Brain Research Bulletin 164 (2020) 198–207

Contents lists available at ScienceDirect

Brain Research Bulletin

journal homepage: www.elsevier.com/locate/brainresbull

Treadmill exercise ameliorates chronic REM sleep deprivation-induced

anxiety-like behavior and cognitive impairment in C57BL/6J mice
Feng Tai a, 1, Che Wang b, 1, Xin Deng c, 1, Ruojin Li b, Zimeng Guo a, Haiying Quan a, *, Song Li d, e, *
a
School of Physical Education, Liaoning Normal University, Dalian, 116029, China
b
Department of Medicinal Chemistry, School of Chemistry and Chemical Engineering, Liaoning Normal University, Dalian, 116029, China
c
Department of Physical Education, Harbin Engineering University, Haerbin, 150001, China
d
Liaoning Provincial Key Laboratory for Research on the Pathogenic Mechanisms of Neurological Diseases, the First Affiliated Hospital, Dalian Medical University,
Dalian, 116011, China
e
Liaoning Provincial Center for Clinical Research on Neurological Diseases, the First Affiliated Hospital, Dalian Medical University, Dalian, 116011, China

A R T I C L E I N F O A B S T R A C T

Keywords: Various sleep disorders have deleterious effects on mental and cognitive performance. Exercise, as an alternative
Exercise therapeutic strategy, exerts beneficial impacts on human health. In the present study, we aimed to evaluate the
REM sleep deprivation effects of 4 weeks treadmill exercise (4W-TE) on anxiety-like behavior and cognitive performance in mice
Anxiety
exposed to 2 months REM sleep deprivation (2M-SD) (20 h per day). Behavioral performance of mice in elevated
Cognition
plus maze test (EPM), open field test (OFT), Y maze test (YM) and Morris water maze test (MWM) was recorded
Neurotransmitter
Neurotrophic factor and analyzed 28 h after the last day of sleep deprivation. After behavioral tests, various neurotransmitters
including norepinephrine (NE), dopamine (DA), serotonin (5-HT) and γ-aminobutyric acid (GABA) in mouse
hippocampus were quantified using high performance liquid chromatography. The hippocampal levels of insulin-
like growth factor-1 (IGF-1) and brain derived neurotrophic factor (BDNF) were further detected using ELISA.
Behavioral data indicated that 2M-SD exposure induced anxiety-like behaviors and cognitive impairment, as
evidenced by the decreased open-arm entries in EPM, reduced central area travels in OFT, declined spontaneous
alteration in YM and prolonged escaping latency in MWM. In addition, 2M-SD exposure increased NE and DA,
decreased 5-HT and GABA, and reduced IGF-1 and BDNF levels in mouse hippocampus. Interestingly, all these
behavioral, neurochemical and neurobiological changes can be ameliorated by 4W-TE training. In summary,
these findings confirm the beneficial impacts of exercise on health and provide further experimental evidence for
future application of exercise as an alternative therapy against the mental and cognitive problems in patients
with sleep disorders.

1. Introduction mental and cognitive disorders, including anxiety, depression and de­
mentia (Palma et al., 2013; Pires et al., 2016; Pineda, 2017; Zhang et al.,
Sleep is an essential biological process for maintaining human health 2017; Zhang et al., 2019; Riemann et al., 2020).
and the stability of emotion and cognition (Gruber and Cassoff, 2014; Sleep plays important roles in regulating the complex interactions
Cruz et al., 2019). Millions of people worldwide experiences sleep dis­ between brain functions and various neurotransmitters, such as
orders or deficiency on a daily basis (Hublin et al., 2001). A noradrenaline (NE), dopamine (DA), serotonin (5-HT) and γ-amino­
population-based study has indicated that, over the past fifty years, sleep butyric acid (GABA) (Stenberg, 2007). Consequently, sleep disturbances
duration in adult and adolescent Americans has decreased by 1.5 – 2 h can induce alterations of neurotransmission in the central nervous sys­
per night, with over 30% reporting sleep of less 6 h per night (Ford et al., tem (CNS), leading to deteriorating neurological or psychological
2015). In older populations, sleep disturbances are even more common. manifestations (Vogel, 1999; Longordo et al., 2009; Zager et al., 2009;
Poor sleep quality and short sleep duration are gaining attention as Fell et al., 2011; Howell and Schenck, 2015; Narwade et al., 2017). For
potential risk factors for the pathogenesis and progression of various example, rapid eye movement sleep disturbance alters

* Corresponding authors.
E-mail addresses: quan-haiying@lnnu.edu.cn (H. Quan), lisong@dmu.edu.cn (S. Li).
1
These authors contributed equally to this work.

https://doi.org/10.1016/j.brainresbull.2020.08.025
Received 19 February 2020; Received in revised form 12 July 2020; Accepted 22 August 2020
Available online 30 August 2020
0361-9230/© 2020 Elsevier Inc. All rights reserved.
F. Tai et al. Brain Research Bulletin 164 (2020) 198–207

