Non-photic phase shifting of the circadian activity rhythm of Syrian hamsters: the relative potency of arousal and melatonin

Brain Research, 591 (1992)20-26 ~:-~1992 Elsevier Science Publishers B.V. All rights reserved 0006-8993/92/$05.00

20

BRES 18077

Non-photic phase shifting of the circadian activity rhythm of Syrian hamsters: the relative potency of arousal and melatonin M.H. Hastings, S.M. Mead, R.R. Vindlacheruvu, F.J.P. Ebling, E.S. M a y w o o d a n d J. G r o s s e Department of Anatomy, Unirersityof Cambridge, Cambridge (UK) (Accepted 14 April 1992)

Key words: Melatonin; Circadian; Entrainment; Arousal; Hamster

This study iavestigated the relative potency of melatonin and arousal as Zeitgebers in the non-photic phase shifting of circadian rhythmicity in the adult Syrian hamster. AnimaE held under dim red light (DD) exhibited robust free-running rhythms of wheel-running activity, Melatonin (1 mg/kg) or ethanolic saline vehicle, delivered manually by subcutaneous injection after removing the animal from its cage, resulted in phase advances of the activity rhythm. This effect was phase dependent, injections at CT 8 and 10 being effective (CT 12 = anticipated activity onset), whereas injection at CT 2, 6, 14 and 20 did not cause a shift. There was no significant difference between the magnitude or timing of phase shifts in response to injections of saline or melatonin. To determine whether the observed shifts were related to arousal of the animals induced by handling, a secLmd group held under DD were fitted with chronic s.c. cannulae so that melatonin solution or vehicle could be delivered remotely at projected CT 10. Neither solution [tad any effect upon the free-running rhythm. However, when these animals received manual s.c. injection of saline or melatonin solution, they exhibited phase advances similar to those observed in Expt. I. These results fail to support the hypothesis that melatonin can excrt a chemically specific, acute phase-shifting action in the adult Syrian hamster. They do, however, demonstrate the potent effect of arousing stimuli upon the circadian clock in this species.

INTRODUCFION The pineal hormone melatonin is secreted with a pronounced circadian rhythmicity, circulating levels being high during subjective night and low during subjective day ~''~. In lower v~rtebrates, this nocturnal signal influences a variety of circadian rhythms, co-ordinating the activity of pineal, retinal and suprachiasmatic oscillators ~'~'-~'.In mammals the melatonin signal acts as a cue to the neuroendocrine axis to direct seasonal rhythmicity in photoperiodic species ~"''~. Melatonin has also been proposed to act as a circadian Zeitgeber in mammals. Large bolus injections of melatonin delivered to pinealectomized, pregnant hamsters synchronize circadian rhythmicity of the foetuses in utero, although the hormone is not the sole agent responsible for circadian entrainment in utero because synchrony between dam and ft~etus can also be established in the abscnce of ma:crnal melatonin ~,2"~.Nevertheless, a role for melatot~.in ir. circadian entrainment seems likely. In

the foetus, high-affinity, specific binding sites for melatonin are present in the suprachiasmatic nucleus (SCN) of the hypothalamus, the principal circadian oscillator :'''27''~4''~. Binding sites for melatonin are also present in the SCN of adult mammals, raising the possibility that the hormone acts as an internal circadian Zeitgeber post-natally "~'~'~s''~'.Single daily injections of melatonin, at doses which would raise serum concentrations by several orders of magnitude above those which occur physiologically, cause acute phase shifts of the circadian activity rhythm of free-running adult rats, whereas repeated injections lock the rhythm into a stable entrainment once the time of activity onset is close to the time of injection 4'5. However, the phase response curve (PRC) to melatonin injections contrasts markedly with the typical profile of PRCs to light pulses", In the rat, most of the circadian cycle is occupied by a dead zone, advances occurring only during a very restricted portion of late subjective day, at a time when endogenous secretion of melatonin

Corn'slxmdem'c." M.H. Hastings, Department of Anatomy, University of Cambridge, Downing Street, Cambridge CB2 3DY, UK. Fax: (44) (223) 333-786.

