Differential effects of duration of sleep fragmentation on spatial learning and synaptic plasticity in pubertal mice.

brain research ] (]]]]) ]]]–]]]

Available online at www.sciencedirect.com

www.elsevier.com/locate/brainres

Research Report

Differential effects of duration of sleep fragmentation on spatial learning and synaptic plasticity in pubertal mice Eli Wallacea, Do Young Kimb, Kye-Min Kimb, Stephanie Chenb, B. Blair Bradenc,d, Jeremy Williamsa, Kalene Jassoa, Alex Garciac, Jong M. Rhoe, Heather Bimonte-Nelsonc,d, Rama Magantia,n a

University of Wisconsin School of Medicine and Public Health, Madison, WI, USA Barrow Neurological Institute/St Joseph’s Hospital and Medical Center, Phoenix, AZ, USA c Arizona State University, Tempe, AZ, USA d Arizona Alzheimer’s Consortium, Phoenix, AZ, USA e Departments of Pediatrics and Clinical Neurosciences, University of Calgary Faculty of Medicine, Calgary, Canada b

art i cle i nfo

ab st rac t

Article history:

Study objective: To examine the differential effects of acute and chronic sleep fragmenta-

Accepted 12 April 2015

tion (SF) on spatial learning and memory, and hippocampal long-term potentiation (LTP) in pubertal mice.

Keywords: Sleep fragmentation Spatial memory Morris water maze LTP Mice Adolescence

Methods: Two studies were performed during which adolescent C57/Bl6 mice were subjected to acute-SF 24 h a day 3 days or chronic-SF for 12 h a day 2 weeks using a programmable rotating lever that provides tactile stimulus with controls housed in similar cages. Spatial learning and memory was examined using the Morris water maze, and longterm potentiation (LTP) was evaluated after stimulation of Schaffer collaterals in CA1 hippocampus post SF. Actigraphy was used during the period of SF to monitor rest-activity patterns. Electroencephalographic (EEG) recordings were acquired for analysis of vigilance state patterns and delta-power. Serum corticosterone was measured to assess stress levels. Results: Acute-SF via tactile stimulation negatively impacted spatial learning, as well as LTP maintenance, compared to controls with no tactile stimulation. While actigraphy showed significantly increased motor activity during SF in both groups, EEG data indicated that overall sleep efficiency did not differ between baseline and SF days, but significant increases in number of wakeful bouts and decreases in average NREM and REM bout lengths were seen during lights-on. Acute sleep fragmentation did not impact corticosterone levels. Conclusions: The current results indicate that, during development in pubertal mice, acuteSF for 24 h a day 3 days negatively impacted spatial learning and synaptic plasticity. Further studies are needed to determine if any inherent long-term homeostatic

Abbreviations: SF, sleep fragmentation; NREM, non rapid eye movement; REM, rapid eye movement n Correspondence to: Department of Neurology, UW Medical Foundation Centennial Building, 1685 Highland Ave, 7th Floor, Madison, WI-53704, USA. Fax: þ1 602 263 0412. E-mail address: [email protected] (R. Maganti). http://dx.doi.org/10.1016/j.brainres.2015.04.037 0006-8993/& 2015 Elsevier B.V. All rights reserved.

Please cite this article as: Wallace, E., et al., Differential effects of duration of sleep fragmentation on spatial learning and synaptic plasticity in pubertal mice. Brain Research (2015), http://dx.doi.org/10.1016/j.brainres.2015.04.037

2


mechanisms in the adolescent brain afford greater resistance to the deleterious effects of chronic-SF. & 2015 Elsevier B.V. All rights reserved.

1.

Introduction

Sleep is a complex, yet fundamental, physiological state necessary for neuronal maturation (Meerlo et al., 2009), plasticity (Peirano and Algarín, 2007) and survival (Dinges and Chugh, 1997) in all animal species. Sleep disruption in childhood and adolescence results in a significant spectrum of adverse neurobehavioral consequences, including reduced learning (Carskadon et al., 1998; Randazzo et al., 1998), abnormal behavior (Paavonen et al., 2009), and mood disturbances (Aronen et al., 2000). Unlike total sleep deprivation or restriction, SF is a problem in many sleep disorders such as obstructive sleep apnea (Carreras et al., 2014), and restless leg syndrome (Trenkwalder and Paulus 2010). In animal models, sleep loss impairs spatial learning and memory in the Morris water maze and radial arm water maze (Smith and Rose, 1996; Youngblood et al., 1999a, 1999b: Kopp et al., 2006; Tartar et al., 2006). Initial studies have focused on either acute sleep deprivation, lasting hours or days, or chronic sleep disruption, the later studies focusing on SF and its effects on learning and memory. No studies have methodically tested the consequences of SF duration within the same model, especially in a maturing brain. Furthermore, there is limited work evaluating the neurobiological effects of sleep alterations in the developing brain. Directly assessing the cognitive effects of multiple forms of sleep disruption via behavioral and electrophysiological assay within a common model is important to better understand the neurobiological relation between sleep loss and cognition. LTP is a form of neural plasticity that has been implicated as a cellular mechanism of memory formation (Malenka and Nicoll, 1997). In vitro studies show that LTP is impaired in various models of sleep deprivation and sleep fragmentation. For example, impaired LTP has been observed with both acute (Patti et al., 2010; Fernandes-Santos et al., 2012; Romcy-Pereira and Pavlides, 2004; Alhaider et al., 2010; Vecsey et al., 2009) and chronic sleep loss (Kim et al., 2005) in adult animals. However, differential effects of sleep fragmentation duration on LTP in the developing brain have not been well investigated. The goal of this study was to assess the effects of acuteand chronic-SF on spatial learning and memory and LTP in pubertal mice. Actigraphy was used to monitor rest-activity patterns and electroencephalograms (EEGs) were obtained for sleep/wake analysis and to validate the actigraphy findings and effect of our SF protocols. We also examined whether stress was a factor in the SF methodology used by measuring serum corticosterone levels.

2.

Results

2.1.

Actigraphy

whereas they had high activity levels during lights-off (Fig. 1A, top).

2.1.1.

2.1.2.

Chronic-SF

In the chronic-SF group, locomotor activity was significantly increased in the lights-on period compared to controls [F (1,18)¼ 4.47, po0.05] (Fig. 1A bottom and 1B bottom), but not different in dark period compared to controls [F(1,19)¼ 1.23, p40.28], suggesting that no recovery sleep was experienced when the lever was not rotating.

2.2.

Morris water maze studies

2.2.1.

Acute-SF

There was a day main effect for distance [F(4,56)¼12.30; po0.0001], with scores decreasing across days, demonstrating learning of the task. Mice given acute-SF exhibited poorer overall performance, showing higher distance scores, collapsed across all 5 testing days, as compared to controls [F (1,14)¼ 5.02; po0.05] (Fig. 2A). For overnight amnesia, neither acute-SF nor control animals increased their swim distance during the overnight interval (data not shown). For the probe trial, assessing memory of the platform location (northeast (NE) quadrant), there was a main effect of Quadrant [F(1,14)¼ 88.142; po0.0001]. We confirmed that each group spent more percent of total swum distance in the target NE quadrant as compared to the opposite southwest (SW) quadrant, showing localization of the platform location [Control: F(1,7) ¼36.949; po0.001; acute-SF: F(1,7)¼ 59.894; po0.0001]. Control and acute-SF animals did not differ in the percent of total swum distance in either the NE or SW quadrant (Fig. 2C).