neurotransmission in four distinct brain regions and results in diseases related symptoms, it is important to investigate the neuro­
mania-like behavior in Swiss albino mice (Siddique et al., 2018). chemical and neurobiological basis of how exercise affects those
Moreover, sleep deficiency also disrupts the expressions of neurotrophic abnormal mood and cognitive performance induced by sleep disorders.
factors in CNS, such as brain derived neurotrophic factor (BDNF) and In the present study, we aim to evaluate the effects of 4 weeks treadmill
insulin-like growth factor-1 (IGF-1), both have been reported to closely exercise (4W-TE) on behavior, neurochemical and neurobiological al­
correlate with cognitive function and mental health (Rusch and Gill, terations in mice subjected to 2 months rapid eye movement (REM)
2015; Chennaoui et al., 2016; Schmitt et al., 2016; Dal-Pont et al., 2019; sleep deprivation (2M-SD) (20 h per day). The Behavioral performance
Mohammadipoor-Ghasemabad et al., 2019). Therefore, restoring the of mice was recorded and analyzed 28 h after the last day of sleep
impaired neurotransmission and neurotrophic status in CNS, via either deprivation. After behavioral tests, various neurotransmitters in mouse
pharmacological therapies or non-pharmacological strategies, may hippocampus were quantified. The hippocampal levels of insulin-like
provide promising tools for the management of sleep-related disorders. growth factor-1 (IGF-1) and brain derived neurotrophic factor (BDNF)
Physical exercise, as a well-accepted and low-cost alternative to were further detected.
disease treatment and prevention, has been reported to positively
modify mood, mobility and memory symptoms of various CNS diseases 2. Materials and methods
(Silva et al., 2015; Roh et al., 2016; Chen et al., 2019). Moreover, NE,
DA, 5-HT and GABA are also major neurotransmitters to be modulated 2.1. Animals
by exercise (Lin and Kuo, 2013; Schoenfeld et al., 2013; Vivar and van
Praag, 2017). Interestingly, recent studies have indicated that exercise One hundred male C57BL/6J mice (2 months of age, 22–25 g body
can also enhance BDNF and IGF-1 level (Kohman et al., 2012; Wrann weight) were obtained from Laboratory Animal Center of Dalian Medi­
et al., 2013; Żebrowska et al., 2018). Given that sleep disturbances may cal University. Animals were housed (5 mice per cage) in standard
lead to anxiety and cognition decline and exercise improves CNS conditions (12 h/12 h light-dark cycle with lights on at 8:00 AM,

Fig. 1. Schematic illustration of experimental

design. All experimental procedures were sum­
marized as Fig. 1a. Non-sleep deprived (NSD)
control animals were kept in their home cages
(8:00 AM to 12:00 PM) or large platform cages
(12:00 PM to 8:00 AM of next day, Fig. 1b, right
panel). Animals in 2M-SD and SD + TE groups
were exposed to REM sleep deprivation using
small platform protocol (from day 1 to 56,
12:00 PM to 8:00 AM of next day, Fig. 1b, left
panel). After 4 weeks sleep deprivation, mice in
SD + TE group were subjected to treadmill
training for 4 weeks (from day 29 to 56, 2:00
PM to 4:00 PM, Fig. 1c). Training control ani­
mals were subjected to treadmill training for 4
weeks (4W-TE) without sleep disturbance.
From day 50 to 56, a battery of behavioral tests,
including Y maze (YM), open field test (OFT),
elevated plus maze (EPM), and Morris water
maze (MWM) were performed (12:00 PM to
2:00 PM, Fig. 1d). Animals were sacrificed on
day 57 and hippocampus was dissected for
HPLC analysis or ELISA.

