Participation of limbic-hypothalamic structures in circadian rhythm of slow wave sleep and paradoxical sleep in the rat

Brain Research, 151 (1978) 255-268 © Elsevier/North-HollandBiomedicalPress

255

PARTICIPATION OF LIMBIC-HYPOTHALAMIC STRUCTURES IN CIRCADIAN RHYTHM OF SLOW WAVE SLEEP AND PARADOXICAL SLEEP IN THE RAT

SADAO YAMAOKA Department of Physiology, Saitama Medical School, 38 Morohongo, Moroyama-cho, Iruma-gun, Saitama, 350-04 (Japan)

(Accepted December 8th, 1977)

SUMMARY The effect of brain lesion or surgical isolation of the neural circuit on SWS and PS circadian rhythm have been studied in female rats under a 14/10 light-dark schedule. Cortical EEGs and dorsal neck EMG were used to monitor SWS, PS and alertness in female rats. Intact and operated controls showed regular 4-5-day vaginal cycles and nocturnal sleep rhythm, but the night PS value on proestrus was lower than in other cycles. Following septal lesion, MPO roof cut, vaginal cycles and SWS rhythm were regularly maintained; however, the PS appearance at night, except during proestrus, increased (night PS peak). These results were similar to those for pinealectomized or ovariectomized female rats. A frontal cut of the MBH produced persistent estrus and disturbed both SWS and PS circadian rhythm. The suprachiasmaticlesioned rats showed persistent estrus and disrupted SWS rhythm, but regularly maintained the circadian PS rhythm. The vaginal cycles and SWS rhythm in the fornical-transected rats were regularly maintained, but the PS rhythm was disturbed during diestrus and showed ultradian rhythm. From these results, it is suggested that the pineal hormone and the gonadal feedback mechanisms may be involved in the night PS peak and this mechanism may involve the septal- and amygdaloid-hypothalamic systems. A different neural mechanism exist for SWS and PS circadian rhythm; SWS rhythm involves the suprachiasmatic-basal hypothalamic system and PS circadian rhythm is related, in part, to the hippocampal-hypothalamic system.

INTRODUCTION Although there is much literature concerned with circadian rhythm and the neural circuitry of hormonal activity, behavior29 and sleep-wakefulnessn, little has been known of the specific neural structures involved in mammals until quite recently. Com-

256 plete or anterior surgical deafferentation of the medial basal hypothalamus or an anterior hypothalamic lesion eliminates the circadian rhythm in the pituitary-adrenal functionT,9,10,22,2v, and more recently Moore and his co-workers described the disruption of the circadian rhythm in adrenal corticosterone 18and N-acetyltransferase activities of the pineal gland 19 following the placement of an anterior deafferentation of the medial basal hypothalamus or of a suprachiasmatic electrolytic lesion. Furthermore, Ibuka et al. reported the loss of circadian rhythm and the slight enhancement of ultradian rhythm in sleep-wakefulness patterns 11. From our previous study, a pinealectomy caused bimodal rhythm in paradoxical sleep during a 24 h period 15, and our unpublished data indicated the dissociation of the circadian rhythm in slow wave sleep (SWS) and in paradoxical sleep (PS) following returning to a light-dark environment from long-term constant illumination (3-5 months). On the basis of these findings, it could be assumed that the regulating mechanism of PS circadian rhythm is different from that of SWS circadian rhythm. Therefore, the present study was designed to investigate the difference in neural mechanisms between SWS and PS circadian rhythms by means of brain lesion and surgical isolation of the neural circuit. MATERIALS AND METHODS

Animals Forty-two female albino rats of the Sprague-Dawley strain (original source: CLEA Japan and Charles River Japan), 10-15 weeks of age and weighing 250-300 g each, were used for this experiment. All animals were bred in our animal quarters and were maintained under controlled humidity (45-55 ~), temperature (24 -4- 1 °C) and lighting conditions. The lights were on from 05:00 to 19:00. Animals were provided free access to regular laboratory food (CE-2: CLEA Japan) and water. Vaginal smears were taken every morning (before 11:00) from the vaginal opening. Animals which exhibited 3 or more consecutive 4- or 5-day estrous cycles were selected for this study.

Experimental groups There were 7 experimental groups: (1), intact control; (2), operated control; (3), septal lesion; (4), horizontal cut above the medial preoptic area at the level of anterior commissure (MPO roof cut); (5), retrochiasmatic frontal deafferentation of the medial basal hypothalamus (frontal cut of the MBH); (6), suprachiasmatic nuclei lesion (SCH lesion); (7), fornical transection. All groups contained 6 animals.