2.2.2. Actigraphy was utilized to examine rest-activity patterns. Control animals had low activity levels during lights-on

Acute-SF

In contrast to controls, acute-SF animals did not show a significant difference between lights-on and -off [control: F(1,12)¼ 9.55, po0.01]; acute-SF: F(1,12)¼ 2.4, p40.14] and there was no statistically significant decline in locomotor activity across the 3 days of SF [F(2,15)¼ 0.52; p40.6]. Moreover, the lack of a significant difference in the locomotor activity counts in the experimental group between lights- on and -off suggests that the acute-SF protocol achieved perturbation of expected rest-activity cycles. While we cannot exclude the possibility that the acute-SF group may have been getting some recovery sleep during the dark phase, the fact that activity counts in the dark-phase did not differ significantly from controls (F(1,12)¼ 3.94, p40.07), suggest that they did not (Fig. 1B, top).

Chronic-SF

There was a day main effect for distance [F(4,68)¼23.51; po0.0001], with scores decreasing across days, demonstrating learning of the



3

Fig. 1 – Actigraphy in control, acute- and chronic-sleep fragmentation groups: (A) Graphs of 24-h activity histograms representing averages across the entire period of recording. The normal statistically significant decrease of locomotor activity in the lights on period seen in controls is abolished in animals undergoing 72-h acute sleep fragmentation (24 h/day 3 days; middle panel). Chronic sleep fragmentation (12 h/day 14 days; bottom panel) maintains significant decrease in lights on actigraphy counts, despite increased activity in the lights on period for the compared to controls. (B) Statistical analysis of average locomotor activity counts of light- vs. dark-phases between groups and lighting condition.


4


Fig. 2 – Morris Water maze data for all groups of animals (n¼ 8 each): (A) Daily inter-trial average swim path length to platform between acute-SF (24 h/day 3 days) and control animals. (B) Daily inter-trial average swim path length to platform between chronic-SF (12 h/day 14 days) and control groups. (C) Percent total distance swum in target (NE) vs. opposite (SW) quadrant between acute-SF and controls during the probe trial. (D) Percent total distance swum in target (NE) vs. opposite (SW) quadrant between chronic-SF and controls during the probe trial.

task. There were no main effects or interactions for chronic-SF [ps40.19] (Fig. 2B). For overnight amnesia, neither chronic-SF nor control animals increased their swim distance during the overnight interval, indicating retention of the platform location overnight (data not shown). For the probe trial, there was a main effect of Quadrant [F (1,17)¼ 50.535; po0.001]. We confirmed that each group spent more percent of total distance swum in the target NE quadrant as compared to the opposite SW quadrant, showing localization of the platform location [Control: F(1,9) ¼27.551; po0.001; chronic-SF: F(1,8) ¼23.545; po0.005]. Control and chronic-SF animals did not differ from each other in the percent of total distance swum in either the NE or SW quadrant (Fig. 2D).

2.3.

Electroencephalography

To confirm our actigraphy findings, we implanted EEG and EMG electrodes in a separate group of animals (n¼ 4 each) to record sleep and wake patterns during acute- and chronic-SF.

2.3.1.

Acute SF

In the acute-SF group, sleep efficiency during lights-on did not vary significantly across all recorded days (Table 1A). However, the number of wakefulness bouts was significantly increased both during lights-on and lights-off on SF days (po0.0001). Furthermore, bouts of both NREM and REM sleep

were much shorter during SF compared to baseline or recovery days (po0.001). In examining the total time spent in different stages, animals had more time spent awake during lights-off especially during days 2 and 3 of SF, and there was no significant difference in time spent in NREM or REM across all days (Fig. 3A). No difference was seen in deltapower between baseline and recovery days (Fig. 3C and E). Sleep latency was also significantly decreased during lightson on fragmentation days 1 and 2 (po0.01) and trended toward shorter latencies on the day 3 lights-on (p¼ 0.09) and lights-off on fragmentation days 1 and 2 (p's¼ 0.0612 and 0.0615, respectively), while sleep latency on the recovery day showed a trend longer (p¼ 0.07) (Fig. 3G).

2.3.2.

Chronic SF

Similarly in the chronic-SF group, sleep efficiency did not differ across fragmentation days during lights-on or -off (Table 1B). The observation that sleep efficiency was consistently greater in lights-on, coupled with its trended decrease during lights-off, suggests that animals did not have recovery sleep during lights-off. The number of wakefulness bouts was significantly higher on SF days compared to baseline day and the average bout length for both NREM and REM sleep was significantly shorter on SF days compared to baseline or recovery day (po0.001) (Table 1B). In examining the total time spent in various vigilance states, no significant differences were seen in time spent in wake, NREM or REM across different days and certainly animals were not sleeping more


5


Table 1 – Sleep bout analysis in acute- and chronic-SF cohorts during light and dark periods. For the acute-SF group (A), bout properties were analyzed for all recorded days, whereas for the chronic-SF group (B), baseline day, day 1, 5, 10, 14 and recovery day data were analyzed. A two-way, repeated measures ANOVA with multiple comparisons was conducted to assess significance across all days of SF; p-values are as denoted in Table 1C. A ASF Light

Sleep Efficiency NREM Avg Bout Length REM Avg Bout length Arousal Count Sleep Efficiency NREM Avg Bout Length REM Avg Bout length Arousal Count

ASF Dark

B CSF Light

Sleep Efficiency NREM Avg Bout Length REM Avg Bout length Arousal Count Sleep Efficiency NREM Avg Bout Length REM Avg Bout length Arousal Count

CSF Dark

C p Values * † ‡ §

Baseline

Day 1

Day 2

Day 3

Recovery

63.9171.63 208.30736.91

54.2671.69 83.1672.43*

54.2871.78 75.6471.58*

82.2373.35 74.4771.13†

60.6975.53 233.73743.71

74.2673.20

34.5673.39‡

40.5272.54†

37.0373.44†

107.7574.44†

94.67715.62 46.8371.95 193.99741.49

260.3374.06§ 35.4470.48 81.8471.16*

251.67724.90§ 30.472.83* 69.7471.70*

275.3377.05§ 16.6272.99§ 71.7175.64*

88.00710.82 38.4274.86 192.17748.50

64.3179.97

34.2673.32†

37.0970.79*

23.78712.50‡

75.1375.66

84.67715.17

166.6777.26†

168.00710.82†

93.00728.58

78.67712.20

Baseline 57.8177.76 259.07734.71

Day 1 50.3771.18 70.93712.56†

Day 5 53.3075.51 108.62728.16*

Day 10 45.8471.48 92.82711.55†

Day 14 55.9473.69 99.16716.36*

Recovery 57.0374.66 218.59737.87

87.2476.16

34.2473.66§

37.0771.1§

38.7771.53§

38.5972.20§

77.1572.64

82.67712.77 40.8075.92 206.52734.49

279.67750.08§ 40.8178.83 176.79714.56

196.67741.74* 34.5974.26 191.98713.29

208.00713.53† 29.5173.72 173.86718.07

222.00725.42† 21.7573.46 230.33722.19

98.67718.77 24.3176.89 253.43767.59

68.6074.01

85.6175.59

79.7575.19

80.2376.35

104.6977.06‡

81.579.79

55.0074.16

71.33716.68

56.0079.54

55.00710.78

38.0076.35

49.00717.61

o0.05 o0.01 o0.001 o0.0001

during lights-off when the lever is not rotating (Fig. 3B). Again, no difference was seen in SWA between baseline and recovery days (Fig. 3D and F).. Average sleep latency was significantly shorter during lights-on on fragmentation days (po0.01) and longer during lights-off on day 14 of the sleep fragmentation protocol (po0.01) (Fig. 3H).