21 MATERIALS AND METHODS

would not occur. Moreover, there is no delay portion to the curve, indicating that daily exposure to the stimulus would not be able to entrain animals with an endogenous period of less than 24 h. These properties of the PRC strongly suggest that melatonin, if it has any role at all in circadian entrainment, does so via non-photic mechanisms, but they also raise serious doubts about the physiological relevance of any circadian effects of melatonin in the adult. The aim of this study was to determine whether melatonin is able to influence the circadian system of the adult Syrian hamster. This species is extremely sensitive to the photoperiodic actions of the hormone in adulthood n3, and to the circadian entraining effect prenatally 7. In addition, the circadian system of the adult hamster is very responsive to non-photic, arousing stimuli nT'2°'2n. It therefore provides an excellent experimental subject in which to explore the physiological actions of melatonin in ,'elation to non-photic entrainment of the circadian system. The approach used was to adopt the acute injection protocol of Armstrong and Redman s in order to map a PRC to melatonin, in order to dissociate the contribution of melatonin from that of arousal associated with handling, in a second experiment melatonin was delivered via remote cannulae, thereby avoiding handling of the animal and so providing the potential to reveal any specific endocrine action.

Adult male Syrian hamsters (Wrights Ltd., Chelmsford, Essex UK) were caged individually with access to a running wheel, under a 16 h light: 8 h darkness photoschedule with water and laboratory chow available ad libitum. This strain of hamster is known to express high-affinity binding sites for melatonin in the SCN during adulthood 3~. The circadian rhythm of wheel running was recorded by closure of micro-switches connected to the wheel and fed via matrices to a Viglen HD40 micro-computer running "Dataquest lit" software (Minimitter Inc., Sunriver, OR, USA). After 2 weeks, the animals were released into constant dim red light (DD) and allowed to free-run. The phase-shifting effect of melatonin solution (1 mg/kg) or ethanolic saline vehicle (200 pA) was tested by removing the animal from its cage, placing it on a top pan balance, and giving it a subcutaneous injection between the scapulae. Injections were arranged to coincide with particular predicted circadian phases, circadian time 12 (CT 12) being defined as activity onset. Animals were immediately returned to their cages and allowed to continue the free-run for at least 7 days in order to determine the magnitude of any subsequent phase shift. In the second experiment, in order to deliver melatonin solution or vehicle without handling the animals, they were fitted with a chronic subcutaneous cannula attached to a fluid swivel, such that an infusion could be made remotely from outside the cage 13, In order to prevent entanglement of the cannula within the wheel, the face of the running wheel was closed off with a perspex disc so that the animal could not enter it, but could still rotate it by running on the outer surface of the wheel. In order to confirm the validity of the traces, the activity of cannulated animals was also monitored using passive infrared detectors (Racal Guardall IR77, MK II, Racal Electronics Ltd., UK) which respond to overall locomotion .bout the cage. The cannula remained empty until the time of injection, when saline vehicle was injected using a 2 ml syringe at a volume of dead-space plus 200 ~l. Most of the injected animals exhibited a very brief behavioural response to the infusion, in the form of a small start followed by a short bout of grooming. For the second injection, the animals received 200 ~l of melatonin solution by displacing the saline in the dead-space with melatonin

Saline 1 , . i i

/ /

?_. I

ii

l i I I

i

i

I i I

1

l

i

i

i

i l

i

i'

l i i

:

...........

V.

mhj

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

~,,....~__..d~

i

i.

i .

li

:...................i . . ~ i

. . . . .

.....i

IJ

1 ........

it.

i

| l l a i l ' L " *ai k J

5

~i

_.

. . . . . . . . . . . . . . . . . . . . i i --Ii " " i "i . . . . . . . . t.[, I I .

I.

mJ,_i

a ...... i ....., . . . . . . . l J - J . - -, . . . . . . . . . . . . .

........

Ik'i. il i,.

.

.

.

.

U. . . . . . . . . .

t ~i ~ ~ ~

41

i

~'

,il

i.

•

,

.

.

• •

b.A...~

......i i

...........

ILk_..i_.i

. . . . . .

