Brain Research, 544 (1991) 42-48 © 1991 Elsevier Science Publishers B V (Blome&cal Dwlslon) 0006-8993/91/$03 50 ADONIS 0006899391164166

42

BRES 16416

Circadian differences in neuronal activity of the suprachiasmatic nucleus in brain slices prepared from photo-responsive and photo-non-responsive Djungarian hamsters Russell R. Margraf, Piotr Zlomanczuk, Lauri A. Liskin and G. Robert Lynch Department of Btology, Wesleyan Umverstty, Mtddletown, CT 06457 (U S A ) (Accepted 9 October 1990) Key words" Suprachmsmatlc nucleus, Circadian rhythm; Brain slice, Dlungarmn hamster, Phodopus sungorus; Photopenod; Neuronal firing rate

The suprachiasmatlc nucleus (SCN) of the hypothalamus, as the putative generator of circadian rhythmicity, plays an important role in mammalian photoinduction To determine if SCN function differs in photo-non-responswe Dlunganan hamsters, we defined the pattern of spontaneous neuronal discharge of single cells from SCN slices in vitro of photo-responswe and photo-non-responswe phenotypes. Responsive hamsters exhibited a peak neuronal discharge rate (4 8 + 0 5 Hz) during the mid day which gradually attenuated through the late day and early night. In non-responsive hamsters, a slmdar discharge rate (5 1 + 0.5) was maintained through the late day and early mght. The delayed decline in spontaneous firing rate of non-responders correlates with their delayed activity onset and delayed nocturnal pineal melatomn pulse These data support the argument that the absence of photoperlodlc adjustments in Phodopus sungorus rests with differences in SCN function INTRODUCTION The Djungarian hamster, Phodopus sungorus, relies on changes in photoperiod to cue seasonal adjustments in reproductive status and thermoregulation. Under short day conditions, this species exhibits body weight loss, increased thermogenic capability, periods of daily torpor, gonadal regression, and molt to a white, winter pelt 9"1° However, not all Djungarian hamsters respond to a short day photoperiod 17. Thus, hamsters can be separated mto photo-responsive (responsive) and photo-non-responsive (non-responsive) phenotypes. Non-responsiveness is not unique to Phodopus and has been described for a number of other species i5. Non-responsive Djungarian hamsters lack the neural/ biochemical mechanisms necessary for a short day melatonin pulse 17. For example, under short day conditions, the nocturnal rtse m pineal melatonin and the onset of wheel-running activity is delayed in non-responsive relative to responsive hamsters. Non-responsive hamsters also exhibit other differences m circadian organization, such as a longer free-running period in constant dark (DD), different phase response curves to light (PRC), and decreased incidence of splitting under constant light 18. Collectively, these differences m circadian organization may reflect differences in function of the

suprachiasmattc nucleus (SCN). Several lines of evidence indicate that the mammalian SCN is critical for initiating and maintaining circadian rhythms. Bilateral lesions of the SCN result in loss of circadian rhythmicity and short day-reduced gonadal regression 14'24'28. Further, grafting of fetal SCN into the third ventricle of previously SCN-lesioned animals indicates that SCN associated neuropeptides, and clock properties can be transplanted from donor to host 3"t2'2°. Daily differences in SCN 2-deoxyglucose metabohsm 16' 22, and ctrca&an pattern in both in vivo and in vitro SCN neuronal actlvtty has been demonstrated 7'11'25 In brain slice studies on the rat and Syrmn hamster, neuronal activity peaks at about the middle of the photophase and reaches a trough midway through the scotophase 7'27 This daily variation has a circadian basts since it persists followmg exposure to D D 26 Even individual cells recorded from SCN explants maintamed in culture express circadian variation that persists with a period slightly less than 24 h 1 The present study characterizes the circadian pattern of m vitro neuronal activity in the SCN of the D j u n g a n a n hamster. Specifically, we examine differences m the SCN firing pattern m brain shces obtained from photoresponswe and photo-non-responswe hamsters.

