BASIC RESEARCH
Sleep Deprivation and Cellular Responses to Oxidative Stress
Anupama Gopalakrishnan, PhD1; Li Li Ji, PhD2; Chiara Cirelli MD, PhD1
1Department
of Psychiatry, 2Department of Kinesiology, University of Wisconsin, Madison
Study Objectives: It has been hypothesized that sleep deprivation represents an oxidative challenge for the brain and that sleep may have a
protective role against oxidative damage. This study was designed to test
this hypothesis by measuring in rats the effects of sleep loss on markers
of oxidative stress (oxidant production and antioxidant enzyme activities)
as well as on markers of cellular oxidative damage (lipid peroxidation and
protein oxidation).
Design: The analyses were performed in the brain and in peripheral tissues (liver and skeletal muscle), after short-term sleep deprivation (8
hours), after long-term sleep deprivation (3-14 days), and during recovery
sleep after 1 week of sleep loss. Short-term sleep deprivation was performed by gentle handling; long-term sleep deprivation was performed
using the disk-over-water method.
Setting: Sleep research laboratory at University of Wisconsin-Madison.
Participants and Interventions: Adult male Wistar Kyoto rats (n = 69)
implanted for polygraphic (electroencephalogram, electromyogram)
recording.
Measurements and Results: Aliquots of brain, liver, or skeletal muscle
homogenate were used to assess oxidant production, superoxide dismutase activity, lipid peroxidation, and protein oxidation. No evidence of
oxidative damage was observed at the lipid and/or at the protein level in
long-term sleep-deprived animals relative to their yoked controls, nor in
the cerebral cortex or in peripheral tissues. Also, no consistent change in
antioxidant enzymatic activities was found after prolonged sleep deprivation, nor was any evidence of increased oxidant production in the brain or
in peripheral tissues.
Conclusion: The available data do not support the assumption that prolonged wakefulness may cause oxidative damage, nor that it can represent an oxidative stress for the brain or for peripheral tissue such as liver
and skeletal muscle.
Key Words: cerebral cortex, rat, disk-over-water method, norepinephrine,
oxidative stress, sleep deprivation
Citation: Gopalakrishnan A; Ji LL; Cirelli C. Sleep deprivation and cellular responses to oxidative stress. SLEEP 2004;27(1):27-35.
INTRODUCTION
Whether sleep deprivation causes oxidative damage, however,
remains unknown, nor it is clear why sleep could protect against oxidative stress. Brain energy metabolism is as high in wakefulness as in rapid
eye movement (REM) sleep, which represents 10% to 30% of total
sleep.5 Moreover, prolonged sleep deprivation in animals is accompanied by a decrease, rather than an increase, in cerebral glucose utilization10 and is followed by an earlier and marked rebound of REM rather
than NREM sleep.11
A few studies have examined signs of oxidative stress after 4 days of
sleep deprivation with the flowerpot technique, which produces relatively selective REM sleep deprivation. Only 1 study12 controlled for
immobilization and isolation stress and found no evidence that REM
sleep deprivation per se causes changes in lipid peroxidation or in
antioxidant defenses. Later studies from the same authors found a
decrease in glutathione levels in the hypothalamus and thalamus of
experimental rats relative to controls,13,14 but the use of single, instead of
multiple, platforms did not allow the authors to tease apart the effects of
sleep loss from those of immobilization stress. Immobilization stress
alone can induce several markers of oxidative damage in the brain and
in peripheral tissues.15,16 Finally, Ramanathan et al17 recently found that
rats deprived of total sleep for 2 weeks show a decrease in the hippocampal and brainstem activity of 1 antioxidant enzyme, Cu-Zn superoxide dismutase. These authors did not assess other antioxidant enzymatic activities, nor did they measure oxidant production or any other
marker of oxidative damage.
The aims of the present study were 4-fold: (1) measure antioxidant
enzyme activity and cellular ROS-RNS production to determine whether
sleep deprivation causes oxidative stress and whether recovery sleep
after sleep loss relieves the oxidative stress, (2) measure lipid peroxidation and protein oxidation in search for direct evidence of cellular damage after sleep deprivation, (3) determine whether sleep deprivation has
differential effects in the brain compared to peripheral tissues, and (4)
compare the effects of short-term sleep deprivation (8 hours) with those
of sustained sleep loss (3-14 days). Preliminary results of this study have
been published in abstract form.18
OXIDATIVE STRESS HAS BEEN INVOLVED IN THE MECHANISMS OF BIOLOGIC AGING, AS WELL AS IN THE PATHOGENESIS OF CANCER, ATHEROSCLEROSIS, DIABETES, AND NEURODEGENERATIVE DISORDERS.1 Oxidative stress occurs whenever there is an imbalance between oxidant production and antioxidant
defenses, either because the former is increased, because the latter are
decreased, or both. At the cellular level, such imbalance can result in
structural damage due to oxidative modifications of proteins, lipids, and
nucleic acids. Major cellular oxidants include reactive oxygen species
(ROS, eg, O2- and H2O2) and reactive nitrogen species (RNS, eg, NO).
Although most ROS are the byproducts of the electron transport chain,
oxidants can also be produced by extramitochondrial sources such as
NADPH oxidases and nitric-oxide synthases.1,2
Several hypotheses about the functions of sleep rest on the assumption that wakefulness represents an oxidative challenge for the brain. It
has been claimed, for instance, that sleep may allow the removal of free
radicals accumulated in the brain during wakefulness.3 Moreover, it has
been proposed that, during sleep, uridine and glutathione may facilitate
the oxidative detoxification of the brain by potentiating GABAergic
transmission and inhibiting glutamatergic transmission, respectively.4
Consistent with these hypotheses, brain energy metabolism, which relies
almost completely on mitochondrial respiration, is higher in wakefulness than in non-rapid eye movement (NREM) sleep.5 Moreover,
peripheral metabolic rate,6 cerebral cortex glutamatergic transmission,7
and extracellular nitric-oxide concentration8,9 are also increased in spontaneous wakefulness and/or sleep deprivation relative to NREM sleep.
Disclosure Statement
This work was funded by the National Institute of Mental Health (R01 MH65135).
Submitted for publication July 2003
Accepted for publication September 2003
Address correspondence to: Chiara Cirelli, MD, PhD, University of Wisconsin,
Madison, Department of Psychiatry, 6001 Research Park Blvd, Madison WI
53719; Tel: 608-263-9236; Fax: 608-263-0265; E-mail: ccirelli@wisc.edu
SLEEP, Vol. 27, No. 1, 2004
27
Cellular Responses to Oxidative Stress—Gopalakrishnan et al
METHODS
rather than in LD conditions to flatten their diurnal sleep and temperature rhythms.19 We found no difference in any of the parameters tested
between rats kept in LD and rats kept in LL. Animal protocols followed
the National Institutes of Health Guide for the Care and Use of
Laboratory Animals and were approved by the University of Wisconsin.
