Compound structure of rodents activity rhythm

Physiology & Behavior, Vol. 57, No. 1, pp. 37-40, 1995 Copyright © 1994 Elsevier Science Ltd Printed in the USA. All rights reserved 0031-9384/95 $9.50 + .00

Pergamon 0031-9384(94)E0108-G

Compound Structure of Rodents Activity Rhythm A. T I C H E R l A N D

I. E. A S H K E N A Z I

Chronobiology Unit, Department of Human Genetics, Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel 69978 R e c e i v e d 16 F e b r u a r y 1993 TICHER, A. AND I. E. ASHKENAZI. Compoundstructureof rodentsactivityrhythm.PHYSIOL BEHAV 57(I) 37-40, 1995.-In the present study an attempt to determine the fine structure of rodents activity rhythm was carried out. To eliminate masking effects which are produced by the active influence of the monitoring system on the behavior of the animal (e.g., running wheel) we designed a passive infra red detection system. Rats were exposed to light-dark 12:12 [LD(I 2:12)] cycles and mice to LD(I 2:12), LD(8:16) and LD(I 6:8) cycles. Multiwave patterns of activity were observed in both rodents genera which differ from each other in the number of activity bouts and the periods of the activity rhythm components. In LD(I 2:12), rats exhibited 2 bouts of activity and 1 bout of rest which were attributed to the presence of 24 and 8 h components. Mice, exposed to the three varying ratios of L to D regimens exhibited 3 bouts of activity and one bout of rest which were attributed to the presence of 24 and 6 h components. The relation of the compound structure to the 24 h rhythm is discussed. Rhythm-components

Mouse

Rat

Infrared detector

Behavior

INTRODUCTION

Motor activity

activity rhythms of photo periodically synchronized rodents possess more then one component; (b) to examine if rodents genera exhibit inherited differences with regard to the number and period of these components and (c) to assess the effects of varying light-dark [LD] cycles on the number, location and periods of rhythm components.

THE suggestion that biological rhythm-period is controlled by a multiple allele or a polygenic system (l) raises the possibility that more then one component (oscillator) can participate in structuring the wave form of a rhythm. Pittendrigh and Daan (9) were among the first to examine this phenomenon by studying the running wheel activity of rats which were exposed to constant light [LL]. They observed that under prolonged exposure to LL, the uni-modal rhythm was split to two components, each of which controlled by separate oscillator, one oscillator synchronized with dawn (the Morning oscillator), and the other with dusk (the Evening oscillator). The process of identifying rhythm components requires long and continuous monitoring. Yet, it is highly important to ensure that the animal-monitoring-system [AMS] will not induce any masking effects, interact or influence the normal behavior of the animal. The use of the running wheel affects rhythm activity patterns of rodents (3,13) and is considered as active detection which masks the inherited structure of the rhythm. It is therefore preferable to use infra red [IR] detectors or other methods which are viewed as passive monitors. An additional attractive feature of IR detection especially suitable for this purpose is the fact that in rodents the spectral sensitivity of the rhythm phase-shifting response is maximal near 500 nm (6,12) while the IR diode has a wavelength of 900 nm and thus will not affect this characteristic of the rhythm. It was Buttner (2) who recommended to modify IR detectors of an alarm system (source of the IR and the sensor in one unit) to measure motor activity. Using the latter idea we assembled a simple IR-AMS specially designed for the fine detection of rodents activity rhythms. The aims of this study were (a) to determine if the

METHODS

The IR Monitoring System A simple digital monitoring system was assembled for the specific aim of this study. It utilized a multiple 8255 I/O peripheral interface card (Racom Electronics, Israel or Computer Boards, USA) which provides 192 parallel I/O programmable channels. The IR detector is composed of an IR diode (MLED930 Motorola or compatible) functioning as a light source and a phototransistor (MRD310 Motorola or compatible) functioning as a target. The lenses of both are convex and powered by the 5 V of the computer bus. The phototransistor is located in a 3 cm long copper pipe to eliminate cage light interference. The sensors are placed at a height of 4 cm, parallel, on a tray and are separated from each other by a distance of 25 cm. The animal cage is placed between the sensors.