Surgical procedures The brain lesion or the surgical isolation of the neural circuit were performed 5-6 weeks before recording under pentobarbital or ether anesthesia. Electrodes for the brain lesion of special knives for surgical isolation of the neural circuits were stereotaxically inserted into the desired brain areas by using the atlas of Albe-Fessard et al. 1. Electrolytic lesions were formed by passing a DC current of 3 mA (in the septal

257

Ant t

...... - - L # "-.L__L--:.

,,;;;" i t

//''

r~

J

::.-.'.:.:. ,,
H+4

-

"---"'~"

, ",

~' " ~ ;

'

'

'

; ~J-~--j~ah~

~ C h

/A/../I

I

D

" /"

,,

I /, 1

-

.':4.h, - ~ , / , ':' ~ . ~ / ~

"

'

. --

Fig. 1. A sagittal diagram of the rat brain, indicating the locations and the extents of the deafferentation and lesions. The deafferentation and lesions. The deafferentations are indicated by thick lines and the lesions by hatched areas. A: septal lesion; B: MPO roof cut; C: fornical transection ; D: frontal deafferentation of MBH; E: suprachiasmatic lesion. Abbreviations used: n. Sept., septal nuclei; DBB, diagonal bundle of Broca; Ca, commissure anterior; Ch, optic chiasma; Fx, fornix; apom, medial preoptic area; aha, anterior hypothalamic area; pv, paraventricular nucleus; vm, ventromedial hypothalamic nucleus; ar, arcuate nucleus; HPC, hippocampus; CC, corpus callosum. lesions) or 0.5 mA (in the SCH lesions) for 30 sec through a pair of insect pin electrodes insulated with epoxylite (Epico 1500, Nihon Ushi) except for their tips under pentobarbital anesthesia (35 mg/kg body weight). The surgical isolation of the neural circuits were performed in ether-anesthetized female rats by using special knives: a modified Hal~sz type knife with a 1.5 mm radius and 2 mm height for the frontal cut of the MBH, an L-shaped knife with a 2.3 mm horizontal blade for the MPO roof cut and a small piece of razor blade 2 mm wide for the fornical transection. All brain lesions and surgical isolations were illustrated in Fig. 1. The operated control group included 3 rats bearing a small unilateral dorsal septal lesion, two rats transected at the middle part of the corpus callosum and one rat bearing a small unilateral anterior hypothalamic lesion. These lesions and deafferentations were confirmed by postmortem histological examination. The stainless steel screw electrodes for the cortical EEG and the Michel's clamp electrodes for the dorsal neck E M G were implanted 10 days before recording experiments. These electrodes were connected to a small IC connector cemented on the skull.

258

EEG PATTERNS FRONTALCORTEX

I

LIMBIC CORTEX Ill

I

II

II "r ~ r ' - ' l " r ' ~ , ~ , ~ r F'- ',- - r ' l r r

. . . . . . . . . . . . .

1T" .~-T r . ~ - ~ ~ ..~,,......

1" Fig. 2. EEG records from frontal cortex and limbic cortex and EMG from dorsal neck muscle during slow wave sleep, paradoxical sleep and alertness, at slow paper speed.

Recording procedures Twelve to twenty hours before recording, animals were transferred to individual plastic boxes (30 x 25 sq.cm and 40 cm height) in a sound-reduced and electrically shielded room with a half-silvered mirror for observation. This room was maintained under the same conditions as the colony room. Food and water were available ad libitum. The IC connectors on the animal's heads were coupled to suspended recording cables and animals were allowed to adapt to the recording conditions for 12 h. The physiological signals were recorded on a 9-channel E E G recorder with multi-purpose amplifiers (ME95-D, Nihon koden). Polygraphic records at a slow paper speed (1 mm/sec) were divided into 3 sleep-wakefulness stages according to characteristic changes in EEG, E M G and somatic activity. These were alertness (A), slow wave sleep (SWS) and paradoxical sleep (PS) (Fig. 2). In all groups, the recordings were continued for 8-10 days. The sums of the duration of each SWS and PS were expressed for each 2 h. The mean appearances of SWS and PS during the same time zone on the same sexual cycle in each group were also calculated and graphed with standard errors. At least 6 determinations were used for each point in each group. RESULTS

(1) Control groups All rats in both intact and operated control groups showed regular 4- or 5-day vaginal cycles. In bihourly mean distribution of SWS and PS, the peak periods of SWS were 06:00-08:00 and 12:00-14:00, and the through period of SWS was 20:0022:00. The peak period of PS was 12:00-14:00 or 14:00-16:00 and the trough period of PS was 04:004)6:00. A small PS peak during the dark phase was observed in the 22:00-24:00 period except on the proestrous day (night PS peak) (Fig. 3). In the control groups, both the SWS and PS values from 06:00-18:00 were significantly (P < 0.05) higher than those from 18:00-06:00 for all vaginal cycles, and both the SWS and PS values from 18:004)6:00 on the day of proestrus were significantly (P < 0.05) less than in other cycles (Table I).

