Chronic Sleep Deprivation Exacerbates Learning-Memory Disability and Alzheimer’s Disease-Like Pathologies in AβPP swe /PS1 Δ E9 Mice

Abstract

Recently, there is an increasing concern over the association between sleep disorders and Alzheimer’s disease (AD). Clinical observations have reported that chronic sleep deprivation (SD) may serve as a risk factor for AD. However, the pathological evidence for this assumption is still lacking. In the present study, we examined the potential impacts of chronic SD on learning-memory and AD-related pathologies in AβPP^swe/PS1^Δ^E9 transgenic (TG) mice and their wild-type (WT) littermates. Results indicated that mice (both TG and WT) exposed to 2-month SD showed an altered amyloid-β protein precursor processing, an elevated level of phosphorylated tau protein, and impaired cognitive performance as compared to non-sleep deprivation (NSD) controls. Moreover, the SD-treated TG mice exhibited more amyloid-β_1-42 production and developed more senile plaques in the cortex and hippocampus than NSD-treated TG mice. In addition, SD caused a striking neuronal mitochondrial damage, caspase cascade activation, and neuronal apoptosis in the hippocampus of both TG and WT mice. More importantly, all these behavioral, neuropathological, and biochemical changes induced by chronic SD were long lasting and were irreversible during a 3-month normal housing condition. Collectively, these results indicate that chronic SD impairs learning and memory, exacerbates AD pathologies, and aggravates the mitochondria-mediated neuronal apoptosis in a long-lasting manner. Our findings provide important experimental evidence to prove that chronic SD is a risk factor for AD.

Keywords

Alzheimer’s disease amyloid-β apoptosis chronic sleep deprivation mitochondria phosphorylated tau protein senile plaques

INTRODUCTION

Alzheimer’s disease (AD) is the most common form of dementia, and it is estimated that the number of patients with AD and other dementias worldwide could reach 66 million by the year 2030 [1]. AD is defined by memory loss, spatial learning disorders, and personality and behavioral changes with insidious onset and fast progression [2]. Only a small portion of AD cases is familial, caused by amyloid-β protein precursor (A βPP), presenilin1 (PS1), and presenilin2 (PS2) genes mutations. In contrast, the etiology for most sporadic AD cases is still unknown, although amyloid-β, tau pathology, and their complicated interactions have been demonstrated to play critical roles in AD pathological progression [3 –5]. Epidemiological studies have identified many potential risk factors for this neurodegenerative disease [6 –8], and increasing evidence suggests that poor sleep quality is among those factors [9 –12]. More importantly, recent mechanistic studies have demonstrated that sleep disorder or circadian rhythm disruption is associated with AD pathologies characterized by amyloid-β deposition and abnormal tau phosphorylation [13, 14]. Clinical studies have shown that sleep deficiency will increase cerebral amyloid-β production [15, 16], and preclinical studies revealed that extended wakefulness causes more amyloid-β plaques deposition in AβPP^swe/PS1^Δ^E9 mouse model of AD [17]. In contrast, good sleep increases the interstitial space of brain to drive the exchange of cerebrospinal fluid with interstitial fluid and promotes the clearance of amyloid-β [16]. Additionally, sleep deprivation (SD) has been associated with GSK3 activation [18], which contributes to tau phosphorylation and the consequent formation of neurofibrillary tangles [19]. Consistently, rodent studies have indicated that 8 weeks chronic SD (4 hours per day) impairs memory, tau metabolism, and synaptic integrity in the 3 × TG mouse model of AD [20].

However, despite these reports on the relationship between sleep and AD, the mechanisms for the SD-caused AD-like pathologies are still not fully demonstrated. In addition, it remains unknown whether the impacts of chronic SD on AD pathologies are transient, reversible, or long-lasting. Therefore, in our present study, we attempt to determine the effects of 2 months of chronic SD on cognitive behaviors and pathological and biochemical progression in AβPP^swe/PS1^Δ^E9 transgenic (TG) mice and their wild-type (WT) littermates. We found that chronic SD causes cognitive impairments in TG and WT mice, changes AβPP processing and increases the level of phosphorylated tau in both TG and WT mice, and aggravates senile plaques deposition in the cortex and hippocampus of TG mice. Furthermore, our results showed that chronic SD can result in mitochondria damage and neuronal apoptosis, which may also be involved in the chronic SD-associated AD pathogenesis. More importantly, all these behavioral, neuropathological, and biochemical changes induced by chronic SD were long lasting and irreversible by 3-month normal housingcondition.

MATERIALS AND METHODS

Mice

The AβPP^swe/PS1^Δ^E9 TG mice and their WT littermates (4 to 4.5 months old) used in this study were bred by male heterozygote and female WT mice. The breeding pairs of mice were purchased from theJackson Laboratory (No. 004462, Bar Harbor, MA, USA). Eighty mice were randomly divided into two batches of 40 mice each and were kept with 12-12 h light-dark cycle. All animal experiment procedures were performed in accordance with the guide for the care and use of laboratory animals by Shanghai Institutes for Biological Sciences of the Chinese Academy of Sciences. All efforts were made to reduce the number of animals and minimize the animals’ suffering.