Experiment Groups and Polygraphic Recordings
Under pentobarbital anesthesia (75 mg/kg, intraperitoneal), adult male
Wistar Kyoto rats (300-450 g, N = 69) were implanted with screw electrodes in the skull to record the electroencephalogram and with silver
electrodes in the nuchal and temporal muscles to record the electromyogram. Transmitters (Barrows, Inc., Sunnyvale, CA) were implanted in
the peritoneum to record peritoneal temperature (Tip). After surgery, rats
were housed individually in sound-proof recording cages where lighting
and temperature were kept constant (LD 12:12, light on at 10 AM, 25°C
± 1°C, food and drink ad libitum). Each day from 10 AM to 10:30 AM,
the rats were gently handled to become familiar with the sleep deprivation procedure (see below). One week after surgery, the rats were connected by means of a flexible cable and a commutator (Airflyte,
Bayonne, NJ) to a Grass electroencephalograph (model 15LT, West
Warwick, RI) and recorded continuously for as many days (7-30 days)
as required to satisfy the criteria for the 4 experimental groups (Table 1).
The electroencephalogram signals were visually scored for 4-second
epochs (SleepSignTM, Kissei Comtec America, Inc., Irvine, CA). In
group 1, the sleeping rats were killed during the light hours (between
4:00 PM and 6:00 PM), at the end of a long period of sleep (> 45 minutes, interrupted by periods of wakefulness < 2 minutes), and after
spending at least 75% of the previous 6 to 8 hours asleep. Short-term
totally sleep-deprived rats (8-hour TSD) were kept awake for most of
their sleeping period (first 8 hours of the light period) by introducing
novel objects in their recording cages. They were killed at the same circadian time as the 8-hour sleep group to assess the effects of behavioral
state independently of circadian factors. In group 2, long-term sleepdeprived rats were kept awake with the disk-over-water (DOW) method
for 3 to 14 days (3- to 14-day TSD) and compared with their yoked controls (YC). The DOW method19 (see below) was used because it represents the most effective and best-controlled system to enforce long-term
sleep deprivation in animals. Sleep loss by the DOW method produces a
series of dramatic physiologic changes that invariably culminate in death
after 3 to 4 weeks.19 Initially (within the first 1-2 days) long-term sleepdeprived rats develop a syndrome characterized by an increase in food
intake, energy expenditure, and heart rate followed by a decrease in body
weight and a decline in body and brain temperature. The sleep-deprivation syndrome and its lethal consequences have also been observed after
selective REM-sleep deprivation, although the pathology associated
with the loss of sleep takes longer to appear, the survival time is longer
(4-5 weeks), and body and brain temperature are not significantly
decreased.20 Since we did not know a priori whether evidence of oxidative stress would occur at an early or at a late stage of sleep deprivation,
we kept rats awake for different periods of time, from 3 to 14 days. The
shortest sleep deprivation with the DOW method was 3 days because, in
our experience, the YC needs some time to learn how and when to sleep
when the disk starts moving and, therefore, its sleep is significantly
reduced for the first 1 to 2 days of the experiment. On the other hand, the
longest sleep deprivation did not exceed 14 days to avoid the occurrence
of terminal effects. More specifically, TSD lasted for 3, 4, 5, 6, 7, 8, 9,
11, or 14 days (n = 2 for each time point except n = 3 for 7 days and n =
1 for 11 days). Group 2 also included several cage controls (CC), which
were housed in the same room where the disk was located and were sacrificed at the same circadian time as the rats on the disk. In group 3,
long-term sleep-deprived rats were allowed to recover sleep for 6 to 8
hours after 1 week of sleep loss (Rec TSD) and were compared with
their YC. One week of sleep deprivation was selected because it is sufficient to induce all the physiologic markers of the sleep deprivation syndrome.19 Rats were then allowed to sleep ad libitum for 6 to 8 hours, the
time period during which the most significant sleep rebound following
long-term sleep deprivation occurs.11 A final group of rats was allowed
to recover sleep for 6 to 8 hours after 1 week of selective REM (paradoxical) sleep deprivation with the DOW (Rec PSD) and were compared
with their YC. Some rats in groups 2 to 4 were kept in constant light
SLEEP, Vol. 27, No. 1, 2004
Sleep Deprivation Procedure
Short-term sleep deprivation in group 1 was enforced by exposing the
rats to novel objects. Every new object was introduced into the cages just
following the first signs of synchronization in the frontal electroencephalogram signal. In groups 2 to 4, long-term sleep deprivation was
performed by the DOW method.19 Briefly, a rat that was to be sleep
deprived and its YC were housed in rectangular Plexiglas cages. A single horizontal 46-cm in diameter disk, which could be rotated in a randomly chosen direction, formed a floor extending 17 cm into each cage.
Under the disk and extending to the cage walls was a rectangular tray
filled with tap water to a depth of approximately 2 cm. When sleep onset
was detected in the sleep-deprived rat, the disk was rotated slowly by a
computerized monitoring system, forcing both rats to walk in a direction
opposite to disk rotation to avoid the water. When the sleep-deprived rat
was spontaneously awake, the disk was stationary and the YC rat was
able to sleep. During baseline, the disk was rotated once per hour for 6
seconds to habituate the rats to rotation. The baseline period continued
until sleep, food intake, body weight, and temperature had stabilized
(usually 3-7 days) in both rats. Cage air temperature was thermostatically maintained at 25ºC ± 1ºC. When TSD was initiated, the disk was
rotated whenever the computerized monitoring system detected sleep
onset in the sleep-deprived rat.
Tissue Preparation
Animals were anesthetized with isoflurane and decapitated, and the
brains and peripheral tissues were removed quickly and frozen in liquid
nitrogen. Small pieces of cerebral cortex, liver, and skeletal muscle were
weighed and homogenized at low speed in a cold buffer containing 0.25
moles of sucrose, 20 mmol KCl, 1 mmol EDTA, and 5 mmol HEPES
(hydroxyethylpiperazine ethanesulfonate) (w/v 1:10, pH 7.4) using a
motor-driven Potter-Elveljem glass homogenizer (Wheaton Science
Products, Millville, NJ). Aliquots of the homogenate were used to assess
lipid peroxidation, superoxide dismutase activity, and oxidant production. To compare mitochondrial integrity, citrate synthase activity was
measured in brain cortical samples using a method described by
Shepherd and Garland.21 Every assay described below was run in duplicate on individual tissues from single animals. In no case were tissues
pooled before running the experiment.