24

Time, h

]li.k ILi.

I

i a

CT 6

. .,

i a

it,, iil

i

*

,

OT 8

.... n i

V •

.

.

.

.

.

.

i

..

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

i

.... i . . . . . . . . . .

llm

I II.

i

i i..

ill

.

.

" ........... ............

i ...............

48

.

........

............ i . . . .

..i ...........

.

L 1

.~.1 -i

..... "

i

.Ilk _.Ill

,

. .... ..,

~ i , ~ i

i

CT

10

I i

i i

i

i~. ~ " , ~', i.

CT 14

.. ....

i

I ..... i

1 ~ .... i

....., ..................F.i' .........

ll¢~,,

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

Nlllm .... i

......i . . . . . . . . . . .

............. i _ i

. . . . -. ~ i l i l 2 1 ? . &

.... i Jk.._

....

i .... i . . . . i . . . . . . . i

Jl

.

i i

,

. . . . . . . .

ilJ k . . ~l_llk.lih . . . . . . . . . I l L lk_ __i ~ . t . . . . . . . . . . . . lh.l,.l. IIL._.3 , ..... ILJi. II . Ja . . . . . . . . .Ib . I _ . h .... li II .

,i

Ii

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

bi

i

,.i,

,

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

1~ ' ili 1.I . i

k

i ,,

,k i |,

/ ii.

..........

" ~U i i i ' l ,~ l i

11

i i

V~i..i

li

I

i

h . i-, ....i

.

i ...... Ill , I

I

..... ~ l , .

lli._i_J

i.

I'-.l

i i

~ ....... i i,~. III i ~ . I

.... I A ~

L I

i

i,

. . . .

•

~

....... , ~ . . . . i . . . . . . . . . . .

I ...........

i

i

a ............................................. , - .

i....~

. . . . . . .

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

am

i,

,i._ ......i b . i .......... < ,ill, ,..l . k . . . . . . . . . . ./l i~i . . . . . . . . . i .... III i V,.

5

.l~

ia

, 61

i

Melatonln

/

.... 24

-7,.' : 48

Time, h

Fig. 1. Representative double-plotted traces of wheel-running activity of individually housed male hamsters held under free-running condiiions and given s.c. injections of either saline or melatonin (1 mg/kg) at one of a range of circadian phases (CT 6-20) as identified by the arrowhead. Solid line on right hand trace denotes eye-fitted estimate of activity onsets used to calculate phase-shift following injection.

22 solution and then injecting a further 200 pl into the cannula, thereby ensuring that they received the full dose of I mg/kg, albeit in a larger volume. The first experiment demonstrated that injections delivered outside the interval of CT 8-10 failed to affect rhythmicity. The infusion study therefore concentrated on the single phase of CT 10 to determine whether non-aroused animals responded to the hormone. it was considered unlikely that infusion would have an effect at phases when injection failed to do so. An independent observer was asked to determine the presence and magnitude of phase shifts by filling a line, by eye, to the activity onsets before and after treatment. Differences between treatments were assessed by one- and two-way ANOVA followed by post-hoc Dunnett's t-tests. Some of the data relating to saline injections in Expt. 1 have been published previously14.

lUtk.k

80

60 U)

.c_

E

40

20 G)

In ra.

.

-2O 0

|

l

I

6

12

18

l

24

Circadian Time, h Fig, 2. Composite phase response curve [mean:l:S.E,M,) to s.c, injections of either melatonin ( I m g / k g ) (closed symbols] or saline (open symbols) for free-running activity rhythms of male hamsters (n = 3 - 8 per group except C T I0, n = 18, N = 78), T h e PRC to saline injection has been reported previously L~,

J

l

JL

I ~ , - - ,LtlL

•-. JL

I

mill

k

U

llllkl,L

m e LL~...,

t

It

*

Lb.

.~

t

S "-~ "~L, L k. din.am*~ dl~ dmama_

m

at*M*

h

a~,,L mm..

dL_ Jm~

.,=

,L

,_, ,.~

12

t,

•

m e,ak= L1

'

D

t

.

•

&dl I k 14, • L i l ~-.k ......