30 SEC

259

INTACT FEMALE

._I•I

RAT

SLOW WAVE SLEEP "

,

l~;i

r

~L ,~'

I

0

12

SEC

18~

0

l

. '

"'

;

I

12

,

:

~,:,

i

~"

~

.

,

V.i;

0

I~

12

12

.'

'

.~

SWS

II

'

"f

~'

i,,, ]2

12

,

I

"~.-

i 'o,"i 12

9

''' ,o

]2

TIRE OF

. PA~OXICAL S~EP

LIGHT

o DAY

i i

DARK

p

~,000

1000

,

,' ".'i ~ ~',,,

i i

2OOO

,

i

=

i

i

.-

i i

i~ .,',.. ~".~"11t',

n.

J !~,'~ I~ i .,ti

i

i

D"

,"

~ i ! "¥" I', \,i/l,,~i

I

i I i

l/'

4

7

0

\A,.J h

I 0

12

12

0

12

0

TII~ OF DAY

Fig. 3. Bihourly distribution of slow wave sleep ( 0 . . . . . O) and paradoxical sleep ( 0 - 0 ) under light-dark schedule in intact female rat. Upper two graphs show an example of consecutive 8-day records. Vertical axes represent 2 h amount and horizontal axes represent time of day. Changes of vaginal smear are indicated in upper horizontal axis of SWS graph,; P, proestrus; E, estrus; DI, diestrus I; DII, diestrus II. Bottom graph shows the mean amounts of bihourly distribution of SWS and PS during same time zone on same vaginal cycles and their standard errors in 6 rats.

(2) Septal lesion All animals showed a brief period of diestrus after surgery (5-8 days) and later recovered to regular 4- or 5-day vaginal cycles. The peak periods of SWS were 06:0008:00, and 10:00-12:00 or 12:00-14:00. The trough periods of SWS were 00 :00-02 :00, 08:00-10:00 and 18:0(020:00. The peak periods of PS were 14:00-16:00 on all vaginal

260 TABLE I The effect o f estrous cycles and brain lesions on 12 h sleep amounts Estrous Slow wave sleep cycles or rat 06:00-18:00 (sec) 18:00-06:00 (sec) no.

Paradoxical sleep 06:00-18:00 (sec)

18:00~96:00 (sec)

lntactcontrol P E DI DII

259434- 873 (11)* 7312±1099 (11)** 239454- 894(11) 12617±1530(10) 261314- 906 (10) 129534-1426 (10) 259214-1058 (9) 148544-1539 (8)

4179±527 33784-302 3744±450 31414-338

(11) 6174-189 (11)** (12) 1424±309 (10) (10) 18904-306 (10) (9) 21354-327 (8)

Operated control

P E DI DII

254544-1256 246524-1112 24895£1553 261054- 993

(7) (8) (7) (8)

94074-1298 117304-1548 12184± 719 143714- 955

(7) (8) (7) (8)

3673±614 3506±372 35384-456 3396±410

(7) (9) (6) (7)

P E DI DII

262634-1383 243514-1211 248864- 796 271244-2648

(7) (10) (8) (8)

13886±2255 150714-1647 138184-1148 136114-1580

(10) (8) (8) (7)

27404-341 30194-332 3183±308 33784-411

(7) 1083i180 (9)** (9) 1934±235 (9) (7) 20774-219 (8) (8) 2014± 178 (7)

MPOroofcut P E DI DII

260324- 943 259584-1371 262584- 687 261194- 691

(6) (6) (8) (7)

120644-1203 (5) 156014- 589 (8) 158214- 728 (7) 151484- 816 (8)

2875i241 31854-525 30504-312 33743 391

(6) (6)§ (8)§ (8)

1307±237 (6)** 28684-454 (8) 2903 :L 352 (8) 2211 4- 134 (8)

Fornicalcut

273704- 668 (5) 26999± 956 (6) 267074- 344 (9) 256304-1249 (6)

