Brain, Behavior, and Immunity 38 (2014) 53–58
Contents lists available at ScienceDirect
Brain, Behavior, and Immunity journal homepage: www.elsevier.com/locate/ybrbi
Short Communication
Increased sleep fragmentation in experimental autoimmune encephalomyelitis Junyun He, Yuping Wang, Abba J. Kastin, Weihong Pan ⇑ Pennington Biomedical Research Center, Baton Rouge, LA 70808, USA
a r t i c l e
i n f o
Article history: Received 6 January 2014 Received in revised form 9 February 2014 Accepted 9 February 2014 Available online 22 February 2014 Keywords: Sleep efficiency Sleep fragmentation Sleep architecture EAE
a b s t r a c t Sleep disturbance in patients with multiple sclerosis is prevalent and has multifactorial causes. In mice with experimental autoimmune encephalomyelitis (EAE), a model of multiple sclerosis, we determined the dynamic changes of sleep architecture and the interactions between sleep changes and EAE symptoms. The changes of sleep patterns were mainly reflected by altered sleep stage distribution and increased sleep fragmentation. Increased waking and decreased non-rapid eye movement sleep occurred after EAE onset and persisted through the symptomatic phase. There also was increased sleep state transition, indicating a reduction of sleep cohesiveness. Furthermore, the extent of sleep fragmentation correlated with the severity of disease. This is the first study of sleep characteristics in EAE mice demarcating specific changes related to the autoimmune disorder without confounding factors such as psychosocial impact and treatment effects. The reduction of sleep efficiency and cohesiveness supports the notion that enhancing sleep might facilitate the recovery of mice from EAE, pertinent to the multimodality treatment of multiple sclerosis. Ó 2014 Published by Elsevier Inc.
1. Introduction Patients with multiple sclerosis (MS) frequently have multiple sleep complaints, such as insomnia, hypersomnolence, sleep apnea, parasomnia, and secondary narcolepsy. Sleep disorders occur in more than 50% of MS patients and contribute to fatigue. Poor sleep may be manifested by a decrease in total sleep time, rapid eye movement (REM) sleep, or sleep efficiency. This is associated with an increase in Stage 1 non-REM (NREM) sleep, as well as changes in sleep stage and arousals (Lunde et al., 2012, 2013; Neau et al., 2012). Sleep disturbance is influenced by MS-related symptoms and adverse effects from immunotherapy and symptomatic medications (Pokryszko-Dragan et al., 2013). Therefore, experimental autoimmune encephalomyelitis (EAE), a widely used animal model of MS, provides a useful tool to determine changes in sleep architecture in the absence of the psychosocial interactions and treatment effects seen in human beings. In animal studies, cytokines and other inflammatory challenges have been associated with hypersomnolence as well as increased sleep fragmentation (Krueger et al., 2007; Kaushal et al., 2012). However, altered sleep patterns in EAE have not been character-
ized. We hypothesize that the development of EAE is associated with disrupted sleep, shown by decreased sleep efficiency and increased sleep fragmentation. This was tested by simultaneous monitoring of disease severity and sleep–wake activities. The results provide a basis for further studies to determine whether sleep treatment helps resolve EAE and the human disease MS. 2. Materials and methods 2.1. Mice and surgery The animal studies were approved by the Institutional Animal Care and Use Committee. Female FVB mice (Jackson Laboratory, Bar Harbor, ME) were group-housed, lights on at 6 am (Zeitgeber time ZT = 0) and lights off at 6 pm (ZT12), and fed ad lib. Headmounts for sleep recordings were placed on mice 7–8 weeks old as described previously (Wang et al., 2013). Mice were allowed to recover for 2 weeks before baseline sleep recording and subsequent EAE induction. 2.2. EAE induction and scoring
⇑ Corresponding author. Address: Blood-Brain Barrier Group, Pennington Biomedical Research Center, 6400 Perkins Road, Baton Rouge, LA 70808, USA. Tel.: +1 225 763 0266; fax: +1 225 763 0261. E-mail address: [email protected] (W. Pan). http://dx.doi.org/10.1016/j.bbi.2014.02.005 0889-1591/Ó 2014 Published by Elsevier Inc.
Two groups of mice were studied: EAE or vehicle control (n = 9–10/group). EAE was induced in 9–10 week old mice on day 0 following an established protocol (Li et al., 2011; Mishra et al.,
54
J. He et al. / Brain, Behavior, and Immunity 38 (2014) 53–58
2013). Each mouse received subcutaneous injection of 100 lg of myelin oligodendrocyte glycoprotein (MOG) fragment 79–96. MOG79–96 was emulsified in 100 ll of complete Freund’s adjuvant (CFA) containing 500 lg of heat-killed Mycobacterium tuberculosis H37RA (DIFCO Laboratories, Detroit, MI). The injection was performed at 3–5 pm (ZT 9–11 h). Pertussis toxin (PT, Sigma, St. Louis, MO) was injected intraperitoneally immediately after induction and again 48 h later. The mice in the vehicle control group underwent the same procedure except that they received CFA + PT only without MOG79–96. Symptoms were monitored at ZT 5–6 h daily by use of a standard EAE scoring system (Pan et al., 1996; Wu et al., 2010, 2013; Mishra et al., 2013), with 0 being symptom-free and 5 being the worst (moribund or dead). 2.3. Sleep recording and scoring Mice with headmounts were individually housed in cylindrical sleep recoding cages, and acclimatized for 3 days before the initiation of recording by connection of the headmount to a preamplifier. Data acquisition and processing were achieved by use of the Pinnacle Serenia System (Pinnacle Technology, Lawrence, KS), as described previously (Wang et al., 2013). Sleep data were acquired for at least 24 h continuously from ZT0, first at baseline before EAE induction, and then again at days 7, 14, 21, and 28 after EAE induction. Manual scoring was performed on 10 s epochs of EEG (0.5–40 Hz) and EMG (10–100 Hz) signals by an experienced sleep researcher, following standard scoring criteria (Veasey et al., 2000; Louis et al., 2004). Hypnograms, the percentage of each sleep stage (wake, NREM, or REM), and sleep bouts were analyzed with Sleep Pro software (Pinnacle Technology). A bout was defined as a minimum of 3 consecutive epochs at a length of 10 s/epoch for a given state. 2.4. Statistical analysis Repeated measures two-way analysis of variance was used to determine the changes of EAE scores over time and the difference between EAE and CFA + PT control groups. Randomized block analysis was performed to determine the effects of EAE and time after induction on sleep architecture, followed by the Bonferroni post hoc test when significant overall changes were identified. Linear regression was used to analyze the correlation between EAE score and sleep fragmentation. Prism GraphPad 5 statistical and graphic program (GraphPad, San Diego, CA) was used to conduct statistical analyses and generate graphs. 3. Results 3.1. EAE mice have decreased sleep efficiency after the onset of disease Both EAE induction and disease duration affected the EAE scores significantly. EAE symptoms occurred around day 12 and persisted throughout the course of the study. There was an initial peak at day 13–14, followed by incomplete recovery, and further worsening of the ascending weakness that reached a plateau on days 25–34 (Fig. 1E and F, inset). The total amount as well as the percentage of wake time during 24 h, representative of sleep efficiency, showed significant changes as a result of EAE and time after induction as shown by two-way ANOVA. During the 24 h baseline recording, there was no difference between the EAE and control groups in the % wake. In the subsequent month, the CFA + PT control mice (n = 9) did not show a significant increase of % wake when sampled weekly. The EAE group (n = 10), however, showed an increase of % wake at day 14, 21, and 28 after induction. The difference between the two groups