Sleep deprivation treatment

Each batch of mice was randomly divided into four groups: sleep-deprived TG mice (TG+SD), non-sleep deprived TG mice (TG+NSD), sleep-deprived WT mice (WT+SD), and non-sleep deprived WT mice (WT+NSD). SD process was established by anadapted multiple platform method [21 –23]. Twenty-four small round platforms (3 cm in diameter, 5 cm in height, and 3 cm from each other) were placed in a water tank. During the SD process, mice were placed on these platforms, and the water tank was filled with water at 26 to 28°C, 1 cm beneath the platform [21]. Mice can move from one platform to another, but would fall into the water if they fell asleep. All conditions of the NSD group were similar with those of SD group, except the 11.5 cm diameter of platform [24]. There were 5 to 6 mice placed in one water tank. Sleep deprivation treatment was given from 12:00 PM to 8:00 AM of the next day, and all mice were placed to their home cages from 08:00 AM to 12:00 PM. The SD process lasted for 2 months. All mice were subjected to behavioral test to evaluate their learning and memory ability after SD treatment. After that, one batch of mice were sacrificed, and the other one batch of mice were housed and maintained in their home cages with normal sleep conditions for 3 months until they were about 10 months old. Behavioral tests were carried out before they were sacrificed.

Behavioral test

The learning and memory ability was evaluated by using the IntelliCage system (NewBehavior AG, Zürich, Switzerland). The IntelliCage apparatus is a high-output cognitive ability testing system and it can monitor specific behaviors of mice all day and night. Furthermore, it is a social environment, so animals can behave as freely as in their home cages [25, 26]. The cage size is 55.0 × 37.5 × 20.5 cm³ (L × W × H). Food is ad libitum available to mice in the cage. There are four operant learning chambers (15.0 × 15.0 × 21.0 cm³) containing sensors equipped in four corners of the cage to detect animals’ behaviors [27 –29]. Each animal would be pre-injected an individual transponder subcutaneously on its neck, so sensors in each corner can recognize, record, and transmit the specific behavioral signals (corner visit, nosepoke, and drinking) of each mouse to the controlling computer. Water bottles were placed in each corner. Mice can visit each corner, triggering access to a water bottle by nosepoke, and drink water. Since mice are nocturnal, we only analyze the data in night period from 18:00 PM to 6:00 AM the next day. The whole test procession was divided into three periods. 1) The first day was adaption period: animals can explore the cage freely, adapt nosepoke and drinking from each corner, and water in every corner was available to all animals. 2) After the adaptation period, the following five days were the place-learning period, in which each mouse was assigned only one available corner to drink. The assigned corner was the least preferred of the mouse during the adaptation period. The mouse can only drink water from this assigned corner, while all other three corners were unavailable. 3) Afterwards, there is a four-day reversal learning period, where a new correct corner would be reassigned to each mouse separately. The number of visits and nosepokes in each corner would be recorded. Animals failing to drink or staying in one corner over 180 s would be excluded [30]. Four ratios were defined and calculated as follows for evaluating animals’ behavioral performance: 1) The incorrect visit ratio was the number of times the tested mice visited incorrect corners out of total visits; 2) The incorrect nosepoke ratio was the number of nosepokes the tested mouse made in incorrect corners out of total nosepokes; 3) The ratio of visiting previous correct corner was the number of visits the mouse entered the correct corner of place learning period out of total visits; 4) The ratio of visiting new correct corner was the number of visits the mouse entered the new assigned correct corner out of total visits.

Immunostaining

After behavioral tests, mice were anesthetized and perfused with ice-cold 0.1 M phosphate-buffered saline (PBS) and 4% paraformaldehyde, and then their brains were removed. The tissues were dehydrated using 15% and 30% sucrose solutions [31]. The tissues were coated by Optimal Cutting Temperature (O.C.T.) Compound (Tissue-Tek, 4583, SAKURA, Torrance, USA), and sliced with a cryostat (Leica CM3050S, Bentheim, Germany). A serial of 10 μm slices was cut coronally from olfactory bulb to cerebellum. Each brain slice was rinsed with 0.1 M PBS, treated with 3% H₂O₂ to eliminate endogenous peroxidase activity, blocked in blocking solution (3% goat serum, 0.05% NaN₃, 0.3% triton in PBS) for 30 min. Then the slice was incubated in 6E10 Monoclonal Antibody (SIG-39300, 1:400, Covance, NJ, USA) or Anti-Tau (phospho T231) antibody (ab151559, 1:200, Abcam, Cambridge, UK) overnight at 4°C [32]. The second antibody for immunofluorescence staining was TRITC conjugated anti-mouse (T6528, 1:1000, Sigma, St. Louis, MO, USA). The second antibody for immunohistochemistry staining was goat anti-rabbit (C0153, 1:200, Westang, Shanghai, China). ABC elite kit (PK-6100, Vector, Burlingame, USA) and diaminobenzydine-H₂O₂ reaction were following the incubation of secondary antibody. Dehydration and clearing were carried out in different concentrations of ethanol solutions and dimethylbenzene. Pictures were captured by the microscope (Olympus IX, Tokyo, Japan) equipped with a DP70 CCD digital camera (Olympus, Tokyo, Japan). Three randomly fields per slice were captured, and area occupied by plaques and p-tau positive neurons was measured and counted on 10 slices per animal. Plaque area was calculated by Image-Pro Plus software (Mediacybernetics, Acton, MA, USA) as previously described [7].

