Sleep rebound in animals deprived of paradoxical sleep by the modified multiple platform method

Brain Research 875 (2000) 14–22 www.elsevier.com / locate / bres

Research report

Sleep rebound in animals deprived of paradoxical sleep by the modified multiple platform method Deborah Suchecki*, Beatriz Duarte Palma, Sergio Tufik ˜ Paulo, Rua Napoleao ˜ de Barros, 925, Vila Clementino, Sao ˜ Paulo, SP 04024 -002, Brazil Department of Psychobiology, Universidade Federal de Sao Accepted 23 May 2000

Abstract The objective of the present study was to assess the sleep rebound of animals exposed to the modified multiple platform method (MMPM), in which cage-mate rats were placed onto narrow platforms (NP56.5 cm in diameter), onto wide platforms (WP514 cm in diameter) or onto a grid (GR). The last two groups were included as environmental controls for the deprivation method. Animals were implanted with bipolar electrodes in the cortex, hippocampus and neck muscle. Baseline sleep was recorded for 6 h, after which the animals were placed in one of the above-mentioned settings for 90 h and their sleep was again recorded. Comparison between baseline and post-GR recordings revealed no sleep differences in these animals. Placement of animals onto WP resulted in augmented sleep time (16%), time spent in PS (199%), duration of PS episodes (177%), sleep efficiency (116%), and in reduced latency to PS (284.8%). Finally, NP animals exhibited a dramatic increase in sleep time (134.3%), time spent in PS (1184.7%), duration of PS episodes (1106%), and in sleep efficiency (134.4%). Moreover, sleep latency (252.2%) and time spent in SWS (212.2%) were reduced. Based on the results of sleep rebound, the data indicated that placement of animals onto narrow platforms in the MMPM was an effective PS deprivation method and the grid should be considered as an adequate environmental control.  2000 Elsevier Science B.V. All rights reserved. Theme: Neural basis of behaviour Topic: Biological rhythms and sleep Keywords: PS deprivation; Methodology; Modified multiple platform method; Polysomnography; Sleep rebound; Rat

1. Introduction Since the first report on paradoxical sleep (PS) deprivation [10], studies in humans and animals have yielded two important findings. First, the longer the period of PS deprivation, the greater the number of interventions necessary to prevent the occurrence of this sleep phase. Second, during the recovery period, PS-deprived animals compensate for the PS they have lost. Therefore, when the animal is again permitted to sleep, the latency to the first PS episode is shortened and the time spent in PS is lengthened [10,18,22,23]. Most of the instrumental methods utilized to induce PS deprivation in animals are modifications of the single *Corresponding author. Tel.: 155-11-539-0155; fax: 155-11-5725092. E-mail address: [email protected] (D. Suchecki).

platform method developed by Jouvet et al. [16] for cats and later, adapted for rats by Cohen and Dement [8]. The method consists of placing one rat onto one narrow platform (6.5 cm in diameter) immersed in water. When the animal enters PS, which is characterized by muscle atonia, it touches the water and wakes up. Control groups usually consist of either cage-control animals or rats placed onto a wide platform (14 cm in diameter) under the exact same environmental conditions. This method has been criticized for submitting the animal to additional adverse stimuli that may, per se, induce several of the effects observed after 96 h of PS deprivation. For instance, both PS deprivation and stress induce hyperthermia [30]. Likewise, PS deprivation and social isolation accelerate murine Lupus erythematosus [19]. In addition, prolonged exposure of animals to this method results in an important activation of the hypothalamic–pituitary–adrenal (HPA) axis [12,26].

0006-8993 / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved. PII: S0006-8993( 00 )02531-2