Animals Twelve-week-old, 20 ICR male mice and 3-wk-old, 4 Wistar male rats were, individually, housed in 40 x 23 x 14 cm clear polystyrene cages. Groups of 4 cages were placed in acoustic and ventilated boxes. 4 mice and 4 rats exposed to a lighting regimen of 12 h light : 12 h dark cycles [LD(12:12)] (light on: 0600), 8

This study is part of Mr. A. Ticher's Ph.D thesis in the Faculty of Medicine, Tel Aviv University. To whom requests for reprints should be addressed. 37

38

TICHER AND ASHKENAZI

mice exposed to LD(8:16) and 8 mice exposed to LD(16:8) (light on: 0500). Light intensity was 100 lx and food and water were freely available throughout the experiment.

to LD(12:i2) were monitored for 24 days. In the second set of experiments (Experiment 2) mice were exposed to two illumination regimen, LD(8:16) and LD(16:8). The detailed experimental designs and analyses are provided in the "results" section.

Experimental Design

RESULTS

Experiment 1

Once the reliability and the accuracy of the IR-AMS was verified, two sets of experiments were designed. In the first set (Experiment 1) the daily motor activity rhythms of mice and rats exposed

In this set of experiments the motor activity patterns of 4 mice and 4 rats exposed to LD(12:12) cycles were monitored for 24 L

A

,

•

.~,itlt

i.i

_

•

ILl.

1~ i

141-i

|

ill

.

Jk

.

.

.

.

.

.

I L ~

.

.

.

ALL

IL

llt

.

&AIk

L

~

~

_..J.dL.

~

.

~

__It

_ .LI

.

_

+ ~_m

-IlL

. ~

J•

t

.

__

~

~

• ..

/.

~.

~

,.It.

.

_

It..

,

I~.

1,..,..

~t

~ _

It,...,t.,+

.L. .

I.

.

_+.~t

.

.

.]'

L

d •

•

.t

.

.

.

LMJ,

•

Jt a.~ ~

~

- ~

,I

,

jti

t

.

.

.

,a_L ,tl. "

.

', :'+t t+t t t ' , ' , ' . t ' . ' , : ' , l 11111 1 1 0 6 12 18 24 6 t2 18 24 T I M E

..

•

k

.1

J

.L~J..

_

.a

.

~,L,a*

bk

LtA._

,

t J,

_

JILl.

i

mr J

•

-

.

t

.

d

_1

j

_ It ~

.It

.,11 . | ~ , l i l i l m _ su _LLlJk.k

. . . . •

.

_

.

.

Idl ,1~ •

_

I [ ..,A " - I

,. .

i.,J

d

i

....... .

--

t

.i,J,

i

IJ

~

_

A..~

-

i~

_,,m,,

~I-LIL

LL

.

_ LJ ,

,

+

L

" +..-" n. .Ja . ~.. ~ , .I

•

I

dl...

_ .

.t . b

.

-L

~.k

+', ', '. t i t ', '. ', ', ', '. I I I I I I I I I I I I

0

E

t2

( H Q U R S )

18

T I M E

200

in k

,,, , _

,,,,,,.A,L

m.,. _ _ . . u . _ lilt . I . _~td J.Jdl.

l

+

IL--~

• . . . .

lit . LJ. L''" . . . . .

.

+ .

..II.IW _ _ ._ l i l t .

- "+

[---

..

--I

_

•

L ~ t

l . k

:': .,,._,a.

J..il~

.BI

,

k

X t .

•

.....

4ti..~l,

.

LaI.~L

Ilia _

tl+LJb. • Mikt+ jh.,.

.

Jl~

_

Jt

.

f.

L

d

_~L

,

Jlm,~

k Jk

L

._

+at,ilk

4

•

•

..

-~i

.L.,..~

.

_

.

L_

.L

~

.+L__... .

. t +

~

+.i.ll.

. t

LL~.t.

IlL-

+ ,-a _

.,l~mL

..,

• .I.~ ~-

i

a

,i ~

~ L ILL

J,.

