Activity of suprachiasmatic and hypothalamic neurons during sleep and wakefulness in the rat

Brain Research, 419 (1987) 279-286 Elsevier

279

BRE 12823

Activity of suprachiasmatic and hypothalamic neurons during sleep and wakefulness in the rat S.F. Glotzbach, C.M. Cornett and H.C. Heller Department of Biological Sciences, Stanford University, Stanford, CA 94305 (U.S.A.) (Accepted 3 February 1987) Key words: Suprachiasmatic nucleus; Hypothalamus; Unit activity; Arousal state

Single unit activity was recorded from the suprachiasmatic nucleus (SCN) and preoptic/anterior hypothalamus (POAH) of unrestrained Wistar rats during sleep and wakefulness. Regularly firing cells, which are abundant in in vitro SCN preparations and have been considered the basis of a central neuronal oscillator, were conspicuously absent in this preparation and in other in vivo studies. Most of the 55 cells recorded in the SCN and POAH were characterized by spontaneous firing rates below 12 Hz and with heterogeneous patterns of changes in frequency with arousal states. In vivo neurophysiologicalstudies of the SCN in which the anesthetic agent urethane is used should consider the effect of different levels of arousal, as indicated by the cortical EEG, in evaluating the relationship between sensory stimulation and single unit activity. INTRODUCTION The suprachiasmatic nuclei (SCN) of the hypothalamus have been the focus of much research in recent years and appear to have a prominent role in the generation and organization of circadian rhythmicity in mammals (see refs. 25, 34 and 48 for extensive reviews). Although lesion 13,21,26,44, electrical stimulation 33 and autoradiographic 31,32,38,39,41 techniques have been invaluable in documenting the importance of the SCN in circadian rhythm control, neurophysiological investigations are necessary to provide a framework for describing cellular mechanisms underlying both the generation of endogenous rhythms and the entrainment of rhythmic activity by the environment. Several electrophysiological approaches have been successful in identifying characteristics of SCN cells. Studies in anesthetized animals have shown that a group of SCN units, particularly in the retinohypothalamic termination in the caudal and ventrolateral portion of the SCN, respond to light or electrical stimulation of the optic nerve and can be considered 'visual'. Features of visual SCN units include

(a) a majority of excitatory vs inhibitory responses, (b) predominant contralateral activation, (c) tonic responses to constant illumination with little adaptation for both light-activated and light-suppressed cells, and (d) very large receptive fields with an apparent absence of an antagonistic center-surround organization characteristic of chiasmatic fibersl0-12,20,28,36. In vitro slice preparations have permitted analysis of pharmacological and biophysical properties of SCN n e u r o n s 19,42,43,46,47,49 and, in addition, have revealed the existence of anatomically distinct cell types such as regularly firing, bursting, oscillating, and irregularly firing neurons 8,43. SCN single units also show a circadian rhythm of firing rate in vitro and in vivo when results are pooled from different preparations with respect to their entrained clock time 7'9'43. Multiple unit activity (MUA) recordings from both anesthetized and unanesthetized rats have also demonstrated a robust circadian rhythm of SCN MUA which persists in constant conditions and is 180° out of phase with M U A rhythms in areas outside of the S C N 14'15'22. This 'phase-reversal' of M U A may be characteristic of nocturnal species since M U A

Correspondence: S.F. Glotzbach, Department of Biological Sciences, Stanford University, Stanford, CA 94305, U.S.A. 0006-8993/87/$03.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)