D. Suchecki et al. / Brain Research 875 (2000) 14 – 22

Further modifications of this method have been proposed as an attempt to reduce the elevated stress response. In 1981, Van Hulzen and Coenen, developed a technique in which one animal was placed into a large water tank, containing seven platforms, thus eliminating the movement restriction experienced in the single platform technique [35]. Still, many stress-related alterations are present, such as thymus atrophy, weight loss, adrenal hypertrophy [7] and augmented activity of the enzyme tyrosine hydroxylase [28]. Nunes and Tufik [24] introduced another modification to the previous method, by depriving several animals together. Ten animals coming from different cages were placed onto 18 narrow platforms, thus avoiding both the social isolation and the movement restriction. However, Suchecki et al. [31] reported that the HPA axis activation of animals placed in this setting is even greater than that of single platform-exposed rats. Since rats establish ranks of hierarchy within the group, the authors interpreted this exacerbated HPA response as a result of the social instability experienced by the animals during the sleep deprivation period. In an attempt to examine this hypothesis, groups of animals placed onto several narrow platforms or onto a stainless steel-made grid as socially unstable groups (similar to Nunes and Tufik [24]) were compared to animals deprived as socially stable groups. The findings show that alterations of stress-related variables, such as ACTH and corticosterone (CORT) plasma levels, relative adrenal weight, food intake and body weight, are attenuated when animals are deprived with familiar cohorts, compared to animals deprived under socially unstable conditions. Under socially stable conditions, the grid, which served as an environmental control, was shown to result in relative adrenal weight similar to that of cage control animals maintained in the same social setting. However, CORT levels were elevated over basal, suggesting that the grid induces stress, albeit to a lesser degree than the narrow platforms [32]. The present report is part of a series of studies, which seeks to establish a PS deprivation method in which the most obvious intervening stressors, i.e., isolation and immobilization, present in the classical platform method are eliminated, thus reducing the activation of the HPA axis. The main purpose of the present study was to examine whether the proposed method is effective to prevent PS occurrence and whether the grid would be an adequate environmental control, by assessing the sleep rebound of animals after 90 h of exposure to the modified multiple platform method (MMPM).

2. Material and Methods

2.1. Subjects Wistar male rats, 3-months-old, bred and raised in the

15

colony of the Department of Psychobiology, were used. After weaning, animals were raised in groups of six, in standard plastic cages (37.5 cm long332.0 cm wide316.0 cm high). The cages were placed in the deprivation room, under controlled temperature (23628C) and light–dark cycle (lights on at 07.00 h and off at 19.00 h), and the animals had free access to food and water.

2.2. Modified multiple platform method ( MMPM) Fourteen narrow (6.5 cm in diameter) or wide platforms (14.0 cm in diameter) or a grid floor (143.5 cm long342.0 cm wide39.0 cm high) were placed inside a water tank made of white tiles (145.0 cm long344.0 cm wide345.0 cm high). The grid is made of stainless steel with rods set 2.3 cm apart from each other, and was included as an attempt to establish an environmental control group for the PS deprivation. On the grid, the animals are allowed to lie down without falling into the water, albeit their tails may do so. The tanks were filled with water until 1 cm of the upper surface of the narrow platforms, of the wide platforms and of the grid.

2.3. Implant of electrodes The surgical procedure followed the guidelines for animal experimentation established by the Ethic Commit˜ Paulo. From each tee from Universidade Federal de Sao cage (containing six rats), only two to three animals were implanted with bipolar electrodes. They were weighed and injected i.p. with a dose of 10% solution of chloral hydrate (0.4 ml / 100 g of body weight). After the animal was anesthetized, xilocaine (2% plus epinephrine) was injected in the dorsal part of the cranium to complete the anesthesia, following the trichotomy on the dorsal region. The animal was attached to a David Kopf’s stereotaxic apparatus and the trepanation was performed on the points where the electrodes were to be inserted in order to record electroencephalographic (EEG) activity. These points were established according to the coordinates of Paxinos and Watson [27]: CA 1 (AP52 3.3; L51.5; H52.6) and CA 3 hippocampal fields (AP52 2.1; L51.5; H53.1) and cortical area A 10 (AP53.5; L52.5). Electrodes implanted in the CA 1 and CA 3 hippocampal fields were used to detect theta rhythm. The electrodes were permanently fixed to the skull with acrylic cement. Additional electrodes were inserted in the neck muscle prior to electromyographic (EMG) recording. Immediately after surgery, the animals were placed in individual transparent acrylic cages, containing sawdust and a top through which food and water were made available, and moved to the sleep recording room, under the same environmental conditions as described above. The animals were allowed seven days to recover from the surgery.

16

D. Suchecki et al. / Brain Research 875 (2000) 14 – 22

D. Suchecki et al. / Brain Research 875 (2000) 14 – 22

2.4. Experimental procedure

17

the onset of recording and the first sleep period; (iv) latency to paradoxical sleep: time lag between onset of sleep and the first episode of PS; (v) sleep efficiency (EFFIC): percentage of TST during the recording time; (vi) total time of SWS: sum of all slow wave sleep periods during the recording; (vii) total time of PS: sum of all PS periods throughout the recording; (viii) Number of awakenings: sum of all periods equal or longer than 1.5 min that the animal remained awake after sleep onset; (ix) number of PS episodes: number of episodes of PS throughout the recording; (x) mean time of each episode of paradoxical sleep: total time of PS divided by the number of PS episodes. Sleep onset was considered when the animal presented cortical synchronization and muscle hypotonia during two consecutive epochs.