Immunoblotting

Mouse frontal cortex and hippocampus were weighed, and sonicated in ice-cold lysis buffer (containing 50 mM Tris, pH 7.4, 150 mM NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS with protease inhibitor phenylmethanesulfonyl fluoride) for two min. The lysate was centrifuged at 13,000 × g, 4°C for 30 min. The supernatant was used for western blotting. The pellet was used to measure the levels of insoluble amyloid-β_1-42 by ELISA. BCA Protein Assay kit (23227, Thermo, Rockford, USA) was used to measure the total protein concentration. The following primary antibodies, including rabbit anti-APP C-terminal monoclonal antibody (A8717, 1:5,000, Sigma, St. Louis, MO, USA), Anti-Tau (phospho T231) antibody (ab151559, 1:200, Abcam, Cambridge, UK), Anti-Cytochrome C antibody (ab133504, 1:20,000, Abcam, Cambridge, UK), PARP (46D11) Rabbit mAb (9532, 1:1000, Cell Signaling, Danvers, USA), Caspase3 polyclonal antibody (19677-1-AP, 1:250, Proteintech, Chicago, IL, USA), mouseanti-actin mAb (A1978, 1:20,000, Sigma, St. Louis, MO, USA), were used as previously described [28 –30]. The second antibody was Anti-mouse IgG, HRP-linked Antibody (7076 S, 1:2,000, Cell Signaling, Danvers, USA) or Anti-rabbit IgG, HRP-linked Antibody (7074 S, 1:2,000, Cell Signaling, Danvers, USA). The target protein bands were analyzed by Image Labtrademark Software (Version 4.1 Bio-Rad, Hercules, CA, Canada).

ELISA assays for amyloid-β_1-42

Hippocampus was weighed, and sonicated inice-cold lysis buffer for 2 min. Lysate was centrifuged at 13,000 × g, 4°C for 30 min. The pellet was mixed with 100 μL 70% formic acid, sonicated and centrifuged at 13,000 × g, 4°C for 20 min. The supernatant was collected for measuring insoluble amyloid-β_1-42 by Human β Amyloid_1-42 ELISA Kit (High-sensitive 292-64401, Wako, Osaka, Japan) as described previously [33]. All experimental procedures were performed according to the manufacturer’s instructions.

Electronic microscopy

Mice were anesthetized and perfused with ice-cold 0.1 M PBS and 4% paraformaldehyde with 1% glutaraldehyde. The hippocampus was removed and samples were prepared as described previously [7]. The 70 nm sections were observed by CM-120 transmission electron microscope (BioTwin, Philips, Holland). All pictures were captured at the magnification of 17,500×. Mitochondria displaying broken crista were regarded as abnormal mitochondria [34]. Ten fields (35 μm²) per mouse were captured randomly, and the number of abnormal mitochondria was counted in a blind manner.

TUNEL assay

The mouse brain slices were prepared as mentioned above. Apoptosis was determined by In Situ Cell Death Detection Kit, Fluorescein (NO. 11684795910, Roche, Mannheim, Germany) as previously described [35]. Detailed protocol was following the manual of the product. Images were captured by the microscope (Olympus IX, Tokyo, Japan) equipped with a DP70 CCD digital camera (Olympus, Tokyo, Japan). Three microscope fields (0.1 mm²) were captured randomly in each slice, and the percentage of TUNEL-positive nuclei in hippocampus pyramidal cell layer was counted and calculated on 6 slices per animal in a blind manner [36].

Statistical analysis

All data were expressed as means±standard error. Student’s t-test, one-way analysis of variance (ANOVA) with SNK post-hoc test and two-way ANOVA were performed by GraphPad Prism (version 5.01, GraphPad software Inc, California, USA). P values less than 0.05 are reported as statistically significant.

RESULTS

Chronic SD impairs place learning and reversal learning abilities

To determine the effects of chronic SD on place learning and reversal learning, the behavioral performances of both AβPP^swe/PS1^Δ^E9 TG mice and WT mice were tested by using the IntelliCage system instantly or 3 months after the termination of SD exposure.

As shown in Fig. 1A to D, during the place learning period, both TG+SD and WT+SD mice made more incorrect visits and nosepokes than TG+NSD and WT+NSD (Fig. 1A, B), respectively, suggesting an impaired learning-memory ability. Moreover, in the reversal learning period, both TG+SD and WT+SD mice made more visits to previously-assigned correct corner, but less visits to the newly assigned correct corner than NSD controls (Fig. 1C, D), suggesting an impaired ability of SD mice to learn a new task.

To determine whether SD has long-lasting effects on learning-memory ability, we re-tested the place learning and reversal learning performance of mice 3 months after the termination of SD. As shown in Fig. 1E to H, during the place learning period, TG+SD mice showed a slightly higher level of incorrect visit and nosepoke ratio than TG+NSD (Fig. 1E, F). Interestingly, the WT+SD mice had a much higher incorrect visit and nosepoke ratio compared with WT+NSD (Fig. 1E, F). In the reversal learning phase, we found that the ratio of visiting previous correct corner was similar between TG+SD and TG+NSD mice (Fig. 1G), while TG+NSD mice showed tendency of preferring to visit the newly assigned correct corner (p = 0.0928) (Fig. 1H). In addition, WT+SD mice still showed a significantly higher ratio of visiting previous correct corner and a lower ratio of visiting newly assigned correct corner than WT+NSD mice (Fig. 1G, H).

Chronic SD increases senile plaques deposition in TG mice

In order to examine the pathological effects of SD, senile plaques were detected instantly (Fig. 2A) or 3 months after SD (Fig. 2B) by immunofluorescence staining in TG mice. Compared with the TG+NSD mice, TG+SD group showed a significant increase of senile plaques deposition in both cortex and hippocampus instantly after SD (Fig. 2C). Quantification analysis showed that plaque area in the cortex and hippocampus of TG+SD mice was about 2.7 and 1.8 folds of that in TG+NSD mice, respectively. Moreover, the plaque area in the cortex and hippocampus of TG+SD mice after 3 months rest remained to be 2.3 and 2.2 folds higher than that of TG+NSD mice (Fig. 2D).

