Effect of prolactin on the sleep-wake cycle in the rat

(1)

Neuroscience Letters, 156 (1993) 117-120 117

NSL 09609

Effect of prolactin on the sleep-wake cycle in the rat

R a c h i d a Roky, Jean-Louis Valatx and Michel Jouvet

Laboratoire de MOdecine Expkrirnentale, 1NSERM U 52, CNRS UA 1195, Universitd Claude Bernard, Lyon (France)

(Received 11 March 1993; Revised version received 16 March 1993; Accepted 16 March 1993) Key words: Prolactin; Sleep; Intra-cerebro-ventricular; Subcutaneous; Circadian rhythm; Rat brain

The purpose of this study was to examine the effects of prolactin on the sleep-wake cycle. Ovine prolactin was injected subcutaneous at a dose of 10/tg/animal or intra-cerebro-ventricularly at doses of 100, 10 and 1 ng/animal. The subcutaneous injections were given during either the diurnal or nocturnal period. Results indicate that oPRL decreases paradoxical sleep duration when injected during the dark period and increases it when injected during the light period. The i.c.v, injections were given only during the dark period and the effects were similar to those obtained with the s.c.

injections. There was no effect on slow-wave sleep duration irrespective of injection time or injection site.

A role of prolactin (PRL) on sleep has been suggested by several observations. Injections of PRL in the pontine cat increase paradoxical sleep (PS) [7]. Injections of vasoactive intestinal peptide (VIP) which stimulates secretion of PRL [1], also produce an increase in PS [11, 15]. Finally, in hypoprolactinemic mutant rats, the circadian rhythm of PS is reversed whereas slow-wave sleep (SWS) remains unchanged [22]. Furthermore, Obal et al.

[11] have found that subcutaneous injections of PRL en- hance PS duration in rabbits.

To investigate the hypothesis that PRL is involved in the circadian regulation of sleep, we injected ovine PRL (oPRL) either subcutaneous (s.c.) during the diurnal or nocturnal period or intra-cerebro-ventricularly (i.c.v.) during the nocturnal period in rats.

22 male OFA Sprague-Dawley rats (IFFA-CREDO, France) (250 g-300 g) were anesthetized with ketamine (150 mg/kg i.p.) and implanted with four EEG electrodes over the parietal and occipital cortex and two EMG electrodes in the dorsal neck muscles. Animals receiving i.c.v, injection were implanted with a stainless steel cannula in the lateral ventricle (1.5 mm lateral to the midline, 0.8 mm posterior to bregrna and 3.5 mm ven- trally) or with a polyethylene tubing in the cisterna magna as described by Sarna et al. [17]. After surgery, the animals were individually housed in a Plexiglas cage at a constant ambient temperature (23 + 1 °C) and a 12

Correspondence: R. Rachida, Laboratoire de Mrdecine Exprrimentale, INSERUM U 52, CNRS UA 1195, Universit6 Claude Bernard, 8 Ave- nue Rockefeller, 69373 Lyon CEDEX 08, France.

h-12 h light-dark schedule (lights on at 07:00, lights off at 19.00) with water and food ad libitum. After 10 days of post-operative recovery, rats were continuously recorded 24 1d24 h for 2 days as a control. Rats then received an injection and were continuously recorded for one more 24-h period. Recordings were visually scored in 30-s ep- ochs. Each epoch was classified as wakefulness (W), SWS and PS (when comprising > 15 s of a given state) according to conventional criteria [9].

Animals were subdivided into three groups. Group 1 received s.c. injections of oPRL consisting of 10/,tg/ani- real dissolved in 0.4 ml of 0.9% NaC1 solution at either 12:00 or 19:00. Group 2 received i.c.v, injections into the lateral ventricle through a cannula. The doses were 100, 10 or 1 ng of oPRL dissolved in 0.2/11 of 0.9% NaCI solution, oPRL was infused for a duration of 1 min and the needle was removed 1 rain after the injection was finished. Group 3 received injections into the cisterna magna through a catheter. The doses were 100, 10 or 1 ng dissolved in 5/11 of saline. A total volume of 15/11 was infused for 20 min. Groups 2 and 3 received i.c.v, injections only during the nocturnal period. All animals were habituated to handling and received several saline injections before PRL administration.

Data were stored and computed with home-developed software. Mean values of sleep duration and numbers of episods were compared by ANOVA, Wilcoxon's test and Student's t test. The dose-effect relationship was obtained by ANOVA. The accepted level of significance was P < 0.05.

S.c. oPRL injection (10/.tg) at 19:00 (Table I) elicited a

(2)

118 TABLE I

E F F E C T OF S.C. I N J E C T I O N OF 10,ug O F oPRL AT 19:00 A N D 12:00 U P O N W, SWS A N D PS D U R A T I O N S A N D EPISODE N U M B E R S (MEAN _+ S.E.M.) ACROSS 7 POST-INJECTION h.