On the two days preceding baseline recording (1 h / day), i.e., five days after the implant of electrodes, the animals were adapted, together with their cage companions, to the water tank, to avoid excessive drops into the water and to maintain the social stability within the group. This period has been shown to be sufficient for the animal to learn to stay on the platform [32]. On the first experimental day, beginning at approximately 09.30 h, the implanted animals were connected to a Nikhon Khoden KQP 223 polygraph for a 6-h period for baseline sleep recording. Immediately after the end of baseline recording, these animals, together with their cage cohorts, were placed into the setting to which they were previously adapted: (i) Grid (GR — Fig. 1, upper panel), (ii) Wide platforms (WP — Fig. 1, middle panel), (iii) Narrow platforms (NP — Fig. 1, lower panel). Exposure to the MMPM began at approximately 15.30 h and animals remained inside the tanks for a 90-h period. Throughout the study, the deprivation room was maintained under controlled temperature (23628C) and light–dark cycle (lights on at 07.00 h and off at 19.00 h). In all settings, food and water were provided ad libitum by placing chow pellets and water bottles on a grid located on top of the tanks (example is given in Fig. 1, middle panel). The water in the tank was changed daily, throughout the study. At the end of the manipulation period, the experimental animals were again recorded for another period of 6 h (post-GR or post-WP or post-NP), also from approximately 09.30 h to 15.30 h.

3. Results

2.5. Sleep parameters

3.1. Grid ( Fig. 2, upper and lower panels)

Sleep was digitally recorded and visually scored by the same experimenter in 30-s epochs. The epochs were classified according to Timo-Iaria et al. [33] as being either wake (W), which is characterized by low-voltage, fast EEG cortical and high EMG activities; slow wave sleep (SWS): continuous high amplitude, slow EEG activity and low EMG activity; or paradoxical sleep (PS), characterized by high EEG cortical activity, presence of regular theta hippocampal rhythm and absence of EMG activity. At the end of the analysis, sleep parameters were quantified by the Polysmith software  . The sleep parameters considered were: (i) Total sleep time (TST): sum of all sleep periods during the recording; (ii) total wake time (TWT): sum of all periods of waking during the recording; (iii) sleep latency: time lag between

Comparison between baseline and post-grid recordings showed no differences in any of the parameters, although a near significance was revealed for the mean duration of paradoxical sleep episodes [t(4)52.45; P50.07].

2.6. Statistical analysis The total number of animals was five / group. In each environment (NP, WP, GR), comparison between baseline and post-manipulation recordings was performed by the paired Student’s t-test. In order to evaluate the magnitude of change induced in each setting, the delta (D) analysis was performed by means of the one-way ANOVA of the difference between baseline and post-MMPM values. Post hoc analysis was carried out by the Duncan Multiple Range Test, with the level of significance set at P#0.05.

3.2. Wide platforms Fig. 3, upper panel, shows that placement of animals onto wide platforms led to a 16% augment in TST [t(4)5 2.9; P,0.05], a 99% increase of the total time of PS [t(4)58.07; P50.002] and of the mean duration of PS episodes [77.1%, t(4)55.32; P,0.006]. Conversely, reduction of the latency to PS [2 84.8%, t(4)52.72; P5 0.005] and of the TWT [234.2%, t(4)52.9; P,0.05] was obtained. In addition, a 16% increase in EFFIC was

Fig. 1. Upper panel: The water tank containing a grid that covers the whole floor and stands 9 cm above it. The rods are placed 2.3 cm apart which permits the animals to lie down and not fall in the water. Note that the tails may be dipped into the water. Middle panel: The water tank containing 12 wide platforms (14 cm in diameter) onto which the animals were placed for 90 h. Note the grid placed on top of the water tank through which food and water are made available to the animals. Lower panel: The water tank containing 14 narrow platforms (6.5 cm in diameter) onto which the animals were placed for 90 h. The tank is filled with water until approximately 1 cm from the platforms’ surface. The animals are, thus, allowed to move from one platform to the other and to interact with their group members.