Correspondence R.R Margraf, Department of Biology, Wesleyan Umversity, Lawn Ave , Middletown, CT 06457, U S A

43 MATERIALS AND METHODS Ammals were derived from a genetically heterogenous stock maintained at Wesleyan University. Hamsters were bred and reared under a 16"8 light-dark cycle (lights on 04 30 h EST) at 21-22 °C. Food (Wayne Lab Blox) and water were available ad hbitum. At 100-110 days of age, 44 animals were singly housed and transferred to a 9:15 light-dark cycle (lights on 08.00 h EST; Sylvania cool-white fluorescent light, 350 lux). Following 12-20 weeks of short day exposure the hamsters were identified as responsive or non-responsive based on molt and gonadal state. Hamsters were dwided into 4 groups: (1) responders taken directly from a 9' 15 hght-dark cycle, (2) responders exposed to 24 h DD, (3) non-responders exposed to 24 h DD pnor to slice preparation, and (4) responders which were maintained under a short day photopenod for 40 weeks and were photorefractory (1 e. resembled long day hamsters) for all measured variables. Coronal slices of the hypothalamus at the level of the opUc chlasm were prepared from responders either during the day prior to actwity onset or during the delay region of the PRC. SCN slices from non-responders were prepared during the 'dead zone' of the PRC at various times throughout the 24 h cycle. During and immediately following decapitation, some hamsters were exposed to less than 30 s of white light. This light pulse had no effect on the SCN firing profile relative to hamsters exposed to dim, red hght Slices (500 ~m) were cut using a vibratome (Series 1000, Technical Products International, St. Louis). The vlbratome reservoir was filled with cold (4 °C) and continuously oxygenated (95% O2: 5% CO2) incubation medium consisting of (mM) NaC1 124, KCI 3 3, K2SO 4 1.24, NaOH 20, glucose 10, CaCI 2 2.5, MgCI2 10, MgSO4 1 0. The pH of all solutions was adjusted to 7.4. In most cases two slices at a time were transferred to a recording chamber (Medical Systems Corp., Greenvdle, NY). Slices were maintained in warmed (34.536.5 °C) and continuously oxygenated (95%O2:5%CO2)mcubaUon medium. The perfusion rate across the shce was 1.5 ml/mm. Slices were allowed to acclimate m the chamber for at least 1 h pnor to recording. The hamster SCN is easily distinguished from other hypothalamic nuclei due to high tissue density and its prominent Iocauon just dorsal to the anterior part of the optic chlasm. Hypothalamlc slices were transillummated using a fiber optic lamp. Extracellular recordmgs were obtained from the SCN using electrodes made from 2.0 mm glass mlcroplpettes (World Preoslon Instruments, New Haven CT 06513) fdled with 3 M NaCl (electrode resistance 2-10 MI2) The microelectrode tip was positioned under visual control through a narrow window at the top of the chamber and slowly advanced through the SCN m 1/zm increments using a hydraulic mlcrodnve (Trent Wells, South Gate, CA) untd an individual umt was tsolated The signal was amplified, filtered (signal to nmse 2 1 or greater) and fed into a window dlscnmmator (manufactured to speofications, Robert Remstatler, University of Seattle, Seattle WA 98195) to distinguish a unit from background nmse. The output from the discriminator was fed into a Tektronix oscilloscope. TTL pulses from the dlscnmmator were also fed into an IBM PC, and firing frequency was determined using a real-time, off-hne histogram analysis system (RHIST, Run Technologies, Los Angeles, CA). Signals were collected m 10 s bins and the frequency was averaged over a minimum of 24 bins once the umt stabdlzed. Indwidual neurons were usually recorded for 10-15 rain Only Type I cells2~' characterized by a stable frequency and dmrnal vanaUon were used for data analysis Although a number of firing patterns exist m the hamster SCN, Type I cells were emphasized here because they exhibit clear orcadtan rhythmicity To monitor locomotor activity, 4 responders and 6 non-responders from the sample were placed m polypropylene cages (24 x 35 x 21 cm) equipped with running wheels (27 cm dmmeter). To facilitate running, each wheel was hned with a fine wire mesh Wheel-running actwity was recorded on a 20 channel event recorder (Esterline-Angus, Indianapohs, IN) Activity charts were prepared for each hamster by pasting 24 h strips below one another After a stable rhythm was established, estimates of phase angle for actiwty

onset and duration of activity (alpha) were calculated for each ammal. Data were analyzed by first calculating the mean firing frequency for all cells recorded in one individual across a 3 h time interval. These individual means were then averaged (+ S.E.M.) across each 3 h bin, and Student's t-test was used to determine differences in discharge rate between treatment groups. Both number of hamsters used to prepare slices (N) and total number of cells recorded (n) are given in the results. All summary data are expressed as a mean (+ S.E.M.).

