Entrainment of Aged, Dysrhythmic Rats to a Restricted Feeding Schedule

Physiology & Behavior, Vol. 60, No. 5, pp. 1205–1208, 1996 Copyright 01996 E S I USA, All rights reserved P 0031-9384/96 $15.00 + ,00

S

ELSEVIER

E

o nA

D

R

t

Ry t a r se

eS

F

E

1

Department of Psychobiology, University of Cah$orniaIrvine, Irvine, CA, USA and Department of Psychiatry and Human Behavior, The Miriam Hospital and Brown Universi@, 164 SummitAvenue, Providence, RI 02906 USA

R

8

Entrainment of aged, dysrhythmic rats to a restricted feeding schedule. P Aged rats often display abnormal circadian activity rhythms; the rhythm amplitude is low and entrainmentto light-darkcycles is irregular. The activity rhythmof youngrats can be entrained by both light and nonphotic cues, specifically food availability. In young rats, entrainment to restricted feeding cycles does not depend on intact suprachiasmatic nuclei, the presumed anatomical location of the light-entrainableoscillator. In this study, aged rats that displayed disruptedentrainmentto light were assessedfor their ability to entminto restricted feedingschedules.Agedrats, youngcontrols, and youngsuprachiasmaticnuclei-lesioned(SCN) rats were placed on a food restriction schedule (FR) for 14days. Food was available for 2 h duringthe light phase of a 12-hlight-darkcycle. Despite the absence of entrainmentto light/dark cycles, both SCN-lesionedand aged groups showedentrainmentto FR, with clear bouts of anticipatoryactivity during a period of complete fooddeprivationfollowing2 weeksof FR. The results suggestthat the dysrhythmiaof agedrats is a result of natural deterioration of a central circadian light-entrainable pacemaker, but that a secondary oscillator entrainable to f c CopyrightO 1996Elsevier Science Inc. s W B

F

T

6

r

1

1

A

1

E

S

A

DAILY feeding schedules can act as zeitgebers to synchronize the circadian system. Early research demonstrated that rats showed a significant increase in activity 2 to 3 h preceding a 2h daytime feeding (15). Rats given food ad lib for several days following a food-restriction schedule (FR), and then deprived of food, continue to exhibit anticipatory behavior prior to the time when they previously received food (2). Early work with foodrestnction schedules suggested the presence in the rat of a multioscillator system ( 1,3,4,19,20). Evidence for a food-entrainable oscillator (FEO) that is separate and seeondary to the lightentrainable oscillatory system located in the suprachiasmatic nuclei (SCN) has been described ( 17). Food-anticipatory behavior persists in SCN-lesioned animals on food restriction (3,17,19,20), implying that the location of the FEO is somewhere other than in the SCN. The exact location remains unknown (7,9,10). The FEO is believed to be anatomically and functionally distinct from the light-entrainable oscillator, but weakly coupled to it (5,6,17,18). The expression of the FEO seems to be enhanced when the light-etttrainable oscillator is compromised, either by exposure to constant light ( 12), or after lesions of the SCN (3,16,17,19,20). Entrainment of the FEO occurs only when food restriction is within the circadian range ( 1,19,20). In rats, 15 to 22 months old, disruptions in timing of food and water intake, pineal melatonin production, body temperature, and activity patterns are present (21 ). Circadian rhythms of loco-

‘

motor activity and drinking behavior show diminished peaks as a function of aging ( 13). Although the underlying cause of this disruption is unknown, research has shown changes in the hypothalamus of the aged rat. The volume of hypothalamic neurons is greater in old female rats than in younger rats, with a corresponding decrease in neuronal density (8). Rats maintained on FR from ages 3 to 20 months demonstrated entrainment to FR at 20 months ( 11). Because food restriction was maintained throughout the life of the animal in this experiment, from young adulthood into old age, the contribution of this history to the ability to entrain to FR in old age cannot be determined. FR was discontinued from 20 to 24 months and, when reinstated, rats required more time to successfully reentrairt and had lower amplitude rhythms than young rats (11 ). However, 24 months of age represents extreme old age in the rat, and it may be difficult to assess reentrainment in the face of other age-related deficits. Therefore, the present study examined whether or not rats exhibiting compromised light-entrainable circadian rhythms can entrain to FR, if FR is induced after the onset of age-associated dysrhythtnia, but prior to extreme old age. METHOD

Young adult Sprague–Dawley female rats aged 4 to 5 months were used for the control and SCN-lesioned groups. Aged rats were 13-18 months old and showed disrupted circadian rhythms

whomrequestsfor reprints shouldbe addressed 1205

c

WALCOTT AND TATE

1206 TABLE 1

u

B

A

MEAN ACTIVITY LEVELS .

Preliminary

Group

Ad Lib

FR

Deprivation

.

5 s

,00,0

,

-

.

G s d b T c

0

:

Y A S m V

a

r m

d

e

m g f o

l

T w s

e

g

c w

s g r

d

d s s

e

4,000

F

(

7.000

1.000

. .. . .

F

p

C

.