W SWS

n Control oPRL Control oPRL

PS

Control oPR L

Injection at 19:00

Duration _+ S.E.M. 9 275 _+ 8 300 _+ 9 123 _+ 7 108 _+ 9 22 _+ 2 12 _+ 2 ~

Episode numbers _+ S.E.M. 9 55 _+ 3 51 _+ 4 56 _+ 4 50 _+ 4 16 -+ 2 9 _+ 28

Injection at 12:00

Duration -+ S.E.M. 8 130 -+ 3 120 _+ 3 237 _+ 3 238 _+ 5 53 _+ 2 62 _+ 38

Episode numbers _+ S.E.M. 8 76 _+ 4 73 _+ 3 79 _+ 4 74 _+ 4 33 _+ 2 32 _+ 2

"P < 0.005, 8p < 0.05. Control, saline injection.

small but significant increase in W (+ 9% P < 0.05). SWS was not significantly changed. The major effect consisted of a significant decrease in PS duration (-55%, P < 0.005) during the 7 post-injection h (Fig. 1A). This decrease was the result of a reduction in the number of PS episodes (-56.4%, P < 0.05) and not from a change in the duration o f PS episodes. There was no overall increase in PS duration during the last 5 h o f the dark period or during the following 12 h o f the light period. PS latency after the injection was increased with respect to control injections but did not reach a level of significance (49 + 10 vs. 28 rain + 8 min).

10/2g o f o P R L injected at 12:00, on the other hand, produced a slight but significant increase in PS duration (+ 17%, P < 0.05) (Fig, 1B). A trend toward an increase in the frequency of PS periods as well as the duration of PS episodes was found. W and SWS were not affected.

There was no significant difference between the results obtained after injections into the lateral ventricle or cis-

terna magna and, therefore, their data were pooled to- gether. At 19:00, o P R L provoked a significant decrease in PS duration which was due to a dimunition in the frequency of PS periods (Table II). This effect was main- tained for up to 10 or 12 h after 10- or 1-ng doses and up to 24 h after a 100-ng dose ( 7 2 + 6 vs. 113_+6 min, P < 0.05). By comparison with s.c. injection, the three doses (100, 10 and 1 ng) elicited during the 7 post-injection h a -50, - 3 7 and - 4 9 % PS reduction, respectively (Fig. 2).

Injections were made at 19:00 since it corresponds to the beginning o f a time period containing the least amount of PS during the circadian cycle. If an effect of the o P R L was to augment PS, as Obal et al. [11] observed, it would be most evident at this time. We obtained, however, a decrease in PS duration. A second set of injections was made at 12:00 when the quantity o f PS is at its highest. Any effect of the injectable substance that would decrease PS expression would be most appar-

TABLE 11

E F F E C T OF 1.C.V. I N J E C T I O N (100, l0 A N D 1 ng) O F oPRL AT 19:00 U P O N W, SWS A N D PS D U R A T I O N S A N D EPISODE N U M B E R S (MEAN _+ S.E.M.) ACROSS 7 POST-INJECTION h.

W SWS PS

n Control oPRL Control oPRL Control oPRL

100-ng dose

Duration + S.E.M. 8 238 -+ 16 Episode numbers + S.E.M. 8 66 -+ 5 10-ng dose

Duration _+ S.E.M. 9 222 _+ 12

Episode numbers -+ S.E.M. 9 69 -+ 5 1-ng dose

Duration + S.E.M. 8 209 -+ 12 Episode numbers _+ S.E.M. 8 72 _+ 4

219 + 24 161 _+ 16 192 + 23 21 + 2 10 + 3 ~

6 6 + 10 68 + 4 6 8 _ + 11 16+ 1 10-+2 b

227 _+ 14 174_+ 12 178 + 13 24 + 2 15 _+ 2

8 3 + 8 7 2 _ + 5 86_+ 8 18_+l 15_+2

240 _+ 20 188 _+ 12 168 _+ 21 23 z 2 12 -+ 3"

63_+ 9 74_+ 5 63_+ 7 17_+1 1 1 + 2 ~

Control, saline injection. "P < 0.02, bp < 0.05.

(3)

119

ent at this time. On the contrary, oPRL injected at 12:00 increased PS duration. This result was in agreement with that obtained by Obal et al. [11] who reported that oPRL enhances PS in the rabbit. Our data, on the other hand, indicate that the amplitude of the effect of PRL administration during the dark period was greater than during the light period.

Administration of oPRL at the beginning of the dark period (19:00) decreased PS duration. This effect was obtained after peripheral as well as central i.c.v, injections.