RESULTS W h e e l - r u n n i n g activity in r e s p o n s i v e h a m s t e r s begins shortly after lights off ( - 0 . 7 9 + 0.64 h) a n d persists for 13.25 + 0.67 h while n o n - r e s p o n s i v e h a m s t e r s b e c o m e active later in t h e night ( - 7 . 5 + 1.05 h) a n d run for only 7.7 + 0.96 h (Fig. 1 A , B for r e p r e s e n t a t i v e data). R e s p o n s i v e h a m s t e r s e x h i b i t e d b o t h b o d y w e i g h t loss and m o l t to a w i n t e r p e l t f o l l o w i n g s h o r t day e x p o s u r e while non-responsive hamsters maintained their body weight and r e t a i n e d a g r e y - b r o w n p e l t ( T a b l e I). O n e g r o u p of r e s p o n d e r s m a i n t a i n e d u n d e r s h o r t day for o v e r 40 w e e k s became photorefractory

and

returned

to a l o n g day

physiological state for all m e a s u r e d v a r i a b l e s (Table I). P h o t o r e f r a c t o r y r e s p o n d e r s c o n t i n u e d to r e s e m b l e short day r e s p o n d e r s in w h e e l r u n n i n g activity w h i c h b e g a n shortly b e f o r e lights off (0.3 + 0.1 h) a n d p e r s i s t e d for

A RESPONDING

B NONRESPONDING

Fig. 1 Representative running wheel activity of two responswe (A) and two non-responswe (B) hamsters after 20 weeks of short day (LD 9:15) exposure. Records were taken for two weeks. Solid bar represents the dark portion of the light:dark cycle

,~

A

"r

>f..)

ml)mlm m I m m

Z

wHim

Iml

B

~5 1o

I

°

li/mill//jj

uJ rr U-

mmmmmlnmnm

1o

Fng. 2. Representative extracellular potential of a single cell from the SCN of Phodopus sungorus Scale: 50/~V/ordlnate divnsion and 1.0 ms/abscissa dwnsmn o'o '

12.7 + 0.5 h. These data are in accordance with previous studies from this lab 17. In this study, most slices were maintained for about 10 h (range 4-30 h). A total of 716 cells were recorded. About 3-4 cells per hour were measured, and approximately 85% of the Type I neurons used for data analysis were collected from the dorsomedial aspect of the SCN. Extracellular action potentials had a duration of 1.5-2.0 ms and an amplitude of 100-150 pV (Fig. 2. Type I cells, which are characterized by regular, short-term interspike intervals and stable frequency histograms (Fig. 3A), were most often recorded from the dorsomedial aspect of the SCN, although these cells are also present in the ventrolateral SCN. Thetr firing rate was significantly higher during the day than at the night in all experimental groups (see below). The daily firtng profile was similar in males and females. Although not used for data analysis, both Type II (spontaneous, trregular firing) and Type III neurons (burstmg activity)26 were also found in the dorsomedial and ventrolateral SCN (Fig. 3B, C, respectively). The spontaneous discharge rate of neurons recorded

TABLE I

The effect of short day (LD 9:15) exposure on mean (+_S E M.) body wetght and molt index for responsive and non-responsive Dlunganan hamsters Body wezght (g) Week 0

Week 12

Molt mdex* Week40

Week12

36.9+06

27.3+10

-

37+0.3

38.7+0.8

40.4+09

-

1 0+0.0

39.3+16

31.2+14

41.6+2.9

39+0.2

Non -responsive 9

Photorefractory 6

~'o

,2~o

I

9Go

TIME ($EGOND$)

Fng 3. Spontaneous single umt dtscharge of a SCN neuron. A: regular finng neuron, Type I B: irregular finng pattern, Type II C' bursting firing pattern, Type III Note Type III firing frequency ns much lower and ceases frequently

from responsive hamsters removed from short day exhibited a diurnal rhythm (Fig. 4). When expressed in 3 h bins, cells recorded during 12-15 h (during midphotophase) had the highest firing rate (5.6 _+ 0.9 Hz), while cells recorded during 0-3 h had the lowest firing rate (0.8 _+ 0.5 Hz; P < 0.05). Similar day/night differences persisted in hamsters removed from DD (Figs. 4, 5A). Mean peak neural activity in DD responsive hamsters was 4.8 + 0.5 Hz

S'

A N

>(.1 z LIJ

4'

O uJ rr ~'

2"

0

i

o

-

~, ~

.