was most apparent during the light phase; there was an effect of EAE [F(1, 59) = 14.1, p < 0.0005] as well as time [F(4, 59) = 4.4, p < 0.005], and there was also a significant interaction [F(4, 59) = 3.2, p < 0.05]. Post-hoc analysis confirmed significant increases on day 14, 21, and 28 after induction of EAE, though the CFA + PT group did not show changes (Fig. 1A). In the dark phase, the % wake only showed a significant overall effect of time (p < 0.05). There was no compensatory recovery sleep that would be reflected by decreased wake percentage (Fig. 1B). The %NREM was also similar at baseline and a week after EAE induction in comparison with the CFA + PT group. While the CFA + PT mice did not show subsequent reduction of NREM sleep, the EAE mice had a significant decrease of %NREM on day 14 (p < 0.05), day 21 (p < 0.005), and day 28 (p < 0.01) in comparison with their own baseline. The decrease was most pronounced during the light phase (Fig. 1C) but not the dark phase (Fig. 1D). By contrast, the %REM did not show significant change, possibly related to large intra-group variation, in either the light phase (Fig. 1E) or dark phase (Fig. 1F). 3.2. EAE is associated with increased sleep fragmentation, the extent of which correlates with symptom severity The number of wake bouts reflects sleep state transition as well as prolonged arousal. It did not show a significant change in the CFA + PT control group. The EAE group had a similar baseline and day 7 waking bout counts as the control group, but there was an increase on day 14 (p < 0.01), day 21 (p < 0.05), and day 28 (p < 0.01) when compared with its own baseline before EAE induction. There was also an increase on day 14 (p < 0.05) and day 28 (p < 0.01) when the EAE group was compared with the same-date CFA + PT controls. When the 24 h interval was analyzed separately in the light and dark phases, wake bouts showed significant changes as a result of EAE and time after induction in the light phase, without significant interactions of the two factors or post hoc difference overtime (Fig. 2A). In the dark phase, there was a significant overall effect of EAE (Fig. 2B). Representative hypnograms of an EAE mouse at baseline, day 14, day 21, and day 28 are shown in Fig. 2C. Whereas CFA + PT alone did not change NREM bouts, EAE mice had more NREM bouts on day 14 and day 28 (p < 0.01) when compared with either baseline or the CFA + PT controls on the same date after induction. NREM bout counts were not different on day 21, a time point corresponding to partial resolution of EAE scores that is shown in the insets of Fig. 1E and F. The significant overall effect of EAE to increase NREM bouts was present in both the light phase (Fig. 2D) and dark phase (Fig. 2E). The frequency of sleep state transition correlated with the severity of EAE. When the number of wake/sleep bouts was plotted against EAE score at 14, 21, and 28 days, there was a linear correlation between EAE score and wake bouts (p < 0.01), as well as between EAE score or NREM bouts (p < 0.01). Increased severity of EAE (higher scores) coincided with an increased number of wake bouts and sleep bouts (Fig. 2F). 4. Discussion We found that sleep disturbance occurred after the onset of EAE and mainly presented with decreased sleep efficiency and increased sleep fragmentation. The extent of sleep fragmentation was positively correlated with the severity of the disease. This is the first study involving sleep and EAE of which we are aware. It addresses the biologically important question how a CNS autoimmune disorder induces dynamic changes of sleep quality. The findings differ from previous reports on sleep architecture in rodents
55
J. He et al. / Brain, Behavior, and Immunity 38 (2014) 53–58 50
80
#
EAE
*
**
**
2
CFA+PT
1 40
2
45
40 1 35
30
0
0 EAE score
EAE score
20 -1
7
14
21
25
28
-1
7
E
80 EAE
21
28
5 CFA+PT
CFA+PT
EAE
2 4
1
50
3 2.5
2
1
0
2.0
EAE clinical score
60
% REM in light phase
70
EAE score
% Wake in dark phase
14
Time after EAE induction (d)
Time after EAE induction (d)
B
EAE score
60
EAE score
% Wake in light phase
CFA+PT
EAE
D % NREM in dark phase
A
CFA+PT EAE
1.5 1.0 0.5 0.0
EAE score
-0.5 4
8
12
16
20
24
28
32
Time after EAE induction (d)
40
-1
7
14
21
0
28
-1
7
14
21
28
Time after EAE induction (d)
C
60
F
CFA+PT
EAE
6 EAE
CFA+PT
1
30
*
#
** ** #
4
2.5 2.0
EAE clinical score
40
% REM in dark phase
50
EAE score
% NREM in light phase
2
2
0
CFA+PT EAE
1.5 1.0 0.5 0.0
EAE score
-0.5 4
8
12
16
20
24
28
32
Time after EAE induction (d)
20 -1
7
14
21
28
Time after EAE induction (d)
0 -1
7
14
21
28
Time after EAE induction (d)
Fig. 1. Sleep efficiency and sleep state distribution during the course of EAE, and their correlation with EAE scores. EAE mice were compared across the duration of EAE with a control group receiving CFA + PT; the relationship of sleep stages (solid line, left y-axis) and EAE scores (dashed line, right y-axis) is shown by superimposed plots. (A) Percentage of wake during light phase was increased at 2–4 weeks after EAE induction than time 0 (⁄p < 0.05, ⁄⁄p < 0.01), and higher than the CFA + PT group on week 3 (#p < 0.05). (B) Percentage of wake during dark phase also tended to show an overall increase though the post hoc test did not show a statistical difference from 0. (C) Reduction of %NREM in the light phase was seen in EAE mice at 2–4 weeks after EAE induction in comparison with its baseline (⁄p < 0.05, ⁄⁄p < 0.01). The CFA + PT control group did not show significant change. The difference between the EAE and CFA + PT groups was significant at week 2 and 3 (#p < 0.05). (D) The overall decrease of %NREM in the dark phase in the EAE mice was paralleled by the CFA + PT mice. (E) There was no statistically significant reduction of %REM in the light phase. (F) The %REM was also unchanged in the dark phase across the time course of EAE. Inset shows that EAE mice started to show symptoms on day 11, were significantly different from the control group by day 13, achieved the first brief peak on day 14–15, and sustained symptoms after day 25.
