Physical Exercise Ameliorates the Cognitive Function and Attenuates the Neuroinflammation of Alzheimer’s Disease via miR-129–5p

Alzheimer’s disease (AD) is one of the major health challenges in the aging population worldwide. It is considered a frequent cause of dementia, and increasingly contributes to global mortality [1]. With the emergence of the aging of the population, the treatment of AD attracts more and more attention and urgently needs to be improved [2]. Neuroinflammation is a common hallmark of the pathogenesis of AD [3]. The continuous neuroinflammation induced by the aberrant activation of microglia leads to the damage and apoptosis of neurons and glial cells [4]. Emerging studies focus on the methods that could inhibit neuroinflammation to improve and exploit novel therapeutic approaches for AD [5].

Physical exercise has been reported to improve the age- or disease-associated atrophy in the brain [6], indicating its protective potential against cognitive decline. Further investigations in the past few years have confirmed that physical exercise has significant protective effects on cognitive function in the elderly [7, 8]. However, understanding the molecular mechanisms responsive for the protective role of physical exercise remains limited. Some inflammatory cytokines, such as interleukin (IL)-1β, IL-6, and tumor necrosis factor (TNF)-α, have been found to be suppressed in AD patients with physical exercise [9], indicating that physical exercise could attenuate the neuroinflammation of AD.

MicroRNAs (miRNAs) are a group of small, noncoding RNAs with an important regulatory function in various cellular and molecular processes [10]. Aberrant expression of miRNAs has been identified in the development of AD, and these functional miRNAs are determined as potential targets for the treatment of this disease [11, 12]. Some of them are involved in the progression of AD by regulating neuroinflammation [13]. MicroRNA-129–5p (miR-129–5p) has been reported to inhibit neuroinflammation in ischemia-reperfusion injury in the spinal cord [14]. However, only a few studies have focused on the role of miR-129–5p in AD progression.

Physical exercise has been reported to improve cognitive dysfunction by regulating miR-132 in AD [15]. A study by Valenti et al. [16]supplied evidence that physical exercise increases the expression of miR-129–5p in progenitor cells promoting osteogenesis. Thus, we deduced that miR-129–5p might also be regulated by physical exercise in AD. To further uncover the molecular mechanisms underlying the protective influence of physical exercise in AD, this study aimed to investigate the role of miR-129–5p in the effects of physical exercise on the cognitive function and neuroinflammation in both AD mice and patients.

APP/PS1 (APPswe/PSEN1dE9) double Tg mice were purchased from the Model Research Center of Nanjing University as the AD animal model. All mice were housed in standard animal facilities with free access to food and water. The 8-month-old mice were randomly divided into 4 groups (n = 12 in each): a control group, a voluntary exercise group (VE), a voluntary exercise + scrambled RNA oligonucleotides negative control group (VE + miR-NC), and a voluntary exercise + miR-129–5p antagomir group (VE + antagomir). The mice in the control group were housed individually in cages with immobilized running wheels, while the mice in the other groups were housed individually in cages with running wheels used for voluntary exercise for 4 weeks. In addition, miR-129–5p antagomir or scrambled RNA (GenePharma, Shanghai, China) was predelivered into the bilateral hippocampus of the mice stereotaxically in the VE + antagomir/miR-NC groups. After the exercise training, 6 mice in each group were randomly selected and euthanized by decapitation, and the hippocampus tissues were collected for further experiments. The remaining mice were used for the Morris water-maze (MWM) evaluation.

A total of 80 AD patients were recruited in the Shandong Provincial Hospital Affiliated to Shandong University from 2015 to 2017. All the patients were diagnosed in accordance with the NINCDS-ADRDA diagnostic criteria for AD [17] and randomly grouped into an exercise group (n = 40) and a matched group (n = 40). The patients in the exercise group received cycling training with a 70% maximum heart rate in the hospital rehabilitation section for 3 months. The matched group underwent health education for 3 months and were asked to not do any aerobic exercise. Blood samples were collected from the participants before and after the study period, and these were centrifuged for serum extraction. The clinical characteristics and the cognitive function of the patients, including age, gender, and the scores for the Minimum Mental State Examination (MMSE), AD Assessment Scale cognition (ADAS-cog), and Neuropsychiatric Inventory Questionnaire (NPI-Q) were recorded and analyzed for subsequent examination.

Total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA, USA), and then reverse-transcribed into cDNA with a PrimeScript RT reagent kit (TaKaRa, Shiga, Japan) following the manufacturer’s instruction. miR-129–5p expression was measured using quantitative real-time PCR (qRT-PCR), which was performed with the SYBR green I Master Mix kit (Invitrogen) on a 7500 REAL-TIME PCR system (Applied Biosystems, USA). The following thermocycling conditions were used: initial denaturation at 95°C for 5 min; 30 cycles at 95°C for 30 s, at 60°C for 30 s, and at 72°C for 20 s; and a final extension at 72°C for 10 min. The final relative expression of miR-129–5p was calculated using the 2^−ΔΔCt method and normalized to U6. The primer sequences were: miR-129–5p, 5′-CAACCTTACCTTTTTGCGGTC-3′ (forward primer) and 5′-TATGCTTGTTCTCGTCTCTG TGTC-3′ (reverse primer); U6, 5′-CTCGCTTCGGCAGCACA-3′ (forward primer) and 5′-AACGCTTCACGAATTTGCGT-3′ (reverse primer).