&. ~

.

&. "7

~

~,

"7

"7

v-

v-

-

,-,

, ,~

,

,

~,-

TIME OF DAY

Responszve 14

[

~o

* A score of 1 ns a grey-brown summer pelt and 6 is a white winter pelt 4. By week 40, all photorefractory hamsters had a grey-brown pelt.

Fig 4 Cnrcadlan vanatnon m mean ( + S.E.M ) dxscharge rate of SCN neurons in slnces prepared from" responsive hamsters taken from a short day photoperlod (open circles) (LD 9 15, N = 14, n = 77), responders removed from D D (filled orcles) (N = 24, n = 351), photorefractory responders (filled triangles) (N = 6, n = 130). The average score for each hamster was used to estnmate the population mean for each 3 h bm. Shces were incubated for a least 1 h m warm (34.5-36.5 °C), oxygenated (95%02: 5%COz) incubation medium p n o r to recording

45 during the mid day (12-15 h bin) and then slowly decreased during late day and early night. SCN electrical activity in non-responsive hamsters express a different firing profile. Neuronal activity reached a similar rate (5.1 + 0.5 Hz) during mid day, but this level persisted through the late day and early night (Fig. 5B). Thus, substantial differences (P < 0.01) in mean SCN electrical activity occur between responders and non-responders during the early night (Fig. 6). The general profile of spontaneous neuronal discharge recorded from photorefractory hamsters resembled LD and DD responders where peak electrical activity was 7.9 + 1.2 Hz (closed triangles, Fig. 4). The daily firing profile of the SCN was not affected by time of slice preparation, given the two time points

L

I A

o

12

10

x

•

8 x~m Oil •

N

2

*

+ • " .-

,

3:

ms)

x :"

.

•

•A o +~11~,.,11

• x

t-

qb~+ .

m m +lldh

.

,

• ~='-,lllb-mll'~l~ x x

~ . ;~

v IN

>0 Z UJ

O

LLI n" 1.1.

I

I

examined here (Fig. 7A). Peak SCN electrical activity of hamsters killed at 14 h or at 20 h occurred during the same 3 h time interval (Fig. 7B). Although perfusate temperature fluctuated between 34.5 and 36.5 °C, this difference had no apparent effect on the firing rate of individual cells. DISCUSSION

The daily SCN firing profile differs in responsive and non-responsive Djungarian hamsters. Spontaneous neural activity recorded from SCN slices prepared from non-responsive individuals is similar to responders during the late night through mid day. Thereafter the two groups yielded highly divergent results. SCN neuronal activity of non-responders exhibited an abnormally prolonged period of rapid discharge which was significantly higher than that exhibited by responders during the late day and early night. This high SCN activity was maintained for about 9 h following the responder mid day peak (Fig. 6). It is not likely that SCN electrical activity in nonresponsive hamsters is just phase delayed by 6 h since wheel running activity offset and the time of increase in SCN electrical activity remain similar between nonresponders and responders. Although photo-non-responsiveness has been reported for a number of rodent species, only scant information is available on physiological mechanisms responsible. Even with little data available, it is apparent that the physiological basis for non-responsiveness is species-specific. For example, in the white-footed mouse, Peromyscus leucopus, non-responsiveness results from a difference in

B

12' •

m

•

(

8

10 g

8

•

o • e

•

6

[] ell

x •~

2

• 6/`

Mx • )~ x , L % , .

-

-

-

,

•

-

=

÷

•

oo

o

. +0 a

o

+

* 'a,~d9R o o*

Q . " o a "t" #

*.~c_

+

x

U .

X

=

UI

ql m o ..~1.

oo o o 0

ii

•

4

oao~

o

m - ,

TIME

•

[11(

am

÷

re

,

tJ. -

-

•

,

-

OF

-

-

,

•

4

0 ill

-

-

,

-

-

-

2

,

DAY

Fig. 5. Circadian differences m s p o n t a n e o u s discharge rate of single SCN n e u r o n s recorded m slices prepared from (A) responsive (N = 24, n = 351) or (B) non-responsive (N = 9, n = 158) hamsters All hamsters were exposed to a short day photoperiod for 12-20 wks and then exposed to constant dark for at least 24 h before m e a s u r e m e n t T h e open bar at the top of the figure represents that time when the hghts would have been on. All cells recorded from the same slice are identified in the figure with a u m q u e symbol. See Fig. 4 caption for additional details.

