Sleep changes in fasting rats

Physiology&Behavior,Vol. 46, pp. 179-184. ©Pergamon Press plc, 1989. Printed in the U.S.A.

0031-9384/89 $3.00 + .00

Sleep Changes in Fasting Rats G. D E W A S M E S , * C. D U C H A M P t

A N D Y. M I N A I R E t

*Laboratoire de Physiologie et de Psychologie Environnementales, Centre National de la Recherche Scientifique Institut National de Recherche et de S6curit6, UMR 32, 67087 Strasbourg Cedex, France i'Laboratoire de Thermor6gulation et Energ6tique de l'Exercice, Centre National de la Recherche Scientifique Unit6 de Recherche Associ6e 1345, Universit6 C. Bernard, 69373 Lyon Cedex 08, France R e c e i v e d 16 D e c e m b e r 1988

DEWASMES, G., C. DUCHAMP AND Y. MINAIRE. Sleep changes in fasting rats. PHYSIOL BEHAV 46(2) 179-184, 1989.--The proportion and the distribution of wakefulness (W) slow-wave sleep (SWS) and paradoxical sleep (PS) were studied in 27-week-old rats over 24 hr periods, both in the fed state and after having been deprived of food for 2 to 3 weeks. In these rodents, prolonged fasting has been characterized by 3 successive metabolic phases which have been found to correspond to changes in protein metabolism. Sleep-waking changes were not studied during the first phase which was often of short duration (24 hr). During the second phase, i.e., when proteins were spared, the 24 hr proportions of W and sleep states remained unchanged. There were, however, profound changes in the daily mean episodic characteristics of each vigilance state (duration and frequency) except in the case of PS. During the phase II, the differences in the day/night proportions observed in each vigilance state were less than in the fed state. This reflected a lowering in the amplitude of their daily rhythms. In contrast, when protein use rose (phase I/I), W was increased sharply at the expense of SWS and PS, the latter being almost completely suppressed. During this last phase, which was also of short duration (by mean 3 days) alertness was greatly enhanced and the rats, which were typically nocturnal when fed, became diurnal. The changes in sleep and wakefulness were examined in relation to their effects on the homeostatic and cyclic components of sleep mechanisms and adaptative strategy to food deprivation in rat. Rat

Fasting

Sleep-waking homeostasy and rhythmicity

IN birds, the metabolic response to prolonged food deprivation has been found to be closely related to the rate of changes in dally sleep amounts and thus was assumed to modulate their control mechanisms (5). Studies on long-term fasting in geese (5) and emperor penguins (6) have shown that sleep state changes are intimately linked with successive metabolic phases, each characterized by important changes in lipid and protein metabolism (14,19). Slow-wave sleep (SWS) has been shown to be rapidly enhanced in association with the early and rapid mobilization of lipids and the reduction in protein use (phase I). This sleep state further increases, but at a slower rate, when lipids are used as the major fuels and the protein stores are conserved (phase II). When fat reserves are eventually near to depletion and proteins are used increasingly (phase III), the SWS episodes are shorter, representing a much lower dally proportion. In mammals, however, the relation between the sleep and metabolic responses to fasting had hitherto been only partially characterized. Indeed, sleep changes have been analyzed either using unsuitable procedures (11), or over too short fasting trials (1, 4, 12, 15, 17, 20), or even in species quite unable to resist starvation (4,8). The present study was thus designed to document this question in the rat, in which the fasting state has already been characterized by three metabolic phases (9) quite comparable to those defined in large birds (14,19).

temperature (25-28°C) with a 12 hr light-dark cycle. They were weighed (accuracy - 1 g) and fed dally (ad lib) with standard chow (Pellets A03, UAR, 91 Villemoisson-Sur-Orge, France). Water was always available.