We also measured the level of insoluble amyloid-β_1-42 in TG mice after SD and found that it was significantly higher in TG+SD than TG+NSD mice, and the difference remained significant even 3 months after SD (Fig. 2E). Consistently, the western blotting data suggested that while total AβPP was similar between TG+SD and TG+NSD mice or between WT+SD and WT+NSD mice (Fig. 3Aa, Ab), the ratio of C99/C83 was much higher in TG+SD mice than in TG+NSD mice (Fig. 3Aa, Ac), and the level of C83 in WT+SD mice was much lower than that in WT+NSD mice (Fig. 3Aa, Ad). In addition, the AβPP level after 3 months of rest still showed no significant difference between TG+SD and TG+NSD mice or between WT+SD and WT+NSD mice (Fig. 3Ba, Bb). However, the ratio of C99/C83 in TG+SD mice remained to be higher than that of TG+NSD mice, and the level of C83 in WT+SD mice was still lower than that in WT+NSD mice, although no significantly statistical difference (Fig. 3Ba, Bc, Bd) was reached. These results imply that the effects of chronic SD are long lasting, which can cause damage to both TG and WT mice.

Chronic SD increases phosphorylated tau

Phosphorylated tau is the main component of neurofibrillary tangles, which are another hallmark of AD. Immunohistochemistry staining showed an increase in the number of phosphorylated tau (Thr 231) positive neurons in both TG+SD and WT+SD mice. In the cortex of TG+SD mice, the distribution of phosphorylated tau positive neurons was pervasive, while in the cortex of TG+NSD and WT+SD mice, the distribution of phosphorylated tau was just localized at the parietal lobe area, and there was no obvious phosphorylated tau staining in the cortex of WT+NSD mice (Fig. 4A). Quantification analysis showed that both TG+SD and WT+SD mice had significantly higher phosphorylated tau level than their controls, respectively (Fig. 4Ba). Consistently, western blotting data revealed that phosphorylated tau (Thr 231) protein level was significantly increased after SD in TG and WT mice as compared with their NSD controls (Fig. 4Ca, Cb).

After 3 months of rest, we re-examined the phosphorylated tau immunostaining level in the cortex of the experimental mice, and found that both TG+SD and WT+SD mice still had more phosphorylated tau (Thr 231) positive neurons than their controls (Fig. 4A, 4Bb). Western blotting revealed that phosphorylated tau (Thr 231) protein level was still higher in TG+SD and WT+SD mice than TG+NSD and WT+NSD mice, respectively. These data indicated that the effect of chronic SD on tau phosphorylation is long-lasting to both TG and WT mice.

Chronic SD causes mitochondria dysfunction and neuronal apoptosis

In order to investigate the mechanism underlying the neuropathological changes after SD, we further examined the morphology of hippocampal neuron mitochondria immediately after SD treatment. Hollow arrows indicated abnormal mitochondria, whereas black arrows indicated normal mitochondria. As shown in these electron microscope pictures, chronic SD induced mitochondrial damage displaying mitochondria pyknotic or swollen with broken crista and vacuole degeneration in both TG (Fig. 5Aa) and WT mice (Fig. 5Ac). In contrast, in the hippocampus of TG+NSD mice (Fig. 5Ab), mitochondria were moderately swollen, but most of the crista was normal. In the hippocampus of WT+NSD mice, the mitochondria were normal in morphology (Fig. 5Ad). Statistically, chronic SD in both TG and WT mice caused nearly 80% of neuronal mitochondria dysmorphism (Fig. 5B). In addition, western blotting showed that the protein levels of cytochrome C (Cyto C), cleaved PARP and cleaved caspase 3 were increased after chronic SD in both TG and WT mice (Fig. 5C). TUNEL staining also showed that chronic SD increased the percentage of apoptotic cells in hippocampus pyramidal cell in both TG and WT mice (Fig. 6A, B).

In order to find out whether the change of mitochondria and apoptosis were long lasting, we conducted the above measurements in TG and WT mice after they received 3-months rest. The data indicated that there were still a significant number of swelled mitochondria with tubular cristae and vacuole degeneration in the hippocampal neurons in TG+SD mice (Fig. 7Aa). In TG+NSD mice (Fig. 7Ab), mitochondria also appeared broken cristae and vacuole degeneration, but not as severe as TG+SD mice. Mitochondria in WT+SD and WT+NSD mice (Fig. 7Ac, Ad) were mostly normal in morphology, and they had much lower level of abnormal mitochondria than TG mice (Fig. 7B). Western blotting showed that protein levels of Cyto C and cleaved PARP were still significantly higher in TG+SD mice even after 3-months rest as compared with other three groups. Level of cleaved caspase 3 was similar among the four groups (Fig. 7C). TUNEL staining showed that TG+SD mice exhibited more apoptotic cells than TG+NSD mice, and both TG groups had much more apoptotic cells than WT ones (Fig. 8A, B).

DISCUSSION

Nowadays an increasing number of people are suffering from sleep deficiency because of job related burdens, social and economic stress, and sleep disorders. It has been reported that people with sleep deficiency had poor learning and memory ability [37]. However, pathological proofs for this correlation are still lacking. In this study, we established chronic SD model using AβPP^swe/PS1^Δ^E9 TG mice and their WT littermates. Our results showed that chronic SD induced long-lasting impairment of learning ability in both TG and WT mice. Pathological and biochemical results showed that chronic SD can alter AβPP processing to form more amyloid-β_1-42 and consistently develop more senile plaques deposition. Furthermore, the phosphorylated tau protein level was significantly increased in both SD-treated TG and SD-treated WT mice. Moreover, chronic SD also caused neuronal mitochondria dysmorphism, increased Cyto C release to cytosol, activated caspase, and promoted neuronal apoptosis, and these may help interpreting the behavioral and pathological defects.