280 rhythms of diurnal species appear to be in phase 35 Despite these advances, there are several reasons why single unit recordings from the SCN and preoptic/ anterior hypothalamus (POAH) in unanesthetized. behaving animals are an essential adjunct to results obtained utilizing M U A . in vitro, and anesthetized preparations. First. anesthetic agents have been shown to abolish the temporal periodicity of regularly firing SCN cells in vitro 46 and therefore preparations involving anesthesia may not give an accurate indication of the percentages of cell types in the SCN and their neuronal characteristics. Second, unanesthetized preparations in which behavioral and physiological mechanisms are intact are necessary to evaluate the functional significance of the different cell types that have been found in the SCN. Since lesmns of the SCN eliminate the circadian organization of sleep and wakefulness 1'13. it is possible that certain SCN cells will show strong correlates to sleep cyclicity. Third, comparisons of single unit recordings from the rat SCN and adjacent hypothalamus may provide a neuronal basis for the phase-reversal in M U A seen in these two loci. Finally, single unit recordings from the adjacent P O A H are of interest considering the possibility of P O A H involvement in circadian body temperature control and evidence suggesting that the functional SCN extends beyond its Nissl-defined borders into the P O A H 3'14"27, MATERIALS AND METHODS

Preparation of electrodes Bundles of 4 etched microwires were used for extraceUular recording of SCN and hypothalamic single unit activity. The procedures for fabricating etched microwires, which combine the advantages of the small tip of a conventional microelectrode with a flexible shaft that permits recording in freely behaving animals, have been previously discussed 4'5. Stainless-steel wire (S.S. 316, 37.5 ~m diameter; California Fine Wire) is cut to 6-cm lengths, one end of which is electrolyticatly etched in 4 N HCI to produce a tip diameter of about 2 .urn. The other end is soldered to a gold-plated contact (Amphenol 220-P02). The electrode is then insulated with a 3-/~m coat of vapor-deposited Parylene (Union Carbide), a xylylene polymer, after a preliminary treatment with an organic silane (Z-6030 Silane; Dow-Corning) to im-

prove the bond between the metal and the Parylene. Selective removal of Parylene from the electrode tip is accomplished by DC arcing. The use and advantage of Parylene as a microelectrode insulator has been discussed in detail 24"37

Animals and surgery Male Wistar rats weighing 250-350 g were used in this study. Animals were housed in an animal room under a 12 L:12 D photoperiod (lights on at 07.00 hi and allowed ad libitum access to food and water. The animals were anesthetized with sodium pentobarbital (Nembutal, 50 mg/kg) or an anesthetic mixture consistmg of Ketamine. 50 mg/kg, Acepromazine. 10 mg/kg, and Rompum. 5 mg/kg, and placed into a Kopf stereotaxic instrument. A microdrive was glued inside a Plexiglas block with its concentric cannulae projecting through the bottom. The microdrive assembly was then positioned stereotaxically so that the cannulae entered the brain on midline 1.3 mm caudal to bregma, and was lowered so that the top of the inner cannula came to rest 5 mm dorsal to the SCN. Either at the time of surgery or the day before the first recording session, a bundle of 4 etched microwires (Z = 1-~8 Mfl), protected by PE 10 tubing, was inserted through the inner cannula so that the initial position of the tips was dorsal to the SCN. Screw electrodes for recording the cortical electroencephalogram (EEG) also served to anchor the microdrive assembly to the skull. Teflon-coated stainlesssteel wire pairs (Medwire S.S. 3t6 5T) were used to record electromyographic (EMG) activity from the neck. Leads from microwires, EEG. and E M G electrodes terminated in Amphenol connector strips. which were glued to the sides of the microdrive assembly. A Plexiglas shell housed and protected the microdrive and microwire bundle. A lateral X-ray at the time of surgery provided preliminary information on microwire position, Furacin (Morton-Norwich) was applied topically after surgery and animals were allowed a 7-10-day postoperative period before being used in experiments.

Protocol and data analysts During recording sessions, the animals were in a small Plexigtas box (16 x 13 x 13 cm) which made visual observation possible. The floor of the box was warmed (30-32 °C) by a water-perfused heating pad

