [1]. The activation of microglial cells and astrocytes in the early stage of AD results in microgliosis, astrogliosis, impairments in amyloid and tau removal, neurotoxin release, and pro-in ammatory cytokine release [31][32][33][34][35]. Previous studies have reported that the frequencies of GFAP-positive astrocytes and Iba-1 microglial cells are increased in the 3xTg-AD hippocampus at 6 and 10 months [36][37][38]. Microglial cells and astrocytes associated with amyloids secrete pro-in ammatory cytokines, such as IL-1β, IL-6, and TNF-α [39]. In various animal models of AD, the entire brain, including the hippocampus and hypothalamus, shows increases in IL-6 and TNF-α at the mRNA and protein levels [40][41][42]. In particular, IL-6 results in the release of a cascade of pro-in ammatory cytokines by microglia and astrocytes [43]. Consistent with previous results, in this study, tau hyperphosphorylation and APP expression were higher, p-Akt and p-GSK3β in the hippocampus were lower, and Aβ-positive cells in CA1 were higher in the AD group than in the control group. These results suggested that the dysregulation of Tau and Aβ increased the expression of TNF-α and IL-6 by activating astrocyte and microglia in the hippocampus. As such, Aβ and tau and phosphorylation are associated with increases in pro-in ammatory cytokine levels via the activation of microglia and astrocytes and contribute to Ca 2+ dysregulation [44], ROS [39], and cell death [45]. Mitochondria provide an important buffer to regulate the calcium concentration during signaling, which is particularly important for excitatory cells, such as neurons [46].
Mitochondrial dysfunction related to Aβ, such as ROS release [47,48] and the disruption of calcium homeostasis [49], is frequently observed in patients with AD and in animal models. The mPTP is associated with Aβ-induced mitochondrial dysfunction, such as a disturbance of intracellular calcium regulation, ROS generation, and the release of pro-apoptotic factors [50]. In previous studies, mitochondria from mice with AD had much lower calcium capacities than those of non-transgenic mouse mitochondria, and the impaired Ca 2+ uptake capacity, starting from 6 months, decreased gradually. This decreased Ca 2+ retention capacity increased the expression of cyclophilin D (cyp-D), a mitochondrial matrix component, and mPTP components, such as voltage-dependent anion channel 1 (VDAC1) of the outer membrane [51,52]. Adenine nucleotide translocator (ANT), a component of the inner membrane, binds to cyp-D, and cyp-D and ANT interact with Aβ. ANT strongly interacts with Aβ to change mPTP regulation and is related to mitochondrial dysfunction [53]. Additionally, H 2 O 2 is an important ROS that induces oxidative damage, and an increase in mitochondrial H 2 O 2 is signi cantly associated with the onset of AD [51,54]. Mitochondrial dysfunction, including increases in mPTP and ROS, can induce increased apoptosis. Increased levels of Cyp-D, VDAC1, and ANT increase apoptosis [51][52][53], and H 2 O 2 penetrates all tissue compartments, triggers oxidative toxicity, and induces apoptosis [55]. Mitochondria play an important role in the regulation of fundamental processes of neuroplasticity [56], and alterations in mitochondrial function are associated with changes in synaptic plasticity [57].
Mitochondrial dysfunction caused by Aβ may be related to neuroplasticity changes. Synapse-related proteins, such as BDNF, PSD95, and synaptophysin, in the hippocampus in the early stage, DCX, and neurogenesis are decreased in various AD animal models [58][59][60][61][62]. In this study, the Ca 2+ retention capacity was lower and H 2 O 2 emission was higher in hippocampal mitochondria in the AD group than in the control group, and mPTP-related proteins were overexpressed, indicating a decline in mitochondrial function. Furthermore, cell death increased, as determined by increases in Bax, a pro-apoptotic factor, cytochrome c, and cleaved caspase-3 and decreases in Bcl-2, an anti-apoptotic factor. Additionally, synapse-related proteins, such as BDNF, PSD95, and synaptophysin, as well as DCX and neurogenesis decreased. In patients with AD, decreased levels of BDNF in the blood and brain and impaired neurogenesis have been observed in the early stages of the disease with decreased cognitive function [63][64][65]. There is a positive correlation between the BDNF concentration and cognitive function [66]. As such, it is believed that Aβ production or aggregation and tau phosphorylation, via direct or indirect pathways, affect neuroin ammation, mitochondrial function, apoptosis, and synapses in various brain regions, including the hippocampus, leading to cognitive decline.