Previous studies on the relationship between sleep and cognition have indicated that sleep can strengthen memory, whereas lack of sleep would cause memory defects [38 –40]. Sleep fragmentation may indicate cognitive impairment, and age-dependent sleep fragmentation is associated with a 1.5-fold increased risk of developing dementia [11]. In our study, we also found that place learning and reversal learning ability were impaired after chronic SD. Place learning and reversal learning are separately mainly related with hippocampus and orbito-frontal cortex [41], and it is the first study demonstrating that reversal learning ability is defected by chronic SD. We believe that SD as a stressor can cause damage to both hippocampus and cortex, which can aggravate dementia. Interestingly, we also found that this impaired learning ability in both TG+SD and WT+SD mice remained even after 3 months of rest with normal sleep. These results indicate that chronic SD can damage the learning ability in the TG AD mouse model and in normal WT mice, and this effect seems to be long lasting.

Interestingly, accumulating clinical evidence has also indicated that sleep disorders may be early indicators of amyloid-β pathology and may actually precede the onset of cognitive symptoms in AD. Self-reported sleep problems, as assessed by composite surveys of sleep-related symptoms, have been associated with an increased risk of future development of dementia within 1–9 years [42 –44]. The senile plaque is the most significant hallmark of AD. The main component of senile plaques is amyloid-β, which is one of the hydrolysates of AβPP. There are two pathways for AβPP processing, the amyloidogenic and non-amyloidogenic pathways [45]. One of the products of the amyloidogenic pathway is C99, while C83 is one of the products of the non-amyloidogenic pathway. C99 shares the first amino acid with amyloid-β [46]. A prospective study has shown that increased rest fragmentation at night appeared to exacerbate apolipoprotein E4-affected dementia risk, amyloid plaque burden, and tau pathology [10]. These clinical studies suggest that alterations in therest-activity pattern, particularly a poor sleep consolidation might portend future development of dementia or even cerebral amyloid-β deposition. Indeed, Sprecher et al. has found that cognitively normal adults with cerebral amyloid-β_1-42 deposition on Pittsburgh Compound B (PiB) amyloid positron emission tomography (PET) imaging were more likely to nap frequently and had significantly worse sleep efficiency than those without PiB-positive plaques [13]. Subsequently, Spira et al. also found that cognitively normal older adults who self-report poor sleep were more likely to have amyloid-β plaque pathology in the precuneus [9]. Additionally, Xie et al. has suggested that sleep might regulate extracellular amyloid-β_1-42 levels by enhancing the removal of amyloid-β_1-42 via glymphatic flow [24]. Consistent with these previous research reports, the present data have demonstrated that chronic SD would increase amyloid-β plaques deposition. Moreover, after 3 months of rest with normal sleep, the SD mice still showed more senile plaques than NSD mice. Additionally, compared with NSD mice, TG+SD mice tended to produce less C83 leading to a relatively high C99/C83 ratio without significant change of AβPP expression, indicating that SD may change the hydrolysis pathway of AβPP to produce more amyloid-β. Our ELISA result is consistent with these findings.

Neurofibrillary tangles are another pathological hallmark consisting of abnormal fibrils of paired helical filaments that are mainly composed of phosphorylated tau proteins [47]. In a normal physiological condition, tau protein plays an important role in the process of microtubules dynamics [48], while under pathological circumstance, such as AD, increased tau phosphorylation would cause damage to the stability of microtubules, such as to the cytoskeleton [49]. Among various phosphorylated-positions of tau, threonine 231 (T231) is a very important site in tau pathology. Phosphorylation on T231 affects the stability of microtubules, and might promote tau hyperphosphorylation at other sites [50, 51]. In our experiments, we have found that both the number of p-tau (T231) positive neurons and p-tau (T231) protein level were increased after chronic SD, and these pathological changes still remained even 3 months after SD termination. More interestingly, the phosphorylated tau level was also increased in WT mice after chronic SD. This finding may imply that SD-induced phosphorylated tau pathology may be a key element leading to cognitive impairment regardless of the predisposition ofAD-related gene mutations.

Aside from alterations in amyloid-β and tau pathologies, several studies have demonstrated that SD can exacerbate neuronal injury via several other mechanisms. For example, SD can induce synaptic injury in TG AD mice [20, 52]. For human volunteers, one-single night of SD led to a 20% increase in cerebrospinal fluid levels of neuron-specific enolase and SB-100, two markers of neuronal injury [53]. SD has also been shown to trigger mitochondrial oxidative stress in wake-promoting neurons of the locus ceruleus, and prolonged SD can diminish the protective impacts of SirT3 signaling and lead to neuronal death [54]. Mitochondrial dysfunction has long been regarded as a trigger or accelerator of neurodegenerative diseases, and it could cause stress and damages of cells, leading to various genes expression changes [55]. The mitochondrial swelling or pyknosis and broken cristae are considered as dysmorphism, and the phenomenon of broken cristae also indicates a mitochondria dysfunction. In our study, we found that mice after chronic SD had striking morphology changes in mitochondria. Pyknosis, swelling, and vacuolation of mitochondria took place in nearly all hippocampal neurons in chronic SD-treated mice. Mitochondria impairment can result in learning and memory defects, cause deposition of misfolding proteins, and enhance oxide stress [56], and subsequently accelerate the progression of AD pathology. In our study, we found that Cyto C, one biomarker for mitochondrial apoptosis, was dramatically increased after SD in both TG and WT mice. The release of Cyto C can activate caspase 3 downstream, and then induce a series of reactions that result in cytoskeleton dissolution and DNA damage [57, 58]. Finally, this mitochondria-mediated apoptosis is increased and the impairment of learning behaviors is exacerbated by SD. Interestingly, we also found that such changes in TG mice were remained even 3 months after SD termination, implying an important role of mitochondrial damage in chronic SD-induced AD pathogenesis.