Lipid Peroxidation
Peroxidative damage to lipids was determined by measuring TBARS
(thiobarbituric acid reactive substances) content in butanol extracts
according to Mihara and Uchiyama.22 Briefly, aliquots of homogenate
(containing 300-400 µg of proteins) were incubated with 1.5 mL of 1%
H3PO4 and 0.5 mL of 0.6% thiobarbituric acid (3:1 v/v) in a boiling
water bath for 15 minutes. The tubes were cooled and mixed with 2 mL
of n-butanol and centrifuged at 2000 x g for 10 minutes at room temperature. Supernatant was removed and read at 520 nm and 535 nm in a
Shimadzu UV-2101 spectrophotometer (Pleasanton, CA). The TBARS
content was also measured after the homogenates were preincubated
with ferrous ions (5 µmol) and ascorbate (1 mmol). The amount of
TBARS produced was calculated using a molar extinction coefficient of
152000·mol-1·cm-1. Lipid peroxidation level was calculated as nmol of
TBARS per milligram of protein. In all assays, protein concentration in
tissue homogenates was determined by the Bio-Rad assay using bovine
serum albumin as the protein standard.
28
Cellular Responses to Oxidative Stress—Gopalakrishnan et al
Protein Oxidation
DCFH is capable of penetrating biomembranes and, therefore, can measure oxidant production in the cytoplasm as well as in intracellular
organelles. The DCFH can be oxidized not only by ROS, but also by
nitric oxide and other low-molecular-weight oxidants. Thus, DCF levels
are a comprehensive indicator of the net ROS/RSN load in the mitochondria as well as in the cytosol.2 The assay buffer contained 130 mmol
KCl, 5 mmol MgCl2, 20 mmol NaH2PO4, 20 mmol Tris-HCl, and 30
mmol glucose (pH 7.4). An aliquot of the homogenate (200-300 µg of
proteins) was incubated at 37°C for 15 minutes in the assay buffer containing 5 µmol DCFH-diacetate (DCFH-DA) dissolved in methanol. The
rate of oxidation of DCFH to DCF was corrected for autooxidation of
DCFH and followed every 10 minutes for 30 minutes at excitation wavelength of 488 nm and emission wavelength of 525 nm using a Hitachi F2000 fluorescence spectrometer (Randolph, MA). The DCFH assay was
run both in the absence and in the presence of 5 mmol sodium azide.
Oxidant production was calculated as nmol DCF formed per milligram
of protein per minute, as determined by a standard curve plotted with
known concentrations of DCF.
Protein oxidation was evaluated by measuring carbonyl formation
using 2,4- dinitrophenylhydrazine (DNPH) as a reagent according to
Levine et al23 with some modification.24 Briefly, 200 mg of frozen tissue
was homogenized in 10 volumes of 5 mmol potassium phosphate buffer
(pH 7.4) containing 0.1% Triton-X and the protease inhibitors leupeptin
(1.0 µg/mL), pepstatin (1.4 µg/mL) and aprotinin (1.0 µg/mL). The
homogenate was centrifuged at 500 x g for 3 minutes, and 900 µL of
supernatant were transferred to a microcentrifuge tube containing 100
µL of 10% streptomycin sulphate (in 50 mmol HEPES, pH 7.4). The
samples were vortexed thoroughly, incubated at room temperature for 15
minutes, and then centrifuged at 6000 x g for 10 minutes at 4°C. The
samples were divided into 3 aliquots (250 µL each) and treated as 2 samples and 1 blank. Samples and blanks were incubated in the dark for 1
hour in the presence of 500 µL of 10 mmol DNPH (in 2 N HCl) and 500
µL of 2 N HCl, respectively. Proteins were precipitated with 250 µL of
40% trichloroacetic acid (TCA) after incubating on ice for 10 minutes
and centrifuging at 12,000 x g for 10 minutes. The supernatant was discarded, and 500 µL of 10% TCA was added to all the tubes and vortexed
thoroughly. After centrifuging (12,000 x g for 10 minutes), the pellet was
washed 3 times with 1 mL each of ethyl acetate:ethanol mixture (1:1),
and excess DNPH were removed by centrifuging at 14,000 x g for 10
minutes after each wash. The final pellet was suspended in 1 mL of 6
mol guanidine HCl (in 20 mmol KH2PO4, pH 2.3), vortexed, and
allowed to dissolve overnight. The absorbance of the samples was measured at 366 nm in a Shimadzu UV-2101 spectrophotometer. Carbonyl
content was calculated as nmol per milligram of protein using a molar
absorption coefficient of 22,000 mol-1·cm-1.
Superoxide Dismutase Activity
Superoxide dismutase (SOD) activity25 was measured as the amount
of tissue extracts necessary to inhibit autooxidation of epinephrine to
adenochrome. In the blanks, the rate of autooxidation of epinephrine (30
mmol in 0.1 N HCl) to colored adenochrome was monitored spectrophotometrically at 320 nm (Shimadzu UV-2101 spectrophotometer)
in 2 mL of assay buffer containing 0.05 mol NaHCO3 buffer, 0.1 mmol
EDTA, pH 10.2. The rate of autooxidation of epinephrine was then monitored in the presence of an aliquot of tissue homogenate containing 20
to 30 µg of proteins. The mixture was incubated at 30°C for 5 minutes,
Oxidant Production
and the reaction was monitored spectrophotometrically for another 3
minutes to calculate the autooxidation rate in the linear range. The SOD
Oxidant production was assayed in tissue homogenates using 2´-7´activity (units per milligram of protein; 1 Unit = 50% inhibition of
dichlorofluorescin (DCFH) as a probe, according to Bejma et al.24 The
epinephrine autooxidation) was measured with and without 1
Table 1—Percentages of wakefulness, NREM sleep, and REM sleep for the mmol KCN.
last 6-8 recording hours before sacrifice in the 4 experimental groups.
Statistical Analysis
Experimental groups
Behavioral state, %,
during last 6-8 recording hours)
Wake
NREM
REM
(1) spontaneous sleep
(6-8 hours)
8h S
(n = 6)
Asleep
24.4 ± 1.4
60.7 ± 1.2
14.9 ± 0.7
short-term
(6-8 hours)
total sleep
deprivation
8h TSD
(n = 7)
Sleep
Deprived
95.9 ± 0.4
4.1 ± 0.4
0.0
(2) long-term
(3-14 days)
total sleep
deprivation
3-14d TSD
(n = 18)
Sleep
Deprived
84.9 ± 2.4
14.7 ± 2.0
0.4 ± 0.1
YC
(n = 16)
Yoked
Controls
64.5 ± 1.0
31.2 ± 0.8
4.3 ± 0.9
CC
(n = 7)
Cage
Controls
...
...
...
(3) long-term
(7 days)
total sleep
deprivation +
recovery sleep
(6-8 hours)
Rec TSD
(n = 5)
Sleep
Deprived
19.3 ± 3.5
31.2 ± 2.2
46.7 ± 4.8
YC
(n = 5)
Yoked
Controls
34.7 ± 4.9
43.0 ± 2.8
22.4 ± 3.7
(4) long-term
(7 days)
REM-selective sleep
deprivation + recovery
sleep (6-8 hours)
Rec PSD
(n = 6)
Sleep
Deprived
25.2 ± 2.1
32.0 ± 4.7
42.7 ± 3.8
YC
(n = 6)
Yoked
Controls
39.9 ± 3.5
49.0 ± 2.4
11.0 ± 2.7
Data are presented as mean ± SEM.