2.4.

LTP

Prompted by the findings of our behavioral tests, we examined the synaptic plasticity in acute hippocampal slices with regard to sleep fragmentation. Only the acute-SF group showed decreased hippocampal LTP maintenance when compared to control animals.

2.4.1.

Acute-SF

In the control group, theta-burst stimulation produced intact hippocampal LTP induction and maintenance (270.5741.9% and 156.9714.3% at peak and 60 min post-TBS, respectively) (8 slices from 4 mice; Fig. 4A). LTP maintenance was markedly impaired in the acute-SF group compared to the control group (120.778.5% at 60 min post-TBS; p ¼0.01, 9 slices from

4 mice) (Fig. 4D), which agrees with our observation that the acute-SF cohort exhibited impaired memory formation in the Morris water maze. Impairment of hippocampal LTP maintenance was observed in acute-SF animals in both stimulation response amplitude and EPSP slope (Fig. 4A and B).

2.4.2.

Chronic-SF

In contrast to the effects of acute-SF, LTP responses were not significantly different between control and chronic-SF groups at hippocampal CA1 synapses (157.5710.3% at 60 min postTBS, 9 slices from 4 mice) (Fig. 4 B and C).

2.5.

Corticosterone levels

We then examined whether stress was a factor in the impairments seen in learning, memory and LTP. Under experimental conditions, there were no differences in corticosterone levels between acute-SF and control animals (31.30672.63 vs 34.36574.29 ng/ml; p ¼0.26) (Fig. 5).


6


Fig. 3 – EEG recordings were acquired for each animal for the each day of sleep fragmentation. Total time spent in each vigilance state was calculated for lights on (hollow bars) and lights off (filled bars) each day of acute-SF protocol (A) 24 h/ day 3 days, n¼ 3) and a selection of days of the chronic-SF protocol (B) 12 h/day 14 days, n¼ 3). In the acute-SF protocol, values varied significantly from baseline for wake (green bars) in lights-on on day 2 and 3 (po0.05) and lights-off on day 3 (po0.0001), NREM (blue bars) during lights-off on day 2 (po0.05) and day 3 (po0.001). (C–F) Power spectral density (PSD) was calculated for baseline and recovery days in consecutive 4-s bins (FFT routine, Hanning window). Values for NREM sleep in the first six hours of baseline day were compared to the NREM sleep in the first hour period on the recovery day which contained at least 50% sleep. In order to compensate for signal strength fluctuations across recordings days, the power of each 0.25 Hz bin was adjusted to the mean 15–20 Hz power in its own window. Here we show the raw adjusted and normalized spectra for acute- (C and E) and chronic-SF (D and F), respectively. Sleep latencies were calculated as percent change from baseline (ΔBaseline/Baseline*100), values are shown for acute-SF (G), where values were significantly shorter for lights-on periods of day 1 and 2 (po0.01), and chronic-SF (H), where sleep latencies during lights-on of all fragmentation days and lights-off of SF day 14 differed significantly from baseline (po0.01). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)


7

Fig. 4 – Differential effects of acute and chronic sleep fragmentation on hippocampal long-term potentiation (LTP) (n¼ 4 animals each) compared to controls. (A) After theta-burst stimulation (TBS) of Schaffer collaterals, hippocampal slices of naïve mice incurred robust induction and maintenance of long-term potentiation (LTP). In hippocampal slices obtained from mice that underwent acute-SF (24 h/day 3 days), EPSP response amplitude (A) and slope (B) at 60 min post-TBS was significantly decreased when compared with control groups (10 slices from 5 mice *, po0.05). (C) Chronic-SF (12 h/day 14 days) had no effect on hippocampal LTP formation post TBS. (D) Summary data of changes in EPSP amplitude 60 min after TBS among experimental groups was qualified by comparing with baseline amplitude. Each vertical bar represents EPSP amplitude7SEM. One way ANOVA followed by Tukey post hoc analysis (C: * po0.05).

Fig. 5 – Corticosterone levels in Control and SD groups: Graph summarizing mean serum levels of corticosterone in acuteSF and control groups of mice. No significant (p ¼0.36) differences were seen between these two groups of animals (N¼ 6 each; each bar represents the mean7S.E.M.).

3.

Discussion

The current findings suggest that acute-SF with tactile stimulation for 24 h a day 3 days in pubertal mice negatively

impacted spatial memory, as well as hippocampal LTP maintenance compared to controls that had no tactile stimulation. By contrast, chronic-SF for 12 h a day 2 weeks was not associated with significant learning, memory or hippocampal LTP impairments, compared to controls in our model. Actigraphy data indicated higher activity counts in experimental groups during the period of SF. EEG data confirmed these actigraphy findings, showing that animals had significantly more wake bouts and shorter average NREM and REM bout lengths during the fragmentation period of both acute- and chronic-SF (Table 1). In addition, the chronicSF group did not compensate for sleep fragmentation as would be indicated by increased sleep efficiency during the dark period when the lever was not rotating (Table 1B). Sleep fragmentation did not affect EEG delta power in our study. A lack of corticosterone difference between the acuteSF and controls indicates that stress was not a confounding factor in the acute-SF methodology used in our model. While there are a few studies examining effects of SF on learning and memory, our study is the first to demonstrate differential effects of acute- and chronic-SF on spatial learning and memory as well as synaptic plasticity within the same model, in a developing brain. Our findings are consistent with studies that showed deleterious effects of acutesleep disruption on learning and memory (Patti et al., 2010; Fernandes-Santos et al., 2012; Vecsey et al., 2009), but this is not universally reported (Oniani, 1984). While early studies focused on selective REM deprivation using the “flower pot