Similar results obtained in both conditions suggest that peripheral PRL penetrates the blood-brain barrier. In- deed, several authors have demonstrated such a penetra- tion in hyperprolactinemic conditions [2, 8] which likely occurs through specific PRL receptors located on cells of the choroid plexus [23]. Thus, peripheral PRL can act directly or indirectly on sleep mechanisms through brain PRL receptors [10, 24].

Several mechanisms may be considered to explain the observed oPRL effect on sleep. First, it has been shown

30

2"

t - O

10.

EL

( ) I ' R I . lt) u g

.I

, i • • ' °

• i ° ~ o ' "

20h 2ih 2:~h 23h Oh ih 2h

70

o P R I . lOiAg B

60 E-- 51)

E c- 40'

2

~ 3 o

E3 03 20.

13._

10,

0 1~1 1,~ 1,5h 16h 17h 18h 191n Fig. I. Effect o f 10/.tg s.c. injection o f o P R L on PS duration across 7 post-injection h (cumulative mean + S.E.M.). A, injection at 19:00; B, injection at 12:00. Continuous lines, oPRL; interrupted lines, saline.

**P < 0.05, ***P < 0.01, ****P < 0.005.

30

c-

E 20

v

g E

"Z3 rJ') I0 13.

oPRL 1 rK:J

•I

2m 2ih 22h ~ Oh ih 2h

30

t.,-

E 2 ~

v

8

' o S3

oPRL 10rig

i ... i

s"

~l~ ° ~

20h 2ih 27.h 23h Oh l'h ;~h

30

oPRL 100ng

-e .I

E Z0 ~ ,,-

v

E ° "

. O •

"13

'° ...i--""

zh~ 2ih 2~h 25h Oh ih ~h

Fig. 2. Effect ofi.c.v, injection o f o P R L at 19:00 on PS duration across 7 post-injection h (cumulative mean + S.E.M.). A, 1 ng; B, 10 ng; C, 100 ng. Continuous lines, oPRL; interrupted lines: saline. *P < 0,05,

**P < 0.02, ***P < 0.01.

that hyperprolactinemia or i.c.v, injections of PRL alter the synthesis of hypothalamic dopamine and proopiomelanocortin [18] which may influence sleep- wake mechanisms [21, 3]. Secondly, it has been demonstrated that VIP increases sleep duration [15]. PRL,

(4)

120

through its negative feedback action on V|P secretion [14, 16], could exert an indirect influence upon sleep mechanisms. Third, hypothalamic neurons containing PRL [5, 6, 19, 20] with numerous projections to the raphe dorsalis has recently been described [12, 13]. This obser- vation suggests that PRL may interact with 5-HT neu- rons in this nucleus as in the preoptic area [4] or in the nucleus dorsomedialis of the hypothalamus [25], struc- tures which are involved in sleep regulation.

The finding that PRL decreases PS during the dark period and increases it during the light period indicates that PRL may act on PS rhythmicity at the level of the suprachiasmatic nucleus where PRL-like fibers were de- scribed [12]. Further experiments including PRL injec- tions in brain areas containing PRL neurons or fibers, are required to elucidate the mechanism by which PRL acts on sleep wake control.

This work was supported by 1NSERM U52. C N R S UA 1195 and D R E T (Grant N" 89203). We thank Markus Schmidt for correcting the English.

1 Abe. H., Engler. D.. Molitch. M . I . , Bollinger-Gruber, J. and Reichlin, S., Vasoactive intestinal peptide is a physiological media- tor of prolactin release in the rat. Endocrinology. I16(1985) 1383 1390.

2 Barbanel, G., Ixart. G.. Arancibia. S. and Assenmacher, 1., Proba- ble extrapituitary source of the immunoreactive prolactin measured in cerebrospinal fluid of unanesthetized rats by push pull cannula- tion of the 3rd ventricle, Neuroendocrinology, 43 ( 1986~ 476 482.

3 Chastrettc, N. and Cespuglio, R., Efl'ects hypnogenes de la des- acetyl<z-MSH et du CLIP (ACTH 18-39) chez le rat, ('.R. Acad.

Sci. (Paris), 301 (1985) 527 530.

4 Clemens. J.A., Gallo. R.V., Whitmoyer, D.I. and Sawyer, C.H., Prolactin responswe neurons in the rabbit hypothalamus, Brain Res., 25 (1971) 371 379.

5 DeVito, W.J., Distribution of immunoreactive prolactin in the male and female rat brain: effects of hypophysectomy and intraventricu- lar administration of colchicine, Neuroendocrinology, 47 {19881 284 289.

6 ttarlan, R.E., Shi'<rs, B.D., Fox. S.R., Kaplove, K.A., Schachter, B.S and Pfaff, D.W,. Distribution and partial characterization of immunoreactive prolactin in the rat brain, Neuroendocrinology, 49 (19891 7 22.