281 to facilitate sleep. Animals were allowed to sleep in the box for 6 h on at least 3 occasions prior to the collection of data. At the time of the experiment, the animal was connected to cables for recording single unit and E E G / E M G activity. Signals from the unit cable (Filotex) were passed through Grass High Impedance probes to Grass P511J preamplifiers (300-10 kHz filtering) and Mentor N-750 spike analysers. Spikes were viewed on a Tektronix 5113 storage oscilloscope and recorded on a Vetter Model A analog tape recorder. Frontal-occipital E E G , EMG, and discriminator outputs were recorded on a Grass Model 7 polygraph at a chart speed of 6 mm/s. A BCD time code generator synchronized records on the polygraph and tape. During an experiment, which typically lasted 30-90 min, the microdrive was advanced in approximately 50-pm increments until single units (>3:1 S/N) were isolated. Most signals were between 100 and 150/xV on a background of 25-40pV. Spike trains were digitized by a PDP 11/34 computer and plotted (HP 7221A) as a function of arousal state. Arousal state was scored according to standard criteria 6, and only samples containing homogeneous segments of wakefulness (AW), nonREM (NREM), and REM were analysed. The diversity of arousal state-related discharge profiles in these units accounts for the large standard deviaton when firing rates of all cells from each area are pooled (Table I). Following the experiments, a small current was passed through one or more of the electrodes to deposit iron in the tissue around the electrode tip. Twenty-micrometer serial sections were taken with a Cryo-Cut 2 (American Optical) microtome and processed through a potassium ferrocyanide solution to produce the Prussian blue reaction with the deposited iron. RESULTS Fifty-five cells were recorded from 24 animals. The locations of the recorded cells, as determined by the positions of the microdrive and blue spot, are depicted in Fig. 1. Sixteen cells were localized within or very near the SCN and were classified as SCN cells; the remaining 39 cells were located outside of the SCN area in the surrounding hypothalamus and were classified as 'extra-SCN' cells. There is evidence that the 'metabolic' borders of the SCN may extend out-

side of the Nissl-defined boundaries14'27. Table I summarizes the distribution of SCN/extraSCN cells in different frequency ranges for each arousal state. Some cells were recorded only while the animal was awake and several were lost before data could be collected over an entire sleep cycle; therefore, the number of cells for each arousal state differs. Almost 90% of the cells recorded during AW had firing rates < 12 Hz, and the frequencies of a majority of these cells were < 4 Hz. The changes in mean firing rate with state for cells < 12 Hz are shown in Fig. 2, and for all SCN cells in Table I1. Of the 37 cells in which at least two arousal states were recorded, 26 (70%; 5 SCN, 21 extra-SCN) showed at least a two-fold change in firing rate between AW and NREM, N R E M and REM, or AW and REM. Two SCN and 3 extra-SCN cells were characterized by increasing firing rate > 2× in REM compared to NREM, while 3 SCN and 9 extra-SCN cells decreased firing rate two-fold or greater in REM compared to NREM. Fig. 2 also suggests that cells with higher firing rates were more state dependent than cells with lower firing rates. Several cells in both the SCN and surrounding hypothalamus showed a dramatic inhibition or complete cessation of firing rate during REM sleep, even during REM-specific phasic activity. Cell no. 301.3, located in the SCN approximately 0.2 mm off of midline and 0.1 mm dorsal to the optic chiasm, was the most profound 'REM-off' cell in this study, decreasing from a spontaneous firing rate of about 35 Hz in N R E M to 0 Hz in R E M sleep. Overall trend analysis indicated that 15 cells (4 SCN; 11 extra-SCN) increased while 18 cells (6 SCN; 12 extra-SCN) decreased firing rates in REM sleep compared to NREM. Many other cells showed increased and decreased firing rates in NREM compared to AW and REM, and one extra-SCN cell (W18.2) was completely inhibited during NREM. This 'NREM-off' cell was located in the medial preoptic area (at A 8.2 mm), 0.5 mm off midline and 1 mm dorsal to the optic chiasm. Cells with invariant interspike intervals (regularly firing cells) which have been reported in the SCN and surrounding hypothalamus in vitro and in vivo were not seen in this study. DISCUSSION It is clear from a variety of studies that the SCN has

282 -

,4 mt ) . . . . v\

x

\

- - -\ ~

/

.

.' I

R~l'----

.~/ .

.

/

--T-"-.

.

.

{

.

kl

"-

(

J

~%

i

,~.j

•

o.'-'.

]

-'"

V M. --""

%(

~..,

,' . . . .