199
F. Tai et al. Brain Research Bulletin 164 (2020) 198–207

temperature 23 ± 1 ◦ C, humidity 50 ± 10 %) with food and water ad × 5) identical sectors (20 × 20 cm) by white stripes. The squares were
libitum. Animal care and procedures were carried out in accordance with subdivided into peripheral and central sector, where the central sector
the Laboratory Animal Care Guidelines approved by the Institutional included the 9 central squares (3 × 3) and the peripheral sector con­
Animal Care Committee at Dalian Medical University. The protocol was tained the squares close to the surrounded wall (20 cm high). The arena
approved by the Institutional Animal Care Committee at Dalian Medical was lit by red-light lamps (60 W). The animals were placed in the central
University (LCKY-2014-29). sector and their activity was video-recorded for 10 min for further
analysis. Motility was scored when an animal crossed a sector border
2.2. Experimental procedures with both its hind-limbs. The following behavioral parameters were
scored: the total frequency of squares crossed, ratio of central squares
All experimental procedures were schematically illustrated in crossed frequency [(central / total) × 100], and ratio of central area
Fig. 1a. Animals were randomized into 4 groups (25 mice per group). duration [(central / total) × 100]. The open field arena was thoroughly
Group 1 was set as non-sleep deprived control group (NSD) with animals cleaned with alcohol between each test (Wang et al., 2014).
kept in their home cages (8:00 AM to 12:00 PM) or large platform cages
(12:00 PM to 8:00 AM of next day, Fig. 1b, right panel) without sleep 2.5.2. Elevated plus maze test
disturbances. Animals in group 2 and group 3 were exposed to REM EPM consists of two open arms (30 × 5 cm) and two closed arms (30
sleep deprivation for 2 months (2M-SD) using small platform protocol × 5 cm, surrounded by 15 cm-high walls), with the two pairs of identical
(from day 1 to day 56, 12:00 PM to 8:00 AM of next day, Fig. 1b, left arms emerging from a central platform (5 × 5 cm) positioned opposite
panel). After 4 weeks sleep deprivation, mice in group 3 were subjected each other. The apparatus was elevated 50 cm above the floor. The test
to treadmill exercise for 4 weeks (SD + TE) (from day 29 to day 56, from was initiated by placing the mouse on the central platform of the maze,
2:00 PM to 4:00 PM, Fig. 1c). Animals in group 4 were subjected to facing one of the open arms, and letting it explore the maze freely for 5
treadmill training for 4 weeks (4W-TE) without sleep disturbance. From min. Mouse behavior was continuously recorded by a video camera
day 50 to 56, a battery of behavioral tests, including open field test placed above the apparatus and analyzed offline by an observer who was
(OFT), elevated plus maze test (EPM), Y maze test (YM) and Morris unaware of the animals grouping. The maze was carefully cleaned with
water maze test (MWM), were performed (from 12:00 PM to 2:00 PM, alcohol and rinsed with water after each test. Behavioral parameters
Fig. 1d). Animals were sacrificed 24 hours after behavioral tests (on day including the frequencies of total arms entries (arm entry = all four paws
57) and the hippocampus was dissected and frozen for high performance into an arm), ratio of open arms entries [(open /total) × 100], time spent
liquid chromatography (HPLC) or enzyme-linked immunosorbent assay in open or close arms and ratio of time spent in open arms [(time in open
(ELISA). Behavioral tests, HPLC analysis and ELISA were performed by arms / session duration) × 100] (Wang et al., 2014).
blinded examiners.
2.5.3. Y maze test
2.3. Sleep deprivation exposure YM task was used to assess basic mnemonic processing by sponta­
neous percent alternation and exploratory activity by total number of
REM sleep deprivation was performed according to our previous arm choices of mice placed into a Y-shape maze apparatus. The arms of
protocol (Qiu et al., 2016). Briefly, twenty-four small round platforms (3 this maze were 21 × 4 cm with 40 cm high walls. Each mouse was placed
cm in diameter, 5 cm in height, and 3 cm from each other) were placed in one of the arms and allowed one 5-min trial of free exploration of the
in a water tank (Fig. 1b, left panel). During 2M-SD process, mice were three arms in the maze. The number of total arm choices and sequence of
placed on these small platforms, and the water tank was filled with arm choices were recorded. Alternation percentage is defined by the
water at 26 to 28 ◦ C, 1 cm beneath the platform. Mice can move from proportion of arm choices that differ from the last two choices. Before
one platform to another, but would fall into the water if they fall asleep. each trial, the maze was cleaned with alcohol to erase any scent cues
All conditions of the NSD control group were similar with those of (Sawmiller et al., 2017).
2M-SD group, except the 11.5 cm diameter of platform (Fig. 1b, right
panel). There were 5 mice placed in one water tank with food and water 2.5.4. Morris water maze
ad libitum. Sleep deprivation treatment was given from 12:00 PM to Spatial learning and memory ability of mice were evaluated by
8:00 AM of the next day, and all mice were placed back to their home MWM test. Briefly, the maze consisted of a circular water tank (120 cm
cages from 08:00 AM to 12:00 PM and gave them an ad libitum sleep diameter, 60 cm height) filled with water (25 ± 1 ◦ C) to a depth of 40
opportunity. cm. Four equally spaced locations around the edge of the pool were used
as start points, which divided the pool into four equal quadrants. An
2.4. Treadmill exercise protocol escape platform (10 cm in diameter) was placed in the middle of one
randomly selected quadrant, 2 cm below the surface of water, and was
All groups were habituated on an eight-channel motor-drive tread­ kept in same position throughout the entire training session. The
mill (YLS-10B, Yuyan Instruments, Shanghai, China). Treadmill exercise training session consisted of 16 trials (60 s each, 4 trials per day, with a
was conducted during daytime from 2:00 PM to 4:00 PM. After three trial interval time of 60 s) for 4 days (day 52 to day 55). Once climbing
days of treadmill familiarization (day 29 to day 31, 10 min/day, 5 rpm), onto the hidden platform, mice were remained there for additional 15 s.
treadmill exercise was performed with a progressive increase from 5 rpm If the mice failed to locate the hidden platform during 60 s training, it
to 10 rpm from day 32 to day 35 and continued at 10 rpm for 3 weeks. was gently guided to the platform and allowed to remain there for 15 s.
The animals in NSD and 2M-SD groups were left in the treadmill, The latency for each mouse to locate the platform was recorded. During
without running, for the same amount of time as the exercise groups the probe test (on day 56), animals were allowed to swim freely in the
(4W-TE and SD + TE). To minimize stress associated with treadmill tank for 60 s, and the time spent in target quadrant, where the hidden
exercise, only gentle tail touching was used to induce mice to run, and platform previously located, was recorded. A visual cue test was con­
no electric or voice stimulant was used. None of the animals were visibly ducted 30 min after the probe test to assess sensorimotor ability and
hurt or died during exercise session. motivation. For this test, the escape platform was set 1 cm above the
water surface and marked with black tape so that the mice could locate
2.5. Behavioral tests the platform using a local visual stimulus rather than relying on spatial
orientation to extra-maze cues. The escape latency was recorded.
2.5.1. Open field test
The open field consisted of a base (100 × 100 cm) divided into 25 (5

200
F. Tai et al. Brain Research Bulletin 164 (2020) 198–207

2.6. Brain sampling and HPLC analysis 2.8. Statistical analyses

On day 57, all mice were anesthetized and hippocampus was Descriptive statistics are presented as mean ± standard error. One
dissected rapidly on ice, homogenized and centrifuged at 12,000 × g for way analysis of variance (ANOVA) followed by post hoc Tukey’s test or
5 min. The supernatant was filtered through 0.22 μm nylon filters before repeated measurement was employed to compare results between
injecting into the HPLC sample injector. Monoamines (DA, NE, 5-HT) multiple groups. Statistical analysis was carried out using GraphPad
and GABA were determined by HPLC with electrochemical detector Prism 7.04. Statistical significance was set at p < 0.05.
according to our previous protocol (Wang et al., 2019). Concentrations
were calculated from the standard curve generated by standard 10–100 3. Results
ng/mL. The values were expressed as percentage of control group.
3.1. Treadmill exercise ameliorates anxiety-like behavior induced by 2M-
2.7. Determination of BDNF and IGF-1 by ELISA SD

Mouse hippocampus was homogenized in lysis buffer and centri­ In EPM test, no significant changes of total arms entries were
fuged at 15,000 × g for 15 min at 4 ◦ C. The supernatants were collected, observed after 2M-SD exposure (p > 0.999, 95% CI -1.676 to 1.756, F (3,
and their protein concentrations were measured with a Bio-Rad protein 96) = 2.484) or 4W-TE training (p = 0.132, 95% CI -3.156 to 0.2758)
assay kit (Bio-Rad, Hercules, CA, USA). BDNF and IGF-1 levels were (Fig. 2a), suggesting an intact locomotor and exploring behavior.
measured using BDNF Emax Immunoassay kit (Promega, Madison, WI) However, those sleep-deprived mice showed decreased entries (p <
and IGF-1 ELISA kit (Abcam, UK) according to the manufacturer’s in­ 0.01, 95% CI 13.46 to 23.68, F (3, 96) = 46.75, Fig. 2b) and traveling
structions. Absorbance at 450 nm was measured with a Victor3 plate time in open arms in EPM (p < 0.01, 95% CI 5.589 to 9.585, F (3, 96) =
reader (PerkinElmer, USA). The quantity of BDNF and IGF-1 was 56.02, Fig. 2d). In addition, sleep deprivation also caused decreased
expressed as percentage of NSD control values. squares crossing (p < 0.01, 95% CI 13.26 to 20.74, F (3, 96) = 69.73,

Fig. 2. Treadmill exercise ameliorates anxiety-like behavior induced by 2M-SD as assessed by EPM. (a) total arms entries; (b) ratio of open arms entries; (c) duration
in close arms; (d) ratio of close arms duration. ** p < 0.01, compared with NSD control group; ## p < 0.01, compared with 2M-SD group.