18

D. Suchecki et al. / Brain Research 875 (2000) 14 – 22

observed in wide platform-exposed animals [t(4)52.92; P,0.05] (Fig. 3, lower panel).

3.3. Narrow platforms Placement of animals onto narrow platforms resulted in a 34.3% increase of TST [t(4)57.71; P,0.002], a 184.7% increase of total time of PS [t(4)513.57; P,0.0002] and 101.3% augment of the mean duration of PS episodes [t(4)55.9; P50.004]. A 52.2% reduction of the sleep latency SLAT [t(4)53.22; P50.03], a 12.2% reduction of total time of SWS [t(4)53.114; P,0.04] and a 58.8% decrease of TWT [t(4)57.71; P,0.002] were obtained (Fig. 4, upper panel). Fig. 4, lower panel, shows a significant increase of EFFIC sleep efficiency [34.4%, t(4)57.77; P,0.002] resulting from the exposure of animals to the MMPM. Neither the number of PS episodes (27.8%; 114.2%; 131.5%) nor the number of awakenings (211.8%; 120.2%; 218.2%) were altered in GR, WP and NP groups, respectively. Comparison among the groups was performed for each sleep parameter by the D between baseline and postMMPM values. The results can be seen in Table 1.

3.4. Total sleep time A significant difference [F(2,12) 56.79; P,0.01] was observed. TST of NP animals was longer than that of GR (P,0.005).

3.5. Sleep latency ANOVA revealed a significant difference on the sleep latency [F( 2,12 ) 53.75; P50.05]. Post hoc analysis showed that animals placed on NP initiated sleep faster than GR animals (P,0.03).

3.6. Sleep efficiency Significant differences were revealed [F( 2,12 ) 56.51; P, 0.02], where the increase of sleep efficiency was more pronounced in NP animals than in GR rats (P,0.005).

3.7. Total wake time The change in total wake time was significantly different among the groups [F( 2,12 ) 56.79; P,0.02]. Animals placed onto NP remained awake for a shorter period of time than GR animals (P,0.005).

Fig. 2. Changes in sleep variables of animals placed onto the grid (GR) for 90 h. Values (mean6S.E.M. of five animals / group) were obtained by 6-h long polysomnographic (PSG) recordings on two occasions: (i) Baseline (white bars) and (ii) Post-GR (black bars). Upper panel shows the parameters (followed by % of variation from baseline): TWT5total wake time (23.3%); TST5total sleep time (2.7%); SLAT5sleep latency (130.3%); LATPS5latency to PS (38.6%); TTSWS5total time in slow wave sleep (27.8%); TTPS5total time in PS (68.4%); MTPS5mean time of PS episodes (79.3%). The parameters EFFIC5sleep efficiency (3.1%); NPSE5number of PS episodes (27.75%); NA5number of awakenings (211.75%) are presented in the lower panel. *Denotes differences between post-GR and baseline values, detected by the paired Student’s t-test.

(P,0.04), which in turn was higher than GR rats (P5 0.05). No differences in latency to PS, total time of SWS, number of episodes of paradoxical sleep, mean duration of PS episodes and number of awakenings were detected among the groups.

3.8. Total time of PS 4. Discussion ANOVA revealed a significant difference among groups [F( 2,12 ) 59.97; P,0.003]. Post hoc analysis showed that total time of PS was higher in NP than in WP animals

The purpose of the present study was to examine whether placement of animals onto narrow platforms or

D. Suchecki et al. / Brain Research 875 (2000) 14 – 22

Fig. 3. Changes in sleep variables of animals placed onto wide platforms (WP) for 90 h. Values (mean6S.E.M. of five animals / group) were obtained by 6-h long PSG recordings on two occasions: (i) Baseline (white bars) and (ii) Post-WP (black bars). Upper panel shows the parameters (followed by % of variation from baseline) TWT (234.2%), TST (16%), SLAT (220.5%), LATPS (284.8%), TTSWS (210%), TTPS (99%) and MTPS (77%), whereas EFFIC (16%), NPSE (14.2%) and NA (20.2%) are displayed in the lower panel. For definition of the sleep parameters, please refer to caption on Fig. 2. *Denotes differences between post-WP and baseline values, detected by the paired Student’s t-test.