139384-1401 178034-1290 16946i1245 154924- 662

36724-527 32454-370 31324-287 30284-454

(5) (6),~ (9)§ (6)§

7684- 38 (6)** 2971 4-149 (10) 2943±326 (6) 2563 4- 336 (6)

Frontalcutof 6 MBH 8 13 73 74 75

238434-1185 (7)§ 247244- 646 (7)§ 251004- 606 (6) 211744-1343 (7)§ 231084- 480 (8) 161584-1547 (8)§

212534- 599 (8)*** 233684- 843 (8)*** 201134- 830 (6)*** 190034- 405 (7)*** 189844- 766 (8)*** 165934-1896 (8)

27594-250 (7)~ 25474-289 (7)§ 24734-230 (6)§ 25174-229 (7)§ 29024-240 (8) 1580~ 229 (8)§

SCH lesion

206354- 203 (8) 237644- 219 (8) 200734- 442 (7)

170584- 287 (8)*** 192554- 479 (8)*** 173884- 404 (8)***

4920_£ 173 (8) 26104- 137 (8)*** 36854-211 (8) 2 8 2 9 i 158 (8)*** 47794-181 (7) 29634-438 (8)***

198794-1093 (7)§

20081 4- 858 (8)***

31994-278 (7)§ 3211 4-325 (8)***

Septallesion

P E DI DII

48 50 51

SCH + MPO lesion 68 * ** *** §

(6) (7) (7) (5)

4394- 46 (8)** 1154--220 (8)

1476±226 (7) 18404-437 (7)

28364-2_320 (8)*** 2939 ± 155 (8)*** 37184-656 (6)*** 3539±439 (8)*** 2210i 169 (8) 20304-227 (8)

Mean ± standard error (days). Differs significantly from other cycles, P < 0.05. Differs significantly from operated control, P < 0.05. No significant difference between 06:00-18:00 and 18:00-06:00.

cycles a n d 2 2 : 0 0 - 2 4 : 0 0 o n the days e x c e p t i n g those o f proestrus. O t h e r t h a n d u r i n g p r o e s t r u s , these p e a k values o f PS b e t w e e n 1 4 : 0 0 - 1 6 : 0 0 a n d 2 2 : 0 0 - 2 4 : 0 0 were n o t significantly different. These n i g h t PS peaks were h i g h e r t h a n those i n c o n t r o l g r o u p s (Fig. 4). T h e 12 h values o f S W S a n d PS in this g r o u p s h o w e d the similar results o b t a i n e d in c o n t r o l g r o u p s ( T a b l e I).

261

Fig. 4. Bihourly distribution of slow wave and paradoxical sleep in septal-lesioned and MPO roof cut female rats. Mean distributions with standard errors were indicated in right graphs. The arrows in the photomicrographs showed the lesions and the cut.

262

(3) MPO roof cut All rats exhibited a regular 4-day vaginal cycle after recovery from surgery. The peak periods of SWS were 06:00-08:00 and 14:00-16:00. The morning SWS peak was higher than the afternoon SWS peak. The PS value from 00 :00-02 :00 except on the proestrous night showed significantly (P < 0.05) higher values than those of the septallesioned group and did not show any significant difference from the PS value during 22:00-24:00 (Fig. 4). Therefore, the night PS appearance in this group was more than in the septal-lesioned group. The 12 h PS value from 06:00-18:00 and from 18:0006:00 did not show any significant difference on the day of estrus and diestrus 1 (Table I). (4) Frontal cut of the M B H After deafferentation, vaginal cycles showed persistent vaginal cornification in 4 rats, and the remaining two rats showed persistent diestrus. The bihourly SWS distribution showed irregular fluctuations and 3-5 peaks were observed during 24 h (ultradian rhythm); the peak intervals were 4-8 h. The bihourly PS distribution also showed irregular fluctuations, but the number of peaks in 24 h was less than for SWS; peak intervals ranged from 4 to 40 h and the time zone of the PS peaks was independent of the light,lark illumination schedule (Fig. 5). Twelve hour values of both SWS and PS did not show any significant difference between the periods of 06:0018:00 and 18:004)6:00 (Table I). (5) SCH lesion Three out of 6 rats had total, localized and bilateral lesions on the suprachiasmatic nuclei; two rats had small unilateral, partial SCH lesions; and one rat had bilateral large anterior hypothalamic lesions. Three SCH-lesioned rats and one large anterior hypothalamic-lesioned rat showed persistent vaginal cornification. The bihourly SWS distribution in SCH-lesioned rats showed irregular fluctuations and multiple peaks during the 24 h period. The SWS peak interval was 4-8 h. However, the bihourly PS distribution showed a circadian rhythm similar to those of control groups. The peak period of PS in SCH-lesioned rats was 12:00-14:00 or 14:00-16:00 (Fig. 5). The large anterior hypothalamic lesion completely disrupted both circadian SWS and PS rhythm and exhibited ultradian rhythm with a 4-10 h peak interval (Fig. 5). (6) Fornical transection The vaginal cycles in all 6 rats showed prolonged diestrus for 8-15 days after surgery and later recovered to regular 4-5-day cycles. The SWS circadian rhythm in this group showed a similar distribution pattern to that in control groups. However, the bihourly PS distribution except on proestrus showed ultradian rhythms with 2-4 peaks. The PS peak intervals on diestrous days were 4-8 h (Fig. 6). The SWS values from 18:004)6:00 were significantly (P < 0.001) less than that from 0.6:00-18:00; however, no significant difference in the PS value in the period of 06:00-18:00 and 18:00436:00 was observed except on proestrus (Table I).