Experimental Procedures When 24 weeks old, the rats were chronically implanted, under pentobarbital anesthesia (50 mg/kg IP), with stainless steel electrodes for recording EEG (from frontal and occipital cortices) and EMG (from nape muscles). Two days were allowed for surgical recovery. Thereafter, the recording leads were connected via a flexible cable and a rotating connector to an eight-channel polygraph (ECEM, model E20zoir-La-Ferd~re, France). Cables and connectors were carefully counterweighted to ensure minimum interference with the animal's behaviour. A further three weeks were allowed for adaptation, at the end of which, one or two successive 24 hr sessions were recorded in the fed rats. Food was removed when the rats were 27 weeks old. In rats, the fasting state is characterized by three successive metabolic phases, based on the turning point in protein breakdown (9). These can be characterized by the rate of change in total daily nitrogen excretion (9), and more conveniently, by that in specific daily mass losses (dm/m.dt) to which it is closely related (Koubi, H., manuscript in preparation). The changes in this last parameter were thus used to delimit the duration of the 3 phases (Table 1 and Fig. 1). Discontinuous 24 hr polygraphic recordings were performed from the very beginning of phase II (2nd day), and ended within the third phase. The second phase, the longest, was further divided into 3 equal subperiods referred to as the beginning (BII), the

METHOD

Animals Seven male Wistar rats (Iffa Credo, Lyon France) were individually caged in a quiet room kept at a thermoneutral ambient 179

180

DEWASMES, DUCHAMP AND MINAIRE

TABLE 1 DATES OF THE FASTING PERIODS AND SUBPERIODS (CALCULATEDFROM dm/m.dt, SEE TEXT AND FIG. 1) AND ACTUAL DATE OF THE RECORDING SESSIONS (MEAN - SEM) Fasting Phase II

Date of occurrence (day) Date of recordings (day)

Subperiod BII (Beginning)

Subperiod MII (Middle)

Subperiod EII (End)

Fasting Phase III

2.3 --- 0.2

9.3 ± 0.9

16.3 ± 1.8

19.1 ± 1.9

3.1 ± 0.4

9.5 ± 0.6

15.7 ± 2.1

20.6 ± 2.2

middle (MII) and the end (EII) of phase II (Table 1). Due to deficiencies in the recording systems, some records had to be disregarded, and consequently sleep-waking data are only presented for 7 fed rats, 7 fasting rats at subperiod BII, 6 fasting rats at subperiod MII and 5 fasting rats at subperiod Eli and at period III. The rats were killed at the end of 3 days spent in period III, without any apparent ill-effect other than body mass loss.

Data Analysis Each 24 hr polygraphic recording was visually scored, over 25-sec periods for three EEG and EMG stages: wakefulness (W) was characterized by low-voltage fast EEG activity and a high EMG amplitude; slow-wave sleep (SWS) was identified by high amplitude slow waves in the EEG; paradoxical sleep (PS) was characterized by a very regular theta rhythm, and a tonic suppression of the EMG. The EEG always enabled discrimination between PS and hypervigilant states whenever EMG was lacking or unreliable, since regarding PS, episodes of hypervigilance were characterized by an irregular theta rhythm of a lower amplitude.

Statistical Methods

used for statistical comparisons. RESULTS

Changes in Body Mass The initial body mass of the fed rats (533 +-27 g) decreased during fasting, at first rapidly (5.9 -+ 0.6 g. 100 g - ].day - 1) during phase I (1-2 days), at a slower rate throughout phase II (2.7---0.3 g. 1 0 0 - 1 . d a y - ] at BII; 2.6"--0.1 g. 100 g - l - d a y - 1 at MII; 2.7--- 0.2 g. 100 g - ] at Eli) and again at an increased rate during phase III (5.3---0.4 g.100 g - l - d a y - I ) . At the end of the experiment the body mass had decreased by 44%.

Sleep Waking Patterns During Phase H of Fasting (BII, MII, Ell) During this 2nd fasting phase the time spent daily in W, SWS and PS did not differ from the fed condition (Fig. 2), but the W and SWS episodes were shorter and more frequent (Table 2). When the data were examined separately during light and dark periods, only moderate changes from the fed state were noted at MII and Eli: the time spent in W was longer in the light and shorter

Values are expressed as means--- S.E.M. Student's t-tests were 100- Fed

ID

800 -

II

6

Ill

O~ m

ta 700-

5

600-

4

75

O

O

II

Ill

Wo SWS • PS o

jP iml

,i

m

"0"0

3

0 /

,.,I

<2 2

~" 400 0 m

---,i .......

I.::-----..~- ......