onto the grid in the modified multiple platform method, as established by Suchecki and Tufik [32] would, respectively, suppress or allow PS, by assessing the sleep of animals during the recovery period. The animals were PS deprived for 90 h because this is approximately the length of deprivation used in ours and several other studies [4,7,9,28,30–32,34]. Although 6 h of sleep recording is insufficient to evaluate sleep regulatory processes, our purpose was to examine the immediate rebound effects of our proposed method. Other studies have used the same or even shorter lengths of sleep recording for the purpose of evaluating sleep rebound [14,25]. The choice to evaluate sleep rebound was three-fold. First, the methodology does not permit an accurate sleep

19

Fig. 4. Changes in sleep variables of animals placed onto narrow platforms (NP) for 90 h. Values (mean6S.E.M. of five animals / group) were obtained by 6-h long PSG recordings on two occasions: (i) Baseline (white bars) and (ii) Post-NP (black bars). Upper panel shows the parameters (followed by % of variation from baseline) TWT (258.8%), TST (34.3%), SLAT (252.2%), LATPS (292.2%), TTSWS (212.2%), TTPS (184.7%) and MTPS (106%), whereas EFFIC (34.4%), NPSE (31.5%) and NA (218.2%) are displayed in the lower panel. For definition of the sleep parameters, please refer to caption on Fig. 2. *Denotes differences between post-NP and basal values, detected by the paired Student’s t-test.

measurement during PS deprivation. Animals cuddle for a significant amount of time, in addition to sleeping on top of each other. Moreover, the grid through which food and water become available would most certainly restrict the movements of the animals inside the water tank (see Fig. 1, middle panel), thus hindering the measurement, and possibly introducing an additional adverse situation (movement restriction). Second, although it has been reported that stress induces sleep rebound, this effect is observed only when the stimulus is applied acutely [5,21,25,29]. In fact, it was shown recently that 4 h of immobilization stress attenuates the sleep rebound previously obtained with 1 h of exposure

D. Suchecki et al. / Brain Research 875 (2000) 14 – 22

20

Table 1 Comparison of the difference between baseline and post-MMPM values of several sleep parameters assessed in animals placed on the grid (GR), on wide (WP) or narrow platforms (NP)a Sleep parameter GR TST (min) SLAT (min) LATSP (min) EFFIC (%) TWT (min) TTSWS (min) TTPS (min) NPSE NA MTPS (min)

4.2617.8 6.4610.7 23.064.5 1.465.1 24.2617.8 221.1616.9 25.3617.1 25.864.6 21.662.1 1.660.7

Group WP 38.3613.2 210.864.8 213.364.9 10.663.7 238.3613.2 22.2613.7 60.567.5* 3.863.7 0.862.1 1.860.4

NP 77.5610.1* 225.267.8* 228.5616.2 21.562.8* 277.5610.1* 220.966.7 98.467.2*† 6.662.9 21.461.0 2.460.4

a

Values are presented as mean6S.E.M. of five animals / group. Different from GR animals. † Different from WP animals. EFFIC5sleep efficiency; LATPS5latency to PS; MTPS5mean time of PS episodes; NA5number of awakenings; NPSE5number of PS episodes; SLAT5sleep latency; TST5total sleep time; TTPS5total time in PS; TTSWS5total time in slow wave sleep; TWT5total wake time. *

[20]. In addition, chronic exposure of animals to a variety of stimuli results in suppressed rebound sleep [1,17] and in some cases result in depression-like sleep architecture [6]. Finally, assessment of PS rebound, although an indirect measurement, is a reliable indicator of PS deprivation. For several years now, it has been shown that rats exhibit a striking sleep rebound in response to PS deprivation [22,23,36]. In addition, the number of interventions necessary to prevent PS occurrence increases as a function of the length of PS deprivation [10,23], indicating a progressive increase for PS propensity. The present results showed that placement of animals onto narrow platforms, for a period of 90 h, induced an exclusive PS rebound, indicating that this sleep phase was significantly suppressed. The rebound was represented by a 184% increase on the total time spent in paradoxical sleep due, mainly, to an increase on the duration, rather than on the number, of PS episodes. An interesting finding of the present study was the significant reduction of total time spent in SWS during the recovery period. Previous studies report a reduction of slow wave activity during a 2- [2] or 4-h period of PS deprivation [11] as well. The reduced time spent in SWS could be attributed to the elevated levels of CORT. Bradbury et al. [3] have recently shown that adrenalectomized rats replaced with CORT pellets near stress levels (9.1 mg / dl) exhibit less SWS than control animals. This result fits nicely with the present data, since rats deprived of PS under the exact same conditions as described above exhibit CORT levels of approximately 10 mg / dl [32]. Curiously, we have recently shown that 18 h (instead of 90 h) of PS deprivation, using the same paradigm as in the present study, result in a much larger CORT secretion (| 20.0 mg / dl) and a 9.4% increase of SWS [25]. A possible explanation for such a dis-