37

. i"-

.

(+1 mfb

/ I'

it

\\~

_ -

t

I\

~

_

LH

'e(' "• I , AHy

'I ]

4.', .ipe %.~__

'

,

.,'

•

f--.,,

~

I1" 't

, . I.

,.,

PaOC. . . .

,,

,'--~

/

I!

~1

i ! ,

,% ,,

,

QY .--,

/ i { ',,Re #' i/

"

//

mfb

N I

3,

,-

----

i

",

¢

I

'i BS T PO -

~ ltltJk,_" • I'1 l"lt

•

\

I

I i ~°

,,'

/

LH

t

AHy

klPl, Jt

,_

. - _.

t

.,,,,,..< ..=,,

.."

,

F

. __

RCh

A 7.2

L/ f )

l

A 7.7

t BAC~'BSTM

V F

ac

,

,

,"4 mch[lll, I ~. I,I~ I I

,

mfb

/

'l.I

ll~pe

/

"-,

,'

I

,

/

-~". tACk . . . . .

:1!

&

'

....

/

v,

, ,,,o

,,

,

..

I

.

MPO '~ ii/

~PSCh

~ _ f - - ~

~.,~

l

I

e

/

/

,,

BSTL

' V~ ,~SHY ~ ~

]

|\

i1: .,,.!

/

/

,,i

I

"~"

MnPO

'..~

.~.

LPO

' _

- ~1

" " --

_ _

..

f

A 8.2

A 8.7

Fig. 1. Location of recorded units in the SCN and adjacent POAH. 3V, third ventricle; AC, anterior commissure: AHy, anterior hypothalamus; BSTPO, bed n. stria terminalis; LPO, lateral preoptic area; MnPO, median preoptic n.; MPO, medial preoptic area: ox. optic chiasm; PaPC, paraventricular n.; Pe. periventrieular n.: PSCh, preoptic suprachiasmatic n.; RCh. retroehiasmatic area: Re. reuniens thalamic n.; SCN. suprachiasmatic n.

TABLE I

Distribution of SCN/extra-SCN cells by firing rate Firing rate (Hz, mean +_ S.D. ): SCN. 7.5 (+ 12.8); extra-SCN. 4.81_ 8.1).

Range (Hz)

Extra-SCN

SCN AW

NREM

REM

AW

NREM

REM

9 (60%) 2 2

5 1 1

6 1 1

28 (76%) 2 3

25 3 0

21 3

Total 0-12 >12

13 (87%) 2 (13%)

7 3

8 2

33 (89%) 4 (11%)

28 0

25 0

n

15

10

10

37

28

25

0-4 4-8 8-12

1

283 12.

12. SCN CELLS (F.R. | ^W < 12 HZ)

10.

10.

/ 8.

8.

8.

6,

4.

4.

-T>c__) Z L~ C~ LL] C~ LJ_

2.L_

\

~I--"~ ^W

I NREM

I REM

2.

0. ^W

NREM

REM

STATE Fig. 2. Changes in firing rates of SCN and extra-SCN cells during wakefulness (AW), non-REM (NREM) and REM sleep.

TABLE

a crucial role in the generation and synchronization

II

of m a m m a l i a n circadian rhythms. However, several

SCNcells (n = 16) Cell no.

WR4.1 W12.1 W12.2 W12.3 W12.4 W15.1 W15.2 301.1 301.2 301.3 301.4 301.5 303.1 304.1 304.2 308.1

lines of evidence indicate that areas outside of the Firing rate (Hz) AW

NREM

REM

4.7 2.7 2.6 10.1 2.9 0.2 2.4 7.1 0.6 49.9 8.4 19.1 0.2 0.9 0.8

3.7 1.3 2.1 6.1 34.9 12.7 32.3 16.4 0.4 1.2 -

9.7 0.5 1.7 5.1 0.0 1.3 46.7 12.4 1.1 1.4 -

SCN, especially in the surrounding P O A H , may also play important roles in circadian rhythm control. First, the P O A H may be involved in the circadian rhythm of body temperature since the P O A H is an important c o m p o n e n t in neural regulation of body temperature 5 and circadian rhythms of body temperature have been reported to persist after total bilateral distruction of the SCN 3. Second, Inouye and Kawamura TM studied M U A rhythms in rats and found that rhythms of M U A recorded from inside the SCN were 180 ° out of phase from M U A rhythms outside of the SCN. In Fig. 7 of their paper it is apparent that some electrode locations which were just outside the Nissl-defined boundaries of the SCN showed M U A rhythms in identical phase with locations within the