8


technique”, later studies were not sleep stage specific in that animals were deprived of both REM and NREM sleep (Youngblood et al., 1997). These studies have shown that acute SD as short as 6 h to as long as 72 h has deleterious effects on learning and memory (Ruskin et al., 2004). In addition, it appears that stress was a factor in some of the earlier, but not later, studies; indeed, later studies showed deleterious effects of acute SD even in the presence of normal corticosterone levels (Ruskin et al., 2006). Chronic sleep restriction had been shown to have a negative impact on spatial learning and memory in some studies (McCoy et al., 2013). Other recent studies have focused on deleterious effects on learning and memory due to sleep interruption or SF, which is seen in sleep disorders such as sleep apnea(Tartar et al., 2006; Ward et al., 2009a; Ward et al., 2009b; McKenna et al., 2007). In all of these studies, sleep interruption and/or SF was achieved via a 24 h treadmill protocol and differed in effect from total sleep deprivation or restriction. While our findings are similar to these in the acute-SF condition, our model of chronic-SF, where only 12 h of sleep were fragmented, was not associated with significant impairments in learning or memory. These contradictory findings may be due to the difference in method and duration of sleep interruption. One recent study examined the effects of chronic SF on spatial learning and memory using methods similar to ours and showed that 2-week chronic-SF was associated with impaired spatial learning and memory, though notably the age at which animals were tested was not mentioned (Nair et al., 2011). While these findings contradict ours, the observation that our model of chronic-SF in pubertal mice did not significantly impair spatial learning and memory is validated by hippocampal LTP maintenance that is similar to controls. There may be several reasons why we saw differences in acute- and chronic-SF groups compared to controls. A major limitation of our study is that, although control animals were housed in cages with common features, they did not receive tactile stimulation. Second, our acute-SF protocol was 24 h a day while the chronic-SF protocol was for 12 h a day, as such, the 24 h SF may be a critical factor in memory and hippocampal LTP impairment. In our study, we found that mice were sleeping between lever rotations, with sleep bouts frequently shorter than 90 s, similar to those reported by others, which allowed maintenance of sleep efficiency despite SF limiting longer sleep bouts, as evidenced by increased wake bout counts during periods of sleep disruption (Gvilia et al., 2011). Moreover, some studies have shown that animals compensate chronic sleep loss with a homeostatic response during recovery periods (Kim et al., 2007) and we speculate that our chronic-SF group may have had better homeostatic responses, given the overall length of sleep fragmentation was 11 days longer than the acute protocol. The chronic homeostatic response may be more robust in a developing brain compared to a mature brain, although this remains to be shown. Other reasons might include the strain of animals studied or differences in the age at which SF started (21 and 32 days for chronic- and acute-SF, respectively), potentially resulting in a differential vulnerability to chronic compared to acute sleep disruption. Only a study directly comparing effects of different forms of sleep disruption, using similar SF protocols, between adolescents and adults can settle the issue of whether adolescents possess better compensatory homeostatic responses to certain forms of sleep perturbations.

EEG analysis of sleep and vigilance states showed that, while the overall sleep efficiency is largely unchanged among SF groups, the number of wake bouts was much higher and average REM bout much shorter during experimental period in all groups, which is similar to findings reported in a study where sleep fragmentation was achieved using an orbital shaker (Li et al., 2014). Additionally, delta-power or slowwave EEG activity (SWA, 0.5–4 Hz) has been used as a defining characteristic of NREM in rodent sleep models and the intensity is expected to increase across periods of wakefulness and decrease during periods of rest. However, this has been shown to evolve across adolescence, becoming most reliable in the later phase of adolescence (Alföldi et al., 1990; Nelson et al., 2013). While SWA can be used as a correlate for the efficacy of sleep deprivation, appearing to be inversely related to the length of sleep disruption in adult animals (Alföldi et al., 1990), it is not consistent in younger animals (Nelson et al., 2013). In our model, both the 72 h acute- and 2 week chronic-SF cohorts had no significant effect in SWA during the first hour of consolidated recovery sleep. Further studies are required to ascertain whether longer sleep fragmentation modalities allow for intra-fragmentation allostatic mechanisms to overcome the detrimental effects of sleep disruption on memory and hippocampal LTP. In the present study, we examined the effects of acute- and chronic-SF on spatial memory. It has been consistently shown that spatial memory is sensitive to sleep loss (Ruskin et al., 2006). However, studies testing working memory and its impairment due to sleep loss have been inconsistent, suggesting that working memory may not always be sensitive to sleep disruption (Ward et al., 2009b; Palchykova et al., 2006). It is noteworthy that impairments due to sleep deprivation also appear to extend to non-hippocampal-based abilities. Indeed, SD negatively impacted non-hippocampal dependent tasks such as object recognition (McCoy et al., 2007), attentional set shifting (Silva et al., 2004) and emotional function (Martinez-Gonzalez et al., 2004). Increasing evidence supports the tenet that SF in adolescence is linked to various neurobehavioral consequences, especially learning and memory deficits (Patti et al., 2010; Fernandes-Santos et al., 2012; Romcy-Pereira and Pavlides, 2004; Alhaider et al., 2010; Vecsey et al., 2009; Kim et al., 2005). Several studies have reported that LTP, the cellular model for memory, is impaired with sleep deprivation from a few hours to 72 h (Kopp et al., 2006; Tartar et al., 2006; Romcy-Pereira and Pavlides, 2004) and with selective REM SD (Davis et al., 2003; McDermott et al., 2003). In our study, non-selective SF within the same model, using consistent SF protocols in a maturing brain showed similar findings with acute sleep fragmentation, but surprisingly we did not find impaired LTP with chronic sleep fragmentation. There are several limitations to our findings. First, our sleep fragmentation method may have allowed too great an interval between lever rotations, allowing animals to sleep. If that is the case, however, it does not explain the difference in results we found between acute- and chronic-SF groups lending to the idea that immature brains may be more resistant to effects of chronic sleep loss due to homeostatic responses that require more time than the 72 h of the acute-SF protocol. Second, all Morris maze trials were conducted after the SF protocol was completed, which may have provided an opportunity for recovery sleep. If the trials had been conducted concurrent with sleep fragmentation, one can



speculate that the opportunity for recovery sleep would have been ablated, mitigating its effect on learning patterns, especially in the Chronic SF. However, this notion of recovery sleep is contested by LTP studies conducted immediately after SF showing results consistent with what we saw in the Morris maze studies. The control group did not have any lever rotations in this study suggesting that controls and experimental groups had different levels of stimulation and exercise that may have impacted their performance in the Morris maze. Our animals were housed in independent cages and in effect, are socially isolated and such social isolation had been shown to be associated with changes in sleep architecture, reduced delta power increases after sleep deprivation (Kaushal et al., 2012)and impaired spatial learning (Quan et al., 2010). We cannot exclude the possibility that social isolation may have contributed to some our findings. Finally, we did not exclude the possibility of SF induced-stress as a contributing factor even though we did not find changes in cortisol levels with the SF protocol used. While an adrenalectomized animal may serve as a proper control, previous studies however showed that sleep deprivation or SF induced changes are independent of stress factor (Ruskin et al., 2006; Guzman-Marin et al., 2007). Future studies examining differences in spatial learning and memory between pubertal and adult animals employing more refined techniques will be valuable to answer some of these questions. Future studies will also be necessary to elucidate the mechanisms of such impaired spatial learning, as well as to understand the differential responses of immature versus mature brains. In conclusion, our data indicate that acute-SF, in adolescent mice impairs spatial learning and memory, an effect concordant with deficient LTP maintenance. Remarkably, this was seen within 72 h of acute-SF. However, similar findings were not observed with chronic-SF. It is noteworthy that actigraphy showed no shifts in the diurnal rest-activity patterns during the period of SF in any of the groups, indicating that neither phase delay in circadian sleep-wake pattern nor recovery sleep outside the periods of sleep fragmentation occurred in our mice. EEG analysis of sleep and vigilance states further confirmed this. Furthermore, EEG sleep bout analysis showed that animals had significantly more wake bouts during the experimental days compared to baseline. A lack of a difference in corticosterone levels suggest that the observed effects are likely not due to sleep alteration induced stress differences between acute-SF and control animals. The current results indicate that acute-SF can be detrimental to memory formation and hippocampal long-term potentiation during maturation. While further studies are needed to expand on these findings and investigate further whether pubertal mice are more resistant to chronic-SF, we propose that pubertal brains may be more resistant to the deleterious effects of chronic SF due to their inherent plasticity and better adaptive responses.