treated with inflammatory cytokines or adjuvant, thereby providing novel insight into regulatory changes specific to EAE. In this study, EAE mice were compared longitudinally with their own baseline, and in parallel with controls receiving CFA and PT that induce inflammation despite the lack of autoimmune disease. CFA has been shown to induce either a reduction or increase of sleep. In rats receiving Freund’s adjuvant in the temporomandibular joint, there is a reduction in sleep efficiency during the subsequent two light periods, delayed sleep onset, and an increased number of awakenings. These changes could be partially reversed
by indomethacin treatment, suggesting that inflammation and inflammatory pain are a major cause of sleep disturbance (Schutz et al., 2003). In a Russian abstract, CFA increases slow-wave and REM sleep in rats. CFA alone leads to a gradual onset of sleepiness that is worse at 3–5 weeks, whereas a combination of CFA and bovine serum albumin precipitates the onset at about 1 week, with symptoms lasting 5 weeks in association with reduced serotonin in the striatum and frontal cortex (Pankova et al., 2001). Results from these two reports indicate a potential biphasic change, with sleep loss during the early stage and hypersomnolence in the later
56
J. He et al. / Brain, Behavior, and Immunity 38 (2014) 53–58
D
CFA+PT
EAE
80
EAE CFA+PT
NREM bouts in light phase
80
2 60
1
20
2
60
1
40
0
0
EAE score
EAE score
0 -1
7
14
21
20
28
-1
7
Time after EAE induction (d)
B
E
CFA+PT
21
28
80 EAE
60
40
1
20 0
CFA+PT
2
EAE score
60
1
EAE score
NREM bouts in dark phase
2
EAE score
Wake bouts in dark phase
14
Time after EAE induction (d)
80 EAE
EAE score
40
EAE score
Wake bouts in light phase
A
40
0
EAE score
0 -1
7
14
21
20
28
-1
Time after EAE induction (d)
C
7
14
21
28
Time after EAE induction (d)
baseline REM NREM WAKE
0
6
12
18
24
F
EAE day14
300
REM
NREM bouts
Number of bouts in 24 h
Wake bouts
NREM WAKE
0
6
12
18
24
EAE day21 REM NREM
200
100
WAKE
0
6
12
18
24
0
EAE day 28
0
REM
1
2 EAE score
3
4
NREM WAKE
0
6
12
18
24
Zeitgeber Time (h)
Fig. 2. Increased sleep fragmentation (solid lines, left y-axis) and correlation with EAE scores (dash line, right y-axis). (A) Wake bouts in the light phase showed a significant overall effect of EAE and days after induction, and a lack of interaction of the two factors by two-way ANOVA. (B) Wake bouts in the dark phase showed a significant overall effect of EAE but not days after induction. (C) In hypnograms of a representative EAE mouse during the course of disease, the increased sleep state transition was most apparent in the dark phase on week 2. (D) NREM bouts in the light phase showed a significant overall effect of EAE. (E) NREM bouts in the dark phase also showed a significant overall effect of EAE but not days after induction. (F) EAE scores correlated with the number of wake bouts (r = 0.63) and NREM bouts (r = 0.62).
course of the inflammation. In our study, CFA + PT did not show a significant effect on sleep architecture during weekly recordings for a month after injection. This may be related to dose, species,
and sex differences, though it is possible that an early change was missed since we did not analyze sleep features immediately after injection. We were not able to monitor within the first week
J. He et al. / Brain, Behavior, and Immunity 38 (2014) 53–58
of induction, as mice were required to remain in an Animal Biosafety Level 2 facility. Sickness behavior as a result of inflammation has been associated with blunted circadian rhythm, increased slow wave sleep, and reduced REM sleep (Dantzer et al., 2008). EAE animals show sickness behavior (Pollak et al., 2003; Campbell et al., 2010). Thus, we set out to test the hypothesis that EAE mice have hypersomnolence. However, contrary to the prediction, the result showed increased waking as well as more fragmented sleep. Based on the literature, there are two main classes of candidate molecules that might contribute to the changes of sleep architecture in EAE mice: proinflammatory cytokines and other inflammatory mediators that may interact and cross the blood–brain barrier (Pan et al., 2011) to exert CNS effects on sleep; and neurotransmitters within the CNS. In general, proinflammatory cytokines tend to be somnogenic whereas anti-inflammatory cytokines inhibit sleep (Zielinski and Krueger, 2011). EAE is associated with production of many cytokines, chemokines, as well as other soluble molecules (Raine, 1994; Eng et al., 1996; El-behi et al., 2010). Higher NREM and suppressed REM activity are induced by the T-cell derived cytokine CD40 ligand and by peripheral injection of tumor necrosis factor a (TNF) (Gast et al., 2013). While these substances promote hypersomnolence, the reduced sleep in the EAE mice suggests that the counterregulatory anti-inflammatory cytokines might be responsible. Nonetheless, sleep fragmentation was a main feature observed in the EAE mice in our study, an effect also exerted by proinflammatory cytokines. We have shown that FVB female mice on day 28 of EAE have higher levels of adenosine in the hippocampus (Mishra et al., 2013). These mice also have higher levels of the inhibitory neurotransmitter GABA and lower levels of the excitatory glutamate in the hippocampus (Pan et al., unpublished observations). Adenosine and GABA are known to promote sleep, whereas acetylcholine and biogenic amines are wake-promoting, although their sleep regulatory effects are dependent on specific CNS regions. EAE mice have dynamic changes of amino acids and biogenic amines, with regional differences and distinct patterns during the course of disease progression. Though several amino acids show an increase at EAE onset and peak but return to baseline during the chronic phase, there is a reduction of glutamate, GABA, and norepinephrine, as well as decreased 5-HT turnover in the late stage (Musgrave et al., 2011a, 2011b). There is defective GABA transmission in EAE (Rossi et al., 2011), whereas GABAergic agents can decrease inflammation and inhibit EAE (Bhat et al., 2010). In parallel, slice electrophysiology shows that EAE is associated with enhanced glutamatergic excitatory postsynaptic currents in the early phase, related to inflammatory cytokines released from infiltrating T cells and from activated microglia (Centonze et al., 2010). The serotonin system is also involved; knockout mice without the 5-HT transporter have less severe EAE, supporting a disease-ameliorating role of 5-HT activity (Hofstetter et al., 2005). Thus, we speculate that decreased GABA transmission and increased serotonergic activity at the peak and chronic stage could result in increased waking, more sleep fragmentation, and attenuated sleep cohesiveness. To fully establish a causal relationship, detailed analyses by use of stereotaxic injection of agonists and antagonists of the candidate molecule or its receptors are needed. The greater changes of sleep architecture and sleep state transition in the light phase are consistent with a blunted circadian rhythm. In C57BL/6 mice at day 2–8 after the onset of EAE symptoms, there is lower amplitude of change of per2 and clock gene expression in the liver and higher serum levels of corticosterone and leptin (Buenafe, 2012). This provides a basis for further characterization of altered circadian rhythms during the course of EAE. Poor sleep quality is well-known to impair immune function in human and animal studies. The deterioration of sleep architecture