C.l Q lm

t

t

S e..da .k k am at t, ill Ud l ,at dt t

} •

,It

I,&

I

~O 10,,

0

Animals which received manual injections of saline or melatonin at CT 10 typically displayed a bout of wheel running soon after being returned to their cage (Fig. 1). The interval of enhanced activity continued into the expected phase of activity onset. On the following circadian cycle, activity onset occurred approximately 45-60 rain earlier than would have been predicted by the pre-injection phase, and in all cases, the advance of activity onset of about 1 h was sustained over the subsequent cycles. The response to injection of both vehicle and melatonin was clearly phase dependent (Figs. 1 and 2). Injections given at CT 2 and CT 6 were not followed by a bout of wheel-running activity

21

t

lit

2Oi

Experiment 1: the effect of manual s.c. injections of melatonin sohaion or saline t'ehicle on free-running actirity rhythms

i

m

I L

RESULTS

38

6 • I

I am

a

I L

A

_

J

1 Is

I

•

•

l

m*.

.,,.

:

.=-

. . . . . . .

24 0

=

L ,4

12

_

I

Jt

a

14

,

24

Time, h Fig. 3. Representative single-plotted traces of wheel-running activity of individually housed male hamsters held under free-running conditions and given s.c. injections of either saline (s) or melatonin (m, 1 mg/kg) at CT 10 in a cross-over design. Note increased wheel-running activity soon after time of injection, denoted by filled circle and sustained phase advances over subsequent cycles. Numbers at head of trace denote animal identification.

and were not effective in shifting the rhythm, the animals continuing their free-run without any overt shift. Injections at CT 8 were as effective as injections at CT 10 in causing a phase advance, whereas injections delivered after activity onset at either CT 14, CT 20 or CT 23 had no phase-shifting effect, even though the animals were active upon return to their cage. The PRC to injections therefore exhibited a highly signifi. cant effect of phase of injection (Phase effect, F -- 35.3, d.f. - 7,76, P < 0.01) but no effect of drug (saline vs. melatonin F = 0.57, n.s.) nor any interaction between drug and phase of injection ( F = 0.01, n,s.). Some animals received injections of both solutions at CT 10 but on separate occasions. They exhibited no difference in their response to melatonin or vehicle (Fig. 3), the 2 solutions producing comparable advance shifts when given to the same animal. Although there was a trend towards a smaller advance shift following a second injection (first injection advance = 64.5 ± 6.3, second injection = 52.5 ± 5.0) this order effect was not significant ( F = 0.95).

Experiment 2: the effect of manual s.c. injections or remote s.c. infusions of melatonin solution or saline vehicle on free-running activity rhythms There was no qualitative difference in the activity traces of cannulated animals recorded from running wheels or from infrared detectors (Fig. 4). Animals which received remote infusions of saline or melatonin solution commonly exhibited a brief start as the fluid

23 lit .Jill

-!

.-

..L=

.L

•a i d .

I=

=

lw

!

........

;

Jill

J,

•

So .....

.

;

__

SO

2Jr

-&--

.....

m,l.

. . . .

,

~

~

j.

SO

ii•m

L I ,

0

L

d,,

0 9

..,

A,.=

iil

=1

=

,hA.

. . . .

(0

•

"=

:~ m

m e_ .

I~

."

-

,11.= . .1,, . lal

=L

".

I,i

.

L.

.

.

t ~A

L |A

"

• S-.

l

me

•

.

J

"= . . . . 181____.._,. . . . . . . 0 12

--

1

I

me

.

.

•

=

i

,

I

.

I , . .

•

i,,

.

J . . [.

i.,

--

•

•

=

J

,.

,-

•

.

.

•

J

i. L =

•

-

=..It= . . .

I.= =

,.

I

.

,

,II

.=j_

a=,

,= .

,

|..

.I.

j .,ll.~a

__

I. n

. =

,m •

. . .d

. ,t

,.

] =

.

a

•,

Ill

d.=

'-'

,. L . ..i •

,.-

ILl.:

-

.

• .=.=

L | L

.t

--"

I I ,.=

.L~L

4

. . . .