The cognitive function of the mice was estimated using the MWM test according to the previously described method [18]. In brief, the mice were subjected to the hidden-platform training for 5 days to examine their learning ability, and then the platform was removed on the 6th day for memory ability evaluation. The trials were monitored by a computerized tracking system (Actimetrics software, Evanston, IL, USA), and the observed performances of the mice, including latency to find the hidden platform, distance traveled, time spent in each quadrant, and number of platform crossings, were recorded and analyzed.

After the exercise training, the brains of the mice were collected and the hippocampi extracted. The hippocampus samples were homogenized with lysis buffer (TBS, NaCl 150 mM, and Triton^TM 1%) and centrifuged to isolate the supernatant. The key proinflammatory cytokines, i.e., IL-1β, IL-6, and TNF-α, in the hippocampus supernatant and patients’ serum specimens, were measured using the ELISA kit (Thermo Fisher Scientific, USA) following the manufacturer’s instruction.

All data are presented as mean ± SD and compared using Student’s t test, the χ² test, or one-way ANOVA. Pearson’s correlation analysis was used to evaluate the correlation between indicators. All the statistical analyses were performed using SPSS v21.0 (SPSS Inc., Chicago, IL, USA) and GraphPad Prism v7.0 (GraphPad Software, Inc., USA). Differences were considered to be statistically significant for values of p < 0.05.

To achieve the goal of voluntary exercise, the AD mice included in this study were housed in individual cages with a differing control status of running wheels. By using a voluntary exercise treatment, we found that the expression of miR-129–5p in the hippocampus of the AD mice was significantly increased compared to that in the mice without exercise (p < 0.001; Fig. 1).

To investigate the functional role of miR-129–5b in the beneficial effect of voluntary exercise in AD mice, its expression was successfully silenced by the treatment of miR-129–5p antagomir (p < 0.001; Fig. 2a). According to the MWM test, the cognitive function of the AD mice was, as expected, improved by the voluntary exercise, indicated by the decreased latency to find the hidden platform, the distances travelled during the test, and the increase in time spent in the target quadrant and the number of crossings to the original platform location (all p < 0.05; Fig. 2b–e). What is worth noting is that the improved cognitive function induced by exercise was significantly abrogated by the knockdown of mIR-129–5p in the AD mice according to the MWM test data (all p < 0.05; Fig. 2b–e).

The proinflammatory cytokines were estimated in the hippocampus of the AD mice. The ELISA results show that the relative levels of IL-1β, IL-6, and TNF-α were all reduced by the voluntary exercise (all p < 0.01; Fig. 3). Furthermore, by the knockdown of miR-129–5p in the AD mice, we found that the exercise-induced attenuated IL-1β, IL-6, and TNF-α levels were all elevated (all p < 0.01; Fig. 3).

The AD patients were randomly classified into the matched and exercise groups. The baseline characteristics of the patients listed in Table 1 show that the patients in the 2 groups had no statistical differences in age, gender, and MMSE, ADAS-cog, and NPI-Q scores (all p > 0.05). After 3 months of aerobic exercise, the cognitive function of the patients was significantly improved in the exercise group, demonstrated by the improved MMSE score and decreased ADAS-cog and NPI-Q scores (all p < 0.01; Table 1). However, the changes in MMSE, ADAS-cog, and NPI-Q scores in the matched group showed no statistical significance (all p > 0.05).

The serum expression of miR-129–5p in AD patients was evaluated before and after the exercise. As shown in Figure 4a, serum miR-129–5p expression was significantly upregulated by the 3 months of aerobic exercise in the exercise group (p < 0.001), but showed no statistically significant change in the matched group after 3 months (p > 0.05). Moreover, compared to the matched group, the serum expression of miR-129–5p was clearly higher in the patients of exercise group after 3 months of exercise (p < 0.001; Fig. 4b).

To verify the role of miR-129–5p in the cognition function and inflammation in the AD patients that undertook exercise, the correlations of miR-129–5p with scores of MMST, ADAS-cog, and NPI-Q, and the levels of IL-1β, IL-6, and TNF-α were analyzed. As shown in Table 2, we observed that miR-129–5p expression was positively correlated with MMST score (r = 0.521, p = 0.004), but negatively correlated with ADAS-cog (r = –0.452, p = 0.012) and NPI-Q (r = –0.499, p = 0.023) scores and with IL-1β (r = –0.602, p = 0.001), IL-6 (r = –0.589, p = 0.003), and TNF-α (r = –0.651, p < 0.001) levels.

