FATIGUE REVERSIBLY REDUCED CORTICAL AND HIPPOCAMPAL ...

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright

Author's personal copy

Neuroscience 161 (2009) 1104 –1113

FATIGUE REVERSIBLY REDUCED CORTICAL AND HIPPOCAMPAL DENDRITIC SPINES CONCURRENT WITH COMPROMISE OF MOTOR ENDURANCE AND SPATIAL MEMORY J.-R. CHEN,a T.-J. WANG,b H.-Y. HUANG,a L.-J. CHEN,c Y.-S. HUANG,a Y.-J. WANGd* AND G.-F. TSENGd*

learning and memory. In human, chronically fatigued patients could display persistent or relapsing unexplainable fatigue lasted for over 6 months (Fukuda et al., 1994). There have been studies exploring the effect of central fatigue on brain structure (e.g. Buchwald et al., 1992; Natelson et al., 1993) and function (e.g. MacHale et al., 2000; Schmaling et al., 2003) and dramatic structural changes such as a substantial and consistent reduction of the gray matter volume of brain have been reported (Lange et al., 1999; Okada et al., 2004; de Lange et al., 2004, 2005). This suggests that fatigue could alter neurons structurally and is in line with the understanding that dendrites, the receiving part of neurons, are dynamic structures that can change in response to environment or stimuli (McEwen, 1994; Tseng and Prince, 1996; Woolley et al., 1997; Wang et al., 2002; Chen et al., 2003, 2004; Deller et al., 2006). In the rat, extended wakefulness reduced cortical concentration of a novel dendritic protein dendrin by 42% (Neuner-Jehle et al., 1996). This also suggests that dendrites could be modulated following the prolonged wakefulness associated with the induction of central fatigue and prompted us to investigate whether and how fatigue alters the dendritic structures of central neurons. To explore this, we chose the recently described rat sleep disturbance–induced central fatigue model in which the extent of the fatigue could be evaluated with weight-loaded forced swimming test (Tanaka et al., 2003). The dendritic structures, including dendritic arbors and spines of corticospinal and hippocampal Cornu Ammonis (CA)1 and CA3 pyramidal neurons were investigated for they are known to be associated with motor coordination and learning and recall of spatial memory, respectively. Neurons were filled with intracellular dye in partially fixed brain slices under visual guidance of a fluorescence microscope. Corticospinal neurons were prelabeled in vivo with retrograde tracer so that they could be filled selectively in isolation. The injected fluorescence dye was converted immunohistochemically into nonfading material following resection of the injected slice into sections. Neurons were then reconstructed three-dimensionally with PC-based software on a microscope fitted with a motorized stage and integrated focusing control and analyzed accordingly for further analysis.

a Department of Veterinary Medicine, College of Veterinary Medicine, National Chung Hsing University, Taichung, Taiwan b Department of Nursing, National Taichung Nursing College, Taichung, Taiwan c

Institute of Anatomy and Cell Biology, College of Medicine, National Taiwan University, Taipei, Taiwan

d

Department of Anatomy, College of Medicine, Tzu Chi University, No. 701, Section 3, Jhongyang Road, Hualien, Taiwan

Abstract—Fatigue could be induced following forced exercise, sickness, heat stroke or sleep disturbance and impaired brain-related functions such as concentration, attention and memory. Here we investigated whether fatigue altered the dendrites of central neurons. Central fatigue was induced by housing rats in cage with 1.5-cm deep water for 1–5 days. Three days of sleep deprivation seriously compromised rats’ performance in weight-loaded forced swimming and spatial learning tests, and 5 days of treatment worsened it further. Combinations of intracellular dye injection and three-dimensional analysis revealed that dendritic spines on retrograde tracer-identified corticospinal neurons and Cornu Ammonis (CA)1 and CA3 pyramidal neurons were significantly reduced while the shape or length of the dendritic arbors was not altered. Three days of rest restored the spine loss and the degraded spatial learning and weight-loaded forced swimming performances to control levels. In conclusion, although we could not rule out additional non-hypothalamic–pituitary– adrenal stress, the apparent fatigue induced following a few days of sleep deprivation could change brain structurally and functionally and the effects were reversible with a few days of rest. © 2009 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: corticospinal, dendrite, fatigue, hippocampus, plasticity, PSD-95.

Modern life stresses many of us and results in fatigue. Fatigue could be induced by infectious, immunological, and neuroendocrine disorders, sleep disturbances and psychological conditions but the etiology remains unclear. Fatigue often results in symptoms related to CNS function (Afari and Buchwald, 2003) including reduced activities and muscle endurance and impaired coordination, concentration, attention,

EXPERIMENTAL PROCEDURES

*Corresponding author. Tel: ⫹886-3-8564641; fax: ⫹886-3-8564641. E-mail addresses: [email protected] (G.-F. Tseng), chris@ mail.tcu.edu.tw (Y.-J. Wang). Abbreviations: CA, cornu ammonis; CFS, chronic fatigue syndrome; DAB, 3,3=-diaminobenzidine tetrahydrochloride; LTP, long-term potentiation; LY, Lucifer Yellow; PB, phosphate buffer; PBS, phosphatebuffered saline; ROD, relative optical density; ROS, reactive oxygen species; TMS, transcranial magnetic stimulation.