57
shown in this study is a consequence of EAE, but would also contribute to neuroimmune dysregulation. In this sense, normalization of sleep by increasing efficiency and reducing fragmentation may facilitate the recovery of mice from EAE. Thus, this first characterization of sleep features in EAE mice opens new directions for the manipulation of sleep to modulate neurological outcome. Acknowledgments AJK receives grant support from NIH DK54880 and DK92245, and WP is supported by NS62291 and the AASM-accredited PBRC Sleep Health Center. We thank Hung Hsuchou and Aurelien Mace for technical assistance and discussions. References Bhat, R., Axtell, R., Mitra, A., Miranda, M., Lock, C., Tsien, R.W., Steinman, L., 2010. Inhibitory role for GABA in autoimmune inflammation. Proc. Natl. Acad. Sci. U.S.A. 107, 2580–2585. Buenafe, A.C., 2012. Diurnal rhythms are altered in a mouse model of multiple sclerosis. J. Neuroimmunol. 243, 12–17. Campbell, S.J., Meier, U., Mardiguian, S., Jiang, Y., Littleton, E.T., Bristow, A., Relton, J., Connor, T.J., Anthony, D.C., 2010. Sickness behaviour is induced by a peripheral CXC-chemokine also expressed in multiple sclerosis and EAE. Brain Behav. Immun. 24, 738–746. Centonze, D., Muzio, L., Rossi, S., Furlan, R., Bernardi, G., Martino, G., 2010. The link between inflammation, synaptic transmission and neurodegeneration in multiple sclerosis. Cell Death Differ. 17, 1083–1091. Dantzer, R., O’Connor, J.C., Freund, G.G., Johnson, R.W., Kelley, K.W., 2008. From inflammation to sickness and depression: when the immune system subjugates the brain. Nat. Rev. Neurosci. 9, 46–56. El-behi, M., Rostami, A., Ciric, B., 2010. Current views on the roles of Th1 and Th17 cells in experimental autoimmune encephalomyelitis. J. Neuroimmune. Pharmacol. 5, 189–197. Eng, L.F., Ghirnikar, R.S., Lee, Y.L., 1996. Inflammation in EAE: role of chemokine/ cytokine expression by resident and infiltrating cells. Neurochem. Res. 21, 511– 525. Gast, H., Muller, A., Lopez, M., Meier, D., Huber, R., Dechent, F., Prinz, M., Emmenegger, Y., Franken, P., Birchler, T., Fontana, A., 2013. CD40 activation induces NREM sleep and modulates genes associated with sleep homeostasis. Brain Behav. Immun. 27, 133–144. Hofstetter, H.H., Mossner, R., Lesch, K.P., Linker, R.A., Toyka, K.V., Gold, R., 2005. Absence of reuptake of serotonin influences susceptibility to clinical autoimmune disease and neuroantigen-specific interferon-gamma production in mouse EAE. Clin. Exp. Immunol. 142, 39–44. Kaushal, N., Ramesh, V., Gozal, D., 2012. TNF-alpha and temporal changes in sleep architecture in mice exposed to sleep fragmentation. PLoS One 7, e45610. Krueger, J.M., Rector, D.M., Churchill, L., 2007. Sleep and cytokines. Sleep Med. Clin. 2, 161–169. Li, Q., Powell, N., Zhang, H., Belevych, N., Ching, S., Chen, Q., Sheridan, J., Whitacre, C., Quan, N., 2011. Endothelial IL-1R1 is a critical mediator of EAE pathogenesis. Brain Behav. Immun. 25, 160–167. Louis, R.P., Lee, J., Stephenson, R., 2004. Design and validation of a computer-based sleep-scoring algorithm. J. Neurosci. Methods 133, 71–80. Lunde, H.M., Aae, T.F., Indrevag, W., Aarseth, J., Bjorvatn, B., Myhr, K.M., Bo, L., 2012. Poor sleep in patients with multiple sclerosis. PLoS One 7, e49996. Lunde, H.M., Bjorvatn, B., Myhr, K.M., Bo, L., 2013. Clinical assessment and management of sleep disorders in multiple sclerosis: a literature review. Acta Neurol. Scand. Suppl., 24–30. Mishra, P.K., Hsuchou, H., Ouyang, S., Kastin, A.J., Wu, X., Pan, W., 2013. Loss of astrocytic leptin signaling worsens experimental autoimmune encephalomyelitis. Brain Behav. Immun. 34, 98–107. Musgrave, T., Benson, C., Wong, G., Browne, I., Tenorio, G., Rauw, G., Baker, G.B., Kerr, B.J., 2011a. The MAO inhibitor phenelzine improves functional outcomes in mice with experimental autoimmune encephalomyelitis (EAE). Brain Behav. Immun. 25, 1677–1688. Musgrave, T., Tenorio, G., Rauw, G., Baker, G.B., Kerr, B.J., 2011b. Tissue concentration changes of amino acids and biogenic amines in the central nervous system of mice with experimental autoimmune encephalomyelitis (EAE). Neurochem. Int. 59, 28–38. Neau, J.P., Paquereau, J., Auche, V., Mathis, S., Godeneche, G., Ciron, J., Moinot, N., Bouche, G., 2012. Sleep disorders and multiple sclerosis: a clinical and polysomnography study. Eur. Neurol. 68, 8–15. Pan, W., Banks, W.A., Kennedy, M.K., Gutierrez, E.G., Kastin, A.J., 1996. Differential permeability of the BBB in acute EAE: enhanced transport of TNF-a. Am. J. Physiol. 271, E636–E642. Pan, W., Stone, K.P., Hsuchou, H., Manda, V.K., Zhang, Y., Kastin, A.J., 2011. Cytokine signaling modulates blood–brain barrier function. Curr. Pharm. Des. 17, 3729– 3740. Pankova, N.B., Popkova, E.V., Vetrile, L.A., Basharova, L.A., Krupina, N.A., 2001. [Alterations in diurnal sleep structure, electrical activity, and neurochemical
58
J. He et al. / Brain, Behavior, and Immunity 38 (2014) 53–58
parameters of the rat brain induced by systemic injection of complete Freund adjuvant and immunization by bovine serum albumin with complete Freund adjuvant. Zh. Vyssh. Nerv. Deiat. Im. I P Pavlova 51, 339–347. Pokryszko-Dragan, A., Bilinska, M., Gruszka, E., Biel, L., Kaminska, K., Konieczna, K., 2013. Sleep disturbances in patients with multiple sclerosis. Neurol. Sci. 34, 1291–1296. Pollak, Y., Ovadia, H., Orion, E., Weidenfeld, J., Yirmiya, R., 2003. The EAE-associated behavioral syndrome: I. Temporal correlation with inflammatory mediators. J. Neuroimmunol. 137, 94–99. Raine, C.S., 1994. The Dale E. McFarlin memorial lecture: the immunology of the multiple sclerosis lesion. Ann. Neurol. 36 (Suppl.), S61–S72. Rossi, S., Muzio, L., De C, V., Grasselli, G., Musella, A., Musumeci, G., Mandolesi, G., De, C.R., Maida, S., Biffi, E., Pedrocchi, A., Menegon, A., Bernardi, G., Furlan, R., Martino, G., Centonze, D., 2011. Impaired striatal GABA transmission in experimental autoimmune encephalomyelitis. Brain Behav. Immun. 25, 947–956.
Schutz, T.C., Andersen, M.L., Tufik, S., 2003. Sleep alterations in an experimental orofacial pain model in rats. Brain Res. 993, 164–171. Veasey, S.C., Valladares, O., Fenik, P., Kapfhamer, D., Sanford, L., Benington, J., Bucan, M., 2000. An automated system for recording and analysis of sleep in mice. Sleep 23, 1025–1040. Wang, Y., He, J., Kastin, A.J., Hsuchou, H., Pan, W., 2013. Hypersomnolence and reduced activity in pan-leptin receptor knockout mice. J. Mol. Neurosci. 51, 1038–1045. Wu, X., Pan, W., He, Y., Hsuchou, H., Kastin, A.J., 2010. Cerebral interleukin-15 shows upregulation and beneficial effects in experimental autoimmune encephalomyelitis. J. Neuroimmunol. 223, 65–72. Wu, X., Mishra, P.K., Hsuchou, H., Kastin, A.J., Pan, W., 2013. Upregulation of astrocytic leptin receptor in mice with experimental autoimmune encephalomyelitis. J. Mol. Neurosci. 49, 446–456. Zielinski, M.R., Krueger, J.M., 2011. Sleep and innate immunity. Front Biosci. (Schol. Ed.) 3, 632–642.