9

z~ o~ 500

i j °~'

so v

I ~'%% I! I • I I i i

25 v

0

_1 m

I r,

300-

o:f

i

25

0

-~,,~ t:,

0 E 0 I.U a.

TIME (days of fasting ) FIG. 1. Decrease in body mass (O) and changes in specific daily mass losses (SDML, stair curve) in a fasting rat. Phase limits are given between dashed vertical lines. SDML, high during phase I, decreased during phase II reaching stable low values, but increasing abruptly during phase III.

o

I

I

5

10

TIME ( d a y s

I I

15

20

2'5

of fasting)

FIG. 2. Daily percentage of W (11), SWS (0) and PS (©) in fed and fasting rat (at subperiods BII, MII, EII and phase III). The mean days of recording and the phase limits are given in Table 1 (Lower and upper part respectively). Fed rats have been compared (Student's paired t-test) to fasting rats at subperiod BII, MII, Eli and at period 11I. The number of rats are respectively N = 7, N = 6, N = 5, and N = 5 for each group. The symbol (**) indicates significant difference from fed condition (p<0.01).

SLEEP IN FASTING RATS

181

TABLE 2 MEAN DALLY DURATION AND FREQUENCY W, SWS AND PS EPISODES, IN FED A N D FASTED (PHASES II AND III) RATS

Fasting Phase II Fed

Subperiod BII (Beginning)

Subperiod MII (Middle)

Subperiod EII (End)

Fasting Phase III

W

Duration (±in) Frequency (No./24 hr)

2.37 +-- 0.10 262 ± 13

2.11 --- 0.11 (ns) 314 --- 22 (*)

1.84 --- 0.10 (t) 341 ± 12 (:~)

1.88 ___ 0.13 (i) 348 _ 11 (1)

4.15 ± 0.54 (*) 252 ___ 18 (ns)

SWS

Duration (±in) Frequency (No./24 hr)

2.58 ± 0.18 278 __. 12

2.06 --- 0.13 (*) 334 - 20 (*)

2.00 ___ 0.11 (*) 354 --- 11 (:~)

1.90 ___ 0.17 (*) 359 _ 12 (t)

1.55 ___ 0.13 (*) 250 _ 18 (ns)

PS

Duration (±in) Frequency (No./24 hr)

1.43 _ 0.03 81 _ 5

1.28 - 0.10 (ns) 88 ± 9 (ns)

1.48 - 0.09 (ns) 74 ± 7 (ns)

1.33 - 0.12 (ns) 84 ± 12 (ns)

1.07 _ 0.34 (ns) 10 ± 3 (~:)

Significant (*p<0.05; tp<0.01; :~p<0.001; Student's paired t-test) and nonsignificant (ns) differences from fed state are indicated. The numbers of rats for each group are given in the legend of Fig. 2.

in the dark, while the opposite was observed for SWS and PS (Table 3). These slight and most often, nonsignificant, changes reflected alterations in episode frequencies and durations (Fig. 3). Although the above light-dark changes in sleep-waking proportions remained modest in absolute term, they led to a signifiWakefulness

L ight

,

o

5

400

~/I i/,~

'sot~--'-"f:" r^i i:;

="

Dark

.__~=

2oo1,,,

/. lO

200

o 5 10 15 20 25 Slow Wave Sleep Dark

i::

,,=

oJi' 5

10

i 20

15

i.

5

10

!,i ;X'5~ ; #

0 j ,, i[ 5

10

15

20

, LJ [! ,; 25

O

15

=

t 20

_=

25

o

=-

Dark ',

F400

i

[300

ii

/i!

0

0

Paradoxical Sleep [ t il

~ 200]' L i g h t '

o

25

e°

,--,!

:' ta

Sleep-Waking Patterns During Phase IH of Fasting

'

15 20 25

Light

2oo

300

cant reduction (p<0.05) in light-dark differences at MII and EII, suggesting a diminution in the amplitude of circadian W, SWS or PS rhythms (Table 3). Their average amplitudes, evaluated according to Borbely (2), were indeed lower in these fasting stages, but were only statistically significant for W or SWS (Table 4). In contrast, the mean hourly level of each vigilance stage did not differ from the fed state throughout phase II (Table 4, Fig. 4). It must be noted that W or SWS hourly percentages were always higher during the first half of the night and the first half of the day, respectively (Fig. 4).