•

i

• ~

.IL,h ,,JL= a= J

a

0

•"

~ i

..

-

_ Jill

tim

-

.

,,,.. J _ , . . ~

|

,

.-

,-

J

.

,

t

•

,.

= L.

,

a

.l.

.

_.

[" ' ii

,

.

a

12

24

24

0

12

24

Time, h Fig. 4. Single-plotted traces of wheel-running activity of individually housed male hamsters held under free-running conditions and given remote s.c. infusions of either saline (s) or melatonin (m, 1 mg/kg) at (.1" 10. Infusion of neither solution caused increased activity nor any phase shift. Traces lir and lw represent activity rhythms of the same animal, as recorded with a passive infrared detector or a running wheel, respectively. Trace 2ir is the rhythm of a second animal recorded with a passive infrared detector.

entered under the skin. However, the animals remained in position and typically returned to sleep. Activity onset was in all cases at the phase predicted by the pro-infusion activity trace. There was no evidence of an advance shift following infusion of either solution at CT 10. However, when the cannulated animals were disconnected from their cannulae and several days later were given a manual injection at CT 10, they exhibited robust phase advances (Fig. 5). There was therefore a highly significant effect of infusion vs. injection (F = 147, d . f . - 1,30, P < 0.01) but no signifi100

T

60 t. l

E J= (n

G)

60

I

40

J=

n

20 m

Sal

Mel

Remote

Sal

Mel Manual

Fig. 5. Phase advances (mean :l: S.E.M.) in free-running activity rhythms produced by s.c. administration of either saline vehicle or melatonin solution (1 mg/kg) delivered either by manual injection ( n - - 7 melatonin, n = 4 saline) or by remote infusion via an indwelling cannula. (n = 11 melatonin, n = 12 saline).

cant difference in the shifts seen following administration of melatonin solution or vehicle ( F - 0 . 6 5 , n.s.). These results demonstrate that the principal agent in causing phase advances following treatment at CT 10 is a stimulus associated with the handling involved in injection, rather than the endocrine stimulus provided by melatonin. DISCUSSION Manual subcutaneous injection of melatonin solution or vehicle caused pronounced phase advances in the activity rhythm of free-running hamsters when the injections were given at CT 8 or 10. Injections at CT 2, 6, 14, 20 or 23 were without effect. There was no difference in the amplitude nor in the timing of the phase advances induced by the 2 solutions. When the compounds were delivered remotely, via a chronic indwelling cannula such that the animals were not handled, neither solution had any effect upon the circadian activity rhythm. These results demonstrate that the arousal associated with handling and s.c. injection is a potent Zeitgeber to hamsters held in continuous dim light. However, the current study provides no evidence of a role for melatonin as a chemically specific, non-photic Zeitgeber in the adult male Syrian hamster. In the adult laboratory rat, single injections of supraphysiological doses of melatonin can phase-shift free-running activity rhythms, and repeated daily injections can entrain the rhythm, provided the endogenous period of the animal is greater than 24 h, entrainment