0

~

,~

&

&

~,

~o

;

~

*

*

(~

~. |

,r|

~

@4 OD

O4 m

*

TIME OF DAY

Fig 6 Daily change m m e a n (_+ S . E . M . ) discharge rate of SCN neurons m slices obtained from responsive (filled circles) or non-responswe (open circles) hamsters. T h e average score for each h a m s t e r was used to estimate the population m e a n for each 3 h bin. Data are the same as presented in Fig. 5. *P < 0.01; **P < 0.001 relative to responders.

46

I

A

o

12" 10" o. 8" •

qlo%

o°o •

o°

•

+

• iI . t . b o B- ,p"e'~ O'-oo *~

A

•

N

,

o.,m.'+. • • . .

0. "~+ ~.

+,

'1" >(3

i"

-.-~.---~,--.:++.--~,...:,...#

o

:""

°" " ' ° ' °

A

A

Z LU ,

O

UJ

8"

IJ= 6"

4"

2"

0

.

t,~

.

Io

.

.

o)

.

I'M

.

in

+o

;.

'

TIME OF DAY Fig. 7

Dady

pattern of SCN neuronal

activity m shces prepared

from responsive hamsters killed at 14 h (open symbols N = 5, n = 83, dead zone of PRC) or 20 h (filled symbols N = 5, n = 138, delay portion of PRC) and maintained for one complete cycle A: spontaneous discharge rate of single neurons Open and closed arrows along the abscissa represent the kill times for dead zone and delay region experimental groups, respectnvely In all cases, peak activity was recorded on the following day The open bar at the top of the figure represents that time when the hghts would have been on B daily change m mean (+ S E M ) discharge rate of SCN neurons Note that the tnme of peak act~vatyis the same between the two groups

the melatonin effector pathway, since both the melatonm and wheel-running rhythms are similar in responsive and non-responmve mice 2"13. Further, melatonin administration fails to elicit photomduction m non-responsive white-footed mice 8". This critical difference in the Peromyscus effector pathway is apparently not due to a defect m melatonm receptor or receptor-effector coupling 29. However, non-responsive Djungarian hamsters exhibit short day adjustments following late afternoon melatonin injections 19. These results not only indicate that nonresponsive hamsters can exhibit a functional response to melatonin, but also that the mechanism responsible for

non-responsiveness may be pre-pineal. This view is supported by data from a related study demonstrating that non-responsive hamsters lack a short day melatonin rhythm. In non-responders, the nocturnal rise in melatonln occurs during late subjective mght 17. This unique melatonin rhythm is paralleled by a short day activity rhythm that has a pronounced negative phase angle and short alpha. Similar activity results were obtained in this study (Fig. 1B). Collectively, these previous studies and the SCN data presented here suggest that a fundamental difference in clock function is responsible for photonon-responsiveness in Dlungarian hamsters. Understanding the causative factors mfluencing these SCN differences is a more difficult issue. One possibility is that they simply reflect differences in the physiological state which exist between the two morphs. For example, non-responsive hamsters are reproductively competent and responders are not. D o differences in physiological state influence SCN function? This scenario seems unlikely in Phodopus since circadian differences for both wheel running activity and melatonin rhythms persist in photorefractory hamsters 17. Further, photorefractory responders, which resemble non-responders m their physiological state (e.g. reproductively competent, greybrown pelt, high body weight, etc.), did not express this modified rhythm of firing rate (compare closed triangles, Fig. 4 and open circles, Fig. 6). We do not know if these differences in daily firing profile occur in the ventrolateral SCN. Although some ventrolateral cells were recorded during this study, they were too few to define a rhythm Gillette 6 recently reported the neuronal rhythm in rats persists only in the ventrolateral SCN, following knife cuts between the dorsomedial and ventrolateral domains. Conversely, bilateral enucleation caused damping of the rat ventrolateral rhythm, but the dorsomedial persisted 26. Given this apparent discrepancy in the rat hterature, additional data from the ventrolateral SCN of responsive and nonresponsive hamsters seems warranted In both responsive and non-responsive hamsters, wheel-running activity is suppressed at times of high SCN unit dnscharge (i.e. running wheel activity is 180 ° out of phase with SCN electrical activity), a characteristic commonly found m nocturnally actwe rodents 21 Data from non-responsive hamsters best dlustrates this relationship, in that the prolonged SCN activity during late day and early night is paralleled by an absence of wheel-runnmg at thin tnme (compare Figs. 1 and 6). SCN glucose utilization and multiple unit activity rhythms always peak during the day even in species with different temporal distributions of overt behavior 21'23, demonstrating that the phase between overt rhythmicity (nocturnal, dmrnal, crepuscular) and SCN activity ns variable. Data