9

Committees (IACUCs) at the Barrow Neurological Institute (BNI) and the University of Wisconsin School of Medicine and Public Health (UWSMPH). A portion of these studies were conducted at BNI and the remainder at UWSMPH.

4.2.

Subjects

Subjects were C57/Bl6 male mice born and raised at Charles River Laboratories (Wilmington, MA). Mice were 21 and 32 days of age at the beginning of the chronic and acute studies, respectively. They had access to food and water ad-libitum and were entrained to a 12/12 light/dark cycle.

4.3.

Sleep deprivation procedure

Mice were randomly assigned to a Control group or one of the SF groups (n ¼8 in each group). All animals were housed in the Smart Cage system (Afasci Inc, Ca) which has a Plexiglas cage with access to food and water, and a plastic lever at the bottom of the unit which is programmed to rotate periodically, each sweep providing tactile stimulation pushing and forcing mice to move and thus awakening them. Control animals were housed in similar cages with plastic lever at the base, but did not receive tactile stimulation. Both SF groups and controls were allowed an adaptation period of 24 h, during which time they habituated to the environment. Then, for acute-SF groups, the lever was programmed to rotate for 10 s out of every 2 min continuously for 72 h. For the chronic-SF group, the rotation protocol was employed 12 h daily from lights-on to lights-off for 2 weeks. All SF procedures were designed such that behavioral testing started during middle adolescence (P35).

4.4.

Actigraphy

All animals used in Morris water maze testing and electrophysiology experiments underwent actigraphic assessment (Minimitter Inc, USA) in which an infrared sensor recorded rest-activity patterns (Ancoli-Israel et al., 2003). Each gross movement of the animal was recorded as an activity count. Counts were grouped in 60-s bins. Actigraphy values are expected to be low during rest periods and high during active periods. Actigraphy provides an indirect measure of sleepwake patterns and can determine large shifts in rest-activity patterns. This is especially important to monitor for changes in circadian rest-activity or sleep-wake patterns due to the sleep fragmentation protocols employed. The activity counts were plotted across light-dark periods as shown in Fig. 1 for all groups. The mean activity counts per minute7standard deviation were calculated for each group across light and dark cycles for the entire duration of the experiment.

4.

Experimental procedures

4.5.

4.1.

Ethics statement

To further confirm our findings on actigraphy, in a separate group of mice (n¼ 4 each), we recorded sleep-wake patterns using electroencephalographic (EEG) monitoring. For EEG electrode implantation surgery, anesthesia was induced with 5% isoflurane and maintained at 1–2% in oxygen flowing at 0.5–1 l per minute. Three stainless steel epidural screws were

All procedures were carried out in accordance with the recommendations made in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health, and were approved by the Institutional Animal Care and Use

Animal surgery


10


placed as electrodes: two over the frontal (Bregma þ1.5 mm and þ1 mm laterally) and parietal cortex (Bregma 3 mm and 1 mm laterally) and one occipital reference (lamda 1 mm at midline). Two stainless steel wire electrodes were placed in the nuchal muscles for electromyography (EMG) recording. The EEG and EMG electrodes were connected to a head cap and secured with dental acrylic. After 472 h recovery, each animal was transferred to individual EEG acquisition chamber and connected to a tether suspended with a slip-ring commutator.

4.6.

EEG recording and analysis

EEG/EMG signals were recorded and digitized with an XL Tech amplifier (XL Tech, USA), sampled at 1024 Hz. The acquired data was converted to an EDF format, uploaded into sleep analysis software (Sirenia Sleep Pro, Pinnacle Technology, Inc.) where sleep/wake patterns were manually scored in 4 s epochs according to EEG/EMG trace waveform (Fig. 6). For the acute-SF group, EEG was scored for all days, whereas representative days were selected for the chronic-SF group (Baseline, 1, 5, 10, 14 and recovery). Using scoring software (Sirenia Sleep Pro) we calculated and compared sleep efficiency, arousal counts and average bout lengths for wake, NREM and REM during lights-on and -off on baseline day, days of SF and recovery day, among the different experimental groups. In addition, we calculated the average latency to sleep onset following three or more consecutive wake bouts during lights-on and -off across analyzed days (Fig. 3 G and H). Twenty-four hour EEG traces were sorted and digitally filtered with a 0.5–70 Hz bandpass filter (EDFbrowser, www.teuniz. net) then analyzed with custom Matlab scripts (Mathworks, Natick, MA) as previously described (Faraguna et al., 2008). EEG power spectra were calculated for each epoch (FFT routine, Hanning window). To compensate for fluctuations of EEG recording quality over time, the power for each 0.25 Hz frequency bin was adjusted to the mean power of 15–20 Hz for each epoch. Adjusted power spectra for NREM in the hour of consolidated sleep (defined as the first hour in which the first four minutes and the entire period contained at least 50% sleep) on the recovery day were normalized to the average adjusted power spectra of NREM in the first six hours of the baseline day.

4.7.

Morris water maze testing

Acute-SF (n¼ 8) and chronic-SF (n¼ 8) were continued until the day testing started (P35). Thus, during the days of Morris maze testing animals in all groups had sleep ad libitum in between trials and when not being tested. Subjects underwent 6 trials per day for 5 days, starting at noon during lightson each day (ZT 6), using a tub (1.8 m diameter) filled with water dyed with non-toxic, white tempura paint. A hidden platform (12 cm in diameter) remained in a fixed location, thereby testing spatial reference memory (Faraguna et al., 2008; Bimonte-Nelson et al., 2006). During testing (Days 1–5) the mouse was placed in a randomly determined start point (North, South, East, or West location) and allowed 60 s to locate the submerged platform centered in the Northeast (NE) quadrant. The trial was terminated either when the mouse

found the platform or after 60 s, whichever occurred first. After each trial, animals were allowed 15 s on the platform, after which time the mouse was placed into a heated cage until the next trial. The approximate inter-test interval was 10 min. Morris maze performance was assessed by swim path length (inches) to the platform. To evaluate whether mice localized the platform to the spatial location, after all test trials on day 5, a 60 s probe trial was given whereby the platform was removed. For each trial, a camera suspended above the maze was used to track path length and latency (Ethovision 3.1, Noldus Instruments). Percent of total distance travelled in the quadrant previously containing the platform (target/NE) quadrant was compared to the quadrant diagonally opposite the platform (SW). Mice that learned the platform location were expected to spend the greatest percent distance in the target quadrant (Morris, 1984; Braden et al., 2010). This procedure was repeated among each of the control and SF groups.