201
F. Tai et al. Brain Research Bulletin 164 (2020) 198–207

Fig. 3b) and traveling duration in the central section of open field (p <
0.01, 95% CI 1.823 to 3.297, F (3, 96) = 36.84, Fig. 3c) without sig­
nificant change of locomotor activity (total squares crossed, Fig. 3a. p =
0.324, 95% CI -1.48 to 7.08, F (3, 96) = 15.8). All these data suggested
that 2M-SD could induce anxiety-like behavior in mice. As expected, this
elevated anxiety level can be ameliorated by 4 weeks treadmill exercise,
evidenced by the increased open arm entries and open arm traveling
time in EPM (p < 0.01, Fig. 2b, c) and the increased central area trav­
eling in OFT (p < 0.01, Fig. 3b, c).

3.2. Treadmill exercise ameliorates cognitive impairment induced by 2M-

SD

In addition to anxiety assessment, behavioral tests were further

performed to evaluate the learning/memory performance of mice. The
data from MWM task indicated that, compared to those NSD control
mice, 2M-SD-treated mice showed an increased latency to find the
hidden platform during training sessions (Fig. 4a) and a decreased target
quadrant percentage in probe test (Fig. 4b), indicating an impaired
learning and memory ability. This impaired cognitive performance
caused by 2M-SD was further confirmed by Y maze test, as shown by the
decreased percentage of spontaneous alternation (Fig. 4e). Interestingly,
while 4W-TE showed no significant impacts on learning and memory
compared to NSD control (p = 0.222, 95% CI 0.1497 to 2.71, F = 82.95,
Fig.4a; p = 0.255, 95% CI -4.339 to 0.7392, F = 89.33, Fig.4b; p = 0.92,
95% CI -3.354 to 5.434, F = 55.93, Fig.4e), 4W-TE ameliorated the 2M-
SD-induced cognition impairment (p < 0.01, Fig. 4a, b and e).
Escape latency in the visual cue test was decreased compared with
day 4 training and showed no significant difference among 4 groups (p =
0.3269, F (3, 96) = 1.166, Fig. 4d), suggesting a normal sensorimotor
ability and motivation. Moreover, consistent with the intact locomotor
performance in EPM (Fig. 2a) and OFT (Fig. 3a), there were no signifi­
cant differences noted in swim speed (p = 0.8613, F (3, 96) = 0.2498,
Fig. 4c) in MWM and total entries in YM (data not shown) among all
groups, suggesting that the nootropic activity of treadmill exercise is not
due to an altered motor ability.

3.3. Treadmill training reverses 2M-SD-induced alterations in

hippocampal neurotransmission

Exercise and sleep both have close interactions with various neuro­
transmitters, which have been reported to be involved in the regulation
of both emotion and cognition. Therefore, we detected neurotransmit­
ters levels in mouse hippocampus, including NE, DA, 5-HT and GABA.
Our data indicated that 4W-TE training significantly elevated DA (p =
0.0374, F (3, 96) = 88.21), 5-HT (p = 0.0282, F (3, 96) = 103.2) and
GABA (p = 0.0127, F (3, 96) = 170.4) levels in hippocampus without
significant alteration of NE (p = 0.404, F (3, 96) = 143.1) (Fig. 5). 2M-
SD exposure caused significant increases of hippocampal NE (p < 0.01,
95% CI -45.74 to -34.5, Fig. 5a) and DA levels (p < 0.01, 95% CI -38.86
to -26.98, Fig. 5b) together with decreased GABA (p < 0.01, 95% CI
31.23 to 41.97, Fig. 5c) and 5-HT levels (p < 0.01, 95% CI 23.47 to
34.61, Fig. 5d), suggesting an altered hippocampal neurotransmission.
All these 2M-SD-induced neurochemical alterations can be ameliorated
by 4 weeks treadmill exercise (p < 0.01, Fig. 5).

3.4. Treadmill exercise reverses 2M-SD-induced hippocampal BDNF and

IGF-1 decline

The hippocampal levels of neurotrophins BDNF and IGF-1 were Fig. 3. Treadmill exercise ameliorates anxiety-like behavior induced by 2M-SD
measured using ELISA. We found that, compared with NSD control mice, as assessed by OFT. (a) total squares crossed; (b) ratio of central squares
the hippocampal levels of BDNF and IGF-1 were both elevated after 4W- crossed; (c) ratio of central area duration. ** p < 0.01, * p < 0.05, compared
TE training (p < 0.01, 95% CI -9.332 to -1.388, F (3, 96) = 2.655, Fig. 6a; with NSD control group; ## p < 0.01, compared with 2M-SD group.
p = 0.0158, 95% CI -11.31 to -0.8546, F (3, 96) = 2.907, Fig.6b). In
contrast, 2M-SD mice showed significantly decreased levels of BDNF and
IGF-1 in hippocampus (p < 0.01, Fig. 6a, b). Four weeks treadmill

202
F. Tai et al. Brain Research Bulletin 164 (2020) 198–207

Fig. 4. Treadmill exercise ameliorates memory impairment induced by 2M-SD as assessed by MWM and Y maze. (a) latency to reach safe platform during training
sessions in MWM; (b) Percentage of time spent in target quadrant during probe test session in MWM; (c) swimming speed in MWM; (d) Escape latency in visual cue
test of MWM; (e) Percentage of spontaneous alternation in YM. ** p < 0.01, compared with NSD control group; # # p < 0.01, compared with 2M-SD group.