263

FRONTAL

DEAFFERENTATION OF MBH

TIR[ ~ NY

•.

+-

-

(

L15i~I

o o IIIII

q

S

).2

0AN 1.6

20 2li TIME OF DAy

LESION OF SCH & ~

s~c

12

0

l

12

0

1~

6

12

O

'12

6

!

12

~

I

12

6

,,

Fig. 5. Bihourly distribution of slow wave and paradoxical sleep in rats bearing frontal deafferentation of MBH, SCH lesion and large anterior hypothalamic lesion (SCH and MPOA). Left graphs indicate the sequential data of sleep records. Right graphs show the mean amounts with standard error of bihourly SWS and PS distribution. The arrows in the photomicrographs showed the lesions and the cut.

264

FORNICAL

TRANSECTION slowWAVESLEEP I

p.

I ]P,I

"

I

LIBgT

DARK I

~A

i:~.

t

"

'l

,,,V

12

0

"~.

' ~;

I

1

|

I

*

liar # FX-6 '

I./':

I

.,'~

,,

'

i

TIREIOF DAY

,,,,-', ..

i,,~

,J V t

: ,'

: I:'/

| ~'

l

:~

0

I

l

I

L. I:

12

tt,,

¢.i~

~'~N.Wq"~CI~ '//

RAT # FX..-tl

SLOWWAVESLEEP

i ~

I t.~

I

;..

: Ii

sec 12

'

I A.x

'I~. "

,

I

'

~\

PAP~X~CALS,L~.P

=/L ,

|

,

'

:W

12

0

:

'

12

,,

.

!

,

',

l~iJi

!i;

i

~ I;

I~.,

"::

/

: ,

t

,

,,

",~

,

I

E,~ °'~'i '"V'

12

0

12

0

12

0

1~)OI ~: / ~ .I]PARADOXI ~ ~,SLEEP CALi J. /~~- '~'] Jj~ TI~.OFnAY Ji LIeIT

sws

J

$EC

5000"

/'(i

if

qO(O)"

" 1

,

30002O00

-!

10(0)

I

0

i

0

!

LIGHT ~ I ~ I j R K 12

0

12

PARADOXICASI LFEP 12

o

TIMOF ED12AY

Fig. 6. Bihourly d i s t r i b u t i o n of s l o w w a v e and paradoxical sleep in fornical-transected female r a t s . U p p e r g r a p h s s h o w the sequential d a t a o f sleep r e c o r d s in two rats. B o t t o m g r a p h s h o w s the m e a n a m o u n t s with s t a n d a r d errors. The s c h e m a illustrates the extent of fornical transection.