7 Jouvet. M., Buda, C., Cespuglio, R., Chastrette, N., Denoyer, M..

Sallanon, M. and Sasfre. J.P., Hypnogenic effects of some hypotha- lamo-pituitary peptides, Clin. Neurophannacol., 9, Suppl., 4 (1986) 465 467

8 Login, I.S. and Macleod, R.M., Prolactin in human and rat serum and cerebrospinal fluid, Brain Res., 132 (1977) 477 483

9 Michel. F., Klein, M.. Jouvet, D. and Valatx, J.L,. Etude poly- graghique du sommeil chez le rat, C.R. Soc. Biol. (Paris). 12 (1961) 2389 2393.

10 Muccioli. G., Ghe, C. and Di ( a r l o . R., Distribution and characterization of prolactin binding sites ill the male and female rat brain:

effects of hypophysectomy and ovariectomy. Neuroendocrinolog}.

53 (1991)47 53.

II Obal, Jr., E., Opp, M.. Cady, A.B., Johannsen, I,. and Kruegcr, J.M., Prolactin. vasointestinal peptide and peptide histidinc methionine elicit selecti~,e increases in REM sleep in rabbits, Brain Rcs., 490 (1989) 292 430.

12 Paut-Pagano, L., Valatx, J.L., Kitahama, K. and Jouvct, M., Neu- tones ii prolactine dans l'hypothalamus dorso-latdral d u r a l Sprague Dawley, C.R. Acad. Sci. (Paris), 309 (19891 369 376, 13 Paut-Pagano, L., Roky, R , Valatx, J.L., Kitahama, K. and Jouvet,

M., Anatomical distribution of prolactin-like imnaunoreacti,,ity in the rat brain. Neuroendocrinology, submitted.

14 Prysor-Jones, R.A., Silverlight. J.J. and Jenkins, J.S.. Hyper- prolactinemia reduces vasoactive intestinal peptide in the anterior pituitary gland of rats, Neurosci. Lett., 80 (1987i 333 338.

15 Riou, F., Cespuglio, R. and Jouvet, M., Endogenous peptides and sleep m the rat. 11 I. The hypnogenic properties of vasoactivc intestinal polypeptide, Neuropeptides, 2 (1982) 265 277,

i6 Sarkar, D.K., Evidence for prolactin feedback actions on hypothalamic oxytocin, vasoactive intestinal polypeptide and dopamine secretion, Neuroendocrinology, 49 (I 989) 520 524.

17 Sarna. G.S.. Hutson. P i t . , Tricklebank, M D . and Curzon, G., De- termination of brain 5-h~droxytryptamine turno\er in freely m o \ - ing rats using repeated sampling of cerebrospinal fluid..I. Neuro- chem., 4() (19831383 388.

18 Selmanoff, M., Shu. ('., Hartman, R.D., Barraclough, C.A. and Petersen, S.L., Tyrosine hydrox)lase and POMC m R N A in the arcuate region are increased by castration and hyperprolactinemia.

Mol. Brain Rcs., ll) (1991) 277 281.

19 Siaud, P., Manzoni, O.. Bahnefrezol, M., Barbanel, G,, Assenma- chef. 1. and Alonso, G., /'he organization of prolactin-like-immunoreactive neurons in the rat central nervous system, Cell Tissue Res.. 255 (19891107 115.

20 Toubeau, G., Desclin, J.. Parmentier, M. and Pasteels. J.L., Cellular localization of a prolactm-like antigen in the rat brain..I Endocri- no1., 83 (19791261 266.

21 [rampus. M., Ferri, N., Monopoli, A. and Ongini, E., The dopam- me DI receptor is inxolved in the regulation of R E M sleep in the rat, lEur. J. Pharmacol.. 194(1991)198 194.

22 Valatx, ,t.k. and Jouvet, M., Les rythmes circadiens du sommeil lent ct du sommeil paradoxal sont en opposition de phase chez les rats alricbes hypoprolactinemiques, C R . Acad. Sci. (Paris), 307 (1988) 789 794.

23 Walsh. R.J,, Slabby, EJ. and Posner, B.I., A receptor mediated mechanism for the transport of prolactin from blood to cerebrospinal lluid, Endocrinolc, gy, 120 (1987) 1846 1850.

24 Walsh~ R J . . Posner, B.I.. Kapriwa, B.M. and Brawer, J.R., Pro- [aclin binding sites m the ral brain, Science, 201 (1978) 1041 1042 25 Yamada. Y., Effects of iontophoretically-applied prolactin on unit

e ctix, ity of the rat brain, Neuroendocrinolog3~. 18 11975) 263 271.