284 SCN, suggesting that the functional SCN extends beyond its Nissl-defined borders into the adjacent P O A H . Third, cytochrome oxidase studies also indicate a disparity between metabolic and Nissl boundaries of the SCN 27. Investigations of P O A H unit activity during sleep have shown changes in discharge rates of P O A H neurons with changes in arousal state. Findlay and Hayward 2 recorded from cells in the P O A H of rabbits during sleep and wakefulness. Of 21 cells in the P O A H , 7 (33%) showed a decrease. 8 (38%) showed an increase, and 6 (29%) showed no significant change from AW to NREM. One anterior hypothalamic cell was totally inhibited in N R E M (NREMoff) similar to cell no. W18.2 in the present study. Four cells from the anterior hypothalamus and medial preoptic area were recorded through a complete sleep cycle: two increased and two decreased from AW to N R E M and 3 out of 4 increased from N R E M to REM. Of 94 cells examined in a number of hypothalamic nuclei, 64% had firing rates in AW < 5 Hz and 94% had firing rates in AW < 12 Hz, which is similar to the 88% of the total cells in the present study which had AW firing rates < 12 Hz. Jacobs et al. 16 reported median values of 0.7.0.8 and 3.2 Hz in AW, N R E M and R E M sleep, respectively, for 3 cat preoptic neurons, while Parmeggiani and Franzini 3° reported mean firing rates of 2.1 + 0.5, 2.3 + 0.9 and 1.3 _ 0.4 Hz for 4 cat anterior hypothalamic neurons. Parmegglani et al. 29 found that in a large sample of cat P O A H neurons (at L2 and L3), the vast majority of cells had firing rates of < 1 Hz in AW decreasing further in N R E M and REM. Glotzbach and Heller 5 reported that 26/30 (87%) of kangaroo rat P O A H cells had firing rates in A W < 12 Hz, and 70% of the cell population increased firing rates at the N R E M - R E M transition The results of these previous studies and the present study have shown that (1) cells in the P O A H are heterogeneous in terms of their changes in firing rate with arousal state, and (2) most P O A H cells have relatively low firing rates. Although the sample size is small, it appears that cells in the SCN share the above characteristics with P O A H cells. Several points are of interest m comparing the present results with data on SCN units in both urethane-anesthetized and in vitro slice preparations. Regularly firing cells (RFCs), which have been re-

ported as a prominent cell type, representing up to 80% of total units recorded in in vitro SCN studies 8A9'42'47. are apparently very rare in the intact ammal. Although firing pattern changes were not quantified in our study by interspike interval analysis, the characteristic pattern of RFCs makes these cells easy to identify by examination of polygraph records. No RFCs were recorded in the present study and only 4 out of 970 cells in the SCN and surrounding P O A H of the urethane-anesthetized rat could be classified as RFCs s. In view of the paucity of RFCs in the intact animal, it is uncertain as to the possible role of RFCs as neuronal oscillating components of central circadian rhythm mechanisms. The high percentage of RFCs could be an artifact of the slice preparation since 100% of cells recorded from the substantia mgra. an area unlikely to be involved in circadian rhythm control, were classified as RFCs 45. The effect of anesthesia on SCN neuronal activity is important to consider for several reasons. First. the addition of urethane or barbiturate anesthesia to the media bathing the SCN slice disrupts the regular firing pattern that is normally present 47. Second. urethane anesthesia is not a homogenous 'state': at least two distinct cortical E E G states are present which alternate with varying periodicities. Moreover. changes in hypothalamic single unit activity may be primarily related to changes in cortical E E G induced by a number of sensory stimuli rather than the sensory stimuli per se 17'18"23. Since in this study the firing rate of many SCN units are arousal state dependent, and since the majority of SCN 'visual' units are excited by photic stimulation delivered to the eyes. it would be valuable to monitor the cortical E E G of urethane-anesthetized animals during the characterization of visual responsiveness of SCN units to see to what extend ~photic' SCN cells are non-specifically influenced by EEG-related changes in arousal. The functional roles of the SCN cells recorded in the present study and their possible relationship to circadian rhythm generating mechanisms are unclear. Schwartz et al.40 have postulated that nonspiking events may form the primary basis of neural mechanisms controlling circadian rhythmicity, based on the fact that (a) the SCN is metabolically active in the fetus before single unit activity is present, and (b) tetrodotoxin infusion into the rat SCN. which blocks spiking activity, causes no change in the free-running