24

6

12

18

24

( H O U R S )

200

C

D

IJJ

100

100

0 0

0

12

18

24

3 0

3 6

0

6

12

H O U R S

18

;24

3 0

3 6

H O U R S

6

F

i:

4

2

2

1

0

O

12

161 TIME

24 ( H O U R S )

6

12

12

1B TIME

24

6

12

( H O U R S )

FIG. I. T w e n t y - f o u r h a c t i v i t y pattern o f m i c e and rats e x p o s e d to L D ( 1 2 : 1 2 ) . ( A ) m o u s e a c t o g r a m ; ( B ) rat a c t o g r a m ; ( C ) chi square p e r i o d o g r a m o f m i c e ; ( D ) chi square p e r i o d o g r a m o f rats; (E) a v e r a g e d daily m o t o r activity o f m i c e ; ( F ) a v e r a g e d daily m o t o r activity o f rats. D o t s and vertical lines d e s c r i b e the m e a n and the S D . C o n t i n u o u s line d e s c r i b e s the c o s i n e fit c u r v e . T i m e units are 15 m i n . T h e lighting s c h e d u l e w a s L D ( 1 2 : 1 2 ) and the b l a c k bars at the b o t t o m o f the plots i n d i c a t e the dark periods.

RODENTS' RHYTHM-COMPONENTS

39

"IA

800

I

~a

1

60O

~400 I.I. O

ae 1

0 0

4

8

12

16

20

24

4

0

6

12

18

24

30

36

0

6

12

18

24

30

30

800

~

600

3 IJJ

>;,2

~400

,,--, LIO

B

200

ae 1

0

12

16

20 TIME

24

4

8

12

(HOURS)

HOURS

FIG. 2. Twenty-four h activity pattern of mice exposed to LD(8:I6) and LD(16:8). (B) chi square periodogram of mice under light regimens of LD(8:I6); (A) averaged daily motor activity of mice under light regimens of LD(8:16); (D) chi square periodogram of mice under light regimens of LD(16:8); (C) averaged daily motor activity of mice under light regimens of LD(16:8). Dots and vertical lines describe the mean and the SD. Continuous line describes the cosine fit curve. Time units are 15 min. The black bars at the bottom of the plots indicate the dark periods.

days. The resulting patterns were analyzed as follows: (a) For each of the examined animals, the prominent periods (tan) were determined from the total series time data by the Chi square periodogram according to the Sokolove-Bushell method (10). (b) The daily activity (DayActivity[i]) was defined as the sum of recorded activity in each day (i). (c) The recorded activity of each 15 min time point in the 24 days (Activity[ij], for i = 1..24 and j = 1..96) was expressed as the percentage of its corresponding daily mean (ActivityPercent[ij] = Activity[ij]/Day Activity[i]*100), (d) The compared patterns of the 4 animals were pooled and averaged to form one 24 h pattern and the best fit curve of that pattern was computed by using the cosinor algorithm (5,7) compatible to most probable tans (harmonies) as determined by Chi square periodogram. Figure IA presents the double plot actograms of one mouse and Fig. 1B that of one rat. The highest observed activity levels were 182 (mouse) and 256 (rat) cutting beams per 15 min. Chi square periodogram analysis produced statistically significant periods of 24, 8 and 6 h for the mice activity pattern (Fig. 1C) and periods of 24 and 8 h for the rats activity pattern (Fig. 1D). The first activity bout of mice occurred l h after the offset of lights (1900 h), the second in the middle of the dark period (2400 h), the third was observed 1 h after light onset (0800 h) (Fig. IE) and in total, the activity spread over 18 h. In rats, one bout of activity occurred 1 h after light offset (1900 h) and the second l h before light onset (0500 h) (Fig. 1F). The cosinor algorithm with the best harmonics cosine functions were fitted to the daily

mean activity of mice and rats (24 and 6 h period and 24 and 8 h, respectively) (Figs. 1E and F, cosine fitted curve).