In conclusion, our study explores the effects of chronic SD on both familial AD model and sporadic one, which may facilitate the understanding of the association between chronic SD and AD [20 , 59] (Fig. 9). One of the highlights of our report is that we found that SD can cause tau phosphorylation alteration and mitochondria damage in both TG and WT mice. Therefore, we believe that chronic SD is not only a risk factor for familial AD, but it also contributes to sporadic AD. Particularly, we have found that chronic SD can cause long-lasting learning-memory deficiency, increase senile plaques deposition, and change the metabolism pathway of AβPP, as well as increase the phosphorylation of tau protein. Furthermore, neuronal mitochondria dysfunction after SD may be also involved in the underlying mechanisms.

Authors’ disclosures available online (http://j-alz.com/manuscript-disclosures/15-0774r2).

Footnotes

ACKNOWLEDGMENTS

This work was supported by Chinese National Sciences Foundation (NO. 81171201, 81370470 and 81430021), the Collaborative Innovation Center for Brain Science, and the Program for Liaoning Innovative Research Team in University (LT2015009).

References

Wortmann

(2012) Dementia: A global health priority - highlights from an ADI and World Health Organization report. Alzheimers Res Ther 4, 40.

Bero

, Yan

, Roh

, Cirrito

, Stewart

, Raichle

, Lee

, Holtzman

(2011) Neuronal activityregulates the regional vulnerability to amyloid-beta deposition. Nat Neurosci 14, 750–756.

Stancu

, Vasconcelos

, Terwel

, Dewachter

(2014) Models of β-amyloid induced Tau-pathology: The long and “folded” road to understand the mechanism. Mol Neurodegener 9, 51.

, Li

, Zhao

, Gottesdiener

, Luo

, Paul

(2014) Tau pathogenesis is promoted by Aβ1-42 but not Aβ1-40. Mol Neurodegener 9, 52.

Rankin

, Sun

, Gamblin

(2007) Tau phosphorylation by GSK-3beta promotes tangle-like filament morphology. Mol Neurodegener 2, 12.

Yang

, Hou

, Xu

, Zhang

, Shi

, Wang

, Hu

, Chen

(2011) Sleep deprivation impairs spatial learning and modifies the hippocampal theta rhythm in rats. Neuroscience 173, 116–123.

Zhang

, Li

, Zhang

, Xie

, Li

, Yang

, Heng

, Du

, Doody

, Le

(2013) Prenatal hypoxia may aggravate the cognitive impairment and Alzheimer’s disease neuropathology in APPSwe/PS1A246E transgenic mice. Neurobiol Aging 34, 663–678.

Matrone

, Djelloul

, Taglialatela

, Perrone

(2015) Inflammatory risk factors and pathologies promoting Alzheimer’s disease progression: Is RAGE the key?. Histol Histopathol 30, 125–139.

Spira

, Gamaldo

, An

, Wu

, Simonsick

, Bilgel

, Zhou

, Wong

, Ferrucci

, Resnick

(2013) Self-reported sleep and β-amyloid deposition in community-dwelling older adults. JAMA Neurol 70, 1537–1543.

10.

Lim

, Yu

, Kowgier

, Scheneider

, Buchman

, Bennett

(2013) Modification of the relationship of the apolipoprotein E ɛ4 allele to the risk of Alzheimer’s disease and neurofibrillary tangle density by sleep. JAMA Neurol 70, 1544–1551.

11.

Lim

, Kowgier

, Yu

, Buchman

, Bennett

(2013) Sleep fragmentation and the risk of incident Alzheimer’s disease and cognitive decline in older persons. Sleep 36, 1027–1032.

12.

, Lucey

, Holtzman

(2014) Sleep and Alzheimer’s disease pathology–a bidirectional relationship. Nat Rev Neurol 10, 115–119.

13.

Sprecher

, Bendlin

, Racine

, Okonkwo

, Christian

, Koscik

, Sager

, Asthana

, Johnson

, Benca

(2015) Amyloid burden is associated with self-reported sleep in nondemented late middle-aged adults. Neurobiol Aging 36, 2568–2576.

14.

Musiek

, Xiong

, Holtzman

(2015) Sleep, circadian rhythms, and the pathogenesis of Alzheimer disease. Exp Mol Med 47, e148.

15.

Jagust

, Mormino

(2011) Lifespan brain activity, β-amyloid, and Alzheimer’s disease. Trends Cogn Sci 15, 520–526.

16.

Slats

, Claassen

, Verbeek

, Overeem

(2013) Reciprocal interactions between sleep, circadian rhythms and Alzheimer’s disease: Focus on the role of hypocretin and melatonin. Ageing Res Rev 12, 188–200.

17.

Kang

, Lim

, Bateman

, Lee

, Smyth

, Cirrito

, Fujiki

, Nishino

, Holtzman

(2009) Amyloid-beta dynamics are regulated by orexin and the sleep-wake cycle. Science 326, 1005–1007.

18.