Animals Thirty male CD®3(SD) IGS rats (BioLasco, Ilan, Taiwan), 250 –350 g body weight were tested. Twenty-five of them were divided into control (n⫽5), fatigue (n⫽15) and fatigue and rest groups (n⫽5). The fatigue group was further subdivided into three sets for different durations of fatigue induction, 1 day (n⫽3), 3 days (n⫽3)

0306-4522/09 $ - see front matter © 2009 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2009.04.022

1104


J.-R. Chen et al. / Neuroscience 161 (2009) 1104 –1113 and 5 days (n⫽9). Six of the 5-day fatigue animals were processed for intracellular dye injection. Three rats each of the 1, 3, and 5 day fatigue-induction group were processed for Western blot analysis of protein expression in their sensorimotor cortex. The fatigue and rest groups were first subjected to 5 days of fatigue treatment followed by 3 days of rest. Since the above animals were sacrificed at separate time points, an additional five rats were subjected to fatigue induction and subsequent rest so that behavioral tests could be collected serially before fatigue induction, 3 and 5 days after fatigue induction and then following 3 days of rest. We followed the method of Tanaka et al. (2003) with slight modification to induce fatigue in rats. To induce fatigue, rats were individually housed in cage filled with water (24⫾1 °C) to a height of 1.5–2 cm. Animals were housed in a temperature (24⫾1 °C), humidity (60%⫾5%), and light (light on at 06:00 h and off at 18:00 h)-controlled environment. Food and water were available ad libitum. Animal care and experiments were approved and conformed to guidelines of the Animal Care and Use Committee of National Chung Hsing University and all efforts were taken to minimize the number of animals used and the suffering during experiments.

Retrograde labeling of corticospinal neurons Animals were anesthetized with 7% chloral hydrate (0.45 ml/kg) injected i.p. and mounted on a stereotaxic device. A midline incision was made over the spinous processes of the C3– 4 vertebrae. The lamina of the C4 vertebra was then removed and the spinal dura mater incised. A total of 1.5 ␮l of the retrograde tracer fluorogold (FG, 2% in distilled water; Fluorochrome, Denver, CO, USA) was injected into the dorsal column bilaterally with a 10-␮l Hamilton microsyringe (Reno, NV, USA) connected through a cannula to an infusion pump (KD Scientific, Holliston, MA, USA). Each injection was performed over a 10-min period and the pipette was left in place for 2 min following each injection and then slowly withdrawn over 4 min. Animals were allowed to survive for at least two weeks before subjecting them to fatigue induction.

Behavioral test We used weight-loaded forced swimming test (Tanaka et al., 2003) in which rats swam with a load of steel rings attached to their tails that weighed approximately 10% of their body weight to assess the extent of the fatigue and the Morris water maze test (Moser et al., 1995; Morris, 1984; Mizunoya et al., 2004; Gibertini et al., 1995) to evaluate the effect of fatigue on memory. For the weight-loaded forced swimming test, the swimming time from the beginning of swimming with weight until the point at which rats could not return to water surface 10 s after sinking was measured. The time, to a maximum of 10 min, was recorded in two trials, separated by at least 30 min, for each rat. The water maze was a circular pool (200 cm in diameter, 60 cm in height) located in a well lit room and filled with water (50 cm height, 23 °C) and contained a platform approximately 2–3 cm below water surface. Distal visual cues arrayed around the room were available for the rats to learn the location of the hidden platform. Rats received six training runs in the water maze before the fatigue experiment to confirm that all rats could find the platform within 30 s and the time and the swimming path for each rat to find the hidden platform were recorded. The performance of each rat following treatment was similarly recorded and normalized to that before fatigue induction and then analyzed. A 30-minute interval was allowed between weight-loaded forced swimming test and Morris water maze. For the 25 rats sacrificed at designated time points, weight-loaded endurance tests were conducted before the Morris water maze. For the five rats for serial behavioral tests, we ran the Morris water maze test before the weight-loaded endurance test at each time point to avoid the possibility that the weight-loaded endurance test

1105

might introduce additional strain to the animals to affect the outcome of spatial memory test.

Intracellular dye injection and subsequent immunoconversion of the injected dye To prepare tissue for intracellular dye injection, rats were deeply anesthetized with chloral hydrate again and perfused with 50 ml of lukewarm saline, followed by fixative containing 2% paraformaldehyde in 0.1 M phosphate buffer (PB), pH 7.3, at room temperature for 30 min. The brain was then carefully removed and the area of interest sectioned with a vibratome into 300-␮m-thick coronal slices immediately. In the present study, we used Lucifer Yellow (LY, Sigma, St. Louis, MO, USA) as the intracellular dye to reveal neuronal dendritic arbors. Slices were pre-treated with 0.1 M PB containing 10⫺7 M 4=,6-diamidino-2-phenyl-indole (DAPI; Sigma) for 30 min to make cell nuclei fluoresce blue under the same filter set that visualized LY as yellow. This enabled us to select individual cells of specific cortical layers for dye injection. To inject neurons, a slice was placed in a dish covered with a thin layer of 0.1 M PB on the stage of a fixed-stage, epifluorescence microscope (BX51, Olympus, Tokyo, Japan). An intracellular micropipette filled with 4% LY in water mounted on a three-axial hydraulic micromanipulator (Narishige, Tokyo, Japan) and a longworking distance objective lens (20⫻) was used to facilitate the selection of corticospinal and CA1 and CA3 hippocampal neurons for dye injection. Negative current generated by an intracellular amplifier (Axoclamp-IIB, Axon, Foster City, CA, USA) was used to inject the LY till all terminal dendrites fluoresced brightly under the scope. Several well-separated neurons could be injected in the area of interest in each slice. Following injection, the slice was removed, rinsed in 0.1 M PB and postfixed in 4% paraformaldehyde in 0.1 M PB for 3 days. They were then rinsed thoroughly in 0.1 M PB, cryoprotected, and carefully sectioned into 60-␮m-thick serial sections with a Leica cryostat (Chen et al., 2003, 2004; Wang et al., 2009). To convert the intracellular dye LY into nonfading material, sections were first preincubated in 1% H2O2 in PB for 30 – 60 min to remove endogenous peroxidase activity. They were then rinsed in phosphate-buffered saline (PBS) three times and incubated for an hour in PBS containing 2% bovine serum albumin and 1% Triton X-100. Sections were then treated in solution containing biotinylated rabbit anti-LY (1:200; Molecular Probes, Eugene, OR, USA) in PBS for 18 h at 4 °C. Following rinses in PBS, sections were incubated with standard avidin– biotin HRP reagent (Vector, Burlingame, CA, USA) for 3 hours at room temperature. They were then reacted at room temperature with a solution containing 0.05% 3,3=-diaminobenzidine tetrahydrochloride (DAB, Sigma) and 0.01% H2O2 in 0.05 M Tris buffer. Reacted sections were mounted onto slides, processed and coverslipped with Permount (Fisher, Fair Lawn, NJ, USA).

