Sleep-Waking Cycle of the Hypophysectomized Rat JEAN-LOUIS VALATX. GUY CHOUVET
Dtparfement de Mtdecine Exptrimentale, Universitt Claude Bernard, 8 Avenue Rockefeller, 69373 Lyon (France)
The role of pituitary hormones in the sleep-waking cycle is not yet clearly defined. The hypophysis is not essential to trigger sleep: in the chronic pontile cat without hypothalamus or hypophysis, paradoxical sleep occurs during the first 5 days of survival (Jouvet, 1965). However, prolonged survival up to 2 months requires chronic administration of corticotropin (ACTH) or vasopressin (ADH). On the other hand, pituitary hormones seem to be effective in the behavior of hypophysectomized rats (de Wied, 1969; Lissak and Bohus, 1972). Acquisition and retention of passive or active avoidance reaction is facilitated by ACTH or ADH administration. Since the first work of Lucero (1970), Leconte et a/. (1973), Smith et aZ. (1972) and Page1 et a/. (1973) have demonstrated a close relation between sleep (paradoxical sleep) and learning. Impairment of learning in the hypophysectomized rat might be due to an alteration of the sleep cycle. Thus, this work has been undertaken to study the sleep-waking cycle of the hypophysectomized rat and the effects of ACTH upon sleep. The problem is to separate specific from non-specific effects of hypophysectomy on sleep mechanisms and learning.
We have studied 30 OFA male rats, 10 weeks old at the time of transaural hypophysectomy, performed by the “Centre d’Elevage des Oncins” (IFFA-CREDO). Under Nembutal anesthesia (30 mg/kg, i.p.), cortical and muscular electrodes were implanted 2 weeks after hypophysectomy. Cerebellar temperature was recorded by means of a thermistor, chronically implanted (Valatx et al., 1973). After surgery animals were placed in a plexiglass cylinder with food and water ad lib. The ambient temperature was constant (24 1 “C) and the lighting schedule consisted of 12 hr light (7.00-19.00) and 12 hr darkness (19.00-7.00). The electrode cable was attached 10 days before the beginning of the electroencephalogram (EEG) recording. References a. 120
After a period of 5 consecutive days to study the baseline of the sleep-waking cycle, a heat test (30°C) was performed during 24 hr. The effect of a long-acting preparation of ACTH (corticotropin Z, Organon) was also tested, and the results were compared to those obtained after treatment with placebo (zn-phosphate). At the end of experimentation, animals were sacrificed. Ablation of the hypophysis was checked and the adrenals were dissected and weighed. Sleep states were scored by visual analysis of 30-sec epochs. Quantitative analysis has been completed by studying the circadian sleep-wakefulness rhythm. By means of spectral analysis methods, the analysis of periodicity within the temporal characteristics of sleep records has been done. For each state of sleep, a time series was computed from the sleep duration within each consecutive hour during 4 or 5 consecutive days (Xi, i = 1, . . . N, i representing the rank of the corresponding hour from the beginning of the experiment). Considering this time series as representing a periodic sampling of the probability of occurrence of a given sleep state versus time, it is then possible to compute the spectral density by means of a Fast Fourier Transform (FFT). The spectrum is divided by the experimental variance of observations (s’) to normalize the results and to allow further comparisons. For each frequency, the following are computed: 2s with
Xi sin io,
2XP (p= 1 , 2 . . . N/2) angular frequency N
If the Xi’s are independent and normally distributed, the numerator in equation (1) has a distribution proportional to x 2 with 2 df(s’x’(2)). The value (1) is then similar to a x2(2)/2. Each value of (1) greater than ~ : , ~ ~ ( 2 ) / 2indicates =3 a correlation between the observations at the corresponding frequency with an error risk of 5 %.