285 period or phase of the drinking rhythm. These results imply that action potentials are unnecessary for pacem a k e r oscillatory mechanisms and m a y be instead involved with coupling of input and output pathways to the p a c e m a k e r . In summary, the SCN cells in this study show changes in firing rate with arousal state similar to P O A H cells. The firing rates of a majority of SCN and P O A H cells are relatively low in the current study and in o t h e r in vivo and in vitro studies; R F C s are abundant in SCN in vitro p r e p a r a t i o n s only, and may not be important in neuronal timekeeping in intact animals. Future studies should consider changes in cortical arousal in urethane-anesthetized preparations when interpreting the relationship between SCN unit activity and the p a r a m e t e r of inter-

est. It would also be valuable to characterize properties of SCN single units in unanesthetized animals during their subjective night to see if group differences in the firing rates of day vs night SCN cells could account for the well-documented SCN M U A rhythm.

REFERENCES

13 Ibuka, N., Inouye, S.-I.T. and Kamamura, H., Analysis of sleep-wakefulness rhythms in male rats after SCN lesions and ocular enucleation, Brain Research, 122 (1977) 33-47. 14 Inouye, S.-I.T. and Kawamura, H., Persistence of circadian rhythmicity in a mammalian hypothalamic 'island' containing the suprachiasmatic nucleus, Proc. Natl. Acad. Sci. U.S.A., 76 (1979) 5962-5966. 15 Inouye, S.-I.T. and Kawamura, H., Characteristics of a circadian pacemaker in the suprachiasmatic nucleus, J. Comp. Physiol., 146 (1982) 153-160. 16 Jacobs, B.L., Harper, R.M. and McGinty, D.J., Neuronal coding of motivational level during sleep, Physiol. Behav., 5 (1970) 1139-1143. 17 Kanosue, K., Nakayama, T., Ishikawa, Y., Hosono, T., Kaminaga, T. and Shosaku, A., Responses of the thalamic and hypothalamic neurons to serotal warming in rats: nonspecific responses?, Brain Research, 328 (1985) 207-213. 18 Komisaruk, B.R, McDonald, P.G., Whitmoyer, D.I. and Sawyer, C.H., Effects of progesterone and sensory stmulation on EEG and neuronal activity in the rat, Exp. Neurol., 19 (1967) 494-507. 19 Kow, L.-M. and Pfaff, D., SCN neurons in tissue slices from ovariectomized rats: electrophysiological and neuropharmacological characterization and the effects of estrogen treatment, Brain Research, 297 (1984) 275-286. 20 Kreisel, B., Contorti, N., Gutnick, M. and Feldman, S., Suprachiasmatic nucleus responsiveness to photic and basal hypothalamic stimulation, Brain Res. Bull., 3 (1978) 707-714. 21 Krieger, D.T., Hauser, H. and Krey, L.C., Suprachiasmatic nuclear lesions do not abolish food-shifted circadian adrenal and temperature rhythmicity, Science, 197 (1977) 398-399. 22 Kubota, A., Inouye, S.-I. and Kawamura, H., Reversal of multiunit activity within and outside the SCN in the rat, Neurosci. Lett., 27 (1981) 303-308. 23 Lincoln, D.W., Correlation of unit activity in the hypothalamus with EEG patterns associated with the sleep cycle, Exp. Neurol., 24 (1969) 1-18. 24 Loeb, G.E., Bak, M.J., Salcman, M. and Schmidt, E.M.,