Western blotting of PSD-95 To determine whether changes of the densities of dendritic spines represent alterations of functional excitatory connections, we measured and compared the amount of PSD-95, a glutamatergic postsynaptic marker (Furuyashiki et al., 1999) in the primary sensorimotor cortex of control and fatigue rats following protocol described earlier (Chen et al., 2009). Briefly, the cortex of interest was rapidly dissected and homogenized immediately at 4 °C in 25 mM Hepes (pH 7.4) containing 0.3 M sucrose, protease inhibitor cocktail, 50 mM sodium vanadate (to inhibit phosphatases) and 0.1 M EDTA. Following centrifugation at 600⫻g for 15 min to remove unbroken cells and nuclei, the postnuclear supernatant was centrifuged at 21,000⫻g to obtain the postmitochondrial supernatant fraction. Protein concentrations in this fraction were determined with Bio-Rad reagents (Hercules, CA, USA). Twenty micrograms of protein was separated on a 15% acrylamide gel


1106

J.-R. Chen et al. / Neuroscience 161 (2009) 1104 –1113

containing sodium dodecyl sulfate, transferred onto a PVDF membrane, and subjected to Western analysis for PSD-95. Mouse anti-PSD-95 (Chemicon, Temecula, CA, USA) diluted (1:1000) in 10 mM Tris, pH 7.4, with 150 mM NaCl and 5% skim milk was used for immunoblotting at 4 °C overnight (Chen et al., 2009). Monoclonal antibody to GAPDH (Chemicon) was used to determine the amount of the internal standard protein. Relative optical densities (ROD) of bands corresponding to the protein of interest were quantified with a densitometer (Image ProPlus, Media Cybernetics, Silver Spring, MD, USA) and presented relative to the ROD of GAPDH for that lane.

Blood cortisol measurement Blood samples (1.5 ml) were collected via tail nick in the afternoon (14:00 h), centrifuged (3000⫻g, 15 min, 4 °C) and the plasma was stored at ⫺20 °C until sent for measurement. Level of cortisol was assayed with an automated chemiluminescence system on an ACS: 180 analyzer (Chiron Diagnostics, East Walpole, MA, USA) with Chiron Diagnostics kit (cat. No. 672304) commissioned by a clinical laboratory (Fushing Reference Laboratory, Taichung, Taiwan).

Data analysis The revealed corticospinal neurons were reconstructed threedimensionally with Neurolucida (MicroBrightField, Williston, VT, USA). The dendritic lengths measured from these tracings were analyzed accordingly. To find out whether fatigue affects the density of dendritic spines, we counted the number of dendritic spines on representative proximal and distal segments of the apical and basal dendrites of each category of neurons studied with a 100⫻ oil-immersion objective lens. For layer V corticospinal neurons of the sensorimotor cortex, analysis of the dendritic segments and the sampling method followed that described before for rat layer V sensorimotor cortical pyramidal neurons (Chen et al., 2003, 2004). For hippocampal CA1 and CA3 pyramidal neurons, basal dendrites were defined as those in the stratum oriens while apical dendrites were on the other side of the cell body layer with a proximal segment in the stratum radiatum and distal segment in the stratum lacunosum-moleculare. Selective segments of the dendrites were reconstructed with a 100⫻ objective and spine densities calculated. Data were expressed as mean⫾SE. Statistical significance between groups was determined using two-tailed Student’s t-test.

RESULTS Increasing the duration of sleep deprivation reduced the swimming time of rats progressively when assessed with weight-loaded forced swimming test, two successive trials separated by at least 30 min (Fig. 1A). Swimming duration was reduced to less than half of that of control following 3 days of sleep disturbance and to approximately a quarter of that of control after 5 days of continued sleep disturbance. The test did not exert additional influence on the animals’ performance as the outcome of the second trial was no different from that of the first trial. At the same time the performance of these animals in the Morris water maze also deteriorated (Fig. 1B). Times to locate the hidden platform increased by almost fourfold after 3 days of fatigue induction and close to sixfold after 5 days of similar treatment while the ability of the animals to swim, the swimming speed measured appeared not to be altered. Three days of rest restored the weight-loaded forced swimming and Morris water maze performances of the 5 day-

fatigue rats to control levels simultaneously (n⫽3) (Fig. 1). Similar results (data not shown) were obtained in those animals with Morris water maze conducted after the weight-loaded forced swimming test. With fixed tissue intracellular dye injection we were able to reveal the dendritic arbors and dendritic spines of the studied neurons for instance the relatively large corticospinal neurons in great detail. Fig. 2 illustrates the overall shape of a representative retrograde tracer-filled corticospinal neuron from normal control (Fig. 2A) and 5 dayfatigue animal (Fig. 2B). The photograph was taken from one of the 60-␮m-thick sections of a 300 ␮m-thick injected slice demonstrating that the majority of the processes of the relatively large pyramidal neuron could be contained within a section if the slice was cut with the brain properly oriented. This enabled us to convincingly reconstruct the dendritic arbor of each neuron from the serial sections of the 300 ␮m-thick injected slice (please see below). Higher magnification of segments of the proximal and distal basal and apical dendrites of the neurons (Fig. 2A1-4 and 2B1-4) shows that dendritic spines were well-filled and could be easily distinguished for subsequent analysis. The dendritic arbors of such injected neurons were reconstructed in three dimensional planes (Fig. 3A) and there was no difference in the overall shape of the neurons of the 5-day fatigue and control rats. The basal, apical and total dendritic lengths of thus analyzed corticospinal neurons of the 5-day fatigue rats (n⫽9) were not different from those of the control corticospinal neurons (n⫽14) (Fig. 3B) neither was there any difference in the dendritic branching profuseness as analyzed with Sholl’s method (not shown). On the effect of fatigue on dendritic spines, Fig. 4A shows that representative distal basal and apical dendritic segments of the corticospinal neurons of 5-day-fatigue rats were covered with fewer spines than control counterparts. The densities of dendritic spines on both the proximal and distal part of the basal and apical dendrites were significantly reduced in rats following 5 days of fatigue by approximately 23%–32% and 3 days of rest restored the dendritic spine densities on the corticospinal neurons of the 5-day-fatigue rats to control levels (Fig. 4A, B). To find out whether fatigue altered the output neurons of CA1 and CA3 hippocampal areas that are known to associate with learning and memory we also filled CA1 and CA3 pyramidal neurons with the same intracellular dye and investigated their dendritic arbors and densities of dendritic spines. Results show that dendritic spines on the proximal and distal part of the apical dendrites of CA1 pyramidal neurons were significantly reduced by 31% following 5 days of fatigue while those on basal dendrites were reduced dramatically by 52% (Fig. 5). On CA3 pyramidal neurons (Fig. 6), spine densities on proximal apical dendrites were significantly reduced by 20% and on distal apical dendrites by 35% while those on basal dendrites were reduced by 32%. Three days of rest again restored the densities of dendritic spines on all segments of the CA1 and CA3 pyramidal neurons to control levels (Figs. 5 and