Yoked controls (YC) were not available for 2 rats sleep-deprived for 3 days.
NREM refers to non-rapid eye movement sleep; REM, rapid eye movement sleep; TSD, total sleep
deprivation; PSD, paradoxical or REM sleep deprivation
SLEEP, Vol. 27, No. 1, 2004
29
For each assay, values from duplicate samples for each animal
were averaged to obtain 1 value point. Mean comparison was done
using nonparametric tests (P < .05). The Mann-Whitney U test was
used for group 1, in which the 8-hour sleep-deprived rats were not
paired to the 8-hour sleeping rats. By contrast, the Wilcoxon
matched-paired test was used for groups 2 to 4, in which the experimental (sleep-deprived) animals were paired to their YC.
RESULTS
Table 1 shows mean values (± SEM) of wakefulness, NREM
sleep, and REM sleep before sacrifice for the 4 experimental
groups. We elected to show percentages of behavioral states for the
last 8 hours because (1) 8 hours represent the entire duration of the
experiment for group 1, (2) they represent the entire duration of
the sleep recovery period for groups 3 and 4, and (3) they are
indicative of any period during the 24-hour cycle for group 2.
Table 1 shows that rats in group 1 were either mostly asleep (8hour sleep) or continuously awake (8-hour TSD) for the last 6 to 8
hours before sacrifice. Table 1 also shows that the DOW method
was effective in enforcing prolonged wakefulness in the long-term
sleep-deprived rats (3- to 14-day TSD) and in allowing their YC to
maintain most of their sleep. During the entire sleep-deprivation
period, the 3- to 14-day TSD rats lost on average (mean ± SEM)
63% ± 4% and 95% ± 4% of their daily baseline values of NREM
and REM sleep, respectively. Their YC, by contrast, lost 28% ±
14% and 42% ± 25% of their daily baseline values of NREM and
REM sleep, respectively. For group 3, the daily percentage of
behavioral states during the 7 days of TSD were similar to those
Cellular Responses to Oxidative Stress—Gopalakrishnan et al
Figure 1—Peroxidative damage to lipids as determined in the rat cerebral cortex by measuring TBARS (thiobarbituric acid reactive substances) content in butanol extracts. The assay was
run under basal conditions (upper panels) and under induced conditions (ie, the assay medium contained oxidizing agents; lower panels). Each bar represents the mean value from duplicate
samples for each animal. Group 1: 8h S = 8 hours of sleep (mean ± SEM, basal 2.3 ± 0.3; induced 13.7 ± 1.3); 8h TSD = 8 hours of total sleep deprivation (1.9 ± 0.3; 14.1 ± 1.6). Group 2:
3-14d TSD = 3-14 days of total sleep deprivation (2.0 ± 0.1; 12.9 ± 0.8); YC = yoked controls (2.1 ± 0.2; 13.9 ± 1.0). Group 3: Rec-TSD, recovery sleep after TSD (2.1 ± 0.4; 18.6 ± 2.4);
YC (2.3 ± 0.1; 19.7 ± 1.9). Group 4: Rec-PSD, recovery sleep after paradoxical sleep (rapid eye movement sleep) deprivation (2.6 ± 0.6; 18.2 ± 0.7);. YC (2.1 ± 0.4; 14.3 ± 2.1). The numbers above the columns in group 2 indicate the duration of the experiment. The Mann-Whitney U test was used for group 1; the Wilcoxon matched paired test was used to compare sleepdeprived (SD) and YC rats in groups 2-4. The YC data were not available for 2 rats sleep deprived for 3 days. Cage controls (CC) for some 3-14d TSD rats are also shown.
reported for animals in group 2 (data not shown). In group 4, the daily
percentage of behavioral states during the 7 days of REM-selective sleep
deprivation were the following: Rec PSD rats, NREM: 28.3% ± 0.8%
(80% of baseline value), REM sleep: 0.86% ± 0.07% (11% of baseline
value); YC rats, NREM: 31.0% ± 0.9% (81% of baseline value), REM
sleep: 5.6% ± 0.3 (70% of baseline value). As expected from previous
studies,11 1 week of TSD (group 3) or REM-selective (group 4) sleep
deprivation caused a significant rebound of REM sleep, which accounted for more than 40% of total recording time during the last 6 to 8 hours
before sacrifice.
Markers of Oxidative Damage in the Brain: Lipid Peroxidation and Protein
Oxidation
The brain may be particularly vulnerable to oxidative stress because
of its high rate of oxygen consumption, high content of polyunsaturated
fatty acids, and low levels of natural antioxidants.26 Since most of the
effects of sleep deprivation in humans are on higher cognitive functions,27,28 the cerebral cortex could be particularly sensitive to the detrimental consequences of sleep loss. Thus, we first searched for markers
of oxidative damage in the cerebral cortex.
Lipid peroxidation due to the reaction of free radicals with lipids is
considered a hallmark of cellular oxidative damage.29 Once established,
such damage can affect the membrane lipid bilayer and, specifically, the
mitochondrial electron transport chain, thus becoming a major cause for
Figure 2—Protein oxidation as evaluated in the rat cerebral cortex by measuring carbonyl
formation (as nmol carbonyl/mg protein) using 2,4- dinitrophenylhydrazine (DNPH) as a
reagent. Each bar represents the mean value from duplicate samples for each animal. 8h S
= 8 hours of sleep; 8h TSD = 8 hours of total sleep deprivation; 3-14d TSD = 3-14 days of
total sleep deprivation (mean ± SEM 3.8 ± 1.2); YC = yoked controls (5.1 ± 1.4); CC = cage
controls. The numbers above the columns in group 2 indicate the duration of the experiment.
The Wilcoxon matched paired test was used to compare sleep-deprived (SD) and YC in
group 2. 8h S, 8h TSD, and CC are only shown for descriptive purposes.
SLEEP, Vol. 27, No. 1, 2004
30
Cellular Responses to Oxidative Stress—Gopalakrishnan et al
Markers of Oxidative Stress in the Brain: Oxidant Production and
Antioxidant Defenses
a further increase in oxidant production. Figure 1 shows the extent of
peroxidative damage to lipids, as measured by malondialdehyde (MDA)
content, for each individual animal in all experimental groups. The same
assay was run both in the absence (Figure 1, basal) and in the presence
(Figure 1, induced) of strong oxidizing agents, to measure the susceptibility of the tissue to a fixed oxidative stress. As shown in the figure,
sleep loss per se did not consistently affect cortical lipid peroxidation
(Figure 1, upper panels) nor the ability of the cerebral cortex to respond
to an oxidative insult (Figure 1, lower panels).