265 DISCUSSION (1) Control female rats on a light-dark schedule Data taken on the circadian changes in sleep-wakefulness patterns of female rats by long-term EEG recording have been reported by a few investigators2,3,6,15,96,34. Our results on SWS and PS circadian rhythm in female rats are mainly in agreement with the findings reported by Colvin et al. 3. Our previous studyn,15 confirmed their results. However, the PS appearance during estrous afternoon is not significantly higher than for other cycles. Further analyses in the present study showed two interesting phenomena: a morning SWS peak (06:00-08:00) without a PS increment and a night PS peak (22:00-24:00) except proestrus. The mechanisms of the first phenomenon will be discussed in future papers in relation to the sexual difference and other factors. The night PS peak may correlate with the gonadal feedback mechanism and the limbic-hypothalamic neural circuit. This point will be discussed partly in this paper. (2) Septal lesion Seggie et al. 25 postulated that septal lesions in rats did not alter the circadian rhythm of plasma corticosterone, but did enhance the adrenal response to stimulation. Our previous studies indicated that intraseptal injection of dexamethasone eliminated both PMS-induced ovulation and diurnal rhythm of plasma corticosterone level on the day following the injection5, and that spontaneous septal unit discharges showed diurnal variationas. Therefore, the septal complex is involved in both adrenal and gonadal activities. In this study, the morning SWS trough was deeper than controls. This may reflect the fact that the septal lesion enhanced response to the stress from taking smears. The night PS peaks were higher than those in controls. A similar phenomenon has been observed in pinealectomized ratslS, 2x, and ovariectomized rats (unpublished data). The pineal body is suspected to have antigonadal activity23 and ovariectomy causes hypergonadotropin secretion due to a loss of negative feedback of gonadal hormones. Therefore the phenomenon of an increased night PS peak in rats with septal lesions may be related to the feedback mechanism of gonadal hormones. (3) MPO roof cut The MPO roof cut did not alter vaginal cycles, but suppressed gonadotropin secretion in adult rats 14, and prevented spontaneous and induced ovulation in acute state 13. The stimulatory effect of estrogen on ovulatory gonadotropin release was replaced by local implantation into the medial amygdala, the bed nucleus of stria terminals or the arcuate nucleus, but not into the MPO 12. In the present study, the MPO roof cut showed the increase in the night PS peak values as similar to the septal lesion. The night PS peak values in this group were more than that in the septaMesioned rats. The 12 h PS values between day and night were not significantly different. Therefore, if the night increase of PS is related to the feedback mechanism of gonadal hormones as mentioned above, it is possible to con-

266 sider that the phenomena in the MPO roof cut animals may be augmented by the isolation of the gonadal feedback pathways through the septal complex and through the amygdaloid complex.

(4) Frontal cut of the MBH Several investigators have demonstrated that surgical interruption of the connections to the MBH disrupts normal circadian rhythmicity in the pituitary adrenal function 4,7,9,~°,32, the circadian rhythm of N-acetyltransferase activity 19 in the pineal body. In our present study, the circadian SWS rhythm was disrupted after frontal deafferentation of the MBH and showed ultradian rhythm. The PS rhythm also demonstrated irregular fluctuations, but did not resemble the SWS rhythm. The PS peak intervals fluctuated from 4 to 40 h and the peak periods were independent of the light-dark schedule. Therefore, the frontal connection of the MBH may be involved in the mechanism of the circadian SWS rhythm and in synchronization mechanism of the PS circadian rhythm with the light-dark schedule. (5) SCH lesions Many investigators have reported that the SCH nucleus is involved in the circadian rhythm of corticosteronO s, N-acetyltransferase activity 19, behaviour 29 and in SWS and PS 11 as mentioned in the introduction. In our present study, the SWS circadian rhythm in the SCH-lesioned rats was eliminated, but the PS circadian rhythm was not disrupted. The findings of Ibuka et al. H are not in agreement with our present results. This difference may be due to the lesion size, since a large anterior hypothalamic lesion completely disrupted both circadian SWS and PS rhythm. Our lesion criteria were almost the same as those of Moore et al. ~9. Therefore, it may be suggested that the SCH nucleus participates in the regulation of circadian SWS rhythm, but not in the circadian PS rhythm. The anatomical retinohypothalamic connections were discussed by several investigators2°,zs, ~0, and Sousa-Pinto and Castro-Correia 2s postulated that the optic afferent to the hypothalamus terminated not only the SCH but also various hypothalamic nuclei. Sawaki 24 also demonstrated that SCH and MPO neurons were influenced by visual stimulation. Therefore, from the present results of the frontal cut and the SCH lesion, it is suggested that the synchronizing mechanism of PS rhythm with the light-dark schedule may exist in extrasuprachiasmatic anterior hypothalamic area. (6) Fornical transection Moberg et a1.17 reported that diurnal ACTH rhythm was eliminated after fornica I transection. Lengv~ri and Hal~sz 16 showed that the ACTH rhythm was eliminated temporarily and resumed 3 weeks after transection. Wilson and Critchilow 31 demonstrated a negative result for the disruption of ACTH rhythm following fornical transection. In our study, the fornical transection partially disturbed the circadian PS rhythm on diestrus, but not on proestrus and estrus, and that, in contrast, the circadian SWS rhythm was regularly maintained in all vaginal cycles. From these results, we feel that the hippocampal-hypothalamic tract may be related to the regulation of circadian