4.8.

LTP

Control animals age-matched to SF animals (acute-SF, chronic-SF) at the end of the SD protocol were sacrificed at zeitgeber time 0 (ZT 0) and acute hippocampal slices (400 μm thickness) were prepared from the brains to measure the change in hippocampal synaptic plasticity. The mice destined for LTP testing did not undergo behavioral testing. The isolated brain was rapidly submerged in ice-cold oxygenated physiological saline (composition in mM: 124 NaCl, 1.3 MgSO4, 3 KCl, 1.25 NaH2PO4, 26 NaHCO3, 2.4 CaCl2, and 10 D-glucose; pH: 7.4). Hippocampal slices were prepared using a vibratome (The Vibratome company, St. Louis, MO), and then stored in an incubation chamber containing physiological saline bubbled with 95% O2/5% CO2 at 35 1C for 1 h. Each slice was transferred to a submersion-type recording chamber on a Zeiss Axioskop FS2 microscope and superfused continuously, at a rate of 2–3 mL per minute with physiological saline. Upon stimulation of Schaffer Collaterals using a MCE-100 bipolar concentric electrode (David Kopf Instrument, Tujunga, Germany), excitatory post-synaptic potentials (EPSPs) in CA1 stratum radiatum were measured with a borosilicate recording electrode (2–6 MΩ tip resistance, backfilled with 2 mM NaCl) connected to a Multiclamp 700A amplifier (Axon Instruments, Foster City, CA, USA). Data were digitized with a Digidata 1322A. After establishing baseline EPSPs amplitudes, theta-burst stimuli (TBS, consisting of 5 trains delivered at 0.2 Hz) were used to provoke LTP, an electrophysiological measure of memory consolidation. LTP is expressed as the percent of the mean baseline EPSP amplitude. Recorded data were filtered at 3 kHz, sampled at 10 kHz using pClamp, and analyzed with Clampfit (Axon Instruments).

4.9.

Corticosterone levels

Corticosterone levels were measured in acute-SF as well their Controls at the end of the SF protocol during lights-on, with blood obtained from tail clips, using liquid chromatographymass spectrometry (LC-MS) techniques. 30 mL of serum were combined with 200 mL of an acetone/acetonitrile 50:50(v/v) mixture. This mixture was shaken for 10 min and then


11


A

200 μV

1 Sec

B

200 μV

1 Sec

C

200 μV

1 Sec

Fig. 6 – Exemplar EEG/EMG templates used for scoring. Primary determining criterion was the EMG. EMG amplitude (bottom waveform in each pair) signified wakefulness/locomotion with high amplitude (A) or sleep/inactivity with low amplitude (B and C). Sleep was further differentiated by EEG waveform. High-amplitude, delta-band waveforms (0.5–4 Hz) were classified as NREM sleep and low-amplitude (B), theta-band waveforms (5–8 Hz) were scored REM (C).

centrifuged for 10 min at 5000 g at 18 1C. The supernatant was transferred to a new tube and dried under vacuum using a Savant Speed Vac under heat. The residues were then resuspended in 50 mL of LC/MS grade methanol and vortexed

of each compound were calculated using MassLynx software by comparison to a standard curve prepared using an internal standard. Standards were prepared as described above and analyzed with LC/MS-MS techniques.

vigorously for 10 min. Corticosterone was separated by high performance liquid chromatography (HPLC) using a Waters C18 column (3 μm 2.1 30 mm) with a Waters alliance HPLC Separation Module (Waters Corporation, Milford, MA) and gradient elution. Quantification was achieved via a tandem mass spectrometer (Waters/Micromass Quattro LC, Waters Corporation) in positive ion mode. All analyses were performed using multiple reaction monitoring. Concentrations

4.10.

Statistical analyses

For behavioral assessments, data were analyzed using an omnibus repeated measures ANOVA, with Treatment as the between variable and Days and Trials as the within variable. In order to test for overnight amnesia of the platform location in the Morris maze, we compared distance scores from the


12


last trial of each day (Trial 6) to the first trial on the following day (Trial 1) (Bimonte-Nelson et al., 2006; Talboom et al., 2008; Acosta et al., 2009; Markham et al., 2002). We examined within-group trial comparisons of the overnight interval (Trial 6 to Trial 1) in order to determine which treatment groups increased their swim distance across the overnight interval, thus showing overnight amnesia of the platform location. The first overnight interval was excluded as we have shown that animals use this time to extend their initial learning curves (Bimonte-Nelson et al., 2006). Differences in actigraphy were measured as mean activity counts7standard deviation, during light and dark cycles and compared using one-way ANOVA. Differences in sleep efficiency, bouts counts and/or average length of wake, NREM and REM sleep were measured with two-way, repeated measures ANOVA with multiple comparisons between baseline and SF days. Differences in SWA means7standard deviation of the first six hours of baseline and first hour of consolidated sleep on recovery day were analyzed using two-tailed, paired t-tests. Percent differences in sleep latencies from baseline values were calculated with two-tailed, paired t-tests. Differences in LTP induction and maintenance between control and SD groups were examined using one-way ANOVA. Differences in corticosterone levels were measured as mean concentration7standard deviation. Groups were compared using t-tests. Alpha was set at 0.05, two-tailed, for all analyses. For all analyses, it was noted that Type I error correction is not necessary with orthogonal planned comparisons (Keppel and Wickens, 2004).

Acknowledgments The study was funded by the Barrow Neurological Foundation. The authors would like to thank the Members of the Cirelli Lab for their advice and support on EEG Analysis.

references

Acosta, J.I., Mayer, L., Talboom, J.S., Tsang, C.W., Smith, C.J., Enders, C.K., Bimonte-Nelson, H.A., 2009. Transitional versus surgical menopause in a rodent model: etiology of ovarian hormone loss impacts memory and the acetylcholine system. Endocrinology 150 (9), 4248–4259. Alfo¨ldi, P., Tobler, I., Borbe´ly, A., 1990. Sleep regulation in rats during development. Am. J. Physiol. 258 (3 Pt 2), R634–R644. Alhaider, I.A., Aleisa, A.M., Tran, T.T., Alzoubi, K.H., Alkadhi, K.A., 2010. Chronic caffeine treatment prevents sleep deprivationinduced impairment of cognitive function and synaptic plasticity. Sleep 33 (4), 437–444. Ancoli-Israel, S., Cole, R., Alessi, C., Chambers, M., Moorcroft, W., Pollak, C.P., 2003. The role of actigraphy in the study of sleep and circadian rhythms. Sleep 26 (3), 342–392. Aronen, E.T., Paavonen, E.J., Fja¨llberg, M., Soininen, M., To¨rro¨nen, J., 2000. Sleep and psychiatric syptoms in shool-age children. J. Am. Acad. Child Adolesc. Psychiatry. 39 (4), 502–508. Bimonte-Nelson, H.A., Francis, K.R., Umphlet, C.D., Granholm, A.C., 2006. Progesterone reverses the spatial memory enhancements initiated by tonic and cyclic oestrogen therapy in middle-aged ovariectomized female rats. Eur. J. Neurosci. 24 (1), 229–242.