Lysine, arginine, and proline residues of enzymatic and nonenzymatic proteins can be oxidized to carbonyl derivatives and thus become
another major determinant of the detrimental effects caused by cellular
oxidative damage. Oxidized proteins undergo physical changes that lead
to their fragmentation, aggregation, and susceptibility to proteolytic
digestion. We measured the extent of protein carbonyl formation, an
index of protein oxidation, in the cerebral cortex of several long-term
sleep-deprived rats and their YC. As shown in Figure 2, cortical protein
oxidation levels did not increase after 9 to 14 days of sleep deprivation.
The absence of visible oxidative damage does not rule out the presence of an oxidative stress during sleep deprivation. It was still possible,
in other words, that there was an increase in oxidant production during
prolonged wakefulness but that such oxidative load was effectively
counteracted by antioxidant defenses. We assessed cortical oxidant production by measuring the rate of oxidation of DCFH (nonfluorescent) to
DCF (fluorescent). To assess the specific contribution of mitochondrial
enzymes to oxidant production, the DCF assay was run both in the
absence and in the presence of sodium azide, a strong inhibitor of mitochondrial complex IV. When sodium azide was added, oxidant production was increased in all animals (Figure 3, lower panels). This result
was expected because sodium azide, by blocking complex IV, prevents
the conversion of O2- to H2O. The DCF levels were significantly higher
in 8-hour TSD rats relative to their sleeping controls but only when sodium azide was present. By contrast, DCF levels were significantly lower
in 3- to 14-day TSD rats relative to their YC in the absence (Figure 3,
upper panels) as well as in the presence (Figure 3, lower panels) of sodi-
Figure 3—Oxidant production (reactive oxygen species [ROS, eg, O2- and H2O2]) as assayed in rat cerebral cortex homogenates using 2´-7´-dichlorofluorescin (DCFH) as a probe. The rate
of oxidation of DCFH to DCF was calculated as nmol DCF formed per milligram of protein per minute. The assay was run in the absence (upper panels) and in the presence (lower panels)
of 5 mmol sodium azide. Each bar represents the mean value from duplicate samples for each animal. Group 1: 8h S = 8 hours of sleep (mean ± SEM, -sodium azide 9.3 ± 0.8; +sodium azide
14.7 ± 0.8); 8h TSD = 8 hours of total sleep deprivation (10.7 ± 0.8; 20.2 ± 1.0). Group 2: 3-14d TSD = 3-14 days of total sleep deprivation (9.1 ± 0.7; 15.5 ± 1.1); YC = yoked controls (11.5
± 1.0; 18.1 ± 1.6). Group 3: Rec-TSD, recovery sleep after TSD (8.3 ± 0.5; 13.7 ± 0.3); YC (9.0 ± 0.8; 15.1 ± 0.8). Group 4: Rec-PSD, recovery sleep after paradoxical sleep (rapid eye movement sleep) deprivation (10.6 ± 1.3; 18.4 ± 2.4);YC (10.5 ± 0.8; 20.8 ± 2.6). The numbers above the columns in group 2 indicate the duration of the experiment. The Mann-Whitney U test
was used for group 1; the Wilcoxon matched paired test was used to compare sleep-deprived (SD) and YC rats in groups 2-4. Cage controls (CC) for some 3-14d TSD rats are also shown.
SLEEP, Vol. 27, No. 1, 2004
31
Cellular Responses to Oxidative Stress—Gopalakrishnan et al
um azide. Thus, oxidant production in the cerebral cortex seems to
increase after short-term sleep deprivation and to decrease after longterm sleep deprivation.
Similarly to the in vivo situation, the DCFH assay was run in the presence of an intact intracellular antioxidant system, which can compete
with DCFH for oxidants. Thus, the calculated oxidation rate of DCFH to
DCF reflects the combined effects of oxidant production minus antioxidant defense capacity. If the antioxidant defense were upregulated as a
consequence of sleep deprivation, the oxidation rate of DCFH to DCF
could still be normal even in the presence of an increased oxidant production. Antioxidant defense systems include nonenzymatic molecules
(eg, glutathione) as well as enzymatic scavengers such as SOD, catalase,
and glutathione peroxidase. In the brain, where catalase and glutathione
peroxidase activities are relatively low,30-32 the primary defense29 is represented by SOD. The SOD, which catalyze the dismutation of O2- to
H2O2, are very sensitive to the redox status of the cell, and changes in
SOD activity are reflective of any oxidative stress. Mitochondrial SOD
is manganese dependent (Mn SOD) and KCN insensitive, while cytosolic SOD is copper or zinc dependent (Cu-Zn SOD) and can be inhibited
by KCN. We measured total, Mn SOD and Cu-Zn SOD activities by
incubating tissue homogenates of cerebral cortex with and without KCN.
As shown in Figure 4, there was no difference between rats that slept for
8 hours relative to rats that were awake for a similar period of time.
Similarly, 3- to 14-day TSD rats did not differ from their YC, although
a trend toward a decrease in Mn SOD activity was observed after 4 to 5
days of TSD. Total and Cu-Zn SOD were significantly increased in RecPSD animals (but not in Rec-TSD rats) relative to their YC.
Figure 4—Superoxide dismutase (SOD) activity in the rat cerebral cortex measured as the amount of tissue extracts necessary to inhibit autooxidation of epinephrine to adenochrome. Total
SOD and manganese-dependent (Mn) SOD were measured in the absence and in the presence of 1 mmol KCN, respectively. The copper or zinc-dependent (Cu-Zn) SOD activity is the difference between total SOD and Mn-SOD activity. Each bar represents the mean value from duplicate samples for each animal. Group 1: 8h S = 8 hours of sleep (mean ± SEM, total 21.7 ±
2.0; Mn SOD 10.2 ± 0.6; Cu-Zn SOD 10.4 ± 1.7); 8h TSD = 8 hours total sleep deprivation (20.5 ± 1.5; 11.6 ± 0.6; 8.8 ± 1.1). Group 2: 3-14d TSD = 3-14 days total sleep deprivation (19.3
± 1.6; 6.7 ± 0.8; 11.9 ± 0.9); YC = yoked controls (21.7 ± 1.5; 7.8 ± 1.0; 14.2 ± 1.6). Group 3: Rec-TSD, recovery sleep after TSD (24.0 ± 2.8; 5.5 ± 1.1; 18.5 ± 1.9); YC (17.1 ± 3.0; 3.4 ±
0.8; 13.6 ± 2.3). Group 4: Rec-PSD, recovery sleep after paradoxical sleep (rapid eye movement sleep) deprivation (22.8 ± 2.5; 12.3 ± 1.5; 10.5 ± 2.2);.YC (19.2 ± 1.9; 13.3 ± 1.7; 5.9 ± 1.0).