,

[200

i,,i

,

,

,,:

x._ ~

5

10

15

20

m

, L

0

25

Since W episodes were much longer (Table 2) their daily proportion was much higher, at the expense of both sleep states (Fig. 2). Due to a reduction in the daily time spent in SWS, and almost complete elimination of PS (Fig. 2), the daily PS to SWS ratio fell (Table 3). While the decrease in SWS was mainly due to shorter episodes, in the case that of PS it was mainly due to a sharp drop in the frequency of occurrence (Table 2), whereas in fed or fasting states the PS episode duration varied little (Table 2). The phase III was also characterized by large changes in the photoperiodic repartition of each vigilance stage. The W percentages were higher during both diurnal and nocturnal periods whereas those of both sleep states were lower, the changes being more marked in the daytime (Table 3). These changes led to a phase inversion and a further decrease in the amplitude of the circadian rhythm for each vigilance state as compared to fed state or phase II of fasting (Fig. 4). Lastly, the daily increase in W was related to a significant increase in the mean hourly level, the inverse being true for the two sleep states (Fig. 4).

TIME (days of fasting)

FIG. 3. Light-dark episode characteristics (episode duration, closed triangles, and episode frequency, open triangles) of W, SWS and PS in fed and fasting rat. Statistically significant differences from fed conditions (p<0.05, p<0.01, p<0.001) are respectively indicated by the symbols (*), (**), (***). The numbers of rats for each group are given in the legend of Fig. 2. Episode frequencies increased significantly during phase II in comparison with fed state both in light and dark during W and SWS states. Episode durations decreased significantly in dark or in light during W and SWS respectively. For other explanations, see caption to Fig. 2.

DISCUSSION

The present findings together with others (1,4) show that the daily SWS amount of adult fasting rats remains stable when they rely primarily on fat reserves and use body proteins at a low rate (phases I and II). This constancy, contrasts, on one hand, with the severe decrease reported in young lean rat and, on the other, with the slight increase found in VMH-lesioned obese rat fasting for four days (4). Thus, when fasting during phase I or II, animals of

182

DEWASMES, DUCHAMP AND MINAIRE

TABLE 3 PERCENTAGE OF TIME SPENT IN W, SWS, PS AND PS/SWS RATIO IN FED AND FASTING (PHASES II AND III) RATS, DURING LIGHT AND DARK, AND LIGHT-DARK DIFFERENCES IN ABSOLUTEVALUES Fasting Phase II Fed

Subperiod BII (Beginning)

Subperiod MII (Middle)

Subperiod El1 (End)

Fasting Phase III

W

Light(L) Dark (D) L-D

26.3 59.5 33.2

± 1.5 ± 2.3 ± 1.2

30.7 59.9 29.1

--+ 2.6 (ns) ± 1.0 (ns) ± 3.4 (ns)

32.1 54.9 22.7

± 1.7 (ns) ± 2.7 (ns) ± 2.8 (*)

36.5 54.3 17.7

± 4.7 (ns) ± 3.3 (ns) --- 3.8 (*)

73.7 68.1 6.9

---4.4 (~:) ± 3.6 (ns) --- 2.4 (:~)

SWS

Light (L) Dark (D) L-D

61.8 ± 1.2 36.4 ± 1.8 25.4 ± 1.2

58.4 36.1 22.2

--- 2.4 (ns) --+ 0.8 (ns) ± 2.8 (ns)

58.1 39.9 18.2

± 2.3 (ns) ± 2.4 (ns) ± 1.9 (*)

54.4 39.9 14.4

± 4.3 (ns) ± 3.6 (ns) ±-3.3 (*)

25.5 30.8 6.7

± 4.1 (t) ± 3.6 (ns) ---2.4 (?)