Braden, B.B., Talboom, J.S., Crain, I.D., Simard, A.R., Lukas, R.J., Prokai, L., Scheldrup, M.R., Bowman, B.L., Bimonte-Nelson, H.A., 2010. Medroxyprogesterone acetate impairs memory and alters the GABAergic system in aged surgically menopausal rats. Neurobiol. Learn. Mem. 93 (3), 444–453. Carskadon, M.A., Wolfson, A.R., Acebo, C., Tzischinsky, O., Seifer, R., 1998. Adolescent sleep patterns, circadian timing and sleepiness at a transition to early school days. Sleep 21 (8), 871–881. Carreras, A., Zhang, S.X., Peris, E., Qiao, Z., Gileles-Hillel, A., Li, R.C., Wang, Y., Gozal, D., 2014. Chronic sleep fragmentation induces endothelial dysfunction and structural vascular changes in mice. Sleep 37 (11), 1817–1824. Davis, C.J., Harding, J.W., Wright, J.W., 2003. REM sleep deprivation-induced deficits in the latency-to-peak induction and maintenance of long-term potentiation within the CA1 region of the hippocampus. Brain Res 973 (2), 293–297. Dinges, D., Chugh, D., 1997. Physiologic correlates of sleep deprivation. In: Kinney, J.M., Tucker, H.N. (Eds.), Physiology, Stress, and Malnutrition: Functional Correlates, Nutritional Intervention. Kinney JM. Lippincott-Raven, New York, pp 1-27. Faraguna, U., Vyazovskiy, V., Nelson, A., Tononi, G., Cirelli, C., 2008. A casual role for BDNF in the homeostatic regulation of sleep. J. Neurosci. 28 (15), 4088–4095. Fernandes-Santos, L., Patti, C.L., Zanin, K.A., Fernandes, H.A., Tufik, S., Andersen, M.L., Frussa-Filho, R., 2012. Sleep deprivation impairs emotional memory retrieval in mice: influence of sex. Prog. Neuropsychopharmacol. Biol. Psychiatry. 38 (2), 216–222. Guzman-Marin, R., Bashir, T., Suntsova, N., Szymusiak, R., McGinty, D., 2007. Hippocampal neurogenesis is reduced by sleep fragmentation in adult rat. Neuroscience 148 (1), 325–333. Gvilia, I., Suntsova, N., Angara, B., McGinty, D., Szymusiak, R., 2011. Maturation of sleep homeostasis in developing rats: a role of pre-optic area neurons. Am. J. Physiol. Regul. Integr. Comp. Physiol 300 (4), R885–R894. Kim, E.Y., Mahmoud, G.S., Grover, L.M., 2005. REM sleep deprivation inhibits LTP in vivo in area CA1 of rat hippocampus. Neurosci. Lett. 388 (3), 163–167. Kim, Y., Laposky, A., Bergmann, B., Turek, F., 2007. Repeated sleep restriction in rats leads to homeostatic and allostatic responses during recovery sleep. Proc. Natl. Acad. Sci 104 (25), 10697–10702. Li, Y., Panossian, L.A., Zhang, J., Zhu, Y., Zhan, G., Chou, Y.T., Fenik, P., Bhatnagar, S., Piel, D.A., Beck, S.G., Veasey, S., 2014. Effects of chronic sleep fragmentation on wake active neurons and the hypercapnic arousal response. Sleep 37 (1), 51–64. Kaushal, N., Nair, D., Gozal, D., Ramesh, V., 2012. Socially isolated mice exhibit a blunted homeostatic sleep response to acute sleep deprivation compared to socially paired mice. Brain Res 1454, 65–79. Keppel, G., Wickens, T.D., 2004. Design and Analysis: A Researcher’s Handbook, fourth ed. NJ: Pearson Prentice-Hall, Upper Saddler River. Kopp, C., Longordo, F., Nicholson, J.R., Lu¨thi, A., 2006. Insufficient sleep reversibly alters bidirectional synaptic plasticity and NMDA receptor function. J Neurosci. 26 (48), 12456–12465. Malenka, R.C., Nicoll, R.A., 1997. Learning and memory: never fear, LTP is here. Nature 390 (6660), 552–553. Markham, J.A., Pych, J.C., Juraska, J.M., 2002. Ovarian hormone replacement to aged ovariectomized female rats benefits acquisition of the morris water maze. Horm. Behav. 42 (3), 284–289. Martinez-Gonzalez, D., Obermeyer, W., Fahy, J.L., Riboh, M., Kalin, N.H., Benca, R.M., 2004. REM sleep deprivation induces changes in coping responses that are not reversed by amphetamine. Sleep 27 (4), 609–617.



McCoy, J.G., Tartar, J.L., Bebis, A.C., Ward, C.P., McKenna, J.T., Baxter, M.G., McGaughy, J., McCarley, R.W., Strecker, R.E., 2007. Experimental sleep fragmentation impairs attentional setshifting in rats. Sleep 30 (1), 52–60. McCoy, J.G., Christie, M.A., Kim, Y., Brennan, R., Poeta, D.L., McCarley, R.W., Strecker, R.E., 2013. Chronic sleep restriction impairs spatial memory in rats. Neuroreport 24 (2), 91–95. McDermott, C.M., LaHoste, G.J., Chen, C., Musto, A., Bazan, N.G., Magee, J.C., 2003. Sleep deprivation causes behavioral, synaptic, and membrane excitability alterations in hippocampal neurons. J. Neurosci 23 (29), 9687–9695. McKenna, J.T., Tartar, J.L., Ward, C.P., Thakkar, M.M., Cordeira, J.W., McCarley, R.W., Strecker, R.E., 2007. Sleep fragmentation elevates behavioral, electrographic and neurochemical measures of sleepiness. Neuroscience 146 (4), 1462–1473. Meerlo, P., Mistlberger, R.E., Jacobs, B.L., Heller, H.C., McGinty, D., 2009. New Neurons in the adult brain: the role of sleep and consequences of sleep loss. Sleep Med. Rev. 13 (3), 187–194. Morris, R.G.M., 1984. Developments of a water-maze procedure for studying spatial learning in the rat. J. Neurosci. Method 11, 47–60. Nair, D., Xiang, S.X., Ramesh, V., Hakim, F., Kaushal, N., Wang, Y., Gozal, D., 2011. Sleep fragmentation induces cognitive deficits via NADPH-oxidase dependent pathways in mouse. Am. J. Respir. Crit. Care Med. 184 (11), 1305–1312. Nelson, A.B., Faraguna, U., Zolla, J.T., Tononi, G., Cirelli, C., 2013. Sleep patterns and homeostatic mechanisms in adolescent mice. Brain Sci 3 (1), 318–343. Oniani, T.N., 1984. Does paradoxical sleep deprivation disturb memory trace consolidation?. Physiol. Behav. 33 (5), 687–692. Paavonen, E.J., Porkka-Heiskanen, T., Lahikainen, A.R., 2009. Sleep quality, duration and behavioral symptoms among 5–6-yearold children. Eur. Child Adolesc. Psychiatry 18 (12), 747–754. Palchykova, S., Winsky-Sommerer, R., Meerlo, P., Du¨rr, R., Tobler, I., 2006. Sleep deprivation impairs object recognition in mice. Neurobiol. Learn Mem. 85 (3), 263–271. Patti, C.L., Zanin, K.A., Sanday, L., Kameda, S.R., FernandesSantos, L., Fernandes, H.A., Andersen, M.L., Tufik, S., FrussaFilho, R., 2010. Effects of sleep deprivation on memory in mice: role of state-dependent learning. Sleep 33, 1669–1679. Peirano, P., Algarıń, C.R., 2007. Sleep in brain development. Biol. Res 40 (4), 471–478. Quan, M.N., Tian, Y.T., Xu, K.H., Zahng, T., Yang, Z., 2010. Post weaning social isolation influences spatial cognition, prefrontal cortex synaptic plassticity and hippocampal potassium ion channels in Wistar rats. Neuroscience 169 (1), 214–222. Randazzo, A., Muehlbach, M.J., Schweitzer, P.K., Walsh, J.K., 1998. Cognitive function following acute sleep restriction in children ages 10–14. Sleep 21 (8), 861–868. Ruskin, D.N., Liu, C., Dunn, K.E., Bazan, N.G., LaHoste, G.J., 2004. Sleep deprivation impairs hippocampal-mediated contextual learning but not amygdale mediated cued-learning in rats. Eur. J. Neurosci. 19 (11), 3121–3124.