Contents lists available at ScienceDirect
Brain, Behavior, and Immunity journal homepage: www.elsevier.com/locate/ybrbi
Short Communication
Increased sleep fragmentation in experimental autoimmune encephalomyelitis Junyun He, Yuping Wang, Abba J. Kastin, Weihong Pan ⇑ Pennington Biomedical Research Center, Baton Rouge, LA 70808, USA
a r t i c l e
i n f o
Article history: Received 6 January 2014 Received in revised form 9 February 2014 Accepted 9 February 2014 Available online 22 February 2014 Keywords: Sleep efficiency Sleep fragmentation Sleep architecture EAE
a b s t r a c t Sleep disturbance in patients with multiple sclerosis is prevalent and has multifactorial causes. In mice with experimental autoimmune encephalomyelitis (EAE), a model of multiple sclerosis, we determined the dynamic changes of sleep architecture and the interactions between sleep changes and EAE symptoms. The changes of sleep patterns were mainly reflected by altered sleep stage distribution and increased sleep fragmentation. Increased waking and decreased non-rapid eye movement sleep occurred after EAE onset and persisted through the symptomatic phase. There also was increased sleep state transition, indicating a reduction of sleep cohesiveness. Furthermore, the extent of sleep fragmentation correlated with the severity of disease. This is the first study of sleep characteristics in EAE mice demarcating specific changes related to the autoimmune disorder without confounding factors such as psychosocial impact and treatment effects. The reduction of sleep efficiency and cohesiveness supports the notion that enhancing sleep might facilitate the recovery of mice from EAE, pertinent to the multimodality treatment of multiple sclerosis. Ó 2014 Published by Elsevier Inc.
1. Introduction Patients with multiple sclerosis (MS) frequently have multiple sleep complaints, such as insomnia, hypersomnolence, sleep apnea, parasomnia, and secondary narcolepsy. Sleep disorders occur in more than 50% of MS patients and contribute to fatigue. Poor sleep may be manifested by a decrease in total sleep time, rapid eye movement (REM) sleep, or sleep efficiency. This is associated with an increase in Stage 1 non-REM (NREM) sleep, as well as changes in sleep stage and arousals (Lunde et al., 2012, 2013; Neau et al., 2012). Sleep disturbance is influenced by MS-related symptoms and adverse effects from immunotherapy and symptomatic medications (Pokryszko-Dragan et al., 2013). Therefore, experimental autoimmune encephalomyelitis (EAE), a widely used animal model of MS, provides a useful tool to determine changes in sleep architecture in the absence of the psychosocial interactions and treatment effects seen in human beings. In animal studies, cytokines and other inflammatory challenges have been associated with hypersomnolence as well as increased sleep fragmentation (Krueger et al., 2007; Kaushal et al., 2012). However, altered sleep patterns in EAE have not been character-
ized. We hypothesize that the development of EAE is associated with disrupted sleep, shown by decreased sleep efficiency and increased sleep fragmentation. This was tested by simultaneous monitoring of disease severity and sleep–wake activities. The results provide a basis for further studies to determine whether sleep treatment helps resolve EAE and the human disease MS. 2. Materials and methods 2.1. Mice and surgery The animal studies were approved by the Institutional Animal Care and Use Committee. Female FVB mice (Jackson Laboratory, Bar Harbor, ME) were group-housed, lights on at 6 am (Zeitgeber time ZT = 0) and lights off at 6 pm (ZT12), and fed ad lib. Headmounts for sleep recordings were placed on mice 7–8 weeks old as described previously (Wang et al., 2013). Mice were allowed to recover for 2 weeks before baseline sleep recording and subsequent EAE induction. 2.2. EAE induction and scoring
⇑ Corresponding author. Address: Blood-Brain Barrier Group, Pennington Biomedical Research Center, 6400 Perkins Road, Baton Rouge, LA 70808, USA. Tel.: +1 225 763 0266; fax: +1 225 763 0261. E-mail address: [email protected] (W. Pan). http://dx.doi.org/10.1016/j.bbi.2014.02.005 0889-1591/Ó 2014 Published by Elsevier Inc.
Two groups of mice were studied: EAE or vehicle control (n = 9–10/group). EAE was induced in 9–10 week old mice on day 0 following an established protocol (Li et al., 2011; Mishra et al.,
54
J. He et al. / Brain, Behavior, and Immunity 38 (2014) 53–58
2013). Each mouse received subcutaneous injection of 100 lg of myelin oligodendrocyte glycoprotein (MOG) fragment 79–96. MOG79–96 was emulsified in 100 ll of complete Freund’s adjuvant (CFA) containing 500 lg of heat-killed Mycobacterium tuberculosis H37RA (DIFCO Laboratories, Detroit, MI). The injection was performed at 3–5 pm (ZT 9–11 h). Pertussis toxin (PT, Sigma, St. Louis, MO) was injected intraperitoneally immediately after induction and again 48 h later. The mice in the vehicle control group underwent the same procedure except that they received CFA + PT only without MOG79–96. Symptoms were monitored at ZT 5–6 h daily by use of a standard EAE scoring system (Pan et al., 1996; Wu et al., 2010, 2013; Mishra et al., 2013), with 0 being symptom-free and 5 being the worst (moribund or dead). 2.3. Sleep recording and scoring Mice with headmounts were individually housed in cylindrical sleep recoding cages, and acclimatized for 3 days before the initiation of recording by connection of the headmount to a preamplifier. Data acquisition and processing were achieved by use of the Pinnacle Serenia System (Pinnacle Technology, Lawrence, KS), as described previously (Wang et al., 2013). Sleep data were acquired for at least 24 h continuously from ZT0, first at baseline before EAE induction, and then again at days 7, 14, 21, and 28 after EAE induction. Manual scoring was performed on 10 s epochs of EEG (0.5–40 Hz) and EMG (10–100 Hz) signals by an experienced sleep researcher, following standard scoring criteria (Veasey et al., 2000; Louis et al., 2004). Hypnograms, the percentage of each sleep stage (wake, NREM, or REM), and sleep bouts were analyzed with Sleep Pro software (Pinnacle Technology). A bout was defined as a minimum of 3 consecutive epochs at a length of 10 s/epoch for a given state. 2.4. Statistical analysis Repeated measures two-way analysis of variance was used to determine the changes of EAE scores over time and the difference between EAE and CFA + PT control groups. Randomized block analysis was performed to determine the effects of EAE and time after induction on sleep architecture, followed by the Bonferroni post hoc test when significant overall changes were identified. Linear regression was used to analyze the correlation between EAE score and sleep fragmentation. Prism GraphPad 5 statistical and graphic program (GraphPad, San Diego, CA) was used to conduct statistical analyses and generate graphs. 3. Results 3.1. EAE mice have decreased sleep efficiency after the onset of disease Both EAE induction and disease duration affected the EAE scores significantly. EAE symptoms occurred around day 12 and persisted throughout the course of the study. There was an initial peak at day 13–14, followed by incomplete recovery, and further worsening of the ascending weakness that reached a plateau on days 25–34 (Fig. 1E and F, inset). The total amount as well as the percentage of wake time during 24 h, representative of sleep efficiency, showed significant changes as a result of EAE and time after induction as shown by two-way ANOVA. During the 24 h baseline recording, there was no difference between the EAE and control groups in the % wake. In the subsequent month, the CFA + PT control mice (n = 9) did not show a significant increase of % wake when sampled weekly. The EAE group (n = 10), however, showed an increase of % wake at day 14, 21, and 28 after induction. The difference between the two groups