Benedetti

, Serretti

, Colombo

, Lorenzi

, Tubazio

, Smeraldi

(2004) A glycogen synthase kinase 3-β promoter gene single nucleotide polymorphism is associated with age at onset and response to total sleep deprivation in bipolar depression. Neurosci Lett 368, 123–126.

19.

Avila

, Lucas

, Perez

, Hernandez

(2004) Role of tau protein in both physiological and pathological conditions. Physiol Rev 84, 361–384.

20.

Di Meco

, Joshi

, Pratico

(2014) Sleep deprivation impairs memory, tau metabolism, and synaptic integrity of a mouse model of Alzheimer’s disease with plaques and tangles. Neurobiol Aging 35, 1813–1820.

21.

Silva

, Abilio

, Takatsu

, Kameda

, Grassl

, Chehin

, Medrano

, Calzavara

, Registro

, Andersen

, Machado

, Carvalho

, Ribeiro Rde

, Tufik

, Frussa-Filho

(2004) Role of hippocampal oxidative stress in memory deficits induced by sleep deprivation in mice. Neuropharmacology 46, 895–903.

22.

Mirescu

, Peters

, Noiman

, Gould

(2006) Sleep deprivation inhibits adult neurogenesis in the hippocampus by elevating glucocorticoids. Proc Natl Acad Sci U S A 103, 19170–19175.

23.

Alzoubi

, Khabour

, Rashid

, Damaj

, Salah

(2012) The neuroprotective effect of vitamin E on chronic sleep deprivation-induced memory impairment: The role of oxidative stress. Behav Brain Res 226, 205–210.

24.

Xie

, Kang

, Xu

, Chen

, Liao

, Thiyagarajan

, O’Donnell

, Christensen

, Nicholson

, Iliff

, Takano

, Deane

, Nedergaard

(2013) Sleep drives metabolite clearance from the adult brain. Science 342, 373–377.

25.

Galsworthy

, Amrein

, Kuptsov

, Poletaeva

, Zinn

, Rau

, Vyssotski

, Lipp

(2005) A comparison of wild-caught wood mice and bank voles in the Intellicage: Assessing exploration, daily activity patterns and place learning paradigms. Behav Brain Res 157, 211–217.

26.

Ogi

, Itoh

, Fushiki

(2013) Social behavior is perturbed in mice after exposure to bisphenol A: A novel assessment employing an IntelliCage. Brain Behav 3, 223–228.

27.

Mechan

, Wyss

, Rieger

, Mohajeri

(2009) A comparison of learning and memory characteristics of young and middle-aged wild-type mice in the IntelliCage. J Neurosci Methods 180, 43–51.

28.

Barlind

, Karlsson

, Bjork-Eriksson

, Isgaard

, Blomgren

(2010) Decreased cytogenesis in the granule cell layer of the hippocampus and impaired place learning after irradiation of the young mouse brain evaluated using the IntelliCage platform. Exp Brain Res 201, 781–787.

29.

Too

, Ball

, McGregor

, Hunt

(2013) A novel automated test battery reveals enduring behavioural alterations and cognitive impairments in survivors of murine pneumococcal meningitis. Brain Behav Immun 13, 00461–00463 .

30.

Karlsson

, Kalm

, Nilsson

, Mallard

, Bjork-Eriksson

, Blomgren

(2011) Learning and activity after irradiation of the young mouse brain analyzed in adulthood using unbiased monitoring in a home cage environment. Radiat Res 175, 336–346.

31.

, Zhang

, Li

, Tang

, Liu

, Yang

, Le

(2013) Autophagy enhancer carbamazepine alleviates memory deficits and cerebral amyloid-beta pathology in a mouse model of Alzheimer’s disease. Curr Alzheimer Res 10, 433–441.

32.

Stygelbout

, Leroy

, Pouillon

, Ando

, D’Amico

, Jia

, Luo

, Duyckaerts

, Erneux

, Schurmans

, Brion

(2014) Inositol trisphosphate 3-kinase B is increased in human Alzheimer brain and exacerbates mouse Alzheimer pathology. Brain 137, 537–552.

33.

Julien

, Tremblay

, Phivilay

, Berthiaume

, Emond

, Julien

, Calon

(2010) High-fat diet aggravates amyloid-beta and tau pathologies in the 3xTg-AD mouse model. Neurobiol Aging 31, 1516–1531.

34.

Hsu

, Sagara

, Arroyo

, Rockenstein

, Sisk

, Mallory

, Wong

, Takenouchi

, Hashimoto

, Masliah

(2000) alpha-synuclein promotes mitochondrial deficit and oxidative stress. Am J Pathol 157, 401–410.

35.

Zhang

, Soto

, Park

, Viswanath

, Kuwada

, Abel

, Wang

(2010) Nuclear receptor SHP, a death receptor that targets mitochondria, induces apoptosis and inhibits tumor growth. Mol Cell Biol 30, 1341–1356.

36.

Wang

, Liu

, Jia

, Zhao

, Cui

, Wang

(2014) Resveratrol protects neurons and the myocardium by reducing oxidative stress and ameliorating mitochondria damage in a cerebral ischemia rat model. Cell Physiol Biochem 34, 854–864.

37.

Mednick

, Nakayama

, Cantero

, Atienza

, Levin

, Pathak

, Stickgold

(2002) The restorative effect of naps on perceptual deterioration. Nat Neurosci 5, 677–681.

38.

Stickgold

(2006) Neuroscience: A memory boost while you sleep. Nature 444, 559–560.

39.

Tworoger

, Lee

, Schernhammer

, Grodstein

(2006) The association of self-reported sleep duration, difficulty sleeping, and snoring with cognitive function in older women. Alzheimer Dis Assoc Disord 20, 41–48.