13

Ruskin, D.N., Dunn, K.E., Billiot, I., Bazan, N.G., LaHoste, G.J., 2006. Eliminating adrenal stress response does not affect sleep deprivation-induced acquisition deficits in water maze. Life Sci. 78 (24), 2833–2838. Romcy-Pereira, R., Pavlides, C., 2004. Distinct modulatory effects of sleep on the maintenance of hippocampal and medial prefrontal cortex LTP. Eur. J. Neurosci. 20 (12), 3453–3462. Smith, C., Rose, G.M., 1996. Evidence for a paradoxical sleep window for place learning in the Morris water maze. Physiol. Behav. 59 (1), 93–97. Silva, R.H., Kameda, S.R., Carvalho, R.C., Takatsu-Coleman, A.L., Niigaki, S.T., Abı´lio, V.C., Tufik, S., Frussa-Filho, R., 2004. Anxiogenic effect of sleep deprivation in the elevated plusmaze test in mice. Psychopharmacology (Berl) 176 (2), 115–122. Talboom, J.S., Williams, B.J., Baxley, E.R., West, S.G., BimonteNelson, H.A., 2008. Higher levels of estradiol replacement correlate with better spatial memory in surgically menopausal young and middle-aged rats. Neurobiol. Learn. Mem. 90 (1), 155–163. Tartar, J., Ward, C.P., McKenna, J.T., Thakkar, M., Arrigoni, E., McCarley, R.W., Brown, R.E., Strecker, R.E., 2006. Hippocampal synaptic plasticity and spatial learning are impaired in a rat model of sleep fragmentation. Eur. J. Neurosci. 23 (10), 2739–2748. Trenkwalder, C., Paulus, W., 2010. Restless legs syndrome: pathophysiology, clinical presentation and management. Nat. Rev. Neurol. 6 (6), 337–346. Vecsey, C.G., Baillie, G.S., Jaganath, D., Havekes, R., Daniels, A., Wimmer, M., Huang, T., Brown, K.M., Li, X.Y., Descalzi, G., Kim, S.S., Chen, T., Shang, Y.Z., Zhuo, M., Houslay, M.D., Abel, T., 2009. Sleep deprivation impairs cAMP signalling in the hippocampus. Nature 461, 1122–1125. Ward, C.P., McCarley, R.W., Strecker, R.E., 2009a. Experimental sleep fragmentation impairs spatial reference but not working memory in Fischer/Brown Norway rats. J. Sleep Res. 18 (2), 238–244. Ward, C.P., McCoy, J.G., McKenna, J.T., Connolly, N.P., McCarley, R.W., Strecker, R.E., 2009b. Spatial learning and memory deficits following exposure to 24 h of sleep fragmentation or intermittent hypoxia in a rat model of obstructive sleep apnea. Brain Res. 1294, 128–137. Youngblood, B.D., Zhou, J., Smagin, G.N., Ryan, D.H., Harris, R.B., 1997. Sleep deprivation by the “flower-pot” technique and spatial reference memory. Physiol. Behav. 61 (2), 249–256. Youngblood, B.D., Smagin, G.N., Elkins, P.D., Ryan, D.H., Harris, R.B., 1999a. The effects of paradoxical sleep deprivation and valine on spatial learning and brain 5-HT metabolism. Physiol. Behav. 67 (5), 643–649. Youngblood, B.D., Smagin, G.N., Elkins, P.D., Ryan, D.H., Harris, R.B., 1999b. The effects of paradoxical sleep deprivation and valine on spatial learning and brain 5-HT metabolism. Physiol. Behav. 67 (5), 643–649.


Sleep fragmentation has differential effects on obese and lean mice.

Effects of PI3Kγ overexpression in the hippocampus on synaptic plasticity and spatial learning.

BAI1 regulates spatial learning and synaptic plasticity in the hippocampus.

From synaptic plasticity to spatial maps and sequence learning.

Hippocampal Insulin Resistance Impairs Spatial Learning and Synaptic Plasticity.

Effects of sleep fragmentation on sleep and markers of inflammation in mice.

Sleep, clocks, and synaptic plasticity.

Hippocampal synaptic plasticity, spatial memory and anxiety.

Effects of Sleep Fragmentation and Chronic Latent Viral Infection on Behavior and Inflammation in Mice.

Differential splicing and glycosylation of Apoer2 alters synaptic plasticity and fear learning.

R-Modafinil exerts weak effects on spatial memory acquisition and dentate gyrus synaptic plasticity.

Effects of dihydrotestosterone on synaptic plasticity of the hippocampus in mild cognitive impairment male SAMP8 mice.

Effects of learning duration on implicit transfer.

Alterations in synaptic plasticity coincide with deficits in spatial working memory in presymptomatic 3xTg-AD mice.

mTORC1-S6K1 inhibition or mTORC2 activation improves hippocampal synaptic plasticity and learning in Angelman syndrome mice.

Sleep and synaptic plasticity in the developing and adult brain.

Role of synaptic structural plasticity in impairments of spatial learning and memory induced by developmental lead exposure in Wistar rats.

Sustained sleep fragmentation induces sleep homeostasis in mice.

Reduced NADPH oxidase type 2 activity mediates sleep fragmentation-induced effects on TC1 tumors in mice.

PS1dE9 mice.

- Mice Lack Bidirectional Synaptic Plasticity in the Dentate Gyrus and Perform Poorly on Spatial Pattern Separation Tasks.

Effects of exercise intensity on spatial memory performance and hippocampal synaptic plasticity in transient brain ischemic rats.

Dynamic learning and memory, synaptic plasticity and neurogenesis: an update.

Effects of acute sleep deprivation on motor and reversal learning in mice.