267 PS rhythm. On the other hand, the fornical-transected rats are still intact in other brain structures, and maintain their regular vaginal cycle and circadian SWS rhythm. Since these periodic activities may influence the PS rhythm in fornical-transected rats, the PS rhythm on proestrus and estrus may show apparently regular circadian activities. As noted above, the circadian SWS and PS rhythm are controlled by different neural circuitry. The SCH nucleus and the SCH-MBH tract may participate, playing important roles in circadian SWS rhythm. However, the SCH nucleus is not involved in circadian PS rhythm. The regulation of the circadian PS rhythm is related to the limbichypothalamic neural circuit and the gonadal-limbic-hypothalamic feedback mechanisms. The hippocampal-hypothalamic tract may be one of the regulating circuit for the circadian PS rhythm, and the tract of the extra-SCH area to the MBH is involved in the synchronizing activity of circadian PS rhythm with light-dark schedule. The septopreoptic or the amygdaloid-preoptic tracts may modulate the circadian PS rhythm in relation to the gonadal feedback mechanisms. ACKNOWLEDGEMENTS The author wishes to express his gratitude to Prof. M. Kawakami, Department of Physiology, Yokohama City University School of Medicine and Prof. K. Uyemura, Department of Physiology, Saitama Medical School, for their valuable advice and encouragement. The author also thanks Misses H. Suzuki, M. Niizeki and K. Horie for their technical assistance. This investigation was supported in part by grants from the Ministry of Education of Japan and the Kudo Science Foundation.

REFERENCES 1 Albe-Fessard, D., Stutinski, F. et Libouban, S., Atlas StHdotaxique du Diencdphale du Rat Blanc, Editions C.N.R.S., Paris, 1971. 2 Branchey, M. and Branchey, L., Sleep and wakefulness in female rats during pregnancy, PhysioL Behav., 5 (1970) 365-368. 3 Colvin, G. B., Whitmoyer, D. I., Lisk, R. D., Walter, D. O. and Sawyer, C. H., Changes in sleepwakefulness in female rats during circadian and estrous cycles, Brain Research, 7 (1968) 173-181. 4 Greer, M. A., Panton, P. and Allen, C. F., Relationship of nycthemeral cycles of running activity and plasma corticosterone concentration followingbasal hypothalamic isolation, Hormon. Behav., 3 (1972) 289-295. 5 Hagino, N. and Yamaoka, S., Neural integration of ACTH adrenal rhythm and ovulatory cycles in rat, Abstr. 1Vth lnt. Congr. EndocrinoL, Washington D.C., 1972, p. 2. 6 Hagino, N. and Yamaoka, S., Role of biorhythm of sleep and wakefulness in the periodicity of gonadotropin secretion in the baboon (non-human primate). In M. Kawakami (Ed.), Biological Rhythms in Neuroendocrine Activity, Igaku Shoin, Tokyo, 1974, pp. 326-337. 7 Halfi.sz, B., Possible role of the hypothalamic hypophysiotrophic area in control of rhythmic functions of the anterior pituitary. In M. Kawakami (Ed.), Biological Rhythms in Neuroendocrine Activity, Igaku Shoin, Tokyo, 1974, pp. 39-53. 8 HalAsz, B. and Pupp, L., Hormone secretion of the anterior pituitary gland after physical interruption of all nervous pathways to the hypophysiotrophic area, Endocrinology, 77 (1965) 553-562. 9 Halfisz, B., Slusher, M. A. and Gorski, R. A., Adrenocorticotrophic hormone secretion in rats after partial or total deafferentation of the medial basal hypothalamus, Neuroendocrinology, 2 (1967) 43-55.