The numbers above the columns in group 2 indicate the duration of the experiment. Note that although there is no significant difference between 3-14d TSD rats and YC in group 2, Mn-SOD
values tend to decrease after 3-5 days of sleep deprivation. Citrate synthase activity, a measure of mitochondrial integrity, was similar in groups 2-4 (data not shown). The Mann-Whitney U
test was used for group 1; the Wilcoxon matched paired test was used to compare sleep-deprived (SD) and YC rats in groups 2-4. The YC were not available for 2 rats sleep deprived for 3
days. Cage controls (CC) for some 3-14d TSD rats are also shown.
SLEEP, Vol. 27, No. 1, 2004
32
Cellular Responses to Oxidative Stress—Gopalakrishnan et al
Peripheral Tissues
to the use of 2 rats per time point, and the fact that we could not know a
priori when oxidative stress would occur, which prompted us to test 9
different lengths of sleep deprivation (from 3 to 14 days). In fact, despite
the small number of animals per time point, Figures 1, 3, and 4 show that
the values of each of the 15 or 18 (depending on the assay) long-term
sleep-deprived rats are well within the range of all of the other experimental animals, were they sleeping, short-term sleep-deprived, or YC
rats. Thus, although the use of a larger number of animals per time point
would have undoubtedly made any conclusion of this paper stronger, we
think it unlikely that the general outcome would have changed.
Previous studies24,33 have demonstrated that strenuous exercise and
aging are associated with an increase in lipid peroxidation, in protein
oxidation in several tissues, or in both. In those experiments, signs of
oxidative damage were also associated with signs of oxidative stress, ie,
with an increase in oxidant production as measured by the DCFH assay.
Here, using this assay, we found no consistent evidence of oxidative
stress due to sleep loss in any of the tissues examined. We did in fact
observe an increase in oxidant production but only in short-term sleepdeprived rats (8-hour TSD) and only in the induced condition, ie, when
mitochondrial respiration was blocked by the presence of sodium azide.
By contrast, ROS production was significantly decreased in rats that
were sleep deprived for up to 14 days (3- to 14-day TSD) relative to their
YC.
We did not find a consistent change in either Mn SOD or Cu-Zn SOD
activity in long-term sleep-deprived rats relative to their YC, nor in rats
allowed to recover sleep after long-term sleep deprivation. We did find
an increase in SOD activity (total and Cu-Zn SOD) in rats allowed to
recover sleep after selective REM deprivation. However, such increase
was not present in rats allowed to sleep after TSD, which had experienced a more profound sleep loss. Recently, Ramanathan et al17 found a
decrease in Cu-Zn SOD activity (Mn-SOD was not measured in that
study) in the hippocampus and brainstem of rats that were sleep deprived
for 2 weeks. Since SOD activity can be inhibited by high levels of oxidants, the authors interpreted this result as an indirect evidence of high
ROS-RNS production in long-term sleep-deprived rats. However, oxidative stress should increase, rather than decrease, SOD activity,34 and
those studies that have reported a decrease in SOD activity also documented other signs of oxidative stress, such as a decrease in glutathione
content,35 a change in other antioxidant enzyme levels,36 or signs of
damage to DNA, lipids, and proteins.37 In our study, Mn SOD activity
levels showed a trend toward a decrease after 4 to 5 days of sleep deprivation. Such a trend, however, did not persist after longer periods of
sleep deprivation and was not associated with high ROS production
(ROS levels were actually decreased in 3- to 14-day TSD rats), nor with
signs of oxidative damage to lipids or proteins. In a previous study,38 we
found no evidence that acute or chronic sleep deprivation causes nucleic-acid fragmentation (see also reference 39). Thus, our data do not seem
to support the hypothesis that prolonged wakefulness represents an
oxidative challenge for the brain or for peripheral tissues such as liver
and skeletal muscle.
Since most of the cellular oxidant production comes from mitochondria, it is important to relate our findings to the studies that measured
metabolic rate during the sleep-wake cycle and after sleep deprivation.
Brain and peripheral tissues respond differently to sleep loss. Peripheral
metabolic rate is increased in insomniacs relative to normal sleepers40
and in normal sleepers on nights of poor sleep relative to baseline
nights.6 Energy expenditure and peripheral metabolic rate are also significantly increased in rats after 1 to 4 weeks of sleep deprivation,19 as
well as in patients with fatal familial insomnia.41 In both humans and
animals, glucose metabolism in many brain regions is higher in wakefulness than in NREM sleep,5,42 and short periods of spontaneous or
forced wakefulness produce an upregulation of subunit I of cytochrome
oxidase, the final key enzyme in the respiratory chain.43 During longterm sleep deprivation, however, cerebral metabolic rate shows no
change or a decrease, relative to normal wakefulness, in animals,10 in
normal human subjects,44 and in patients with fatal familial insomnia.45
To determine whether prolonged sleep loss could differentially affect
brain and peripheral tissues, lipid peroxidation, protein oxidation, and
oxidant production were measured in liver and skeletal muscle (hind
limb) of 3- to 14-day TSD rats and their YC. As summarized in Table 2,
there were no differences between experimental groups in any of the
variables examined.
DISCUSSION
In this study we have measured the effects of sleep loss on markers of
oxidative stress (oxidant production and antioxidant enzyme activities)
as well as on markers of cellular oxidative damage (lipid peroxidation
and protein oxidation). The analyses were performed in the brain and in
peripheral tissues, after short-term sleep deprivation (8 hours) and longterm sleep deprivation (3 to 14 days) and during recovery sleep after sustained sleep loss. Long-term sleep deprivation was performed using the
DOW method, which delivers the same mild stimulus (forced locomotion) to both the YC and the experimental rat but times the stimulus in
order to significantly restrict sleep only in the latter.19 As seen in Table
1, long-term sleep-deprived rats lost most of their sleep, while their YC
did not. The fact that YCs were forced to move as much as the sleepdeprived rats is relevant for the present study because exercise per se can
cause oxidative stress, at least in skeletal muscles.33
Our results show no evidence of oxidative damage at the lipid or at the
protein level in sleep-deprived animals, neither in the brain nor in
peripheral tissues. Neurons are potentially more prone to oxidative stress
relative to other cell types.26 Indeed, we found that the presence of oxidizing agents increased lipid peroxidation levels by 7- to 9-fold in the
cerebral cortex but not in the muscle or in the liver. However, such
increase was similar in sleep-deprived and control animals, suggesting
that sustained periods of wakefulness do not represent an additional
oxidative challenge. As discussed in the Methods section, the experimental design of our study is a compromise between the labor-intensive
nature of the long-term sleep deprivation experiments, which limited us
Table 2—Measures* of lipid peroxidation, protein oxidation, oxidant
production and SOD enzymatic activities in peripheral tissues of longterm sleep deprived rats and their yoked controls.