1107

Fig. 1. The effect of sleep deprivation–induced fatigue on rats measured with weight-loaded forced swimming (A) and Morris water maze (B) tests. Results illustrated were from five rats with Morris water maze test conducted before the weight-loaded forced swimming test at each time point to avoid acute motor strain. Rats were tested before fatigue induction (control), 3 (F3d) and 5 days (F5d) after fatigue treatment and 5-day fatigue treatment followed by 3 days of rest (F⫹R). Swimming times and swimming path lengths in A and B after fatigue induction were normalized to the measurements obtained before treatment (control). Short bar in the histogram represents SE of the mean of the group. * P⬍0.05 between the marked and each corresponding control; ⫹ P⬍0.05 between the marked and 3 day-fatigue group; # between the marked and 5 day-fatigue group (unpaired two-tailed Student’s t-test).

6). Fatigue however, did not affect the gross appearance or length of the dendritic arbors of these two categories of neurons (not shown). To find out whether loss of dendritic spines represents functional downregulation, we measured the amount of PSD-95, a protein known to associate with glutamatergic excitatory postsynaptic densities, as indication of the efficacy of excitatory synaptic connection. Sensorimotor cortex where corticospinal neurons concentrated was chosen for this analysis because of its relative abundance quantitywise and easiness to procure as compared to CA1 and CA3 areas of the hippocampus. The amount of PSD-95 in

this cortical area was found to have reduced by 49% following 5 days of fatigue as compared to that of the control rats, respectively (Fig. 4C). To find out whether fatigue changes the level of blood corticosteroid, which may also affect dendritic morphology, we examined the cortisol concentration in the blood of control and fatigue rats sampled right before being sacrificed for morphological observation. Blood cortisol concentration in control rats, was 1.13⫾0.25 ␮g/ml (n⫽3) while it was 1.32⫾0.3 ␮g/ml (n⫽3), 1.07⫾0.33 ␮g/ml (n⫽3) and 1.39⫾0.39 ␮g/ml (n⫽3) for the 1, 3, and 5-day fatigue rats, respectively.


1108


Fig. 2. Representative intracellular dye-filled corticospinal neurons from control (A) and 5 day-fatigue rats (B). The photograph was taken from one of the serial sections, 60-␮m-thick, of the injected slice in which the injected fluorescence dye was immunohistochemically converted to nonfading DAB reaction product. (A1–A4, B1–B4) High magnifications of the corresponding dendritic segments marked in A and B to show that dendritic spines are well-labeled. Scale bar⫽100 ␮m for A and B and 10 ␮m for A1–A4 and B1–B4.

DISCUSSION In the literature, studies of fatigue focused mainly on changes of blood amino acids, 5-HT and glucose during fatigue-induced behavioral impairments (Chao et al., 1992; Glaser and Kiecolt-Glaser, 1998; Katafuchi et al., 2003; McDermott et al., 2003, 2006; Tanaka et al., 2003; Mizuno et al., 2007). Here we documented for the first time that fatigue readily affects the structure of central neurons. A few days of sleep deprivation in our model reduced the dendritic spines on corticospinal and hippocampal CA1 and CA3 pyramidal neurons and these changes were reversible following 3 days of rest. The concomitant reduction of the glutamatergic excitatory postsynaptic marker protein PSD-95 is consistent with the association of dendritic spines with glutamatergic synapses and argues that fatigue downregulated cortical as well as hippocampal pyramidal neuronal excitatory connections. Loss of cortical neuronal dendritic spines following sleep deprivation–induced fatigue is also in line with the report that prolonged wakefulness decreases dendrin, a protein associated with dendritic spines (Neuner-Jehle et al., 1996). In central neurons, dendritic spines are dynamic even at rest (Brown