47 from this study indicate that other aspects of overt rhythmicity, such as alpha and rho, may rely on the duration of the daytime peak in SCN activity. The neural physiology of SCN in responsive Djungarian hamsters appears to be similar to the rat and Syrian hamster. Shibata's Type I, II, and III neurons 26 were found in the dorsomedtal and ventrolateral SCN. The daily firing profile for responsive Djungarian hamsters (Figs. 4 and 6) is also similar to the rat and Syrian

hamster 5,7,25,27. An important concern is whether time of sacrifice might produce instantaneous phase shifts in SCN function due to neural activation associated with decapitation or transecting the optic nerves. Such a phase-shifting effect on the SCN has been reported in rats 5". In her study, the direction of this phase shift was the same as that produced by light, although the amplitude was considerably greater (e.g. a 4.5 h phase advance occurred at ct 19.5). To test the possible effect of kill time in Djunga-

rian hamsters, slices were prepared from responsive hamsters killed at either ct 9 (dead zone for light) or ct 14 (light-induced 3.5 h phase delay). In this study, we found no difference in peak activity when hamsters were killed at these two circadian times (Fig. 7A,B). Furthermore, Newman and Hospod 16 did not observe differences in slice 2-deoxyglucose uptake from rats killed at two different circadian times, ct 8 and ct 19. The brain slice preparation has been useful in establishing that the SCN discharge profile is a dynamic process which reflects differences in photomduction in this species. The technique should prove equally valuable in future investigations which address the neurophysiological basis for these differences.

Acknowledgements We thank Woifgang Puchalski and Mary Harrington for their help. Supported m part by NSF 8719120 and a grant from Wesleyan Umverslty to G.R.L

REFERENCES 1 Bos, N.P.A and Mlrmlran, M , Clrcadmn rhythms m spontaneous neuronal discharges of the cultured suprachlasmat~c nucleus, Brain Research, 511 (1990) 158-162. 2 Carlson, L . L , Zimmermann, A. and Lynch, G.R., Geographic differences for delay of sexual maturahon m Peromyscus leucopus, effects of photoperiod, pinealectomy and melatonm, Btol Reprod., 41 (1989) 1004-1013 3 Earnest, D J , Sladek, C D., Gash, D.M and Wlegand, S J , Speclfioty of orcadian funcuon in transplants of the fetal suprachlasmatlc nucleus, J. Neurosci, 9 (1989) 2671-2677 4 FIgala, J , Hoffmann, K. and Goldau, G , Zur Jahrespenod~k belm Dsunganschen Zwerghamsters, Phodopus sungorus, Pallas Oecologta, 12 (1973) 89-118. 5 Gillette, M U., The suprachlasmatlc nucler circadian phaseshifts induced at the t~me of hypothalamic shce preparation are preserved m vitro, Brain Research, 379 (1986) 176-181. 6 Gillette, M U., Locahzation of a circadian pacemaker to the ventrolateral suprachlasmatlc nucleus (SCN), Soc Res Btol Rhythms, Amelia Island, FL, 1990, p 74 (abstract). 7 Green, D.J. and Gdlette, R , Circadian rhythm of firing rate recorded from single cells m the rat suprachlasmatlc brain shce, Brmn Research, 245 (1982) 198-200. 8 Heath, H W. and Lynch, G R., IntraspeclfiC ddferences for melatonin-mduced reproductive regression and the seasonal molt m Peromyscus leucopus, Gen Comp Endocrmol ,48 (1982) 289-295 9 Heldmaler, G and Stemlechner, S , Seasonal control of energy requirements for thermoregulatlon m the Djunganan hamster (Phodopus sungorus), hvmg in natural photopenod, J Comp Phystol, 142A (1981) 429-437. 10 Hoffmann, K , The influence of photopenod and melatonln on testis size, body weight, and pelage colour m the Djunganan hamster (Phodopus sungorus), J Comp Physiol, 85 (1973) 267-282 11 Inouye, S. and Kawamura, H , Pers,stence of orcadian rhythmicity in a mammahan hypothalamlc hsland' containing the suprachlasmatlc nucleus, Proc. Nail Acad Sc:. U S.A, 76 (1979) 5962-5966 12 Lehman, M . N , Sdver, R , Gladstone, W . R , Kahn, R M., Gibson, M. and Blttman, E L , Clrcadmn rhythmloty restored by neural transplant Immunocytochemical characterization of