In the isolated hypophysectomized rat, motor activity during wakefulness was reduced; sleep behavior (posture) was unchanged; EEG patterns of sleep were normal. One
SLEEP IN THE HYPOPHYSECTOMIZED RAT
Fig. 1. Polygraphic samples of slow-wave sleep (SWS) and paradoxical sleep (PS) from normal rat (1) and hypophysectomized rat (2-5). Note important muscular activity during SWS (2) and numerous eye movements during PS ( 5 ) in hypophysectornized rat. Note narcoleptic episode ( 3 4 ) characterized by direct transition from wakefulness to paradoxical sleep. EEG = electroencephalogram; EMG = electromyogram; EOG = electro-oculogram.
month after hypophysectomy, episodes of narcolepsy were observed: paradoxical sleep occurred directly after wakefulness or a few seconds of slow-wave sleep (SWS) (Fig. 1). Cerebellar temperature presented approximately the same fluctuations during sleep-waking cycle as in control rats but the mean level was over 2°C lower than normal (34-35 "C versus 36.5-37.5 "C).
( I ) Sleep-waking cycle at 25'C We have observed some variations among series of hypophysectomized rats, but 15 overall sleep duration was always reduced. The total sleep time per day was 604 min versus 715 18 min in control rats. SWS duration was reduced (- 10 %) less than paradoxical sleep (PS) ( - 50 %). Thus the PS/SWS ratio was smaller (PS/SWS = 8) than in control animals (PS/SWS = 14.4) (Fig. 2). The mean duration of the PS phase tends to increase while the number of phases is drastically reduced. The diminution of sleep time was preferentially observed during daytime. The day-night difference of PS duration was not as important (15 min) as in the control rat (50 min). The circadian sleep rhythm is completely altered (Fig. 3). Referem c s
Fig. 2. Paradoxical sleep (PS) duration during night (black bars) and during day (white bars) at 25°C (A) and 30°C (B) ambient temperature in normal rat (1) and in hypophysectomized rat, 30 days (2) and 60 days (3) after ablation of the hypophysis.
Fig. 3. Spectral analysis of slow-wave sleep (SWS) and paradoxical sleep (PS) rhythm of normal rat (A) and hypophysectomized rat (B). The computed values (A2s B2x/2sZ) in arbitrary units 0 5 = 3, i.e., (ordinates) are plotted against the frequency (abscissae). Dotted lines represent ~ ~ o . (2)/2 the limit of confidence of the spectrum amplitude. Note the lack of circadian rhythm (24 hr) in hypophysectomized rat.
Some subjects presented progressive alterations in sleep with a maximal diminution 2 months after ablation of the hypophysis (Fig. 2). This maximal diminution was often observed at the beginning of EEG recording (2-3 weeks after hypophysectomy).
( 2 ) Efects of ambient temperature (30 "C) Twenty four hours of exposure at 30 "C increased sleep duration (SWS and PS) as in normal rats. The augmentation started within the first several hours of heat exposure. Return to 25°C immediately stopped this alteration. The PS variation (+ 66 %) was more important than that of SWS ( + 5 %) and was due to the increase in the number of episodes. The mean duration of each episode remained unchanged.
SLEEP IN THE HYPOPHYSECTOMIZED RAT
TABLE I ADRENAL WEIGHT (mg) -~~ -..
Number of rats
( 3 ) EfSeect of ACTH Corticotropin Z was always administered at 18:OO hr ( 5 injections of 3 I.U./2 days) by the subcutaneous route. Some rats received placebo. The augmentation of adrenal weight after the fifth injection was a criterion for effectiveness of the injections (see Table I). PS duration increased progressively after the second or third injection of ACTH. The maximum was observed after the fourth injection (80 min verws 45 min before injection). The narcolepsy syndrome seemed to be unaffected by ACTH. Cerebellar temperature increased progressively with ACTH administration (up to 36 "C) and then returned gradually to 34°C when ACTH injections were stopped.