Liver
YC
Muscle
3-14d TSD
YC
3-14d TSD
Lipid peroxidation,
nmol TBARS
per mg protein
basal
9.86 ± 1.44
induced
11.69 ± 1.81
10.35 ± 1.81
9.83 ± 1.11
4.10 ± 0.14
5.52 ± 0.49
3.94 ± 0.19
4.88 ± 0.19
Protein oxidation,
carbonyl
per mg protein
4.92 + 0.89
6.91 + 1.17
15.76 + 2.59
16.36 + 1.35
Oxidant production,
nmol DCF
per mg protein
per minute
- sodium azide 3.77 + 0.25
+ sodium azide 5.42 + 0.37
4.57 + 0.35
6.42 + 0.81
12.23 + 0.86
22.17 +1.62
12.99 + 0.77
23.64 + 2.93
SOD, units
per mg protein
Total SOD
Mn SOD
Cu-Zn SOD
15.32 + 1.65
5.93 + 1.15
9.38 + 2.04
47.22 + 4.05
14.81 + 4.19
32.43 + 2.84
48.45 + 3.06
13.45 + 2.33
34.99 + 4.36
14.68 + 1.78
5.06 + 1.61
9.72 + 1.77
YC refers to yoked controls; 3-14d TSD, rats that were totally sleep deprived for 3-14 days;
TBARS, thiobarbituric acid reactive substances; DCF, 2´-7´-dichlorofluorescin; SOD,
superoxide dismutase; Mn SOD, manganese-dependent superoxide dismutase; Cu-Zn
SOD, copper and zinc-dependent superoxide dismutase
*All measurements were performed in tissues from single animals. Since in all cases sleepdeprived rats were not significantly different from their controls (Wilcoxon matched paired
test), data are shown as mean ± SEM.
SLEEP, Vol. 27, No. 1, 2004
33
Cellular Responses to Oxidative Stress—Gopalakrishnan et al
Thus, the relative decrease in brain metabolic activity during prolonged
wakefulness may play a crucial role in avoiding any potential increase in
oxidant production and may at least partially explain why oxidant production was decreased, rather than increased, after long-term sleep
deprivation.
Additional protection against oxidative stress during sleep deprivation
could be provided by the noradrenergic system of the locus coeruleus.
Norepinephrine can inhibit lipid peroxidation in vitro46 and can promote
the survival of dopaminergic neurons by directly reducing oxidant production.47 The firing of the noradrenergic cells of the locus coeruleus is
high in wakefulness, low in NREM sleep, and absent in REM sleep.48
The activity of noradrenergic neurons has not been measured during prolonged sleep loss. However, transcript levels of aryl sulfotransferase, the
main enzyme responsible in rodents for the catabolism of catecholamines, increase progressively during sleep deprivation,49 suggesting that prolonged wakefulness may be accompanied by elevated locus
coeruleus activity. To test the hypothesis that norepinephrine can protect
sleep-deprived animals from oxidative stress, we are currently measuring markers of oxidative damage in rats subjected to long-term sleep loss
following selective lesioning of the locus coeruleus.
Obstructive sleep apnea (OSA) is a frequent condition associated with
upper-airway obstruction during sleep, which causes both sleep disruption and intermittent hypoxia. Patient with OSA show significant daytime cognitive and behavioral deficits, some of which can persist after
the OSA is treated, suggesting the presence of an irreversible brain
injury. The relative contribution of blood-gas abnormalities and sleep
disruption to the cognitive deficits and the neural injury is still unclear.50
A rat model of intermittent hypoxia without sleep fragmentation has
recently been established. Experiments with this animal model show that
episodic hypoxia alone is sufficient to cause corticohippocampal apoptosis and cognitive impairment,51 and preliminary data suggest that an
increase in oxidant production may be responsible for the irreversible
neural damage.52 Our findings are compatible with this evidence and
suggest that any potentially irreversible damage in the brain of patients
with OSA is more likely due to abnormalities in blood-gas composition
than to sleep loss.
1998;9:2853-6.
14. D'Almeida V, Hipolide DC, Lobo LL, de Oliveira AC, Nobrega JN, Tufik S. Melatonin
treatment does not prevent decreases in brain glutathione levels induced by sleep deprivation. Eur J Pharmacol 2000;390:299-302.
15. Liu J, Wang X, Shigenaga MK, Yeo HC, Mori A, Ames BN. Immobilization stress causes oxidative damage to lipid, protein, and DNA in the brain of rats. FASEB J
1996;10:1532-8.
16. Radak Z, Sasvari M, Nyakas C, et al. Single bout of exercise eliminates the immobilization-induced oxidative stress in rat brain. Neurochem Int 2001;39:33-8.
17. Ramanathan L, Gulyani S, Nienhuis R, Siegel JM. Sleep deprivation decreases superoxide dismutase activity in rat hippocampus and brainstem. NeuroReport 2002;13:138790.
18. Gopalakrishnan A, Ji LL, Cirelli C. Oxidative stress and cellular damage after sleep
deprivation. Sleep 2003;26:A10.
19. Rechtschaffen A, Bergmann BM. Sleep deprivation in the rat: an update of the 1989
paper. Sleep 2002;25:18-24.
20. Shaw PJ, Bergmann BM, Rechtschaffen A. Effects of paradoxical sleep deprivation on
thermoregulation in the rat. Sleep 1998; 21:7-17.
21. Shepherd D, Garland PB. The kinetic properties of citrate synthase from rat liver mitochondria. Biochem J 1969;114: 597-610.
22. Mihara M, Uchiyama M. Determination of malonaldehyde precursor in tissues by thiobarbituric acid test. Anal Biochem 1978;86:271-278.
23. Levine RL, Garland D, Oliver CN, et al. Determination of carbonyl content in oxidatively modified proteins. Methods Enzymol 1990;186:464-78.
24. Bejma J, Ramires P, Ji LL. Free radical generation and oxidative stress with ageing and
exercise: differential effects in the myocardium and liver. Acta Physiol Scand
2000;169:343-51.
25. Sun M, Zigman S. An improved spectrophotometric assay for superoxide dismutase
based on epinephrine autooxidation. Anal Biochem 1978;90:81-9.
26. Gutteridge JM, Halliwell B. Free radicals and antioxidants in the year 2000. A historical
look to the future. Ann N Y Acad Sci 2000;899:136-47.
27. Horne JA. Why we sleep. The functions of sleep in humans and other mammals. Oxford
University Press, Oxford; 1988:319.
28. Van Dongen HPA, Maislin G, Mullington JM, Dinges DF. The cumulative cost of additional wakefulness: dose-response effects on neurobehavioral functions and sleep physiology from chronic sleep restriction and total sleep deprivation. Sleep 2003;26:117-26.
29. Yu BP. Cellular defenses against damage from reactive oxygen species. Physiol Rev
1994;74:139-162.
30. Damier P, Hirsch EC, Zhang P, Agid Y, Javoy-Agid F. Glutathione peroxidase, glial cells
and Parkinson's disease. Neuroscience 1993;52:1-6.
31. Ku HH, Sohal RS. Comparison of mitochondrial pro-oxidant generation and anti-oxidant defenses between rat and pigeon: possible basis of variation in longevity and
metabolic potential. Mech Ageing Dev 1993;72:67-76.