being maintained by daily advance shifts 4'5. In the adult hamster, the interpretation of the response to equivalent single injections of melatonin was confounded ['y the observation that phase advances followed treatment with saline vehicle at CT 8 and CT 10. Although such an effect of a 'control' procedure might not be anticipated, careful inspection of the data presented in the studies with rats reveals a comparable but less marked advancing effect of saline injections delivered at CT 10. This raises two related questions. First, what is the contribution of handling to the acute shifting effect of timed injections, and second, does melatonin exert any chemically specific, entraining action that is independent of the effects of handling? In the hamster, handling is clearly a very potent stimulus: handling and removal from the cage associated with manual injections acted as a very strong non-photic Zeitgeber, producing robust phase advances to the overt rhythm of activity which did not occur in animals receiving solutions via remote infusion. Non-photic shifts comparable to those described here have been reported in the hamster following exposure to a wide range of arousing and/or activityinducing stimuli, such as oestrous odours, enforced wheel running and the administration of Triazolam 2-''3~. Whether the shift is caused by the increased locomotion associated with handling, or by a central state of arousal acting independently of the expressed activity is not clear :(},-''. Activity appears to play some role, because some arousal-induced shifts can be prevented if animals are restrained ",4, and in the present study, injections which phase shifted the rhythm were typically foUowed by an immediate increase in locomotor activity. However, the circadian timing of the activity at~0ears to affect the shifting response, insofar as animals injected at Cl" 14 and 20 were also active upon return to their cage, having commenced their nocturnal activity bout, but they did not exhibit any advance shift. It is now clear that non-photic shifts of the overt activity rhythm are accompanied by an equivalent advance in the phase of the light-entrainable oscillator, as determined by the effect of a subsequent light pulse upon free-running rhythmicity and the expression of immediate-early gene,~ , thin the SCN~4'~".)~ thus confirming that non-photiu cues do have access to the central oscillator. Non-photic entrainment has also been described in the adult laboratory rat in the context of schedules of restricted feeding .~", and i+: bt++r,a:++ in relation to social cues ¢', However, a recent study which employed schedules of enforced activity in rats concluded that 'the mechanism for entrainment by activity schedules clearly exists in rats, but.., activity is a very weak zeitgeber in this species.'~'. Should this be

the case, it could explain why only a small proportion of rats injected with saline in the studies of Armstrong and Redman 5 showed phase shifts: a result in contrast to those obtained with Syrian hamsters in the present study. Sensitivity to arousal may therefore be an important interspecific variable in considering non-photic entrainment, and one effect of melatonin in the rat may be to amplify the response to arousing stimuli, either by enhancing the perceived intensity of the stimulus, or by altering the tonic level of arousal such that the animal is more responsive to the same intensity of stimulus. Alternatively, melatonin may influence the coupling between the light-entrainable oscillator and a proposed second oscillatory system entrained by non-photic cues -+(). Both interpretations predict that treatment with melatonin in the absence of arousing cues would not cause phase shifts of the rhythm. This was certainly the case in the hamster and it awaits experimental test in the rat. This suggested role for melatonin in entrainment of the adult circadian system differs from that of a true Zeitgeber. It proposes that the hormone acts as a gating mechanism, regulating the sensitivity of the clock to non-photic stimuli. This effect could certainly be mediated at the level of the SCN, which, in rats and the strain of hamster used in the present study, contain high affinity melatonin binding sites throughout adult life "~7.Melatonin may also regulate the response of the circadian system to photic cues, insofar as repeated injections of hormone alter the rate of re-entrainment following an advance shift in the photoschedule 4.s,'.~. Moreover, the administration of exogenous melatonin to human subjects is reported to ameliorate the subjective effects of jet-lag, subsequent to eastwards transmeridinal traveP and to enhance the rate of re-entrainment of certain circadian measures following an acute advance shift in the photoschedule 2u, The phasing of the rhythm of secretion of endogenous melatonin also seems to be particularly responsive to treatment with exogenous hormone in both sighted and blind humans 3>)'28'2'), and in rats 12. It remains to be determined whether these effects of melatonin are a consequence of an interaction between the hormone and mechanisms of arousal, particularly influences of the hormone upon the sleep wake cycle in terms of sleep quality and latency, or whether they represent the effect of a true Zeitgeber acting directly upon the clock. Regardless of the precise mechanism of action of melatonin, the available evidence on the PRC demonstrates that the hormone could not exert a physiologically relevant action in the adult. Its secretion during subjective night would never coincide with the sensitive

25

phase of the PRC restricted to late subjective day. The PRC expressed in the rat may represent the vestige of a foetal response. In utero, the circadian system of the foetus is independent of that of the dam, and so prior to synchronization, maternal melatonin could potentially be encountered by the foetus at the sensitive phase and thereby shift the foetal clock. The vestigial nature in the adult is emphasized further by the requirement for supraphysiological doseages, if melatonin is to be effective. Whether or not such high doses, given repeatedly over several weeks would have a weak entraining influence in the adult hamster remains to be determined. It is clear, however, that an understanding of the neural basis to the entraining action of melatonin awaits identification of the neural substrates mediating non-photic entrainment in mammals and their relationship to identified areas of highaffinity melatonin binding sites. Acknowledgements. This work was supported by the Medical Research Council (UK)via project Grant G8912713N to M.H.H. and a MRC Research Training Fellowship to FJ.P.E.R.R.V. was supported by vacation scholarships from the Welleome Trust, King's College Cambridge and Queens' College, Cambridge. The Authors are grateful to Mr. J Bashford and colleagues of the Audio-visual Media Unit, Department of Anatomy, University of Cambridge for technical assistance.