PS

Light (L) Dark (D) I_,-D

11.9 4.1 7.8

10.9 4.0 6.9

± 0.7 (ns) ± 0.3 (ns) ± 0.9 (ns)

9.8 5.2 4.8

± 0.9 (*) ± 0.6 (ns) ± 0.9 (*)

9.1 5.8 3.4

± 0.9 (ns) -- 0.5 (ns) ± 1.2 (*)

0.8 1.1 0.6

± 0.4 (~:) ± 0.3 (*) _+ 0.2 (:~)

PS/SWS

24 hr Light (L) Dark(D)

+-- 0.8 ± 0.5 +-- 0.4

0.16 ± 0.01 0.19 ± 0.01 0.11 ± 0.01

0.16 ± 0.01 (ns) 0.19 ± 0.01 (ns) 0.11 ± 0.01 (ns)

0.16 ± 0.01 (ns) 0.17 --- 0.02 (ns) 0.13 ± 0.01 (ns)

0.16 ±- 0.02 (ns) 0.17 ± 0.02 (ns) 0.15 ± 0.03 (ns)

0.03 ± 0.01 (?) 0.03 ± 0.01 (t) 0.03 + 0.01 (*)

Significant (*p<0.05; tp<0.01; :~p<0.001; Student's paired t-test) and nonsignificant (ns) differences from fed state are indicated. The numbers of rats for each group are given in the legend of Fig. 2.

the same species exhibit quite opposite SWS responses, while undergoing comparable metabolic adjustments (9). However, these discrepancies seem more apparent than real when initial energy stores, modalities of fuel mobilization and use, and the ability to save proteins, all of which partly modulate tolerance to fasting, are taken into account. Indeed, SWS increases slightly in obese rats in which protein saving appears to be efficient (9). Conversely, the SWS state decreases greatly in young lean rats or mice with high proteolytic rates throughout the fasting trials (3,9). The significant increase in SWS reported in acute fasting humans (15), or large long-term fasting birds, such as geese and emperor penguins (5,6) in which protein saving appears efficient (10, 14, 19) and which tolerate this nutritional state quite well, support this finding. Thus, in view of the SWS response to fasting, the 27-week-old rat could be considered as an intermediate case. It is difficult to relate PS changes to metabolic tolerance during the first two fasting phases. At the beginning of fasting the sleep state was indeed found not to vary both in the present study, and in the VMH-lesioned obese rat. Furthermore PS was found to be either lower of unchanged in acutely fasting humans (12,14). Nevertheless, SWS and PS occurrences were strongly inhibited when protein use was high, i.e., when the 27-week-old rat entered phase III, or in young lean rats and mice (4,8) both of which always maintained a high proteolytic rate throughout fasting (3,9). The PS state was found to be suppressed when protein synthesis was inhibited (7,18), but promoted when it was stimulated (21). The relationship between the changes in the rat's daily sleep amount and their lipid and protein metabolism during fasting suggests, as in large birds (5,6), that the homeostatic component of sleep could be metabolically modulated. A fortiori, such modulating influences may also be involved in eliciting new episode characteristics. During both phase I and phase II, the sleep-waking patterns were more fragmented, due to shorter and more frequent W and SWS episodes. In view of these broken sleep-waking patterns and also the shortening of the motor-activity episodes, Borbely (1) suggests that in fasting rats, the major adaptive mechanisms may be represented by adjustments in behavioural episodes in order to increase the chances of finding food with a minimum energy expenditure.

In the present experiment, the episodic characteristics of each vigilance level were further altered during fasting phase III. But the lengthening in W episodes and the consequent increase in the daily W proportion were the most prominent changes during this fasting period. Alertness was actually reinforced and, since in the rat this increase in W is concomitant with a large increase in spontaneous running (13), these changes in W may express food

Wakefulness

10075 A

50

F ,d

E li

m

25 v iii

00

F: III

Ell

.--"4 ::

.....

12

o

24

12

2L. 0

12

24 I

12

24

1

12

2/.

12

24

Slow Wave ( • ) and Paradoxical SLeep ( o )

100 I.lU 0 Z

M ii

i

F,'d :: ~,

,,

75 50 25

:

:1

..........

:

1

0 12

24 0

12

24 q

12

24

12

24 0

TIME ( hours )

FIG. 4. Hourly levels (in percent) of the 24 hr rhythms of W, SWS (O) and PS (O) in fed and fasting rats (at subperiods BII, MI/, EII and phase IN). Horizontal dashed lines represent the mean hourly level for the vigilance stage. Stair curves represent the hourly data, averaged for 4 hr, i.e., the beginning, the middle, the end of the day, or of the night.