1 Eastman, C.I., Mistlberger, R.E. and Rechtschaffen, A., SCN lesions eliminate circadian temperature and sleep rhythms in the rat, Physiol. Behav., 32 (1984) 357-368. 2 Findlay, A.L.R. and Hayward, J.N., Spontaneous activity of single neurones in hypothalamus of rabbits during sleep and waking, J. Physiol. (London), 201 (1969) 237-258. 3 Fuller, C.A., Lydic, R., Sulzman, F.M., Albers, H.E., Tepper, B. and Moore-Ede, M.C., Circadian rhythm of body temperature persists after suprachiasmatic lesions in the squirrel monkey, Am. J. Physiol., 241 (1981) R385-R391. 4 Glotzbach, S.F., Gerber, R.L. and Heller, H.C., ParyleneC etched microwires for chronical unicellular recording in unrestrained animals, Proc. 7th ISOB, (1982) 172-175. 5 Glotzbach, S.F. and Heller, H.C., Changes in the thermal characteristics of hypothalamic neurons during sleep and wakefulness, Brain Research, 309 (1984) 17-26. 6 Glotzbach, S.F. and Heller, H.C., Central nervous regulation of body temperature during sleep, Science, 194 (1976) 537-539. 7 Green, D.J. and Gillette, R., Circadian rhythm of firing rate recorded from single cells in the rat suprachiasmatic brain slice, Brain Research, 245 (1982) 198-200. 8 Groos, G.A. and Hendricks, J., Regularly firing neurones in the rat suprachiasmatic nucleus, Experientia, 35 (1979) 1597-1598. 9 Groos, G. and Hendricks, J., Circadian rhythms in electrical discharge of rat suprachiasmatic neurones recorded in vitro, Neurosci. Lett., 34 (1982) 283-288. 10 Groos, G. and Mason, R., Maintained discharge of rat suprachiasmatic neurons at different adaptation levels, Neurosci. Lett., 8 (1978) 59-64. 11 Groos, G.A. and Mason, R., The visual properties of rat and cat suprachiasmatic neurons, J. Comp. Physiol., 135 (1980) 349-356. 12 Groos, G.A. and Meijer, J.H., Effects of illumination on suprachiasmatic nucleus electrical discha~-ge, Ann. N.Y. Acad. Sci., 453 (1985) 134-146.

ACKNOWLEDGEMENTS W e thank Carolyn R a d e k e for her assistance in many phases of this research, and Dr. B a r b a r a Snapp for critically evaluating the manuscript. This work was supported by N A S A N A G W - 7 0 and N I H NS21619 awarded to S . F . G . and N I H NS21918 awarded to H . C . H .