268 10 Hakisz, B., Vernikos-Danellis, J. and Gorski, R. A., Pituitary ACTH content in rats after partial or total interruption of neural afferents to the medial basal hypothalamus, Endocrinology, 81 (1967) 921-924. 11 Ibuka, N., Inoue, S. T. and Kawamura, H., analysis of sleep-wakefulness rhythm in male rats after suprachiasmatic nucleus lesions and ocular enucleation, Brain Research, 122 (1977) 33-47. 12 Kawakami, M., Kimura, F. and Konda, N., Role of forebrain structure in the regulation of gonadotropin secretion. In T. C. Anand Kumar (Ed.), Neuroendocrine Regulation of Fertility, Kager, Basel, 1976, pp. 101-113. 13 Kawakami, M. and Terasawa, E., Acute effect of neural deafferentation on timing of gonadotropin before proestrus in the female rat, Endocr. Jap., 19 (1972) 449459. 14 Kawakami, M. and Terasawa, E., Role of limbic forebrain structures on reproductive cycles. In M. Kawakami (Ed.), Biological Rhythm in Neuroendocrine Activity, lgaku Shoin, Tokyo, 1974, pp. 197-219. 15 Kawakami, M., Yamaoka, S. and Yamaguchi, T., Influence of light and hormones upon circadian rhythm of slow wave and paradoxical sleep. In S. ltoh, K. Ogata and H. Yoshimura (Eds.), Advances in Climatic Physiology, Igaku Shoin, Tokyo, 1972, pp. 344-366. 16 Lengv~ri, I. and Hakisz, B., Evidence for a diurnal fluctuation in plasma corticosterone levels after fornix transection in the rat, Neuroendocrinology, 11 (1973) 191-196. 17 Moberg, G. P., Scapagnini, U., De Groot, J. and Ganong, W. F., Effect of sectioning the fornix on diurnal fluctuation in plasma corticosterone levels in the rat, Neuroendocrinology, 7 (1971) 11-15.

18 Moore, R. Y. and Eichler, V. B., Loss of a circadian adrenal corticosterone rhythm following suprachiasmatic lesions in the rat, Brain Research, 42 (1972) 201-206. 19 Moore, R. Y. and Klein, D. C., Visual pathways and the central neural control of a circadian rhythm in pineal serotonin N-acetyltransferase activity, Brain Research, 71 (1974) 17-33. 20 Moore, R. Y. and Lenn, N. J., A retinohypothalamic projection in the rat, J. comp. NeuroL, 146 (1972) 1-14. 21 Mouret, J., Coindet, J. et Chouvet, G., Effet de la pin6alectomie sur les 6tats et rythmes de sommeil du rat male, Brain Research, 81 (1974) 97-105. 22 Palka, Y., Coyer, D. and Critchlow, V., Effect of isolation of medial basal hypothalamus on pituitary-adrenal and pituitary-ovarian functions, Neuroendocrinology, 5 (1969) 219-255. 23 Reiter, R. J., Comparative physiology: pineal gland, Ann. Rev. Physiol., 35 (1973) 321-333. 24 Sawaki, Y., Retinohypothalamic projection: electrophysiological evidence for the existence in female rats, Brain Research, 120 (1977) 336-341. 25 Seggie, J., Shaw, B., Uhlir, I. and Brown, G. M., Baseline 24-hour plasma corticosterone rhythm in normal, sham-operated and septally-lesioned rats, Neuroendocrinology, 15 (1974) 51-61. 26 Sieck, G. C., Ramaley, J. A., Harper, R. M. and Taylor, A. N., Sleep-wakefulness changes at the time of puberty in the female rat, Brain Research, 116 (1976) 346-352. 27 Slusher, M. A., Effects of chronic hypothalamic lesions on diurnal and stress corticosteroid levels, Amer. J. Physiol., 206 (1964) 1161-1164. 28 Sousa-Pinto, A. and Castro-Correia, J., Light microscopic observation on the possible retinohypothalamic projection in the rat, Exp. Brain Res., 11 (1970) 515-527. 29 Stephan, F. K. and Zucker, I., Circadian rhythm in drinking behavior and locomotor activity of rats are eliminated by hypothalamic lesions, Proc. nat. Acad. Sci. (Wash.), 69 (1972) 1583-1586. 30 Szent~.gothai, J., Flerk6, B., Mess, B. and Hal~sz, B., The Hypothalamic Control of the Anterior Pituitary, 3rd Ed., Academia Kiado, Budapest, 1968, p. 59. 31 Wilson, M. and Critchlow, V., Effect of fornix transection or hippocampectomy on rhythmic pituitary-adrenal function in the rat, Neuroendocrinology, 13 (1973/74) 29-40. 32 Wilson, M. and Critchlow, V., Absence of a circadian rhythm in persisting corticosterone fluctuations following surgical isolation of the medial basal hypothalamus, Neuroendocrinology, 19 (1975) 185-192. 33 Yamaoka, S. and Hagino, N., Spontaneous septal neuron activity in the rat, Brain Research, 67 (1974) 147-152. 34 Yokoyama, A., Ramirez, V. D. and Sawyer, C. H., Sleep and wakefulness in female rats under various hormonal and physiological conditions, Gen. comp. Endocr., 7 (1966) 10-17.