SLEEP IN FASTING RATS

183

TABLE 4 AVERAGE AMPLITUDE AND MEAN HOURLY LEVEL (BOTH IN PERCENT) OF CIRCADIAN RHYTHMS, DURING FED CONTROL PERIOD AND DURING PHASE II OF FASTING (AT SUBPERIODS BII, MII AND Eli) Fasting Phase II Fed

Subperiod BII (Beginning)

Subperiod MII (Middle)

Subperiod Ell (End)

W

Amplitude Hourly mean

39.4 ___ 1.1 42.9 + 1.9

35.1 --- 3.7 (ns) 45.3 ± 1.0 (ns)

27.9 --- 2.4 (t) 43.5 --- 1.7 (ns)

24.4 ± 4.4 (*) 45.4 ___ 3.6 (ns)

SWS

Amplitude Hourly mean

25.1 -¢- 1.7 49.0 ± 1.5

25.4 ± 1.9 (ns) 47.1 ± 1.1 (ns)

19.7 +-- 1.9 (*) 49.0 ± 2.1 (ns)

18.3 ± 3.6 (ns) 47.1 ± 3.6 (ns)

PS

Amplitude Hourly mean

51.1 ± 2.8 7.5 ± 0.8

53.0 ± 2.4 (ns) 7.5 + 0.3 (ns)

35.8 ± 5.9 (ns) 7.5 ± 0.5 (ns)

29.6 ± 6.2 (ns) 7.4 ± 0.5 (ns)

Significant (*p<0.05; tp<0.01; Student's paired t-test) and nonsignificant (ns) differences from fed state are indicated. The numbers of rats for each group are given in the legend of Fig. 2.

foraging behaviour. The immediate reversal of this high level of running elicited during phase III by presenting food favours this hypotheses (Koubi, H., manuscript in preparation). Thus, the metabolism of fasting rat seems to modulate processing in CNS structures controlling the sleep-waking production and their related homeostatic component. It could also intervene in mechanisms underlying their cyclical characteristics. Such influences are particularly obvious when the rat enters the ultimate, third, phase. Increased protein use clearly modifies the amplitude, the mean hourly level and the phase reached in each vigilance stage, and, in the latter case, the rat, which is normally nocturnal, readily becomes diurnal. Moreover, it has been shown that, during phase II, the amplitude of the circadian rhythms could vary while the homeostatic component remained stable.

This last finding is of primary importance, since, together with other data on sleep-deprived suprachiasmatic-lesioned rat (16,22), it suggests that the presence of a circadian rhythm is not a prerequisite for the maintenance of sleep homeostasis. It also suggests that the homeostatic component of sleep may be morphologically and functionally distinct from the circadian component (22). ACKNOWLEDGEMENTS The authors thank Mr. B. Marchand for his technical assistance, Mrs. A. Brillant and G. Grosthor for their secretarial assistance in the preparation of the manuscript and Mrs. J. Saini and Mr. J. Pengelly-Bennet for the correction of this manuscript. This study was supported by grants from the Department of Biology Humaine, Universit6 Lyon 1.