286

25

26

27

28

29

30

31

32

33

34 35

36

Parylene as a chronically stable, reproducible microelectrode insulator, IEEE Trans. Biomed. Eng., 34 (1977) 121-128. Moore-Ede, M.C., Sulzman, F.M. and Fuller, C.A., The Clocks That Time Us, Harvard Univ. Press, Cambridge, MA, 1982, 448 pp. Mosko, S. and Moore, R.Y., Neonatal SCN ablation: absence of functional and morphological plasticity, Proc. Natl. Acad. Sci. U.S.A., 75 (197816243-6246. Murakami, D.M. and Fuller, C.A., The development of cytochrome oxidase activity within the SCN of the cat, Soc. Neurosci. Abstr., 11 (1985) 386. Nishino, H., Koizumi, K. and Brooks, C.M., The role of suprachiasmatic nuclei of the hypothalamus in the production of circadian rhythm, Brain Research, 112 (1976) 45-59. Parmeggiani, P.L., Azzaroni, A., Cevolani, D. and Ferrari, G., Polygraphic study of anterior hypothalamic-preoptic neuron thermosensitivity during sleep, Electroencephalogr. Clin. Neurophysiol., 63 (1986) 289-295. Parmeggiani, P.L. and Franzini, C., On the functional significance of subcortical single unit activity during sleep, Electroencephalogr. Clin. NeurophysioL, 34 (1973) 495-508. Reppert, S.M. and Schwartz, W.J., Maternal coordination of the fetal biological clock in utero, Science, 220 (1983) 969-971. Reppert, S.M. and Schwartz, W.J., The suprachiasmatic nuclei of the fetal rat: characterization of a functional circadian clock using X4C-labeled deoxyglucose, J. Neurosci., 4 (1984) 1677-1682. Rusak, B. and Groos, G., Suprachiasmatic stimulation phase shifts rodent circadian rhythms, Science, 215 (19791 1407-1409. Rusak, B. and Zucker, I., Neuronal regulation of circadian rhythms, Physiol. Rev., 59 (1979) 449-526. Sato, T. and Kawamura, H., Circadian rhythms in multiple unit activity inside and outside the SCN in the diurnal chipmunk ( Eutamias sibiricus) , Neurosci. Res., 1 (1984) 45-52. Sawaki, Y., Suprachiasmatic nucleus neurones: excitation and inhibition mediated by the direct retino-hypothalamic projection in female rats, Exp. Brain Res., 37 (1979) 127-138.

37 Schmidt. E.M., Parylene as an electrode insulator: a review, J. Electrophysiol. Tech., 10 (t983) 19-29. 38 Schwartz, W.J. and Gainer, H., Suprachiasmatic nucleus: use of 14C-labeled deoxyglucose uptake as a functional marker. Science, 197 119771 1089-1091, 39 Schwartz. W.J., Davidson. L,C. and Smith. C.B.. In vivo metabolic activity of a putative circadian oscillator, the rat suprachiasmatic nucleus, l. Comp. ~'eurol. 189 (1980l 157-167. 40 Schwartz. W.J.. Gross. R.A. and Morton. M.T.. The suprachiasmatic nuclei contain a tetrodotoxin-resistant c~rcadian pacemaker, Soc. Neurosci. Abstr.. 1 l (1985) 818. 41 Schwartz, W.J., Reppert, S.M., Eagan, S. and Moore-Erie. M.C.. In vivo metabolic activity of the suprachiasmatic nuclei: a comparative study, Brain Research. 274 (19831 184-187. 42 Shibata. S.. Liou. S.Y. and Ueki. S. Effect of amino acids and monoamines on the neuronal activity of SCN in hypothalamic slice preparations, Jpn, J. Pharmacol.. 33 (19831 1225-1231. 43 Shibata. S., Shiratsuchi, A.. Liou. S.Y, and Ueki. S.. The role of calcium ions in circadian rhythm of suprachiasmatic nucleus neuron activity in rat hypothalamic slices, Neurowt. Lett.. 52 (1984) 18t-184, 44 Stephan, F.K. and Zucker, I.. Circadian rhythms in drinking behavior and locomotor activity of rats are eliminated by hypothalamic lesions. Proc. Natl. Acad. Sci. U.S.A. 69 (19721 1583-1586. 45 Thomson. A.M., Regular firing patterns of cells in the SCN and substantia nigra maintained in vitro: dependence on extracellular calcium, J. Physiol. (London). 338 (1983141 46 Thomson. A.M., Slow, regular, discharge m suprachiasmatic neurones is calcium dependent, in slices of rat brain. Neuroscience, 13 (19841 761-767. 47 Thomson. A.M., West. D.C. and Vlachonikolis, I.G., Regular firing patterns of suprachiasmatic neurons maintained tn vitro, Neurosci. Lett.. 52 (19841329-334. 48 Turek. F.W.. Circadian neural rhythms in mammals, Am. Rev, Physiol., 47 (1985] 49-64. 49 Wheal. H.V. and Thomson. A.M,, The electrical properties of neurones of the rat SCN recorded intracellularly in vitro. Neuroseienee. 13 (1984) 97-104