Altered gamma oscillations have been observed in several brain regions in various neurological and mental disorders, including a decrease of spontaneous gamma synchronization in patients with AD and a decrease in gamma power in several AD mouse models [67][68][69][70]. Although gamma oscillations could not be measured in this study, they are associated with functional decline in AD, as established in an APP/PS1 model of AD [71]. Gamma oscillations in the hippocampus are degraded according to the concentration and time of Aβ [72], and damaged mitochondrial function abolishes gamma oscillations in the hippocampal network [73]. In a transgenic mouse model of AD, theta-gamma coupling was found to be defective in the subiculum, the major output area of the hippocampus [74]. Recent work has shown that 40-Hz light icker stimulation, a non-invasive treatment method, reduces Aβ and phosphorylated tau in various AD animal models by entraining gamma oscillations through the visual cortex. This improves spatial learning and memory by reducing the progression of the degenerative state of neurons, improving synaptic function, enhancing neuronal protective factors, reducing DNA damage, and reducing the in ammatory response of microglia [19,20,75].
Furthermore, 40-Hz auditory stimulation boosts hippocampal function by gamma entrainment, and combined visual and auditory stimulation improves cognitive function by reducing amyloid pathology [21]. In this study, Aβ and phosphorylated tau in the hippocampus were attenuated in the 40-Hz light icker group, as were Iba-1-positive microglial cells and GFAP-positive astrocytes. Long-term 40-Hz light icker stimulation through the visual cortex stabilized gamma oscillations and alleviated Aβ, tau, and in ammatory responses in the hippocampus.
In the hippocampus, the mitochondrial Ca 2+ retention capacity and alleviated H 2 O 2 emissions may have improved cognitive function via the inhibition of apoptosis, increased neurogenesis, and improved neuroplasticity, including increased synaptic protein expression. Exercise, another non-invasive intervention, has a positive effect on brain function and is a candidate lifestyle intervention for reducing the incidence of dementia and AD [76]. Physical exercise decreases Aβ and tau levels and increases memory, and high-quality exercise is associated with decreased Aβ in the plasma and brain of patients with AD [77][78][79]. Additionally, exercise is considered a powerful tool to prevent neuroin ammation and prevent the decline of cognitive function [80].
Liu et al. [81] showed that exercise increases the expression of Akt/GSK3β pathway members in the 3xTg-AD model, decreases Aβ deposition, and decreases Iba-1-positive microglial cells and GFAP-positive astrocytes in the hippocampal DG. Exercise at the early stage of AD has a protective effect on cognitive function by reducing Aβ plaques and GFAP-positive astrocytes in the hippocampus and increasing neurogenesis [82]. Exercise with anti-in ammatory effects inhibits pro-in ammatory cytokines, such as TNF-α, IL-6, and IL1β, in the tau-transgenic hippocampus [83].
In this study, the AD exercise group showed increases in Akt/GSK3β pathway expression, decreases in tau and Aβ, and decreases in excessive Iba-1-positive microglial cells and GFAP-positive astrocytes, as well as decreases in TNF-α and IL-6 expression in the hippocampus. Exercise induces an antioxidant defense system, which reduces ROS levels, and adaptation to long-term regular exercise directly decreases ROS production, reduces oxidative damage [80], and may improve mitochondrial function. In brain regions, including the hippocampus, cortex, and cerebellum, increased mitochondrial Ca 2+ accumulation and calcium-induced mPTP opening by exercise increased resistance to exercise and reduced apoptosis, suggesting that exercise has a protective effect against mitochondrial degeneration and cell death [84,85]. Furthermore, exercise inhibits the overexpression of various membrane-related proteins, such as ANT1/2, cyp-D, and VDAC1, and H 2 O 2 , an ROS marker [85]. Therefore, in an AD transgenic and systemic model, Bax, cytochrome c, caspase 3, and caspase 9 levels are decreased and Bcl-2 levels are increased after exercise [86,87]. Brain mitochondrial function increases in a BDNF concentration-dependent manner [88], and BDNF expression increases in response to exercise [89,90]. Neurotrophin is associated with learning, memory, neuronal activity, and plastic responses [91,92]. Decreases in synapse-related proteins, such as PSD95 and synaptophysin, in the hippocampus in the early stage of AD are also increased by exercise [93,94]. Exercise, especially through increased neurogenesis with decreased Aβ in the AD hippocampus and increased BDNF, improves cognitive function [95].
In the exercise group in this study, mitochondrial function improved with an increased Ca 2+ retention capacity, decreased mPTP proteins, including ANT1/2, VDAC1, and cyp-D, decreased H 2 O 2 emissions in the hippocampal mitochondria, and reduced cell death via decreased Bax, cytochrome c, and caspase-3 and increased Bcl-2. Additionally, BDNF, synaptophysin, PSD95, and neurogenesis were increased. In early AD, exercise seems to have a protective effect on cognitive function by improving the neuroplasticity of the mitochondria and hippocampus.
Under speci c environmental conditions, such as 40-Hz light icker, a non-invasive method, exercise seemed to have a greater effect. A limitation of this study was the inability to analyze gamma