was most apparent during the light phase; there was an effect of EAE [F(1, 59) = 14.1, p < 0.0005] as well as time [F(4, 59) = 4.4, p < 0.005], and there was also a significant interaction [F(4, 59) = 3.2, p < 0.05]. Post-hoc analysis confirmed significant increases on day 14, 21, and 28 after induction of EAE, though the CFA + PT group did not show changes (Fig. 1A). In the dark phase, the % wake only showed a significant overall effect of time (p < 0.05). There was no compensatory recovery sleep that would be reflected by decreased wake percentage (Fig. 1B). The %NREM was also similar at baseline and a week after EAE induction in comparison with the CFA + PT group. While the CFA + PT mice did not show subsequent reduction of NREM sleep, the EAE mice had a significant decrease of %NREM on day 14 (p < 0.05), day 21 (p < 0.005), and day 28 (p < 0.01) in comparison with their own baseline. The decrease was most pronounced during the light phase (Fig. 1C) but not the dark phase (Fig. 1D). By contrast, the %REM did not show significant change, possibly related to large intra-group variation, in either the light phase (Fig. 1E) or dark phase (Fig. 1F). 3.2. EAE is associated with increased sleep fragmentation, the extent of which correlates with symptom severity The number of wake bouts reflects sleep state transition as well as prolonged arousal. It did not show a significant change in the CFA + PT control group. The EAE group had a similar baseline and day 7 waking bout counts as the control group, but there was an increase on day 14 (p < 0.01), day 21 (p < 0.05), and day 28 (p < 0.01) when compared with its own baseline before EAE induction. There was also an increase on day 14 (p < 0.05) and day 28 (p < 0.01) when the EAE group was compared with the same-date CFA + PT controls. When the 24 h interval was analyzed separately in the light and dark phases, wake bouts showed significant changes as a result of EAE and time after induction in the light phase, without significant interactions of the two factors or post hoc difference overtime (Fig. 2A). In the dark phase, there was a significant overall effect of EAE (Fig. 2B). Representative hypnograms of an EAE mouse at baseline, day 14, day 21, and day 28 are shown in Fig. 2C. Whereas CFA + PT alone did not change NREM bouts, EAE mice had more NREM bouts on day 14 and day 28 (p < 0.01) when compared with either baseline or the CFA + PT controls on the same date after induction. NREM bout counts were not different on day 21, a time point corresponding to partial resolution of EAE scores that is shown in the insets of Fig. 1E and F. The significant overall effect of EAE to increase NREM bouts was present in both the light phase (Fig. 2D) and dark phase (Fig. 2E). The frequency of sleep state transition correlated with the severity of EAE. When the number of wake/sleep bouts was plotted against EAE score at 14, 21, and 28 days, there was a linear correlation between EAE score and wake bouts (p < 0.01), as well as between EAE score or NREM bouts (p < 0.01). Increased severity of EAE (higher scores) coincided with an increased number of wake bouts and sleep bouts (Fig. 2F). 4. Discussion We found that sleep disturbance occurred after the onset of EAE and mainly presented with decreased sleep efficiency and increased sleep fragmentation. The extent of sleep fragmentation was positively correlated with the severity of the disease. This is the first study involving sleep and EAE of which we are aware. It addresses the biologically important question how a CNS autoimmune disorder induces dynamic changes of sleep quality. The findings differ from previous reports on sleep architecture in rodents
55
J. He et al. / Brain, Behavior, and Immunity 38 (2014) 53–58 50
80
#
EAE
*
**
**
2
CFA+PT
1 40
2
45
40 1 35
30
0
0 EAE score
EAE score
20 -1
7
14
21
25
28
-1
7
E
80 EAE
21
28
5 CFA+PT
CFA+PT
EAE
2 4
1
50
3 2.5
2
1
0
2.0
EAE clinical score
60
% REM in light phase
70
EAE score
% Wake in dark phase
14
Time after EAE induction (d)
Time after EAE induction (d)
B
EAE score
60
EAE score
% Wake in light phase
CFA+PT
EAE
D % NREM in dark phase
A
CFA+PT EAE
1.5 1.0 0.5 0.0
EAE score
-0.5 4
8
12
16
20
24
28
32
Time after EAE induction (d)
40
-1
7
14
21
0
28
-1
7
14
21
28
Time after EAE induction (d)
C
60
F
CFA+PT
EAE
6 EAE
CFA+PT
1
30
*
#
** ** #
4
2.5 2.0
EAE clinical score
40
% REM in dark phase
50
EAE score
% NREM in light phase
2
2
0
CFA+PT EAE
1.5 1.0 0.5 0.0
EAE score
-0.5 4
8
12
16
20
24
28
32
Time after EAE induction (d)
20 -1
7
14
21
28
Time after EAE induction (d)
0 -1
7
14
21
28
Time after EAE induction (d)
Fig. 1. Sleep efficiency and sleep state distribution during the course of EAE, and their correlation with EAE scores. EAE mice were compared across the duration of EAE with a control group receiving CFA + PT; the relationship of sleep stages (solid line, left y-axis) and EAE scores (dashed line, right y-axis) is shown by superimposed plots. (A) Percentage of wake during light phase was increased at 2–4 weeks after EAE induction than time 0 (⁄p < 0.05, ⁄⁄p < 0.01), and higher than the CFA + PT group on week 3 (#p < 0.05). (B) Percentage of wake during dark phase also tended to show an overall increase though the post hoc test did not show a statistical difference from 0. (C) Reduction of %NREM in the light phase was seen in EAE mice at 2–4 weeks after EAE induction in comparison with its baseline (⁄p < 0.05, ⁄⁄p < 0.01). The CFA + PT control group did not show significant change. The difference between the EAE and CFA + PT groups was significant at week 2 and 3 (#p < 0.05). (D) The overall decrease of %NREM in the dark phase in the EAE mice was paralleled by the CFA + PT mice. (E) There was no statistically significant reduction of %REM in the light phase. (F) The %REM was also unchanged in the dark phase across the time course of EAE. Inset shows that EAE mice started to show symptoms on day 11, were significantly different from the control group by day 13, achieved the first brief peak on day 14–15, and sustained symptoms after day 25.