crepancy is that during the prolonged (90 h) PS deprivation procedure, animals could at least exhibit SWS and, as reviewed by Friess et al. [13], there is a blunted secretion of CORT during this sleep phase. Thus, it is possible to assume that in the work of Palma et al., animals were also deprived of SWS [25]. A surprising result was the lack of statistical significance in the reduction of latency to PS. In absolute numbers, latency to PS decreased from 29.7635.8 to 1.160.65 min. However, due to the large variability and the small number of animals used, this difference was probably masked. When this data was transformed to log, in order to reduce the variability within the groups, the difference between baseline and post-NP became statistically significant [t(4)54.98; P,0.008]. Several animals were lost or discarded due to the surgical procedure or by virtue of a poor EEG signal, which hindered the sleep staging. Thus, we chose to use a smaller sample size to avoid submitting an excessive number of animals to an aggressive procedure (such as the stereotaxic surgery) once we realized that it was sufficient to indicate statistical differences between baseline and post-manipulation values. Placement of animals on wide platforms, which is proposed to be an environmental control for the PS deprivation procedure, was also shown to produce sleep deprivation, albeit with a smaller magnitude. Thus, all PS-related parameters, such as reduced latency to PS, increase of total time spent in PS and of mean duration of PS episodes were observed when the animals remained on wide platforms for a 90-h period. However, when the magnitude of change between baseline and post-manipulation recordings was considered the only difference observed between grid and wide platform groups was in regard to the total time spent in PS. Most of the results obtained with the wide platforms were intermediate between the grid and narrow platforms values. Landis [18] reported that during the PS deprivation procedure, animals placed on wide platforms present approximately 50% of PS suppression. Porkka-Heiskanen and co-workers [28] report a PS suppressing effect of the narrow, but not the wide platform. However, a closer inspection of their data show that on the 1st and 3rd days of a 72-h period of PS deprivation wide platform-exposed animals exhibited approximately 50% of PS suppression, compared to the baseline recording. Curiously, several studies exploring the behavioural and physiological consequences of PS deprivation report similar effects produced by narrow and wide platform-exposed animals. For instance, lipid peroxidation is augmented and total glutathione content is reduced in animals exposed to the single platform technique, regardless of the size of the platform [9]. Administration of 10 mg / kg of apomorphine to animals submitted to the single platform technique for 96 h results in increased aggressive behaviour in narrow and, to a lesser degree, in wide platform-placed rats [34]. Hamdi and co-workers [15] showed that after a 96-h period spent on both wide or narrow platforms there is a

D. Suchecki et al. / Brain Research 875 (2000) 14 – 22

reduction of D 2 striatal dopaminergic receptors as well as increased D 1 / D 2 ratio. The levels of norepinephrine are similarly altered in the parietal cortex and the anterior hypothalamus of rats placed either on narrow or wide platforms. Under certain circumstances, hyperphagia, a typical PS deprivation-induced behaviour, is also observed in both groups [4]. Collectively, these and the present data indicate that wide platforms also produce PS deprivation, and therefore may not be the most adequate environmental control. Finally, minimal sleep changes were observed after placement of animals onto the grid, indicating that this procedure may serve as an adequate environmental control. In this case, animals were supposedly exposed to a similar environment (presence of water, similar amount of light and temperature inside the tank), without having the possibility to fall into the water. Therefore, the lack of sleep alterations during the post-manipulation period indicates the appropriateness of this control. It should be noticed, however, that the near significant increase in the mean time of PS episodes resulted from an increase of total time of PS (68.4%) and decrease of the number of PS episodes (27.8%), different from what was observed in NP- (184.7% and 31.5%, respectively) and WP-exposed animals (99% and 14.2%, respectively). In summary, the present |findings indicate that the placement of socially stable groups onto narrow platforms in the modified multiple platform method is efficient to produce a selective deprivation of the paradoxical phase of sleep. In addition, the grid appears to be a suitable environmental control since it results in virtually no changes in sleep patterns during the recovery period. These results may represent an important methodological improvement, insofar as the major intervening stimuli present in the single platform method seem to be now under control. Obviously, it is virtually impossible to eliminate the stress from methods of PS deprivation, since prevention to enter this sleep phase appears to be an adverse situation per se. Therefore, this method may provide a new perspective to explore more specific aspects of the effects of PS deprivation on the organism.