40.

Sperling

, Johnson

(2012) To sleep, perchance to delay dementia. Arch Neurol 69, 118–120.

41.

Dias

, Robbins

, Roberts

(1996) Dissociation in prefrontal cortex of affective and attentional shifts. Nature 380, 69–72.

42.

Hahn

, Wang

, Andel

, Fratiglioni

(2014) A change in sleep pattern may predic Alzheimer’s disease. Am J Geriatr Psychiatry 22, 1262–1271.

43.

Sterniczuk

, Theou

, Rusak

, Rockwood

(2013) Sleep disturbance is associated with incident dementia and mortality. Curr Alzheimer Res 10, 767–775.

44.

Potvin

, Lorrain

, Forget

, Dubé

, Grenier

, Préville

, Hudon

(2012) Sleep quality and 1-year incident cognitive impairment in community-dwelling older adults. Sleep 35, 491–499.

45.

Zhang

, Thompso

, Zhang

, Xu

(2011) APP processing in Alzheimer’s disease. Mol Brain 4, 3.

46.

LaFerla

, Green

, Oddo

(2007) Intracellular amyloid-beta in Alzheimer’s disease. Nat Rev Neurosci 8, 499–509.

47.

Zhang

, Song

, Liu

, Kang

, Kwon

, Duong

, Seyfried

, Hu

, Liu

, Wang

, Cheng

, Sun

, Yu

, Levey

, Ye

(2014) Cleavage of tau by asparagine endopeptidase mediates the neurofibrillary pathology in Alzheimer’s disease. Nat Med 20, 1254–1262.

48.

Brandt

(1996) The tau proteins in neuronal growth and development. Front Biosci 1, 118–130.

49.

Spillantini

, Goedert

(2013) Tau pathology and neurodegeneration. Lancet Neurol 12, 609–622.

50.

Darios

, Muriel

, Khondiker

, Brice

, Ruberg

(2005) Neurotoxic calcium transfer from endoplasmic reticulum to mitochondria is regulated by cyclin-dependent kinase 5-dependent phosphorylation of tau. J Neurosci 25, 4159–4168.

51.

Lin

, Cheng

, Liang

, Ko

, Lo

, Lu

(2007) The binding and phosphorylation of Thr231 is critical for Tau’s hyperphosphorylation and functional regulation by glycogen synthase kinase 3beta. J Neurochem 103, 802–813.

52.

Rothman

, Herdener

, Frankola

, Mughal

, Mattson

(2013) Chronic mild sleep restriction accentuates contextual memory impairments, and accumulations ofcortical Aβ and pTau in a mouse model of Alzheimer’sdisease. Brain Res 1529, 200–208.

53.

Benedict

, Cedernaes

, Giedraitis

, Nilsson

, Hogenkamp

, Vågesjö

, Massena

, Pettersson

, Christoffersson

, Phillipson

, Broman

, Lannfelt

, Zetterberg

, Schiöth

(2014) Acute sleep deprivation increases serum levels of neuron-specific enolase (NSE) and S100 calcium binding protein B (S-100B) in healthy young men. Sleep 37, 195–198.

54.

Zhang

, Zhu

, Zhan

, Fenik

, Panossian

, Wang

, Reid

, Lai

, Davis

, Baur

, Veasey

(2014) Extended wakefulness: Compromised metabolics in and degeneration of locus ceruleus neurons. J Neurosci 34, 4418–4431.

55.

Amadoro

, Corsetti

, Atlante

, Florenzano

, Capsoni

, Bussani

, Mercanti

, Calissano

(2012) Interaction between NH(2)-tau fragment and Abeta in Alzheimer’s disease mitochondria contributes to the synaptic deterioration. Neurobiol Aging 33, 833e1–25.

56.

Bros

, Millward

, Paul

, Niesner

, Infante-Duarte

(2014) Oxidative damage to mitochondria at the nodes of Ranvier precedes axon degeneration in ex vivo transected s. Exp Neurol 261, 127–135.

57.

Sugawara

, Fujimura

, Morita-Fujimura

, Kawase

, Chan

(1999) Mitochondrial release of cytochrome c corresponds to the selective vulnerability of hippocampal CA1 neurons in rats after transient global cerebral ischemia.RC. J Neurosci 19, 39.

58.

Ding

, He

, Li

(2004) Activation of cell death pathway after a brief period of global ischemia in diabetic and non-diabetic animals. Exp Neurol 188, 421–429.

59.

Roh

, Huang

, Bero

, Kasten

, Stewart

, Bateman

, Holtzman

(2012) Disruption of the sleep-wake cycle and diurnal fluctuation of beta-amyloid in mice with Alzheimer’s disease pathology. Sci Transl Med 4, 150ra122.

Chronic Sleep Deprivation Exacerbates Learning-Memory Disability and Alzheimer’s Disease-Like Pathologies in AβPP swe /PS1 Δ E9 Mice

Abstract

Keywords

INTRODUCTION

MATERIALS AND METHODS

Mice

Sleep deprivation treatment

Behavioral test

Immunostaining

Immunoblotting

ELISA assays for amyloid-β1-42

Electronic microscopy

TUNEL assay

Statistical analysis

RESULTS

Chronic SD impairs place learning and reversal learning abilities

Chronic SD increases senile plaques deposition in TG mice

Chronic SD increases phosphorylated tau

Chronic SD causes mitochondria dysfunction and neuronal apoptosis

DISCUSSION

Footnotes

ACKNOWLEDGMENTS

References

ELISA assays for amyloid-β_1-42