REFERENCES I Arendt, J., Melatonin and the human circadian system. In A. Miles, D.R.S. Philbrick and C. Thompson (Eds.), Melatonin: Clinical Pervla,ctit,~:~, Oxford University Press, Oxford, 1988, pp. 43-61. 2 Arendt, J., Mammalian pineal rhythms, Adt,. Pineal Res., 3 (1985) 161-214. 3 Arendt, J., AIdhous, M., English, J., Marks, V., Marks, M. and Folkard, S., Some effects of jet-lag and their alleviation by melatonin, Erlconomics, 30 (1987) 1379-1393. 4 Armstrong, S.M., Melatonin and circadian control in mammals, F,xperientia, 45 (1989)932-938. 5 Armstrong, S.M. and Redman, J., Melatonin administration: effects of rodent circadian rhythms, in D. Evered and S. Clark (Eds.), Photoperiodism, Melatonin and the Pineal Gland, Ciba Foundation Symposium No. 117, Pitman, London, 1985, pp.

188-207, 6 Aschoff, J., Fatranoska, M. and Giedke, F., Human circadian rhythms in constant darkness, entrainment by social cues, Science, 171 (1971)213-215. 7 Davis, F.C. and Mannion, J , Entrainment of hamster pup circadian rhythms by prenatal melatonin injections to the mother, Am. J. Physiol., 255 (1988) R43¢-R4~. 8 Elliott, J.A., Circadian rhythms and photoperiodic time measurement in mammals, Fed. Prec., 35 (1976) 2339-2346. 9 Folkard, S., Arendt, J., AIdhous, M. and Kennett, H., Melatonin stabilizes sleep onset time in a blind man without entrainment of cortisol or temperature rhythm, Neurosci. Lett., 113 (1990) 193198. 10 Goldman, B.D. and Darrow, J.M., The pineal gland and mammalian photoperiodism, Neuroendocrinoiogy, 37 (1983) 386-396. 11 Hastings, M.H., Neuroendocrine rhythms, Pharmacol. Ther., 50 (1991) 35-71.