13

14

15

16

17

18

19

20

21

22

23 24

25

the graft and its integration with the host brain, J. Neurosct., 7 (1987) 1626-1638 Lynch, G R., Heath, H.W., Sullivan, J.K. and Tamarkin, L., Dally melatonin rhythms m photoperiod sensmve and insensitive white-footed mice (Peromyscus leucopus). In G Brown and S Wainwright (Eds), The Pineal Gland and its Endocrine Role, Oxford Umversity Press, New York, 1982, pp. 127-133. Moore, R.Y. and Elchler, V B , Loss of circadian adrenal corticosterone rhythm following suprachlasmatic nucleus lesions m the rat, Bram Research, 42 (1972) 201-206 Nelson, R J , Photoperiod-nonresponsive morphs: a possible variable m mlcrotlne population-density fluctuations, Am Nat, 130 (1987) 350-369. Newman, G.C and Hospod, EE., Rhythm of suprachlasmatlc nucleus 2-deoxyglucose uptake in vitro, Bram Research, 381 (1986) 345-350 Puchalskl, W and Lynch, G R., Evidence for differences in the orcadlan orgamzaUon of hamsters exposed to short-day photoperiod, J. Comp. Phystol, 159A (1986) 7-11. Puchalski, W and Lynch, G . R , Characterization of circadian function m Djunganan hamsters msensmve to short day photoperiod, J Comp Physzol., 162A (1988) 309-316. Puchalskl, W., Khman, R.M and Lynch, G.R., Differenual effects of short day pretreatment on melatomn-mduced adjustments in Dlunganan hamsters, Life Sci., 43 (1988) 1005-1012 Ralph, M R , Foster, R G. Davis F.C. and Menaker, M., Transplanted suprachlasmatlc nucleus determines orcadian pcnod, Science, 247 (1990) 975-978 Sato, T and Kawamura, H , Circadian rhythms in multiple unit activity Inside and outside the suprachmsmatlc nucleus m the diurnal chipmunk (Eutamtas stbincus), Neurosct Res., 1 (1984) 45-52 Schwartz, W.J. and Garner, H , Suprachiasmatlc nucleus, use of 14C-labeled deoxyglucose uptake as a functional marker, Sclence, 197 (1977) 1089-1091. Schwartz, W.J., Reppert, S . M , Eagan S M and Moore-Ede, M C., In VlVOmetabolic actwlty of the suprachiasmatlc nuclei a comparative study, Brain Research, 274 (1983) 184-187 Stephan, F K and Zucker, I , Circadian rhythms in drinking behavior and locomotor actwlty of rats are ehmmated by hypothalamlc lesions, Proc Natl. Acad Sct U.S.A., 69 (1972) 1583-1586 Sh~bata, S , Oomura, Y , Klta, H and Hatton, K., Circadian

48 rhythmic changes of neuronal activity in the suprachlasmat~c nucleus of the rat hypothalamic slice, Brain Research, 247 (1982) 154--158 26 Shlbata, S., Llou, S Y., Uekl, S. and Oomura, Y , Influence of environmental light-dark cycle and enucleation on actwlty of suprachiasmatic neurons in slice preparations, Brain Research, 302 (1984) 75-81. 27 Shibata, S and Moore, R.Y, Electrical and metabohc activity of suprachlasmatlc nucleus neurons in hamster hypothalamlc

shces, Brain Research, 438 (1988) 374-378. 28 Stetson, M.H. and Watson-Whltmyre, M , Nucleus suprachlasmatlcus the biological clock m the hamster? Scwnce, 191 (1976) 197-199 29 Weaver, D . R , Carlson, L L and Reppert, S.M., Melatonm receptors and signal transductlon in melatonin-sensitive and melatomn-msensltive populations of white-footed mice (Peromyscus leucopus), Brain Research, 506 (1990) 353-357