et al., 2007) and have been shown to be swiftly and dynamically modulated by many factors such as changes in environment (Horner, 1993), normal cyclic changes of gonadal hormones (Chen et al., 2009), cortical compression (Chen et al., 2003, 2004), insults such as axonal injury (Tseng and Hu, 1996; Tseng and Prince, 1996), deafferentation (Tailby et al., 2005; Deller et al., 2006), and ischemic injury (Brown et al., 2007), drug or hormonal treatments (Gould et al., 1990; Woolley and McEwen, 1992, 1994; McEwen, 1994; Kolb et al., 1997; Woolley et al., 1997; Woolley and Schwartzkroin, 1998; Cooke and Woolley, 2005; Chen et al., 2009) and aging (Wang et al., 2009). Stress has been reported to induce atrophy of the apical dendrites of hippocampal CA3 pyramidal neurons (Watanabe et al., 1992). Indeed, chronic administration of corticosterone, which could have increased during stress, also resulted in dendritic atrophy of hippocampal pyramidal neurons (Woolley et al., 1990). In light of this, we also measured the blood cortisol concentration of our control and 5 day-fatigue rats and found that blood cortisol did not appear to be affected following 1, 3 or 5 days of fatigue induction, the longest that we used. This is in line with earlier studies that fatigue does not affect the plasma cortisol level of rats (Tobler et al., 1983; Neuner-Jehle et al., 1996) and argued against the possibility that the observed dendritic changes were caused by alterations of adrenocortical function secondary to fatigue. Reduction of corticospinal neuronal excitatory synapses following fatigue suggests that fatigue could compromise motor execution. Clinical studies of chronic fatigue syndrome (CFS) show that CFS subjects have decreased amplitude of motor-evoked potentials compared with normal subjects from transcranial magnetic stimulation (TMS) of the motor cortex during exercise, and that the diminished muscle force could be significantly restored in CFS subjects by a TMS-induced muscle twitch (Sacco et al., 1999). In addition, CFS subjects fatigue faster than normal subjects during sustained contraction of the wrist, and they have less of an increase in motor cortical excitability during exercise as compared with normal subjects, and a larger depression of motor cortical excitability when fatigued (post-exercise depression) (Samii et al., 1996a,b). Starr et al. (2000) also demonstrated that the threshold of TMS to evoke motor end plate potentials from the first dorsal interosseous muscle was significantly higher in CFS than in control subjects. These studies demonstrated that motor fatigue in CFS was not due to changes of muscle, nerve or lower motoneuron functions, but rather, reflected an altered excitability of the motor cortex, consistent with our finding that fatigued rats swam at speed comparable to control animals in the Morris water maze test when spatial memory was apparently compromised. Our finding that corticospinal neurons lost some of their dendritic spines in the fatigue status could serve as a structural substrate underlying the decline of excitability of the motor cortex. In the present experiments, we found that 5 days of sleep disturbance induced fatigue in association with impairment of spatial learning and dendritic spine loss of CA1 and CA3 pyramidal neurons. This is consistent with the



1109

Fig. 3. Representative three-dimensionally reconstructed corticospinal neurons from control (A) and 5-day fatigue rat (A=). Apical, basal and total dendritic lengths of 14 such reconstructed control neurons and 9 5-day fatigue neurons were measured and analyzed as shown in B. Scale bar⫽150 ␮m.

consensus that hippocampus is critical to many cognitive/ memory processes and sensitive to sleep loss (Blissitt, 2001; Guan et al., 2004; Ruskin et al., 2004). Studies found that sleep deprivation induced by a variety of methods affects the long-term potentiation (LTP) of the synaptic transmission of hippocampus (Campbell et al., 2001; McDermott et al., 2003, 2006; Davis et al., 2003; RomcyPereiro and Pavlides, 2004). Since synaptic interaction in the hippocampus is crucial for the encoding and storing of memories and compromise of synaptic connection sleep deprivation–induced loss of dendritic spines is likely to be responsible for the observed cognitive impairment. The fact that NMDA receptor activity modulates dendritic spine formation and stability (Halpain et al., 1998; Nagerl et al., 2004; Alvarez et al., 2007) and that 3 days of sleep deprivation caused impairment of LTP on CA1 pyramidal neurons in association with decreased surface expression of NMDA receptor (McDermott et al.. 2003, 2006), suggests that short-term, 3–5-day sleep disturbance could downscale hippocampal function by decreasing NMDA receptor expression and consequently reduction of synaptic contacts and densities of dendritic spines. The proposition however remains to be tested. Alternatively, the oxidative stress produced by sustained neural activity during fatigue induction could have contributed to the fatigue-induced reduction of dendritic spines on corticospinal and CA1 and CA3 pyramidal neurons. Oxidative stress is the result of unregulated production of reactive oxygen species (ROS), such as hydrogen peroxide, nitric oxide, superoxide and the highly reactive hydroxyl radicals. Central neurons are specialized cells that require a great quantity of energy to maintain their function. High oxygen consumption, relatively low antioxidant levels and low regenerative capacity render brain

tissue more susceptible to oxidative damage. Experiments show that brain glucose uptake was significantly reduced in rats subjected to 5 days of fatigue (Tanaka et al., 2003) and the uptake was improved in association with the recovery of fatigue after 1 day of rest (Mizokawa et al., 2003). Since glucose is the source of ATP production of cells, decreased glucose uptake in cortical neurons following fatigue could have led to the accumulation of ROS, which could react with a multitude of molecules to initiate neuronal damages and in addition inhibition of Na⫹,K⫹-ATPases (Mark et al., 1995). ROS catabolism in neurons of the fatigued animals was expected to require more ATPs, and neurons should accurately share the energy for the maintaining of neural activity and ROS catabolism. In the process of fatigue induction, activated neurons might be caught in a vicious circle and led eventually to the reduction of neural activity which among others could have resulted in the withdrawal of synapses and consequently loss of spines that we observed. In light of this, the possibility of glial involvement such as astrocytes, which closely associate with synapses (Liu et al., 2006) and microglial reaction that involves in ROS production (Ling et al., 2001) need to be further explored. Loss of dendritic spines nevertheless, may protect neurons from excessive activityinduced cell damages or death. Methodology-wise, in the present study we employed intracellular dye injection in semi-fixed brain slices (Wang et al., 2002; Chen et al., 2003, 2004) to reveal the dendritic arbors of the studied pyramidal neurons. This allowed us to study specifically identified corticospinal neurons and CA1 or CA3 pyramidal neurons. Neurons, well-spaced apart, could be individually filled with no time constraint since the tissue has already been partially fixed. With proper orientation, we were able to preserve most, if not all, of the