REFERENCES 1. Borbely, A. A. Sleep in the rat during food deprivation and subsequent restitution of food. Brain Res. 124:457-471; 1977. 2. Borbely, A. A.; Huston, J. P.; Waser, P. G. Control of sleep states in the rat by short light-dark cycles. Brain Res. 95:89-101; 1975. 3. Cuendet, G. S.; Loten, E. G.; Cameron, D. P.; Renold, A. E.; Marliss, E. B. Hormone-substrate responses to total fasting in lean and obese mice. Am. J. Physiol. 228:276-283; 1975. 4. Danguir, J.; Nicolaldis, S. Dependance of sleep on nutrients, availability. Physiol. Behav. 22:735-740; 1979. 5. Dewasmes, G.; Cohen-Adad, F.; Koubi, H.; Le Maho, Y. Sleep changes in long-term fasting geese in relation to lipid and protein metabolism. Am. J. Physiol, 247(Regnl. Integ. Comp. Physiol. 16): R663-R671; 1984. 6. Dewasmes, G.; Buchet, C.; Geloen, A.; Maho, Le, Y. Sleep changes in emperor penguins during fasting. Am. J. Physiol. 256(Regnl. Int. Comp. Physiol. 25):R476-R480; 1989. 7. Drucker-Colin, R.; Zamora, J.; Bernal-Pedraza, J.; Sosa, B. Modification of REM sleep and associated phasic activities by protein synthesis inhibitors. Exp. Neurol. 63:458-467; 1979. 8. Faradji, H.; Cespuglio, R.; Valatx, J. L.; Jouvet, M. Effets du je0ne sur les phdnomdnes phasiques du sommeil paradoxal chez la souris. Physiol. Behav. 23:539-546; 1979. 9. Goodman, M. N.; Reed Larsen, P.; Kaplan, M. M.; Aoki, T. T.; Young, V. R.; Ruderman, N. B. Starvation in the rat. II. Effect of age and obesity on protein sparing and fuel metabolism. Am. J. Physiol. 239(Endocrinol. Metab. 2):E277-E286; 1980. 10. Goschke, H.; Stahl, M.; Tholen, H. Nitrogen loss in normal and obese subjects during total fast. Klin. Wochenschr 53:605-610; 1975. 11. Jacobs, B. L; McGinty, D. J. Effects of food deprivation on sleep and

wakefulness in the rat. Exp. Neurol. 30:212-222; 1971. 12. Karacan, I.; Rosenbloom, A. L.; Londono, J. H.; Sails, P. J.; Thornby, J. I.; Williams, R. L. The effect of acute fasting on sleep and the sleep-growth hormone response. Psychosomatics 14:33-37; 1973. 13. Koubi, H.; Dewasmes, G.; Goffart, M.; Frutoso, J.; Le Maho, Y. Evolution de l'activit6 motrice spontan6e et de la perte de masse corporelle au cours du je0ne prolong6 chez le rat. J. Physiol. (Paris) 79:28A; 1984. 14. Le Maho, Y.; Vu Kan Kha, H.; Koubi, H.; Dewasmes, G.; Girard, J.; Ferre, P.; Cagnard, M. Body composition, energy expenditure and plasma metabolites in long-term fasting geese. Am. J. Physiol. 241(Endocrinol. Metab. 4):E342-E351; 1981. 15. McFadyen, U. M.; Oswald, I.; Lewis, S. A. Starvation and human slow-wave sleep. J. Appl. Physiol. 35:391-394; 1973. 16. Mistlberger, R. E.; Bergmann, B. M.; Waldenar, W.; Rechtschaffen, A. Recovery sleep following sleep deprivation in intact and suprachiasmatic nuclei-lesioned rats. Sleep 6:217-233; 1983. 17. Parker, D. C.; Rossman, L. G.; Vander Laan, E. F. Persistence of rhythmic human growth-hormone release during sleep in fasted and isocalorically fed normal subjects. Metabolism 21:241-252; 1972. 18. Petitjean, F.; Sastre, J. P.; Bertrand, N.; Cointy, C.; Jouvet, M. Suppression du sommeil paradoxal par le chlorampbenicol chez le chat. Absence d'effet du thiampbenicol. C R Soc. Biol. 169:1236-1239; 1975. 19. Robin, J. P.; Frain, M.; Sardet, C.; Groscolas, R.; Le Maho, Y. Protein and lipid utilization during long-term fasting in emperor penguins. Am. J. Physiol. 254(Regul. Int. Comp. Physiol. 23): R61-R68; 1988.

184

20. Ruckebusch, Y.; Gaujoux, M. Sleep patterns of the laboratory cat. Electroencephalogr. Clin. Neurophysiol. 41:483--490; 1976. 21. Shapiro, C.; Girdwood, P. Protein synthesis in rat brain during sleep. Neuropharmacology 20:457--460; 1981.

DEWASMES, DUCHAMP AND MINAIRE

22. Tobler, I.; Borbely, A. A.; Gross, G. The effect of sleep deprivation on sleep in rats with suprachiasmatic lesions. Neurosci. Lett. 42: 49-54; 1983.