203
F. Tai et al. Brain Research Bulletin 164 (2020) 198–207

Fig. 5. Treadmill training reverses 2M-SD-induced alterations in hippocampal levels of neurotransmitters. Hippocampal levels of NE (a), DA (b), GABA (c) and 5-HT
(d) were determined by HPLC. * p < 0.05, ** p < 0.01, compared with NSD control group; ## p < 0.01, compared with 2M-SD group.

training ameliorated the declined hippocampal levels of BDNF and IGF-1 wheel running reduced the anxiety level of both GFAP-TK mice and wild
in 2M-SD mice (increase 30% and 27% respectively compared with 2M- type control mice on EPM and novelty-suppressed feeding test
SD group, p < 0.01, Fig.6). (Schoenfeld et al., 2016). Moreover, previous clinical trials also revealed
positive impacts of exercise on anxiety in human (Stonerock et al.,
4. Discussion 2015). A meta-analysis has suggested that exercise is effective in
improving anxiety symptoms in people with a current diagnosis of
Exercise has been indicated to be beneficial to human health. In the anxiety (Stubbs et al., 2017). Interestingly, in our present study, we
present study, we found that 4 weeks treadmill exercise ameliorated found four weeks mild treadmill training exerted anxiolytic effect in
anxiety-like behavior and cognitive impairment induced by 2-month mice exposed to chronic REM sleep deprivation. These data further
chronic REM sleep deprivation in mice. In addition, our data further confirmed the beneficial impact of exercise on anxiety under
indicated that all these emotional and cognitive benefits of exercise stress-related conditions. In contrast to these findings, Zielinski et al.,
might be due to the neurochemical and neurotrophic modulations in found that moderate sleep deprivation protocol did not deteriorate
mouse hippocampus. anxiety-related behavior and even exerted an anxielytic tendency as
Previous studies have demonstrated a beneficial effect of exercise on assessed by EPM (open arm percentage 8.70 ± 1.17 in normal sleep
basal levels of anxiety in rodents. For examples, Fulk et al. found that control group v.s. 14.07 ± 2.74 in Sleep restriction + sedentary group)
treadmill training reduces anxiety-like behaviors in rats on EPM and (Zielinski et al., 2013). Furthermore, the exercise protocol used in Zie­
OFT (Fulk et al., 2004). Schoenfeld et al. also revealed that five weeks of linski et al.’s study failed to ameliorate the impacts of REM sleep

204
F. Tai et al. Brain Research Bulletin 164 (2020) 198–207

Fig. 6. Treadmill exercise reverses 2M-SD-induced BDNF and IGF-1 decline in hippocampus. BDNF (a) IGF-1 (b) were determined by ELISA. ** p < 0.05, * p < 0.05,
compared with NSD control group; # # p < 0.01, compared with 2M-SD group.

deprivation on anxiety-like behaviors. These differences might be due to arousal (Volkow et al., 2008), the DA content remains unaffected in
the protocols discrepancies in both sleep deprivation and exercise striatum (Lara-Lemus, 1998) or decreased in hippocampus (Wen et al.,
training. In Zielinski et al.’s study, using a 2 × 2 experimental design, 2013) in chronic sleep deprived rats. Our data revealed that after 8
they aimed to evaluate the beneficial effects of exercise (1 h per day, 6 weeks sleep deprivation, mice showed an increased DA content in hip­
days per week for 11 weeks) on the cognition and anxiety-like behavior pocampus, which is consistent with the elevated NE concentration
and further explore the detrimental impacts of chronic but relative mild (Fig. 5a,b). The reasons for these conflict findings compare with those by
sleep deprivation (11 weeks, 4 h per day). Comparing with this Wen et al. are possibly due to the different experimental design of sleep
well-designed study by Zielinski et al., our present study aimed to test deprivation (18 h daily for 3 weeks v.s. 20 h daily for 8 weeks) or
the potential ameliorating impacts of treadmill exercise (4 weeks, 2 h different animal species (rats v.s. mice). Much more interestingly, in our
per day) on behavioral alterations induced by REM sleep deprivation (8 present study, we found that sleep-deprivation-caused DA elevation can
weeks, 20 h per day). While Zielinski et al. performed sleep deprivation be blocked by 4W-TE, whereas 4W-TE alone also increased DA (Fig.5b).
and exercise training in a parallel way, we conducted the treadmill The exact neurochemical mechanisms underlying the modulation of
training and sleep deprivation protocol in a therapeutic manner, i.e. 4 exercise on sleep-deprivation-induced DAergic alterations need further
weeks after the initiation of sleep deprivation exposure (Fig. 1). exploration. In contrast to the elevated NE and DA, 2M-SD significantly
Despite the positive impacts on emotion regulation, previous studies decreased the hippocampal levels of 5-HT and GABA, both inhibitory
have also indicated a beneficial effect of exercise on cognition. Meta- neurotransmitter that has a close correlation with sleep control (Got­
analyses of randomized controlled trials have revealed that exercise tesmann, 2002; Monti, 2011; Cespuglio, 2018; Hernández-Vázquez
programs might play an important role in cognition and activities of et al., 2019). Interestingly, serotonin and GABA have been reported to
daily living in patients with dementia (Li et al., 2019). Ichinose et al. serve as key modulator of anxiety (Lydiard, 2003; Kalueff and Nutt,
have also demonstrated that both intermittent and continuous exercise 2007; Zangrossi and Graeff, 2014; Liu et al., 2019) and cognition
may improve cognitive function (Ichinose et al., 2020). Additionally, (Strasser et al., 2016; Ma et al., 2018; Schmidt-Wilcke et al., 2018;
previous studies have further indicated that exercise protects against Chakraborty et al., 2019). Therefore, the altered neurotransmission
cognitive decline in rodents under chronically stressed condition (Kwon including NE, DA, 5 H T and GABA in hippocampus induced by 2M-SD
et al., 2013; Lee et al., 2018). In consistent with these findings, in our may contribute the anxiety- and amnesia-like behavior. Moreover,
present study, we found that 4 weeks treadmill exercise ameliorated the moderate exercise can ameliorate these neurochemical alterations in
impaired learning and memory ability in sleep-deprived mice. mouse hippocampus induced by sleep deprivation, providing neuro­
Considering the important roles of hippocampus in cognition and chemical interpretations for the beneficial impacts of exercise on
emotion (Femenía et al., 2012; Bannerman et al., 2014; Anacker and cognitive and mood disorders in individuals suffering sleep
Hen, 2017), we further determined the levels of various neurotrans­ disturbances.
mitters and neurotrophic factors levels in mouse hippocampus, to Despite the neurochemical changes, sleep disorders have been re­
further explore the neurochemical and neurobiological mechanisms ported to disrupt neurotrophic factors expressions in CNS, such as BDNF
underlying the anxiety and amnesia-like behaviors induced by chronic and IGF-1, both have been closely correlated with cognitive function and
sleep deprivation and the ameliorating effects of treadmill exercise. We mood disorders. Dal-Pont et al. found that 36 h sleep deprivation
found that 4W-TE training elevated DA, GABA and 5-HT concentrations. decreased the BDNF, NGF, and GDNF levels in the frontal cortex and
In addition, 2M-SD exposure in mice increased NE and DA but decreased hippocampus of mice and induced manic-like behavior (Dal-Pont et al.,
5-HT and GABA levels in hippocampus, whereas treadmill training can 2019). Mohammadipoor-Ghasemabad et al. suggested that sleep depri­
ameliorate all these neurochemical alterations (Fig. 5). NE, as a pivotal vation can decrease hippocampal BDNF expression together with an
stress hormone and monoamine neurotransmitter, has been associated impaired spatial cognition in ovariectomized rats (Mohammadipoor-­
with general arousal and stress-responses, such as stress-related mood Ghasemabad et al., 2019). Sleep deprivation protocol in Zielinski et al.’s
disorders (Kalk et al., 2011; Liu et al., 2018) and stress-associated study showed a tendency of declined BDNF expression in CA1 region of
amnesia (Tully and Bolshakov, 2010; Shansky and Lipps, 2013). mouse hippocampus (Zielinski et al., 2013). Consistent with these pre­
Importantly and consistent with our present findings, previous studies vious findings, our data confirmed the impacts of sleep deprivation on
have also documented that sleep deprivation can increase NE level in the the expressions of neurotrophic factors. This declined neurotrophic level
brain (Gulyani and Mallick, 1993; Das and Mallick, 2008; Daniele et al., in hippocampus after chronic stressful stimuli may result in an impaired
2017), which can be reversed by exercise (Daniele et al., 2017). The neurogenesis and synaptic plasticity, which consequently leading to
effects of sleep deprivation on dopaminergic systems remain elusive. mood disorders (Jiang and Salton, 2020) and cognitive damage (Toda
While one night sleepless increase DA level in human brain to maintain and Gage, 2018). Interestingly, recent studies have indicated that