1110


rolucida software to reconstruct the dendritic arbors of the studied neurons through the serial sections obtained from each injected slice in the three-dimensional plane. This improved the data significantly as compared to conventional two-dimensional camera lucida reconstruction (Chen et al., 2003; Wang et al., 2009) as Z-axis information was integrated. We were able to compared the three-dimensionally reconstructed dendritic arbors of the studied neurons and found that short-term fatigue of 5 days affects neither the dendritic length nor the dendritic branching profuseness of corticospinal neurons. Finally, on the issue of what stages of sleep was affected in our sleep deprivation model, although we did not record EEG, it appeared that REM with the loss of muscle tone was most likely to be strongly affected as the animals

Fig. 4. Changes of the dendritic spines of corticospinal neurons following fatigue. (A) Different representative dendritic segments of corticospinal neurons of the control, 5-day fatigue, and 5-day fatigue followed by 3-day rest rats. Changes of spine densities were analyzed in B. Number of proximal basal, distal basal, proximal apical and distal apical dendritic segments was 37, 48, 35 and 44 for control, 35, 27, 33 and 31 for 5 day-fatigue and 31, 38, 46, and 35 for the 5-day fatigue and 3-day rest group. (C) Representative immunoblot of PSD-95 in the sensorimotor cortex of control and 5-day fatigue rats. GAPDH is the internal standard. The results analyzed are shown in the right in which the amount of PSD-95 (OD reading) was normalized to that of the GAPDH of each lane. * P⬍0.05 between the marked and its corresponding control; # P⬍0.05 between the marked and its corresponding 5-day fatigue group, two-tailed Student’s t-tests. Scale bar⫽10 ␮m.

dendritic arbors for instance of the relatively large corticospinal neurons close to completeness in a 300 ␮m-thick brain slice. For analysis, we adopted the PC-based Neu-

Fig. 5. The effect of fatigue on the dendritic spines of hippocampal CA1 pyramidal neurons. (A) Representative dendritic segments of the CA1 pyramidal cells of control, 5-day fatigue and 5-day fatigue followed by 3-day rest rats. Changes of spine densities were analyzed in B. Number of basal, proximal apical and distal apical dendritic segments analyzed was 31, 24 and 33 for control, 27, 28 and 36 for 5 day-fatigue and 36, 47, and 50 for 5-day fatigue and 3-day rest group. * P⬍0.05 between the marked and its corresponding control, # P⬍0.05 between the marked and its corresponding fatigue group, two-tailed Student’s t-tests. Scale bar⫽10 ␮m.



1111

densities on cortical as well as hippocampal pyramidal neurons. Although we could not exclude other stressing complications in our sleep deprivation model, the concurrent nature suggests that dendritic alteration may underlie the behavioral deficits seen following sleep disturbance and could represent a target for fatigue prevention studies. Although alterations of both behaviors and structures following 5 days of fatigue were reversible following 3 days of rest suggesting that it could merely reflect an adapted response of the brain neurons to environmental challenges, it remains to be ascertained whether extended fatigue affects cortical neuronal structure in a rest-reversible manner. Acknowledgments—This work was supported by grants from the National Science Council of Taiwan to J.-R. Chen (NSC 97–2313B-005-045) and T.-J. Wang (NSC-95–2320-B-438-001) and from the Tzu-Chi University (TCIRP-95003) to T.-J. Wang and G.-F. Tseng.

REFERENCES

Fig. 6. The effect of fatigue on the dendritic spines of hippocampal CA3 pyramidal neurons. (A) Representative dendritic segments of the CA1 pyramidal cells of control, 5-day fatigue and 5-day fatigue followed by 3-day rest rats. Analyses of the changes of dendritic spine densities are shown in B. Number of basal, proximal apical and distal apical dendritic segments analyzed was 29, 31 and 37 for control, 25, 40 and 24 for 5 day-fatigue and 31, 29, and 34 for 5-day fatigue and 3-day rest group. * P⬍0.05 between the marked and its corresponding control, # P⬍0.05 between the marked and its corresponding fatigue group, two-tailed Student’s t-tests. Scale bar⫽10 ␮m.

had to maintain antigravity muscles to keep their bodies from the water underneath, consistent with earlier reports that a similar model affects paradoxical sleep (McDermott et al., 2003; Yang et al., 2008). Slow-wave sleep was probably disrupted as well for the animals could not sit or lie down to sleep in our model.

CONCLUSION Central fatigue could follow demanding physical conditions and result in impaired performances. In this study, we showed that 5 days of sleep disturbance induced fatigue and resulted in reduced motor and spatial memory performances of rats. At the same time, brain neurons responded to this challenge with reduction of dendritic spine