12 lilnerova, H., Trentini, G.P. and Maslova, L., Melatonin accelerates re-entrainment of the circadian rhythm of its own production after an 8-h advance of the light-dark cycle, J. Comp. Physiol., 166 (1989) 97-102. 13 Maywood, E.S., Buttery, R.C., Vance, G.H.S., Herbert, J. and Hastings, M.H., Gonadal responses of the male Syrian hamster to programmed infusions of melatonin are sensitive to signal duration and frequency but not to signal phase nor to lesions of the suprachiasmatic nuclei, Biol. Reprod., 43 (1990) 174-182. 14 Mead, S.M., Ebling, F.J.P., Maywood, E.S., Humby, T., Herbert, J. and Hastings, M.H., A non-photic stimulus caL,ses instantaneous phase-advances of the light-entrainable circadian oscillator of the Syrian hamster, but does not induce the expression of c-fos in the suprachiasmatic nuclei, J. Neurosci., 12 (1992) 2516-2522. 15 Menaker, M., Hudson, D.J. and Takahashi, J.S., Neural and endocrine components of circadian clocks in birds. In B.K. Foilett and D.E. Follett (Eds.), Biological Clocks in Seasonal Reproductive Cycles, Scientechnica, Bristol, 1981, pp. 171-184. 16 Mistlberger, R.E., Effects of daily schedules of forced activity on free-running rhythms in the rat, J. Biol. Rhythms, 6 (1991) 71-80. 17 Mrosovsky, N., Phase response curves for social entrainment, J. Comp. Physiol., 162 (1988) 35-46. 18 Mrosovsky, N., Nonphotic enhancement of adjustment to new light-dark cycles: masking interpretation discounted, J. Biol. Rhythms, 4 (1989) 365-370. 19 Mrosovsky, N., Double-pulse experiments with non-photic and photic phase-shifting stimuli, J. Biol. Rhythms, 6 (1991) 167-179. 20 Mrosovsky, N., Reebs, S.G., Honrado, G.I. and Salmon, P.A., Behavioural entrainment of circadian rhythms, Experientia, 45 (1989) 696-702. 21 Mrosovsky, N. and Salmon P.A., A behavioural method for accelerating re-entrainment of rhythms to new light-dark cycles, Nature, 330 (1987) 372-373. 22 Mrosovsky, N. and Salmon, P.A., Triazolam and phase-shifting acceleration re-evaluated, Chronobiol. Int., 7 (1990) 35-41. 23 Puchalski, W. and Lynch, G.R., Daily melatonin injections affect the expression of circadian rhythmicity in Djungarian hamsters kept under a long-day photoperiod, Neuroendocrinology, 48 (1988) 28O-286. 24 van Reeth, O. and Turek, F.W., Stimulated activity mediates phase shifts in the hamster circadian clock induced by dark pulses or benzodiazepines, Nature, 339 (1989) 49-51. 25 Reppert, S.M., Rivkees, S.A. and Weaver, D.R., Prenatal function and entrainment of a circadian clock. In S.M. Reppert (Ed.), The De~elopment of Circadian Rhythmicity and Photoperiodism in Mammals, Perinatology Press, NY, 1989, pp. 25-44. 26 Reppert, S.M., Weaver, D.R., Rivkees, S.A. and Stopa, E.G., Putative melatonin receptors in a human biological clock, Sci. ence, 242 (1988) 78-81. 27 Rusak, B. and Zucker, l., The neural regulation of circadian rhythms, Physiol. Ret,., 159 (1979)449-526. 28 Sack, R.L., Lewy, AJ., Blood, M.L., Stevenson, J. and Keith, L.D., Melatonin administration to blind people: phase advances and entrainment, J. Biol. Rhythms, 6 (1991)249-261. 29 Samel, A., Wegmann, H.-M., Vejvoda, M., Maass, H., Gundel, A. and Schutz, M., Influence of melatonin treatment on human circadian rhythmicity before and after a simulated 9-h time shift, J. Biol. Rhythms, 6 (1991) 235-248. 30 Stephan, F.K., Forced dissociation of activity entrained to T cycles of food access in rats with suprachiasmatic lesions. J. Biol. Rhythms, 4 (1989) 467-479. 31 Turek, F.W., The effects of stimulated physical activity on the circadian pacemaker of vertebrates, J. Biol. Rhythms, 4 (1989) 135-147. 32 Underwood, H., The pineal and melatonin: regulators of circadian function in lower vertebrates, Experientia, 46 (1990) 120-128. 33 Vanacek, J., Pavlik, A. and lllnerova, H., Hypothalamic melatonin receptor sites revealed by autoradiography, Brain Res., 435 (1988) 359-363. 34 Weaver, D.R.~ Namboodiri, M.A.A. and Reppert, S.M., lodinated melatonin mimics melatonin action and reveals discrete binding sites in foetal brain, FEBS Lett., 228 (1988) 123-127.

26 35 Weaver, D.R., Rivkees, S.A. and Reppert, S.M., Autoradiographic localization of melatonin receptors in rodent brain by in vitro autoradiography, J. Neurosci., 9 (1989) 2581-2590. 36 Williams, L.M., Martinoli, M., Titchener, L.T. and Pelletier, G., Ontogeny of central melatonin binding sites in the rat, Endocrinology, ! 28 (1991) 2083-2090.

37 Williams, L.M., Morgan, P.J., Hastings, M.H., Lawson, W., Davidson, G. and Howell, H.E., Melatonin receptor sites in Syrian hamster brain and pituitaw. Localization and characterization using [t~l]iodomelatonin, I. Neuroendocrinol., I (1989)315320.