205
F. Tai et al. Brain Research Bulletin 164 (2020) 198–207

exercise can enhance BDNF and IGF-1 level (Kohman et al., 2012; Wrann Femenía, T., et al., 2012. Dysfunctional hippocampal activity affects emotion and
cognition in mood disorders. Brain Res. 1476, 58–70.
et al., 2013; Żebrowska et al., 2018). All these prior works further
Ford, E.S., Cunningham, T.J., Croft, J.B., 2015. Trends in Self-Reported Sleep Duration
supported our present data that exercise can reverse sleep among US Adults from 1985 to 2012. Sleep. 38, 829–832.
deprivation-induced anxiety-like behavior and amnesia possibly via Fulk, L.J., et al., 2004. Chronic physical exercise reduces anxiety-like behavior in rats.
restoring impaired neurotrophic level. Int. J. Sports Med. 25, 78–82.
Gottesmann, C., 2002. GABA mechanisms and sleep. Neuroscience. 111, 231–239.
Gruber, R., Cassoff, J., 2014. The interplay between sleep and emotion regulation:
5. Conclusion conceptual framework empirical evidence and future directions. Curr. Psychiatry
Rep. 16, 500.
Gulyani, S., Mallick, B.N., 1993. Effect of rapid eye movement sleep deprivation on rat
In conclusion, our present study showed that exercise can reduce brain Na-K ATPase activity. J. Sleep Res. 2, 45–50.
anxiety-like behavior and ameliorate cognitive decline induced by Hernández-Vázquez, F., Garduño, J., Hernández-López, S., 2019. GABAergic modulation
chronic REM sleep deprivation. These present findings add to the un­ of serotonergic neurons in the dorsal raphe nucleus. Rev. Neurosci. 30, 289–303.
Howell, M.J., Schenck, C.H., 2015. Rapid Eye Movement Sleep Behavior Disorder and
derstanding of behavior and brain alterations in connection with sleep Neurodegenerative Disease. JAMA Neurol. 72, 707–712.
disorders and provide experimental evidence for the therapeutic po­ Hublin, C., et al., 2001. Insufficient sleep–a population-based study in adults. Sleep. 24,
tential of exercise against sleep-related disorders. 392–400.
Ichinose, Y., et al., 2020. Comparison of the Effects of Continuous and Intermittent
Exercise on Cerebral Oxygenation and Cognitive Function. Adv. Exp. Med. Biol.
Fundings 1232, 209–214.
Jiang, C., Salton, S.R., 2020. Neuroprotective roles of neurotrophic growth factors in
mood disorders. In: Gozes, Illana, Levine, Joseph (Eds.), Neuroprotection in Autism,
This work was supported by Science and Technology Project of Schizophrenia and Alzheimer’s Disease. Academic Press, pp. 145–172, 2020.
Liaoning Department of Education, China, Natural Science Foundation Kalk, N.J., Nutt, D.J., Lingford-Hughes, A.R., 2011. The role of central noradrenergic
of Liaoning Department of Science and Technology, China dysregulation in anxiety disorders: evidence from clinical studies.
J. Psychopharmacol. 25, 3–16.
(20180550306 and 2019-ZD-0467), Liaoning Provincial Project for
Kalueff, A.V., Nutt, D.J., 2007. Role of GABA in anxiety and depression. Depress.
Innovative Talents, China, the Talent Program of the First Affiliated Anxiety. 24, 495–517.
Hospital Dalian Medical University, China (2017YC005). Kohman, R.A., et al., 2012. Wheel running attenuates microglia proliferation and
increases expression of a proneurogenic phenotype in the hippocampus of aged mice.
Brain Behav. Immun. 26, 803–810.
CRediT authorship contribution statement Kwon, D.H., et al., 2013. Exercise ameliorates cognition impairment due to restraint
stress-induced oxidative insult and reduced BDNF level. Biochem. Biophys. Res.
Feng Tai: Conceptualization, Investigation, Methodology, Writing - Commun. 434, 245–251.
Lara-Lemus, A., 1998. Biochemical effects induced by REM sleep deprivation in naive
original draft. Che Wang: Data curation, Funding acquisition, Investi­ and in D-amphetamine treated rats. Neurobiology. 6, 13–22.
gation, Methodology, Writing - original draft. Xin Deng: Data curation, Lee, H., et al., 2018. Intermittent intense exercise protects against cognitive decline in a
Investigation, Methodology. Ruojin Li: Data curation, Investigation, similar manner to moderate exercise in chronically stressed mice. Behav. Brain Res.
345, 59–64.
Methodology. Zimeng Guo: Data curation, Investigation, Methodology. Li, X., et al., 2019. Effectiveness of Exercise Programs on Patients with Dementia: A
Haiying Quan: Conceptualization, Project administration, Supervision, Systematic Review and Meta-Analysis of Randomized Controlled Trials. Biomed.
Validation, Writing - review & editing. Song Li: Conceptualization, Res. Int. 2019, 2308475.
Lin, T.W., Kuo, Y.M., 2013. Exercise benefits brain function: the monoamine connection.
Formal analysis, Funding acquisition, Project administration, Resources, Brain Sci. 3, 39–53.
Software, Supervision, Validation, Visualization, Writing - original draft, Liu, Y., et al., 2019. Dysfunction in Serotonergic and Noradrenergic Systems and Somatic