Afari N, Buchwald D (2003) Chronic fatigue syndrome: a review. Am J Psychiatry 160:221–236. Alvarez VA, Ridenour DA, Sabatini BL (2007) Distinct structural and ionotropic roles of NMDA receptors in controlling spine and synapse stability. J Neurosci 27:7365–7376. Blissitt PA (2001) Sleep, memory, and learning. J Neurosci Nurs 33:208 –215. Brown CE, Li P, Boyd JD, Delaney KR, Murphy TH (2007) Extensive turnover of dendritic spines and vascular remodeling in cortical tissues recovering from stroke. J Neurosci 27:4101– 4109. Buchwald D, Garrity D, Pascualy R, Kith P, Ashley RL, Wener MH, Kidd PG, Katon WJ, Russo JE (1992) Chronic fatigue syndrome. Toxicol Ind Health 8:157–173. Campbell IG, Guinan MJ, Horowitz JM (2001) Sleep deprivation impairs long-term potentiation in rat hippocampal slices. J Neurophysiol 88:1073–1076. Chao CC, DeLaHunt M, Hu S, Close K, Peterson PK (1992) Immunologically mediated fatigue: a murine model. Clin Immunol Immunopathol 64:161–165. Chen JR, Wang YJ, Tseng GF (2003) The effect of epidural compression on cerebral cortex: a rat model. J Neurotrauma 20:767–780. Chen JR, Wang YJ, Tseng GF (2004) The effects of decompression and exogenous NGF on compressed cerebral cortex. J Neurotrauma 21:1640 –1651. Chen JR, Yan YJ, Wang TJ, Chen LJ, Wang YJ, Tseng GF (2009) Gonadal hormones modulate the dendritic spine densities of primary cortical pyramidal neurons in adult female rat. Cereb Cortex doi:10.1093/cercor/bhp048. Cooke BM, Woolley CS (2005) Gonadal hormone modulation of dendrites in the mammalian CNS. J Neurobiol 64:34 – 46. Davis CJ, Harding JW, Wright JW (2003) REM sleep deprivationinduced deficits in the latency-to-peak induction and maintenance of long-term potentiation within the CA1 region of the hippocampus. Brain Res 973:293–297. de Lange FP, Kalkman JS, Bleijenberg G, Hagoort P, van der Meer JW, Toni I (2005) Gray matter volume reduction in the chronic fatigue syndrome. Neuroimage 26:777–781. de Lange FP, Kalkman JS, Bleijenberg G, Hagoort P, van der Werf SP, van der Meer JW, Toni I (2004) Neural correlates of the chronic fatigue syndrome—an fMRI study. Brain 127:1948 –1957. Deller T, Bas Orth C, Vlachos A, Merten T, Del Turco D, Dehn D, Mundel P, Frotscher M (2006) Plasticity of synaptopodin and the spine apparatus organelle in the rat fascia dentata following entorhinal cortex lesion. J Comp Neurol 499:471– 484. Fukuda K, Straus SE, Hickie I, Sharpe MC, Dobbins JG, Komaroff A (1994) The chronic fatigue syndrome: a comprehensive approach


1112


to its definition and study. International Chronic Fatigue Syndrome Study Group. Ann Intern Med 121:953–959. Furuyashiki T, Fujisawa K, Fujita A, Madaule P, Uchino S, Mishina M, Bito H, Narumiya S (1999) Citron, a Rho-target, interacts with PSD-95/SAP-90 at glutamatergic synapses in the thalamus. J Neurosci 19:109 –118. Gibertini M, Newton C, Friedman H, Klein TW (1995) Spatial learning impairment in mice infected with Legionella pneumophila or administered exogenous interleukin-1-beta. Brain Behav Immun 9: 113–128. Glaser R, Kiecolt-Glaser JK (1998) Stress-associated immune modulation: relevance to viral infections and chronic fatigue syndrome. Am J Med 105:S35–S42. Gould E, Woolley CS, Frankfurt M, McEwen BS (1990) Gonadal steroids regulate dendritic spine density in hippocampal pyramidal cells in adulthood. J Neurosci 10:1286 –1291. Guan Z, Peng X, Fang J (2004) Sleep deprivation impairs spatial memory and decreases extracellular signal-regulated kinase phosphorylation in the hippocampus. Brain Res 1018:38 – 47. Halpain S, Hipolito A, Saffer L (1998) Regulation of F-actin stability in dendritic spines by glutamate receptors and calcineurin. J Neurosci 18:9835–9844. Horner CH (1993) Plasticity of the dendritic spine. Prog Neurobiol 41:281–321. Katafuchi T, Kondo T, Yasaka T, Kubo K, Take S, Yoshimura M (2003) Prolonged effects of polyriboinosinic: polyribocytidylic acid on spontaneous running wheel activity and brain interferon-alpha mRNA in rats: a model for immunologically induced fatigue. Neuroscience 120:837– 845. Kolb B, Cote S, Ribeiro-da-Silva A, Cuello AC (1997) Nerve growth factor treatment prevents dendritic atrophy and promotes recovery of function after cortical injury. Neuroscience 76:1139 –1151. Lange G, DeLuca J, Maldjian JA, Lee H, Tiersky LA, Natelson BH (1999) Brain MRI abnormalities exist in a subset of patients with chronic fatigue syndrome. J Neurol Sci 171:3–7. Ling EA, Ng YK, Wu CH, Kaur C (2001) Microglia: its development and role as a neuropathology sensor. Prog Brain Res 132:61–79. Liu PH, Yang LH, Wang YJ, Tseng GF (2006) Proximity of lesioning determines response of facial motoneurons to peripheral axotomy. J Neurotrauma 23:1857–1873. MacHale SM, Lawrie SM, Cavanagh JT, Glabus MF, Murray CL, Goodwin GM, Ebmeier KP (2000) Cerebral perfusion in chronic fatigue syndrome and depression. Br J Psychiatry 176:550 –556. Mark RJ, Hensley K, Butterfield DA, Mattson MP (1995) Amyloid beta-peptide impairs ion-motive ATPase activities: evidence for a role in loss of neuronal Ca2⫹ homeostasis and cell death. J Neurosci 15:6239 – 6249. McDermott CM, La Hoste GJ, Chen C, Musto A, Bazan NG, Magee JC (2003) Sleep deprivation causes behavioral, synaptic, and membrane excitability alterations in hippocampal neurons. J Neurosci 23:9687–9695. McDermott CM, Hardy MN, Bazan NG, Magee JC (2006) Sleep deprivation-induced alterations in excitatory synaptic transmission in the CA1 region of the rat hippocampus. J Physiol 570:553–565. McEwen BS, Woolley CS (1994) Estradiol and progesterone regulate neuronal structure and synaptic connectivity in adult as well as developing brain. Exp Gerontol 29:431– 436. Mizokawa S, Tanaka M, Matsumura A, Nozaki S, Watanabe Y (2003) Recovery from fatigue: changes in local brain 2-[18F]fluoro-2-deoxy-D-glucose utilization measured by autoradiography and in brain monoamine levels of rat. Neurosci Lett 353:169 –172. Mizuno K, Tanaka M, Nozaki S, Yamaguti K, Mizuma H, Sasabe T, Sugino T, Shirai T, Kataoka Y, Kajimoto Y, Kuratsune H, Kajimoto O, Watanabe Y (2007) Mental fatigue-induced decrease in levels of several plasma amino acids. J Neural Transm 114:555–561. Mizunoya W, Oyaizu S, Hirayama A, Fushiki T (2004) Effects of physical fatigue in mice on learning performance in a water maze. Biosci Biotechnol Biochem 68:827– 834.