treated with inflammatory cytokines or adjuvant, thereby providing novel insight into regulatory changes specific to EAE. In this study, EAE mice were compared longitudinally with their own baseline, and in parallel with controls receiving CFA and PT that induce inflammation despite the lack of autoimmune disease. CFA has been shown to induce either a reduction or increase of sleep. In rats receiving Freund’s adjuvant in the temporomandibular joint, there is a reduction in sleep efficiency during the subsequent two light periods, delayed sleep onset, and an increased number of awakenings. These changes could be partially reversed
by indomethacin treatment, suggesting that inflammation and inflammatory pain are a major cause of sleep disturbance (Schutz et al., 2003). In a Russian abstract, CFA increases slow-wave and REM sleep in rats. CFA alone leads to a gradual onset of sleepiness that is worse at 3–5 weeks, whereas a combination of CFA and bovine serum albumin precipitates the onset at about 1 week, with symptoms lasting 5 weeks in association with reduced serotonin in the striatum and frontal cortex (Pankova et al., 2001). Results from these two reports indicate a potential biphasic change, with sleep loss during the early stage and hypersomnolence in the later
56
J. He et al. / Brain, Behavior, and Immunity 38 (2014) 53–58
D
CFA+PT
EAE
80
EAE CFA+PT
NREM bouts in light phase
80
2 60
1
20
2
60
1
40
0
0
EAE score
EAE score
0 -1
7
14
21
20
28
-1
7
Time after EAE induction (d)
B
E
CFA+PT
21
28
80 EAE
60
40
1
20 0
CFA+PT
2
EAE score
60
1
EAE score
NREM bouts in dark phase
2
EAE score
Wake bouts in dark phase
14
Time after EAE induction (d)
80 EAE
EAE score
40
EAE score
Wake bouts in light phase
A
40
0
EAE score
0 -1
7
14
21
20
28
-1
Time after EAE induction (d)
C
7
14
21
28
Time after EAE induction (d)
baseline REM NREM WAKE
0
6
12
18
24
F
EAE day14
300
REM
NREM bouts
Number of bouts in 24 h
Wake bouts
NREM WAKE
0
6
12
18
24
EAE day21 REM NREM
200
100
WAKE
0
6
12
18
24
0
EAE day 28
0
REM
1
2 EAE score
3
4
NREM WAKE
0
6
12
18
24
Zeitgeber Time (h)
Fig. 2. Increased sleep fragmentation (solid lines, left y-axis) and correlation with EAE scores (dash line, right y-axis). (A) Wake bouts in the light phase showed a significant overall effect of EAE and days after induction, and a lack of interaction of the two factors by two-way ANOVA. (B) Wake bouts in the dark phase showed a significant overall effect of EAE but not days after induction. (C) In hypnograms of a representative EAE mouse during the course of disease, the increased sleep state transition was most apparent in the dark phase on week 2. (D) NREM bouts in the light phase showed a significant overall effect of EAE. (E) NREM bouts in the dark phase also showed a significant overall effect of EAE but not days after induction. (F) EAE scores correlated with the number of wake bouts (r = 0.63) and NREM bouts (r = 0.62).
course of the inflammation. In our study, CFA + PT did not show a significant effect on sleep architecture during weekly recordings for a month after injection. This may be related to dose, species,
and sex differences, though it is possible that an early change was missed since we did not analyze sleep features immediately after injection. We were not able to monitor within the first week
J. He et al. / Brain, Behavior, and Immunity 38 (2014) 53–58
of induction, as mice were required to remain in an Animal Biosafety Level 2 facility. Sickness behavior as a result of inflammation has been associated with blunted circadian rhythm, increased slow wave sleep, and reduced REM sleep (Dantzer et al., 2008). EAE animals show sickness behavior (Pollak et al., 2003; Campbell et al., 2010). Thus, we set out to test the hypothesis that EAE mice have hypersomnolence. However, contrary to the prediction, the result showed increased waking as well as more fragmented sleep. Based on the literature, there are two main classes of candidate molecules that might contribute to the changes of sleep architecture in EAE mice: proinflammatory cytokines and other inflammatory mediators that may interact and cross the blood–brain barrier (Pan et al., 2011) to exert CNS effects on sleep; and neurotransmitters within the CNS. In general, proinflammatory cytokines tend to be somnogenic whereas anti-inflammatory cytokines inhibit sleep (Zielinski and Krueger, 2011). EAE is associated with production of many cytokines, chemokines, as well as other soluble molecules (Raine, 1994; Eng et al., 1996; El-behi et al., 2010). Higher NREM and suppressed REM activity are induced by the T-cell derived cytokine CD40 ligand and by peripheral injection of tumor necrosis factor a (TNF) (Gast et al., 2013). While these substances promote hypersomnolence, the reduced sleep in the EAE mice suggests that the counterregulatory anti-inflammatory cytokines might be responsible. Nonetheless, sleep fragmentation was a main feature observed in the EAE mice in our study, an effect also exerted by proinflammatory cytokines. We have shown that FVB female mice on day 28 of EAE have higher levels of adenosine in the hippocampus (Mishra et al., 2013). These mice also have higher levels of the inhibitory neurotransmitter GABA and lower levels of the excitatory glutamate in the hippocampus (Pan et al., unpublished observations). Adenosine and GABA are known to promote sleep, whereas acetylcholine and biogenic amines are wake-promoting, although their sleep regulatory effects are dependent on specific CNS regions. EAE mice have dynamic changes of amino acids and biogenic amines, with regional differences and distinct patterns during the course of disease progression. Though several amino acids show an increase at EAE onset and peak but return to baseline during the chronic phase, there is a reduction of glutamate, GABA, and norepinephrine, as well as decreased 5-HT turnover in the late stage (Musgrave et al., 2011a, 2011b). There is defective GABA transmission in EAE (Rossi et al., 2011), whereas GABAergic agents can decrease inflammation and inhibit EAE (Bhat et al., 2010). In parallel, slice electrophysiology shows that EAE is associated with enhanced glutamatergic excitatory postsynaptic currents in the early phase, related to inflammatory cytokines released from infiltrating T cells and from activated microglia (Centonze et al., 2010). The serotonin system is also involved; knockout mice without the 5-HT transporter have less severe EAE, supporting a disease-ameliorating role of 5-HT activity (Hofstetter et al., 2005). Thus, we speculate that decreased GABA transmission and increased serotonergic activity at the peak and chronic stage could result in increased waking, more sleep fragmentation, and attenuated sleep cohesiveness. To fully establish a causal relationship, detailed analyses by use of stereotaxic injection of agonists and antagonists of the candidate molecule or its receptors are needed. The greater changes of sleep architecture and sleep state transition in the light phase are consistent with a blunted circadian rhythm. In C57BL/6 mice at day 2–8 after the onset of EAE symptoms, there is lower amplitude of change of per2 and clock gene expression in the liver and higher serum levels of corticosterone and leptin (Buenafe, 2012). This provides a basis for further characterization of altered circadian rhythms during the course of EAE. Poor sleep quality is well-known to impair immune function in human and animal studies. The deterioration of sleep architecture