Writing - review & editing. Symptoms in Psychiatric Disorders. Front. Psychiatry. 10, 286.
Liu, Y., Zhao, J., Guo, W., 2018. Emotional Roles of Mono-Aminergic Neurotransmitters
in Major Depressive Disorder and Anxiety Disorders. Front. Psychol. 9, 2201.
Declaration of Competing Interest Longordo, F., Kopp, C., Lüthi, A., 2009. Consequences of sleep deprivation on
neurotransmitter receptor expression and function. Eur. J. Neurosci. 29, 1810–1819.
Lydiard, R.B., 2003. The role of GABA in anxiety disorders. J. Clin. Psychiatry. 64 (Suppl
The authors report no declarations of interest. 3), 21–27.
Ma, S., et al., 2018. Dual-transmitter systems regulating arousal, attention, learning and
References memory. Neurosci. Biobehav. Rev. 85, 21–33.
Mohammadipoor-Ghasemabad, L., et al., 2019. Hippocampal microRNA-191a-5p
Regulates BDNF Expression and Shows Correlation with Cognitive Impairment
Anacker, C., Hen, R., 2017. Adult hippocampal neurogenesis and cognitive flexibility -
Induced by Paradoxical Sleep Deprivation. Neuroscience. 414, 49–59.
linking memory and mood. Nat. Rev. Neurosci. 18, 335–346.
Monti, J.M., 2011. Serotonin control of sleep-wake behavior. Sleep Med Rev. 15,
Bannerman, D.M., et al., 2014. Hippocampal synaptic plasticity, spatial memory and
269–281.
anxiety. Nat. Rev. Neurosci. 15, 181–192.
Narwade, S.C., Mallick, B.N., Deobagkar, D.D., 2017. Transcriptome Analysis Reveals
Cespuglio, R., 2018. Serotonin: its place today in sleep preparation, triggering or
Altered Expression of Memory and Neurotransmission Associated Genes in the REM
maintenance. Sleep Med. 49, 31–39.
Sleep Deprived Rat Brain. Front. Mol. Neurosci. 10, 67.
Chakraborty, S., et al., 2019. Serotonergic system, cognition, and BPSD in Alzheimer’s
Palma, J.A., Urrestarazu, E., Iriarte, J., 2013. Sleep loss as risk factor for neurologic
disease. Neurosci. Lett. 704, 36–44.
disorders: a review. Sleep Med. 14, 229–236.
Chen, Z., et al., 2019. Effects of Treadmill Exercise on Motor and Cognitive Function
Pineda, O.L., 2017. Are sleep disorders associated with cognitive decline? Neurology 88
Recovery of MCAO Mice Through the Caveolin-1/VEGF Signaling Pathway in
e42-e43.
Ischemic Penumbra. Neurochem. Res. 44, 930–946.
Pires, G.N., et al., 2016. Effects of experimental sleep deprivation on anxiety-like
Chennaoui, M., et al., 2016. Sleep extension increases IGF-I concentrations before and
behavior in animal research: Systematic review and meta-analysis. Neurosci.
during sleep deprivation in healthy young men. Appl. Physiol. Nutr. Metab. 41,
Biobehav. Rev. 68, 575–589.
963–970.
Qiu, H., et al., 2016. Chronic Sleep Deprivation Exacerbates Learning-Memory Disability
Cruz, T., et al., 2019. Sleep quality and memory function in healthy ageing. Neurologia.
and Alzheimer’s Disease-Like Pathologies in AβPP(swe)/PS1(ΔE9) Mice.
https://doi.org/10.1016/j.nrl.2018.10.001 (in press).
J. Alzheimers Dis. 50, 669–685.
Dal-Pont, G.C., et al., 2019. Effects of lithium and valproate on behavioral parameters
Rusch, H.L., Gill, J.M., 2015. Effect of Acute Sleep Disturbance and Recovery on Insulin-
and neurotrophic factor levels in an animal model of mania induced by paradoxical
Like Growth Factor-1 (IGF-1): Possible Connections and Clinical Implications.
sleep deprivation. J. Psychiatr. Res. 119, 76–83.
J. Clin. Sleep Med. 11, 1245–1246.
Daniele, T.M.D.C., et al., 2017. Effects of exercise on depressive behavior and striatal
Riemann, D., et al., 2020. Sleep, insomnia, and depression. Neuropsychopharmacology.
levels of norepinephrine, serotonin and their metabolites in sleep-deprived mice.
45, 74–89.
Behav. Brain Res. 332, 16–22.
Roh, J.H., et al., 2016. Treadmill exercise ameliorates intracerebral hemorrhage-induced
Das, G., Mallick, B.N., 2008. Noradrenaline acting on alpha1-adrenoceptor mediates
depression in rats. J. Exerc. Rehabil. 12, 299–307.
REM sleep deprivation-induced increased membrane potential in rat brain
Sawmiller, D., et al., 2017. Beneficial effects of a pyrroloquinolinequinone-containing
synaptosomes. Neurochem. Int. 52, 734–740.
dietary formulation on motor deficiency, cognitive decline and mitochondrial
Fell, M.J., et al., 2011. N-(4-((2-(trifluoromethyl)-3-hydroxy-4-(isobutyryl)phenoxy)
dysfunction in a mouse model of Alzheimer’s disease. Heliyon. 3, e00279.
methyl)benzyl)-1-methyl-1H-imidazole-4-carboxamide (THIIC), a novel
Schmidt-Wilcke, T., et al., 2018. GABA-from Inhibition to Cognition: Emerging Concepts.
metabotropic glutamate 2 potentiator with potential anxiolytic/antidepressant
Neuroscientist. 24, 501–515.
properties: in vivo profiling suggests a link between behavioral and central nervous
system neurochemical changes. J. Pharmacol. Exp. Ther. 336, 165–177.