Experiment 2 In this set of experiments one group of 8 mice was exposed to LD(8:16) and another group of 8 mice to LD(16:8). The activity was monitored for 14 days and the recorded activity of each time point was expressed as percentage of the corresponding daily mean (Fig. 2A, LD(8:16), Fig. 2C, LD(16:8)). Chi square periodogram analyses are exhibited in Fig. 2B for LD(8:16) and in Fig. 2D for LD(16:8). In LD(8.'16) and LD(16:8), the mice exhibited three bouts of activity, similar to LD(12:12) activity patterns. In LD(8:16) mice (Fig. 2A) the first bout of activity occurred 2 h after the offset of lights (0700 h), the second was in the middle of the dark period (1400 h) the third was observed at light onset (2100 h) and in total, the activity was spread over 19 h. In mice exposed to LD(16:8) (Fig. 2C), one bout of activity occurred 1 h after light offset (2200 h), the second was at the end of the dark period (0400 h), the third 5 h after light onset (1000 h) and in total, the activity was spread over 17 h. Chi square periodogram analysis produced statistically significant periods of 24, 8 h for the LD(8:16) mice activity pattern (Fig. 2B) and periods of 24 and 6 h for the LD(16:8) activity pattern (Fig. 2E). When these periods were integrated in the cosinor algorithm with the best harmonics cosine functions and fitted to the dally mean activity of mice (Figs. 2A and 2C) a significant fitness was ob-

40

TICHER AND ASHKENAZI

tained for the 24 and 8 h periods (R% 0.87, p < 0.0001, Fig. 2A) and for 24 and 6 h periods (R% 0.68, p < 0.0001, Fig. 2C). DISCUSSION Studies of rodents activity patterns, using running wheel detection, showed that rhythm components have been observed only when the animals were exposed to LL (9) or when the normal environmental cues were interrupted (8). In this study we observed the presence of a compound structure in activity rhythms of rodents even when the animals were entrained by LD cycles (8,16, 12:12, and 16:8). Furthermore, the activity-rhythm structures of mice and rats different from each other with regard to the number of activity bouts and components periodicities. Both rodent genera exhibited two activity bouts during the dark period, and the mice had in addition a third bout of activity at the beginning of the light period (Figs. 1E and F). Multiwave patterns have been previously described for the rhythms of other physiology variables of rodents. The dual modal pattern observed in rats motor activity (Figs. IF) is similar to the one observed for their drinking behavior pattern under identical experiment conditions (4) and that observed in mice (Figs. 1E) is similar to their Carbon Dioxide emitted pattern (11). It is possible that many rhythms possess a compound structure which is not detectable by certain monitoring systems or superimposed by masking effects induced by the detecting methodology. We assume that in the present study the elucidation of the components was enabled by the use of the passive IR detectors. Does the ratio of L to D periods affect the structure of the compound rhythm? In LD(12:12) Chi square periodogram analysis produced statistically significant periods of 24, 8 and 6 h for mice and periods of 24 and 8 for rats. However, the daily mean of activities (Figs. 1E and IF) includes four bouts (three of activity and one of rest) in mice and three bouts (two of activity and one of rest) in rats. When the various periodicities were fitted to the daily mean activity of LD(12:12) mice by use of the best harmonies cosine function the 6 h period had a better fitting in harmony then the 8 h one. Consequently we concluded that under LD(12:12) the dominant periods of the mice activity rhythm are