The sleep-wakjng cycle of the hypophysectomized rat is markedly disturbed in our experimental conditions from 3 weeks to 2 months after ablation of the hypophysis. In order to obtain a stable baseline of sleep, it was not possible to start recordings earlier than 3 weeks after ablation, the time taken for recovery from surgery and implantation and habituation to the recording cable. Alteration of the circadian rhythm of PS has been observed by Kawakami et al. (1972). However our results showed an important reduction of PS duration at 25°C. This variation might be due to hypothermia observed in hypophysectomized rats. Therefore heat exposure (30 "C) increases central temperature and provokes an augmentation of PS duration as in normal rats. Neurochemical mechanisms of the action of heat exposure on the sleep cycle are not yet clearly understood (see refs. in Valatx et a/.,1973). On the other hand, the action of ACTH upon sleep might be explained by nonspecific effects. The progressive action of ACTH on sleep and cerebral temperature could be the consequence of improvement of health due to the increased adrenal activity. The physiological doses used in our experiments (3 I.U./2 days) alter sleep and temperature in hypophysectomized rats but not i n normal animals. However, heat exposure and ACTH have a common specific effect: the preferential increase of paradoxical sleep. The utilization of ACTH analogs without hormonal References a. 120
activity could solve this problem. Narcolepsy occurs 1 month after hypophysectomy. This delay might be due to the degeneration of nerve fibers of the hypophysis or hypothalamus coming from the brain stem. In fact, surgical technique for ablation of the hypophysis might destroy a small part of the hypothalamus.
Hypophysectomy provokes a diminution of total sleep time in the albino rat. The reduction of paradoxical sleep is more important than that of slow-wave sleep. Heat exposure (30 "C) immediately reverses this diminution. ACTH administration progressively increases the duration of paradoxical sleep. In both cases cerebral temperature increases. These findings seem to be partly due to non-specific effects on sleep mechanisms. These results indicate that the basic mechanisms of sleep are not altered i n the hypophysectomized rat, but only the regulation or triggering of sleep.
This work was supported by INSERM (U.52), CNRS LA162 and DRME No. 73198. The generous supply of ACTH by Organon Comp., Oss, The Netherlands, is gratefully acknowledged. The authors are very indebted to Mrs. Luce Paut for technical assistance.
JOUVET,M. (1965) Etude de la dualite des &tatsde sommeil et des mecanismes de la phase paradoxale. In Aspects Anatomo-fonctionnels de la Physiologie du Sommeil, CNRS, Paris, pp. 397449. KAWAKAMI, M., YAMAOGA, S. AND YAMAGUCHI, T. (1972) Influence of light and hormones upon
circadian rhythm of EEG slow wave and paradoxical sleep. In Advances in Climatic Physiology, S. ITOH,K. OGATAAND H. YOSHIMURA (Eds.), Igaku Shoin, Tokyo, pp. 349-366. LECONTE, P., HENNEVIN, E. ET BLOCH, V. (1973) Analyse des effets d'un apprentissage et de son niveau d'acquisition sur le sommeil paradoxal consecutif, Bruin Res., 49, 367-379. LISSAK,K . AND BOHUS,B. (1972) Pituitary hormones and avoidance behavior of the rat. In/. J . Psychobiol., 2, 103-1 15. LUCERO, M. (1970) Lengthening of REM sleep duration consecutive to learning in the rat. Brain Res., 20, 319-322. PAGEL, J., PEGRAM, V., VAUGHAN, S., DONALDSON, P. AND BRIDGERS, W. (1973) The relationship of REM sleep with learning and memory in mice. Behav. Biol., 9, 383-388. C. T., KITAHAMA, K., VALATX, J.-L. ET JOUVET,M. (1972) Sommeil paradoxal et apprentissage SMITH, chez deux souches consanguines de souris. C.R. Acad. Sci. (Paris), 275, 1283-1286. VALATX, J.-L., ROUSSEL,B. ET CURE,M. (1973) Sommeil et temperature cerebrate du rat au cows de l'exposition chronique en ambiance chaude. Brain Res., 55, 107-1 22. WIED,D. DE (1969) Effects of peptides hormones on behavior. In Frontiers in Neuroendocrinology, W. F. GANONGAND L. MARTINI (Eds.), Oxford Univ. Press, New York, N.Y., pp. 97-140.