Physical exercise has been reported to serve a protective role in cognitive function in the elderly with neurodegenerative diseases, including AD [19]. However, the understanding of the underlying mechanisms remains limited. This study used AD mice and patients to provide a novel sight into the molecular mechanisms involved in the protective effect of physical exercise in AD. After 4 weeks of voluntary exercise, we found significantly increased expression of miR-129–5p in the hippocampus of AD mice. By in vivoregulation of miR-129–5p, the improved cognitive function and inhibited inflammatory responses induced by exercise were obviously abrogated by the silence of miR-129–5p. In AD patients, serum expression of miR-129–5p was also found to be elevated by 3 months of aerobic exercise and it correlated with the levels of cognitive function indicators and inflammatory cytokines.

There is increasing evidence of the beneficial effects of physical exercise to protect against or delay the neurodegeneration that occurs in Parkinson’s disease (PD) and AD [20]. For example, Otsuka et al. [21]investigated the effect of preconditioning exercise on the brain damage in focal brain ischemia rats and reported that exercise played a neuroprotective role by enhancing astrocyte proliferation and angiogenesis as well as reducing neuronal apoptosis and oxidative stress. Jang et al. [22]suggested that exercise protects against the neuronal degeneration in PD by attenuating neuroinflammation. Koo et al. [23]also analyzed the neuroprotective effect of exercise in PD and indicated the mediating role of the TLR2/MyD88/NF-κB signaling pathway in the protective influence. As a frequent neurodegenerative disease in the elderly, AD has also been reported to be improved by physical exercise [24, 25]. These previous studies indicated that physical exercise could significantly ameliorate cognitive function and attenuate neuroinflammatory responses in AD [26]. However, there are only a few reports on the molecular mechanisms underlying the effect of exercise in AD.

Deregulated miRNAs have been identified as pivotal molecules in the pathogenesis of various human diseases, including AD [27]. Emerging studies have found that some miRNAs are mediators in the neuroprotective effect of physical exercise. For example, Kou et al. [28]reported that swimming could improve brain aging by regulating autophagy impairment and abnormal mitochondrial dynamics by inhibiting miR-34a. In AD mice, miR-132 was found to mediate the effect of voluntary exercise on cognitive function [15]. In our study, the AD mice were subjected to voluntary exercise and a significant increase in the expression of miR-129–5p was obtained. A previous study by Valenti et al. [16]has reported the regulatory effect of physical exercise on the expression of miR-129–5p in progenitor cells. Moreover, miR-129–5p plays a protective role in ischemia-reperfusion injury in spinal cord by inhibiting neuroinflammation [14]. Thus, we considered that miR-129–5p might also be involved in the neuroprotective effect of exercise in AD.

Given the improved cognitive function and inhibited inflammatory response in AD with physical exercise in the literature [29], the MWM test was conducted for the evaluation of cognition, and the proinflammatory cytokines in the hippocampus were examined for the analysis of neuroinflammation. By regulating the expression of miR-129–5p in AD mice, we found that the exercise-induced improved cognitive function and the reduced inflammatory responses were reversed by the silence of miR-129–5p. Thus, we considered that physical exercise might improve cognition and ameliorate neuroinflammation by promoting miR-129–5p. To further confirm the role of miR-129–5p in AD, the serum expression of miR-129–5p in AD patients was measured after 3 months of aerobic exercise. Similar results were found, i.e., that serum miR-129–5p was elevated by exercise in AD patients. Correlations of miR-129–5p with cognitive function indicators and proinflammatory cytokines were also found. All these data suggested that miR-129–5p serves as a potential mediator for the regulatory effect of physical exercise on the cognitive function and inflammatory responses in AD. A recent study by Zeng et al. [30]gave evidence of miR-129–5p as a functional molecule in the pathogenesis of AD by alleviating nerve injury and inflammatory response through targeting SOX6. However, whether SOX6 is involved in the protective effect of exercise in AD is not clear, and this warrants further investigation to enrich the molecular mechanisms for physical exercise in AD treatment.

In conclusion, our study found elevated expression of miR-129–5p in AD mice and patients after physical exercise. Serum miR-129–5p is correlated with the exercise-induced changes in cognitive function and inflammation in AD patients. The knockdown of miR-129–5p abrogated the neuroprotective effect of exercise on cognition and neuroinflammation in AD mice. This study provides a novel insight into the molecular mechanisms underlying the effect of physical exercise in AD, and miR-129–5p may become a novel therapeutic target for AD treatment.

The animal experiments were carried out in accordance with the guidelines of the National Institutes of Health and Association for Assessment and Accreditation of Laboratory Animal Care and approved by the Institutional Animal Care and Research Advisory Committee of Shandong University, Jinan, Shandong. Approval for the study was obtained from the Ethics Committee of Provincial Hospital Affiliated to Shandong University. Written informed consent was obtained from all the patients.

The authors have no conflicts of interest to declare.

No funding was received.

All authors made substantial contributions to conception and design, analysis, and interpretation of data. Z.L., Q.C., and J.L. performed the experiments regarding AD mice and AD patients. All authors were involved in drafting the manuscript and revising it and they all read and approved the final version.

Zhen Li and Qi Chen contributed equally to this work.