24 and 6 h and those of rats are 24 and 8 h. The dissection of the activity patterns of mice subjected to LD(16:8) and LD(8:16) showed also the presence of four bouts, three of activity and one of rest. However, Chi square periodogram analysis produced statistically significant periods of 24 and 8 h for the LD(8:16) (Fig. 2B) and 24 and 6 h for the LD(16:8) (Fig. 2D). Integrating the length of the total activity pattern in the computations reveals that when the total activity period is shorter than 18 h, the spectrum analyses will yield a significant component of 6 h period while if it greater than 18 h, the spectrum analyses will yield a significant component of 8 h period. Thus, it seems that the activity rhythms of rodents are structured by an assembly of components. Different genera exhibit inherited differences with regards to the periodicities of these components. It is tempting to assume that the period of the components (6 or 8 h) reflects the endogenous control of the activity rhythms and it is not affected by varying the LD ratios while the structure of the 24 h rhythm is affected by the exogenous illumination regimens (the length of the dark period). The output (structure and amplitude) of the combination between these entities (24 and 6 or 8 h) structures the activity pattern of the animal and can explain some of differences in the activity wave patterns obtained under various lighting conditions. The present observations (a) support the notion that 24 h rhythms are compound and structured by more than one component (oscillator); (b) emphasize the need to use passive detection methods to elucidate the presence of the short period components even in entrained conditions; (c) elucidated differences among rodents genera with regard to components number and periodicity; and (d) stress the necessity to monitor activity patterns under varying LD cycles and to account for the total period of activity in the computation of the periodicities. ACKNOWLEDGEMENTS We are grateful to Prof. 1. Gozes for her assistance, to Mr. B. Starobinets for his help with the electronic devices, to Drs. A. Mattes and D. Buttner, Germany, for the statistical algorithms, and to C. J. Abraham for his comments. This research was supported by Racom Electronics, Israel.

REFERENCES 1. Ashkenazi, I. E.; Reinberg, A.; Bicakova-Rocher, A.; Ticher, A. The genetic background of internal desynchronization of circadian rhythms in healthy human adults. The Am. J. Hum. Genet. 52:12501259; 1993. 2. Buttner, D. The use of passive infrared detectors to measure locomotor activity of laboratory animals. Z. Versuchstierkd 31:137-141 ; 1988. 3. Edgar, D. M.; Kiiduff, T. S.; Martin, C. E.; Dement, W: C. Influence of running wheel activity on free-running sleep/wake and drinking circadian rhythms in mice. Physiol. Behav. 50(2):373-378; 1991. 4. Johnson, R. F.; Johnson, A. K. Light-dark cycle modulates drinking to homeostatic challenges. Am. J. Physiol. 259:R1035-R1042; 1990. 5. Mattes, A.; Witte, K.; Hohmann, W.; Lemmer, B. PHARMFIT--A nonlinear fitting program for pharmacology. Chronobiol. Int. 8:460476; 1991. 6. McGuire, R. A.; Rand, W. M.; Wurtman, R. J. Internment of the body temperature rhythm in rats: Effect of color and intensity of environmental light. Science 181:956-957; 1973. 7. Minors, D. S.; Waterhouse, J. M. Circadian rhythms in general practice and occupational health. In: Arendt, J.; Minors, D. S.; Water-

8. 9. 10. 11.

12. 13.

house, J. M., eds. Biological rhythms in clinical practice. London: Wright Press; 1989:207-224. Mrosovsky, N.; Janik, D. Behavioral decoupling of circadian rhythms. J. Biol. Rhythms. 8:57-65; 1993. Pittendrigh, C. S.; Daan, S. A. A functional analysis of circadian pacemakers in nocturnal rodents. V. Pacemaker structure: A clock for all seasons. J. Comp. Physiol. (A). 106:333-335; 1976. Sokolove, P. G.; Bushell, W. N. The chi square periodogram: its utility for analysis of circadian rhythms. J. Theor. Biol. 72:131 - 160; 1978. Stupfel, M.; Perramon, A.; Gourlet, V.; Thierry, H.; Lemercerre, C.; Davergne, M.; Monvoisin, J.-L.; Da Silva, J. Light-dark and societal synchronization of respiratory and motor activities in laboratory mice, rats, guinea-pigs and quails. Comp. Biochem. Physiol. 70A:265-274; 1981. Takahashi, J. S.; DeCoursey, P. J.; Bauman, L.; Menaker, M. Spectral sensitivity of novel photoreceptive system of mammalian circadian rhythms. Nature 308:186-188; 1984. Welsh, D. K.; Richardson, G. S.; Dement, W. C. Effect of running wheel availability on circadian patterns of sleep and wakefulness in mice. Physiol. Behav. 43(6):771-777; 1988.