Morris R (1984) Developments of a water-maze procedure for studying spatial learning in the rat. J Neurosci Methods 11:47– 60. Moser MB, Moser EI, Forrest E, Andersen P, Morris RG (1995) Spatial learning with a minislab in the dorsal hippocampus. Proc Natl Acad Sci U S A 92:9697–9701. Nagerl UV, Eberhorn N, Cambridge SB, Bonhoeffer T (2004) Bidirectional activity-dependent morphological plasticity in hippocampal neurons. Neuron 44:759 –767. Natelson BH, Cohen JM, Brassloff I, Lee HJ (1993) A controlled study of brain magnetic resonance imaging in patients with the chronic fatigue syndrome. J Neurol Sci 120:213–217. Neuner-Jehle M, Denizot JP, Borbely AA, Mallet J (1996) Characterization and sleep deprivation-induced expression modulation of dendrin, a novel dendritic protein in rat brain neurons. J Neurosci Res 46:138 –151. Okada T, Tanaka M, Kuratsune H, Watanabe Y, Sadato N (2004) Mechanisms underlying fatigue: a voxel-based morphometric study of chronic fatigue syndrome. BMC Neurol 4:14. Romcy-Pereiro R, Pavlides C (2004) Distinct modulatory effects of sleep on the maintenance of hippocampal and medial prefrontal cortex LTP. Eur J Neurosci 20:3453–3462. Ruskin DN, Liu C, Dunn KE, Bazan NG, La Hoste GJ (2004) Sleep deprivation impairs hippocampus-mediated contextual learning but not amygdala-mediated cued learning in rats. Eur J Neurosci 19: 3121–3124. Sacco P, Hope PA, Thickbroom GW, Byrnes ML, Mastaglia FL (1999) Corticomotor excitability and perception of effort during sustained exercise in the chronic fatigue syndrome. Clin Neurophysiol 110: 1883–1891. Samii A, Wassermann EM, Ikoma K, Mercuri B, Hallett M (1996a) Characterization of postexercise facilitation and depression of motor evoked potentials to transcranial magnetic stimulation. Neurology 46:1376 –1382. Samii A, Wassermann EM, Ikoma K, Mercuri B, George MS, O’Fallon A, Dale JK, Straus SE, Hallett M (1996b) Decreased postexercise facilitation of motor evoked potentials in patients with chronic fatigue syndrome or depression. Neurology 47:1410 –1414. Schmaling KB, Lewis DH, Fiedelak JI, Mahurin R, Buchwald DS (2003) Single-photon emission computerized tomography and neurocognitive function in patients with chronic fatigue syndrome. Psychosom Med 65:129 –136. Starr A, Scalise A, Gordon R, Michalewski HJ, Caramia MD (2000) Motor cortex excitability in chronic fatigue syndrome. Clin Neurophysiol 111:2025–2031. Tailby C, Wright LL, Metha AB, Calford MB (2005) Activity-dependent maintenance and growth of dendrites in adult cortex. Proc Natl Acad Sci U S A 102:4631– 4636. Tanaka M, Nakamura F, Mizokawa S, Matsumura A, Nozaki S, Watanabe Y (2003) Establishment and assessment of a rat model of fatigue. Neurosci Lett 352:159 –162. Tobler I, Murison R, Ursin R, Ursin H, Borbely AA (1983) The effect of sleep deprivation and recovery sleep on plasma corticosterone in the rat. Neurosci Lett 35:297–300. Tseng GF, Hu ME (1996) Axotomy induces retraction of the dendritic arbor of adult rat rubrospinal neurons. Acta Anat (Basel) 155: 184 –193. Tseng GF, Prince DA (1996) Structural and functional alterations in rat corticospinal neurons after axotomy. J Neurophysiol 75:248 –267. Wang YJ, Chen JR, Tseng GF (2002) Fate of the soma and dendrites of cord-projection central neurons after proximal and distal spinal axotomy: an intracellular dye injection study. J Neurotrauma 19:1487–1502. Wang TJ, Chen JR, Wang YJ, Tseng GF (2009) The cytoarchitecture and soma-dendritic arbors of the pyramidal neurons of aged rat sensorimotor cortex: an intracellular dye injection study. Neuroscience 158:776 –785.


J.-R. Chen et al. / Neuroscience 161 (2009) 1104 –1113 Watanabe Y, Gould E, McEwen BS (1992) Stress induces atrophy of apical dendrites of hippocampal CA3 pyramidal neurons. Brain Res 588:341–345. Woolley CS, McEwen BS (1992) Estradiol mediates fluctuation in hippocampal synapse density during the estrous cycle in the adult rat. J Neurosci 12:2549 –2554. Woolley CS, McEwen BS (1994) Estradiol regulates hippocampal dendritic spine density via an N-methyl-D-aspartate receptor-dependent mechanism. J Neurosci 14:7680 –7687. Woolley CS, Schwartzkroin PA (1998) Hormonal effects on the brain. Epilepsia 39 (Suppl 8):S2–S8.

1113

Woolley CS, Gould E, McEwen BS (1990) Exposure to excess glucocorticoids alters dendritic morphology of adult hippocampal pyramidal neurons. Brain Res 531:225–231. Woolley CS, Weiland NG, McEwen BS, Schwartzkroin PA (1997) Estradiol increases the sensitivity of hippocampal CA1 pyramidal cells to NMDA receptor-mediated synaptic input: correlation with dendritic spine density. J Neurosci 17:1848 –1859. Yang RH, Hu SJ, Wang Y, Zhang WB, Luo WJ, Chen JY (2008) Paradoxical sleep deprivation impairs spatial learning and affects membrane excitability and mitochondrial protein in the hippocampus. Brain Res 1230:224 –232.

(Accepted 9 April 2009) (Available online 17 April 2009)