Sleep EEG stages and growth hormone levels in endogenous and exogenous hypercortisolemia or ACTH elevation D. T. KRIEGER - Division of Endocrinology, Department of Medicine, Mount Sinai School of Medicine, City University of New York, New York City, N.Y. (U.S..4.) Studies of nocturnal sleep EEG stages and plasma growth hormone (GH) and cortisol levels (used as indices of central nervous system (CNS) function) were performed in subjects with endogenous or exogenous elevation of plasma corticosteroid or ACTH levels, and in patients with hypothalamic tumors. These studies were designed to determine if there was any evidence of altered CNS function that was unique to Cushing’s disease, independent of any effects of hypercortisolemia per se. The findings of the same marked reduction of sleep EEG stages 111-IV and of the nocturnal GH rise in 4 treated (remission 5 months to 2 years) and 6 untreated patients with Cushing’s disease (clinically active 2 months to 10 years duration); in 4 “eu-corticoid” patients with hypothalamic tumors; and the presence of normal sleep EEG stages and nocturnal G H rise in a patient with Cushing’s syndrome 16 months following removal of an adrenal adenoma, as well as normal sleep EEG stages and lessened decrements in nocturnal G H rise in 7 patients receiving chronic prednisone therapy (15-60 mg daily for 4 months to 10 years) support the concept that altered CNS function may be involved in the pathophysiology of Cushing’s disease. The finding of suppression of stage 111-IV sleep in the patient with the adrenal adenoma and lack of such suppression in patients receiving chronic prednisone therapy suggests differential effects of exogenous and endogenous steroids. The presence of normal percentages of stage 111-IV sleep following the removal of the adrenal adenoma suggests, however, that cortisol excess also plays some role in the observed sleep EEG and G H changes. The presence of normal sleep EEG stages in 4 patients with Cushing’s disease who have subsequently developed Nelson’s syndrome of 1-9 years duration additionally suggests a role of this peptide in the genesis of sleep EEG stages.
Antidiuretic hormone secretion during sleep in adult men
ROBERT T. RUBIN,RUSSELL E. POLAND, FERNANDO RAVESSOUD, PAULR. GOUINAND BARBARA B. TOWER - Department of Psychiatry and Pharmacology, U.C.L.A. School of Medicine, Harbor General Hospital Campus, Torrance, Cal$90509 ( U . S . A . ) Decreased urine volume and increased osmolality have been noted in older men
(with indwelling urethral catheters) in close relationship to REM sleep episodes*. The postulated mechanism was ADH release from the posterior pituitary during REM sleep. However, the specific study of sleep-related ADH release has awaited the recent development of a sensitive ADH radioimmunoassay’. We studied 8 normal young adult men on two consecutive nights with blood sampling every 20 min from 23:OO to 07:OO according to an established protocol3. Water restriction began at 19:OO. ADH was released episodically in all subjects. However, in contrast to the postulated mechanism of REM-activated ADH release, there was no increase of ADH during REM sleep. For the 8 subjects, Kendall’s coefficient of concordance for wake + stage I, stage 11, stage 111 + IV, and stage REM was only 0.08. Thus, the REM-related decreases in urine volume noted in the earlier study’ do not appear to be related to ADH release, although that study was done in older men, and this one in young men. 1 MANDELL, A . J., CHAFFEY, B., BRILL,P., MANDELL, M. P., RODNICK, J., RUBIN,R . T., AND SCHEFF, R., Science, 151 (1966) 1558-1560. W. R., ROSENBLOOM, A. A., AND FISHER,D. A., J . din. Endocr. Merab., 38 (1974) 2 SKOWSKY, 278-287. 3 RURIN,R. T., GOUIN,P. R . , KALES,A., AND ODELL,W. D., Psychosom. Merl., 35 (1973) 309-321.