32. Marklund SL, Westman NG, Lundgren E, Roos G. Copper- and zinc-containing superoxide dismutase, manganese-containing superoxide dismutase, catalase, and glutathione
peroxidase in normal and neoplastic human cell lines and normal human tissues. Cancer
Res 1982;42:1955-61.
33. Ji LL. Exercise-induced modulation of antioxidant defense. Ann N Y Acad Sci
2002;959:82-92.
34. Benzi G, Moretti A. Age- and peroxidative stress-related modifications of the cerebral
enzymatic activities linked to mitochondria and the glutathione system. Free Radic Biol
Med 1995;19:77-101.
35. Candelario-Jalil E, Al-Dalain SM, Castillo R, Martinez G, Fernandez OS. Selective vulnerability to kainate-induced oxidative damage in different rat brain regions. J Appl
Toxicol 2001;21:403-7.
36. Erakovic V, Zupan G, Varljen J, Radosevic S, Simonic A. Electroconvulsive shock in
rats: changes in superoxide dismutase and glutathione peroxidase activity. Mol Brain
Res 2000;76:266-74.
37. Christen Y. Oxidative stress and Alzheimer disease. Am J Clin Nutr 2000;71: S621-9.
38. Cirelli C, Shaw PJ, Rechtschaffen A, Tononi G. No evidence of brain cell degeneration
after long-term sleep deprivation in rats. Brain Res 1999a;840:184-93.
39. Hipolide DC, D'Almeida V, Raymond R, Tufik S, Nobrega JN. Sleep deprivation does
not affect indices of necrosis or apoptosis in rat brain. Int J Neurosci 2002;112:155-66.
40. Bonnet MH, Arand DL. 24-Hour metabolic rate in insomniacs and matched normal
sleepers. Sleep 1995;18:581-8.
41. Plazzi G, Schutz Y, Cortelli P, et al. Motor overactivity and loss of motor circadian
rhythm in fatal familial insomnia: an actigraphic study. Sleep 1997;20:739-42.
42. Ramm P, Frost BJ. Regional metabolic activity in the rat brain during sleep-wake activity. Sleep 1983;6:196-216.
43. Cirelli C, Tononi G. Differences in gene expression between sleep and waking as
revealed by mRNA differential display. Mol Brain Res 1998;56:293-305.
44. Wu JC, Gillin JC, Buchsbaum MS, Hershey T, Hazlett E, Sicotte N, Bunney WE Jr. The
effect of sleep deprivation on cerebral glucose metabolic rate in normal humans assessed
with positron emission tomography. Sleep 1991;14:155-62.
45. Cortelli P, Perani D, Parchi P, et al. Cerebral metabolism in fatal familial insomnia: relation to duration, neuropathology, and distribution of protease-resistant prion protein.
Neurology 1997;49:126-33.
46. Liu J, Mori A. Monoamine metabolism provides an antioxidant defense in the brain
against oxidant- and free radical-induced damage. Arch Biochem Biophys
1993;302:118-27.
47. Troadec JD, Marien M, Darios F, et al. Noradrenaline provides long-term protection to
dopaminergic neurons by reducing oxidative stress. J Neurochem 2001;79:200-10.
48. Aston-Jones G, Bloom FE. Activity of norepinephrine-containing locus coeruleus neurons in behaving rats anticipates fluctuations in the sleep-waking cycle. J Neurosci
1981;1:876-86.
ACKNOWLEDGMENTS
We thank Drs. Ruth M. Benca and Giulio Tononi and several members of the laboratory for useful comments on the manuscript.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Droge W. Free radicals in the physiological control of cell function. Physiol Rev
2002;82:47-95.
Murrant CL, Reid MB. Detection of reactive oxygen and reactive nitrogen species in
skeletal muscle. Microsc Res Tech 2001;55:236-48.
Reimund E. The free radical flux theory of sleep. Med Hypotheses 1994;43: 231-3.
Inoué S, Honda K, Komoda Y. Sleep as neuronal detoxification and restitution. Behav
Brain Res 1995;69:91-6.
Maquet P. Functional neuroimaging of normal human sleep by positron emission tomography J Sleep Res 2000;9: 207-31.
Bonnet MH, Berry RB, Arand DL. Metabolism during normal, fragmented, and recovery sleep. J Appl Physiol 1991;71:1112-8.
Bettendorff L, Sallanon-Moulin M, Touret M, Wins P, Margineanu I, Schoffeniels E.
Paradoxical sleep deprivation increases the content of glutamate and glutamine in rat
cerebral cortex. Sleep 1996;19:65-71.
Leonard CS, Michaelis EK, Mitchell KM. Activity-dependent nitric oxide concentration
dynamics in the laterodorsal tegmental nucleus in vitro. J Neurophysiol 2001;86:21592172.
Williams JA, Vincent SR, Reiner PB. Nitric oxide production in rat thalamus changes
with behavioral state, local depolarization, and brainstem stimulation. J Neurosci
1997;17:420-427.
Everson CA, Smith CB, Sokoloff L. Effects of prolonged sleep deprivation on local rates
of cerebral energy metabolism in freely moving rats. J Neurosci 1994;14:6769-78.
Rechtschaffen A, Bergmann BM, Gilliland MA, Bauer K. Effects of method, duration,
and sleep stage on rebounds from sleep deprivation in the rat. Sleep 1999;22:11-31.
D'Almeida V, Hipolide DC, Azzalis LA, Lobo LL, Junqueira VB, Tufik S. Absence of
oxidative stress following paradoxical sleep deprivation in rats. Neurosci Lett
1997;235:25-8.
D'Almeida V, Lobo LL, Hipolide DC, de Oliveira AC, Nobrega JN, Tufik S. Sleep deprivation induces brain region-specific decreases in glutathione levels. NeuroReport
SLEEP, Vol. 27, No. 1, 2004
34
Cellular Responses to Oxidative Stress—Gopalakrishnan et al
49. Cirelli C, Tononi G. Changes in gene expression in the cerebral cortex of rats after shortterm and long-term sleep deprivation. Sleep 1999b;22 Suppl 1:113.
50. Beebe DW, Gozal D. Obstructive sleep apnea and the prefrontal cortex: towards a comprehensive model linking nocturnal upper airway obstruction to daytime cognitive and
behavioral deficits. J Sleep Res 2002;11:1-16.
51. Gozal D, Daniel JM, Dohanich GP. Behavioral and anatomical correlates of chronic
episodic hypoxia during sleep in the rat. J Neurosci 2001;21:2442-50.
52. Liu R, Luo C, Goldbart A, et al. Increased production of reactive oxygen species is associated with intermittent hypoxia-mediated cortical neural cell death. Soc Neurosci Abstr
2002;225.7.
SLEEP, Vol. 27, No. 1, 2004
35
Cellular Responses to Oxidative Stress—Gopalakrishnan et al