Acknowledgements ˜ Fundo de This study was supported by Associac¸ao ` Incentivo a Psicofarmacologia (AFIP). Deborah Suchecki was a post-doctoral fellow and Beatriz Duarte Palma, a ˜ de Amparo a` graduate student fellow from Fundac¸ao ˜ Pesquisa do Estado de Sao Paulo (FAPESP), grants [ 97 / 04857-9 and 96 / 12545-4, respectively. The authors would like to thank Dr. Jose´ N. Nobrega for the original idea of placing a grid inside the water tank as an ˆ environmental control, to Dr. Vania D’Almeida for her ˜ Padilha for the comments on the manuscript and to Joao care of animals.

21

References [1] J. Adrien, C. Dugovic, P. Martin, Sleep-wakefulness patterns in the helpless rat, Physiol. Behav. 49 (1991) 257–262. [2] J.H. Benington, M.C. Woudenberg, H.C. Heller, REM-sleep propensity accumulates during 2-h REM-sleep deprivation in the rest period in rats, Neurosci. Lett. 180 (1994) 76–80. [3] M.J. Bradbury, W.C. Dement, D.M. Edgar, Effects of adrenalectomy and subsequent corticosterone replacement on rat sleep state and EEG power spectra, Am. J. Physiol. 275 (1998) R555–R565. [4] J.W. Brock, S.M. Farooqui, K.D. Ross, S. Payne, C. Prasad, Stressrelated behavior and central norepinephrine concentrations in the REM sleep-deprived rat, Physiol. Behav. 55 (1994) 997–1003. [5] R. Cespuglio, S. Marinesco, V. Baubet, C. Bonnet, B. El Kafi, Evidence for a sleep-promoting influence of stress, Adv. Neuroimmunol. 5 (1995) 145–154. [6] S. Cheeta, G. Ruigt, J. van Proosdij, P. Willner, Changes in sleep architecture following chronic mild stress, Biol. Psychiat. 41 (1997) 419–427. [7] A.M.L. Coenen, E.L.J.M. Van Luijtelaar, Stress induced by three procedures of deprivation of paradoxical sleep, Physiol. Behav. 35 (1985) 501–504. [8] H.B. Cohen, W.C. Dement, Sleep: changes in threshold to electroconvulsive shock in rats after deprivation of ‘paradoxical’ phase, Science 150 (1965) 1318–1319. ´ [9] V. D’Almeida, D.C. Hipolide, L.A. Azzalis, L.L. Lobo, V.B.C. Junqueira, S. Tufik, Absence of oxidative stress following paradoxical sleep deprivation in rats, Neurosci. Lett. 235 (1997) 25–28. [10] W.C. Dement, The effect of dream deprivation, Science 131 (1960) 1705–1707. [11] T. Endo, B. Schwierin, A.A. Borbely, I. Tobler, Selective and total sleep deprivation: effect on the sleep EEG in the rat, Psychiat. Res. 66 (1997) 97–110. [12] P. Fradda, W. Fratta, Stress-induced sleep deprivation modifies corticotropin releasing factor (CRF) levels and CRF binding in rat brain and pituitary, Pharmacol. Res. 35 (1997) 443–446. [13] E. Friess, K. Wiedeman, A. Steiger, F. Holboer, The hypothalamic– pituitary–adrenal system and sleep in man, Adv. Neuroimmunol. 5 (1995) 111–125. ´ ´ L. Beltran-Parrazal, ´ ´ [14] F. Garcıa-Garcıa, A. Jimenez-Anguiano, A. ´ ´ Manipulations during forced Veja-Gonzalez, R. Drucker-Colın, wakefulness have differential impact on sleep architecture, EEG power spectrum, and Fos induction, Brain Res. Bull. 47 (1998) 317–324. [15] A. Hamdi, J. Brock, K. Ross, C. Prasad, Effects of rapid eye movement sleep deprivation on the properties of striatal dopaminergic system, Pharmacol. Biochem. Behav. 46 (1993) 863–866. ´ [16] D. Jouvet, E. Vimont, F. Delorme, M. Jouvet, Etude de la privation ´ selective de la phase paradoxale de sommeil chez le chat, Compt. Rend. Soc. Biol. 158 (1964) 756–759. [17] G.J. Kant, R.H. Pastel, R.A. Bauman, G.R. Meininger, K.R. Maughan, T.N. Robinson, W.L. Wright, P.S. Covington, Effects of chronic stress on sleep in rats, Physiol. Behav. 57 (1995) 359–365. [18] C. Landis, Altered sleep patterns with the platform method of REM sleep deprivation in rats, Sleep Res. 25 (1996) 469. [19] V.P.M. Lourenzi, A. Gabriel Jr., G. Nunes Jr., E. Atra, S. Tufik, REM sleep deprivation and social isolation accelerate autoimmune disease in mice, Sleep Res. 22 (1993) 338. [20] S. Marinesco, C. Bonnet, R. Cespuglio, Influence of stress duration on the sleep rebound induced by immobilization in the rat: a possible role for corticosterone, Neuroscience 92 (1999) 921–933. [21] P. Meerlo, B.J. Pragt, S. Daan, Social stress induces high intensity sleep in rats, Neurosci. Lett. 225 (1997) 41–44. [22] W. Mendelson, R.D. Guthrie, G. Frederick, R.J. Wyatt, The flower pot technique of rapid eye movement (REM) sleep deprivation, Pharmacol. Biochem. Behav. 2 (1974) 553–556. [23] B. Morden, G. Mitchell, W. Dement, Selective REM sleep depriva-