206
F. Tai et al. Brain Research Bulletin 164 (2020) 198–207

Schmitt, K., Holsboer-Trachsler, E., Eckert, A., 2016. BDNF in sleep, insomnia, and sleep Vogel, G.W., 1999. REM sleep deprivation and behavioral changes. In: Mallick, B.N.,
deprivation. Ann. Med. 48, 42–51. Inoue, S. (Eds.), Rapid eye movement sleep. Marcel Dekker, New York, pp. 355–366.
Schoenfeld, T.J., et al., 2016. Anxiolytic Actions of Exercise in Absence of New Neurons. Volkow, N.D., et al., 2008. Sleep deprivation decreases binding of [11C]raclopride to
Hippocampus. 26, 1373–1378. dopamine D2/D3 receptors in the human brain. J. Neurosci. 28, 8454–8461.
Schoenfeld, T.J., et al., 2013. Physical exercise prevents stress-induced activation of Wang, C., et al., 2014. Chronic mild stress-induced changes of risk assessment behaviors
granule neurons and enhances local inhibitory mechanisms in the dentate gyrus. in mice are prevented by chronic treatment with fluoxetine but not diazepam.
J. Neurosci. 33, 7770–7777. Pharmacol. Biochem. Behav. 116, 116–128.
Shansky, R.M., Lipps, J., 2013. Stress-induced cognitive dysfunction: hormone- Wen, X.S., et al., 2013. The regulation of SKF38393 on the dopamine and D1 receptor
neurotransmitter interactions in the prefrontal cortex. Front. Hum. Neurosci. 7, 123. expression in hippocampus during chronic REM sleep restriction. CNS Neurosci.
Siddique, S.A., et al., 2018. Evaluation of neurotransmitter alterations in four distinct Ther. 19, 730–733.
brain regions after rapid eye movement sleep deprivation (REMSD) induced mania- Wrann, C.D., et al., 2013. Exercise induces hippocampal BDNF through a PGC-1α/FNDC5
like behaviour in Swiss albino mice. Neurochem. Res. 43, 1171–1181. pathway. Cell Metab. 18, 649–659.
Silva, B.A., et al., 2015. A 20-week program of resistance or concurrent exercise improves Zager, A., et al., 2009. Modulation of sickness behavior by sleep: the role of
symptoms of schizophrenia: results of a blind, randomized controlled trial. Braz. J. neurochemical and neuroinflammatory pathways in mice. Eur.
Psychiatry. 37, 271–279. Neuropsychopharmacol. 19, 589–602.
Stenberg, D., 2007. Neuroanatomy and neurochemistry of sleep. Cell. Mol. Life Sci. 64, Zangrossi, H. Jr., Graeff, F.G., 2014. Serotonin in anxiety and panic: contributions of the
1187–1204. elevated T-maze. Neurosci. Biobehav. Rev. 46, 397–406.
Stonerock, G.L., et al., 2015. Exercise as treatment for anxiety: Systematic review and Żebrowska, A., et al., 2018. Brain-derived neurotrophic factor, insulin like growth factor-
analysis. Ann. Behav. Med. 49, 542–556. 1 and inflammatory cytokine responses to continuous and intermittent exercise in
Strasser, B., Gostner, J.M., Fuchs, D., 2016. Mood, food, and cognition: role of patients with type 1 diabetes. Diabetes Res. Clin. Pract. 144, 126–136.
tryptophan and serotonin. Curr. Opin. Clin. Nutr. Metab. Care. 19, 55–61. Zhang, F., et al., 2017. The missing link between sleep disorders and age-related
Stubbs, B., et al., 2017. An examination of the anxiolytic effects of exercise for people dementia: recent evidence and plausible mechanisms. J. Neural. Transm. 124,
with anxiety and stress-related disorders: A meta-analysis. Psychiatry Res. 249, 559–568.
102–108. Zhang, F., et al., 2019. Alteration in sleep architecture and electroencephalogram as an
Toda, T., Gage, F.H., 2018. Review: adult neurogenesis contributes to hippocampal early sign of Alzheimer’s disease preceding the disease pathology and cognitive
plasticity. Cell Tissue Res. 373, 693–709. decline. Alzheimers Dement. 15, 590–597.
Tully, K., Bolshakov, V.Y., 2010. Emotional enhancement of memory: how Zielinski, et al., 2013. Influence of chronic moderate sleep restriction and exercise
norepinephrine enables synaptic plasticity. Mol. Brain. 3, 15. training on anxiety, spatial memory, and associated neurobiological measures in
Vivar, C., van Praag, H., 2017. Running Changes the Brain: the Long and the Short of It. mice. Behav Brain Res. 250, 74–80.
Physiology (Bethesda) 32, 410–424.

207