57
shown in this study is a consequence of EAE, but would also contribute to neuroimmune dysregulation. In this sense, normalization of sleep by increasing efficiency and reducing fragmentation may facilitate the recovery of mice from EAE. Thus, this first characterization of sleep features in EAE mice opens new directions for the manipulation of sleep to modulate neurological outcome. Acknowledgments AJK receives grant support from NIH DK54880 and DK92245, and WP is supported by NS62291 and the AASM-accredited PBRC Sleep Health Center. We thank Hung Hsuchou and Aurelien Mace for technical assistance and discussions. References Bhat, R., Axtell, R., Mitra, A., Miranda, M., Lock, C., Tsien, R.W., Steinman, L., 2010. Inhibitory role for GABA in autoimmune inflammation. Proc. Natl. Acad. Sci. U.S.A. 107, 2580–2585. Buenafe, A.C., 2012. Diurnal rhythms are altered in a mouse model of multiple sclerosis. J. Neuroimmunol. 243, 12–17. Campbell, S.J., Meier, U., Mardiguian, S., Jiang, Y., Littleton, E.T., Bristow, A., Relton, J., Connor, T.J., Anthony, D.C., 2010. Sickness behaviour is induced by a peripheral CXC-chemokine also expressed in multiple sclerosis and EAE. Brain Behav. Immun. 24, 738–746. Centonze, D., Muzio, L., Rossi, S., Furlan, R., Bernardi, G., Martino, G., 2010. The link between inflammation, synaptic transmission and neurodegeneration in multiple sclerosis. Cell Death Differ. 17, 1083–1091. Dantzer, R., O’Connor, J.C., Freund, G.G., Johnson, R.W., Kelley, K.W., 2008. From inflammation to sickness and depression: when the immune system subjugates the brain. Nat. Rev. Neurosci. 9, 46–56. El-behi, M., Rostami, A., Ciric, B., 2010. Current views on the roles of Th1 and Th17 cells in experimental autoimmune encephalomyelitis. J. Neuroimmune. Pharmacol. 5, 189–197. Eng, L.F., Ghirnikar, R.S., Lee, Y.L., 1996. Inflammation in EAE: role of chemokine/ cytokine expression by resident and infiltrating cells. Neurochem. Res. 21, 511– 525. Gast, H., Muller, A., Lopez, M., Meier, D., Huber, R., Dechent, F., Prinz, M., Emmenegger, Y., Franken, P., Birchler, T., Fontana, A., 2013. CD40 activation induces NREM sleep and modulates genes associated with sleep homeostasis. Brain Behav. Immun. 27, 133–144. Hofstetter, H.H., Mossner, R., Lesch, K.P., Linker, R.A., Toyka, K.V., Gold, R., 2005. Absence of reuptake of serotonin influences susceptibility to clinical autoimmune disease and neuroantigen-specific interferon-gamma production in mouse EAE. Clin. Exp. Immunol. 142, 39–44. Kaushal, N., Ramesh, V., Gozal, D., 2012. TNF-alpha and temporal changes in sleep architecture in mice exposed to sleep fragmentation. PLoS One 7, e45610. Krueger, J.M., Rector, D.M., Churchill, L., 2007. Sleep and cytokines. Sleep Med. Clin. 2, 161–169. Li, Q., Powell, N., Zhang, H., Belevych, N., Ching, S., Chen, Q., Sheridan, J., Whitacre, C., Quan, N., 2011. Endothelial IL-1R1 is a critical mediator of EAE pathogenesis. Brain Behav. Immun. 25, 160–167. Louis, R.P., Lee, J., Stephenson, R., 2004. Design and validation of a computer-based sleep-scoring algorithm. J. Neurosci. Methods 133, 71–80. Lunde, H.M., Aae, T.F., Indrevag, W., Aarseth, J., Bjorvatn, B., Myhr, K.M., Bo, L., 2012. Poor sleep in patients with multiple sclerosis. PLoS One 7, e49996. Lunde, H.M., Bjorvatn, B., Myhr, K.M., Bo, L., 2013. Clinical assessment and management of sleep disorders in multiple sclerosis: a literature review. Acta Neurol. Scand. Suppl., 24–30. Mishra, P.K., Hsuchou, H., Ouyang, S., Kastin, A.J., Wu, X., Pan, W., 2013. Loss of astrocytic leptin signaling worsens experimental autoimmune encephalomyelitis. Brain Behav. Immun. 34, 98–107. Musgrave, T., Benson, C., Wong, G., Browne, I., Tenorio, G., Rauw, G., Baker, G.B., Kerr, B.J., 2011a. The MAO inhibitor phenelzine improves functional outcomes in mice with experimental autoimmune encephalomyelitis (EAE). Brain Behav. Immun. 25, 1677–1688. Musgrave, T., Tenorio, G., Rauw, G., Baker, G.B., Kerr, B.J., 2011b. Tissue concentration changes of amino acids and biogenic amines in the central nervous system of mice with experimental autoimmune encephalomyelitis (EAE). Neurochem. Int. 59, 28–38. Neau, J.P., Paquereau, J., Auche, V., Mathis, S., Godeneche, G., Ciron, J., Moinot, N., Bouche, G., 2012. Sleep disorders and multiple sclerosis: a clinical and polysomnography study. Eur. Neurol. 68, 8–15. Pan, W., Banks, W.A., Kennedy, M.K., Gutierrez, E.G., Kastin, A.J., 1996. Differential permeability of the BBB in acute EAE: enhanced transport of TNF-a. Am. J. Physiol. 271, E636–E642. Pan, W., Stone, K.P., Hsuchou, H., Manda, V.K., Zhang, Y., Kastin, A.J., 2011. Cytokine signaling modulates blood–brain barrier function. Curr. Pharm. Des. 17, 3729– 3740. Pankova, N.B., Popkova, E.V., Vetrile, L.A., Basharova, L.A., Krupina, N.A., 2001. [Alterations in diurnal sleep structure, electrical activity, and neurochemical
58
J. He et al. / Brain, Behavior, and Immunity 38 (2014) 53–58
parameters of the rat brain induced by systemic injection of complete Freund adjuvant and immunization by bovine serum albumin with complete Freund adjuvant. Zh. Vyssh. Nerv. Deiat. Im. I P Pavlova 51, 339–347. Pokryszko-Dragan, A., Bilinska, M., Gruszka, E., Biel, L., Kaminska, K., Konieczna, K., 2013. Sleep disturbances in patients with multiple sclerosis. Neurol. Sci. 34, 1291–1296. Pollak, Y., Ovadia, H., Orion, E., Weidenfeld, J., Yirmiya, R., 2003. The EAE-associated behavioral syndrome: I. Temporal correlation with inflammatory mediators. J. Neuroimmunol. 137, 94–99. Raine, C.S., 1994. The Dale E. McFarlin memorial lecture: the immunology of the multiple sclerosis lesion. Ann. Neurol. 36 (Suppl.), S61–S72. Rossi, S., Muzio, L., De C, V., Grasselli, G., Musella, A., Musumeci, G., Mandolesi, G., De, C.R., Maida, S., Biffi, E., Pedrocchi, A., Menegon, A., Bernardi, G., Furlan, R., Martino, G., Centonze, D., 2011. Impaired striatal GABA transmission in experimental autoimmune encephalomyelitis. Brain Behav. Immun. 25, 947–956.
Schutz, T.C., Andersen, M.L., Tufik, S., 2003. Sleep alterations in an experimental orofacial pain model in rats. Brain Res. 993, 164–171. Veasey, S.C., Valladares, O., Fenik, P., Kapfhamer, D., Sanford, L., Benington, J., Bucan, M., 2000. An automated system for recording and analysis of sleep in mice. Sleep 23, 1025–1040. Wang, Y., He, J., Kastin, A.J., Hsuchou, H., Pan, W., 2013. Hypersomnolence and reduced activity in pan-leptin receptor knockout mice. J. Mol. Neurosci. 51, 1038–1045. Wu, X., Pan, W., He, Y., Hsuchou, H., Kastin, A.J., 2010. Cerebral interleukin-15 shows upregulation and beneficial effects in experimental autoimmune encephalomyelitis. J. Neuroimmunol. 223, 65–72. Wu, X., Mishra, P.K., Hsuchou, H., Kastin, A.J., Pan, W., 2013. Upregulation of astrocytic leptin receptor in mice with experimental autoimmune encephalomyelitis. J. Mol. Neurosci. 49, 446–456. Zielinski, M.R., Krueger, J.M., 2011. Sleep and innate immunity. Front Biosci. (Schol. Ed.) 3, 632–642.