22

[24]

[25]

[26]

[27] [28]

[29]

D. Suchecki et al. / Brain Research 875 (2000) 14 – 22 tion and compensation phenomena in the rat, Brain Res. 5 (1967) 339–349. G.P. Nunes Jr., S. Tufik, Validation of the modified multiple platform method (MMP) of paradoxical sleep deprivation in rats, Sleep Res. 22 (1994) 339. B.D. Palma, D. Suchecki, S. Tufik, Differential effects of acute footshock and cold on the sleep of rats, Brain Res. 861 (2000) 97–104. ´ V. Patchev, K. Felszeghy, L. Koranyi, Neuroendocrine and neurochemical consequences of a long-term sleep deprivation in rats: similarities to some features of depression, Homeostasis 33 (1991) 97–108. G. Paxinos, C. Watson (Eds.), The Rat Brain in Stereotaxic Coordinates, 2nd Edition, Academic Press, San Diego, 1986. T. Porkka-Heiskanen, S.E. Smith, T. Taira, J.H. Urban, J.E. Levine, F.W. Turek, D. Stenberg, Noradrenergic activity in rat brain during rapid eye movement sleep deprivation and rebound sleep, Am. J. Physiol. 268 (1995) R1456–R1463. C. Rampin, R. Cespuglio, N. Chastrette, M. Jouvet, Immobilisation stress induces a paradoxical sleep rebound in rat, Neurosci. Lett. 126 (1991) 113–118.

[30] M.L.V. Seabra, S. Tufik, Sodium diclofenac inhibits hyperthermia induced by paradoxical sleep deprivation: the possible participation of prostaglandins, Physiol. Behav. 54 (1993) 923–926. ´ [31] D. Suchecki, L.L. Lobo, D.C. Hipolide, S. Tufik, Increased ACTH and corticosterone secretion induced by different methods of paradoxical sleep deprivation, J. Sleep Res. 7 (1998) 276–281. [32] D. Suchecki, S. Tufik, Social stability attenuates the stress in the modified multiple platform method for paradoxical sleep deprivation in the rat, Physiol. Behav. 68 (2000) 309–316. ˜ W.R. Schmidek, K. Hoshino, C.E.L. [33] C. Timo-Iaria, N. Negrao, Menezes, T.L. Rocha, Phases and states of sleep in the rat, Physiol. Behav. 5 (1970) 1057–1062. [34] S. Tufik, C.J. Lindsey, E.A. Carlini, Does REM sleep deprivation induce a supersensitivity of dopaminergic receptors in the rat brain?, Pharmacology 16 (1978) 98–105. [35] Z.J.M. Van Hulzen, A.M.L. Coenen, Paradoxical sleep deprivation and locomotor activity in rats, Physiol. Behav. 27 (1981) 741–744. [36] E.L.J.M. Van Luijtelaar, A.M.L. Coenen, Paradoxical sleep depriva¨ tion in rats: effects on rebound sleep, in: W.P. Koella, E. Ruther, H. Schulz (Eds.), Sleep 1984, Gustav Fischer Verlag, Stuttgart, 1985, pp. 394–396.