,,000

(

t s d p

as

,.000

s i

a

1 FR+

F

prior to the experiment. All animals were individually housed in 26 cm X 42 cm cages in a temperature- and light-controlled room. The lights were on 12 h per day (LD 12:12), from 0600 to 1800 h. Locomotor activity was measured via an intrapentoneally implanted radio transmitter (Mini-mitter, Inc., Sunriver, OR). The data were collected in 10-min bins by a radio receiver placed under the cage of each animal and transmitted to a personal computer programmed with a data collection and analysis system (Dataquest IIIm; Datasciences, Inc., Minneapolis, MN). The level of gross motor activity was compared between groups of rats, ensuring that the health of the aged rats were not severely compromised prior to evaluation of the circadian pattern of activity. The period and amplitude of a circadian rhythm in baseline data prior to food restriction was analyzed by power spectrum analysis. Comparison between groups of mean maximum power spectrum values was done by analysis of variance (ANOVA) followed by a Tukey’s protected t-test. Graphic representations of activity data (actograms) were generated for each animal, using the circadian data analysis software (Dataquest III*”).

D

D .

L

k-

,.

&

D

0

4

8

12

T

F a l h L (

a

r

r p

p 2w f

f

i

r

p 2d

T

s s

p p

( B

e o

s

i

T

t

r a

0

(

f t 3d r s

~.

24

(

y

( c

d

r

1

f

p

f o b

I’2n6

d a

f A p a

d e

a N

SCN Lesions Animals were deeply anesthetized with sodium pentobarbital prior to stereotaxic placement of lesion electrodes. Stereotaxic coordinates were AP – 1.3 mm, lateral ~ 0.3 mm, DV 9.5 mm ( 14). Electrodes were size 00 stainless steel insect pins, insulated with EpoxyliteTM, with a 0.5-mm uninsulated tip. Bilateral electrolytic lesions were created using a Grass Lesion Maker. Radio transmitters were implanted at the time of stereotaxic surgery. Animals were allowed to recover for 1 week before behavioral data was collected. Food Restriction Baseline activity data were collected from rdl animals for lne week, and then the animals were placed on a food-restriction schedule (FR) for 14 days. During food restriction, rats had daily access to rat chow from 1300 to 1500 h (lights on at 0600 h). Rats had unlimited access to water throughout the experiment. On day 15, food was returned to ad lib availability for 3 days. On day 18, food was removed for 48 h (deprivation), and returned to ad lib on day 20. Animals continued to receive food ad lib for 3 days (days 20–22). On day 23, food was again removed for 48 h, with the experiment terminating on day 25. The groups tested on this schedule included: young, food-restricted rats (YFR, n = 5); young, SCN-lesioned rats (SCN, n = 6); and aged rats (AGED, n = 11). Young, nonFR rats (YC, n = 5), fed ad lib throughout the FR phase, but deprived and fed ad lib with the other groups, were also included to control for effects

of food deprivation. Data from the 2 deprivation periods were analyzed for the presence of activity anticipatory to the previous time of food presentation. Activity during the light phase was analyzed for the presence of an anticipatory bout. A bout was defined as activity with a duration of at least 100 min and an amplitude of at least the mean level of activity of the preceding 24 h. If a bout of activity was present, the onset time was calculated. After the experiment was completed, all lesioned rats were sacrificed by cardiac perfusion and the brains examined histologically to confirm the presence of SCN lesions. RESULTS

Young animals lost an average of 10.5% of their body weight during food restriction, and aged animrds lost an average of 7.5%. Mean activity levels were calculated for all groups during each of the following: the l-week baseline data collection period; the 2-week FR period; the 2 3-day ad lib feeding periods; the 248h food deprivation periods. There were no significant differences in the mean activity levels for any of the groups during any phase of the experiment (Table 1). Prior to food restriction, young SCN-intact animals had robust circadian activity rhythms. In Fig. IB, activity data for a representative young control animal prior to food restriction is shown in the area labeled baseline. A pronounced circadian activity rhythm is present, with a period of 24

1207

FOOD ENTRAINMENT OF AGED RATS

1

q r 1

A 1

1

1

J

1

1!7 I I!

11ii t QIMl

n

I

d

1 o

2

1 T

(

h, as indicted by power spectrum analysis for the baseline data (Fig. IB, top). All young animals showed entrainment to food restriction, as demonstrated during the period of deprivation by anticipatory bouts of activity prior to the 1300-h feeding (Fig. 2a). The behavioral pattern displayed by the animal depicted in Fig. 1 was typical of the group. At the beginning of the food restriction period, young intact rats demonstrated activity prior to and coincident with the time of food presentation (Fig. IB, FR arrow). The group mean onset time of this anticipatory bout on 3 consecutive days during week 2 of FR was 0938 ~ 12 rnin, with a mean bout length of 285 ? 12 min. By the second week of FR, 3 of the 5 young animals experienced a noticeable loss in nocturnal activity. One of these animals displayed a brief bout of activity at lights out, followed by marked stretches of inactivity. The other 2 showed no nocturnal activity bouts, leaving only the food anticipatory bout as the major activity expressed by the animal. Following FR, food was provided ad lib for 3 days, followed by 48 h of food deprivation (Fig. IB, AD LIB). All 5 young FR animals exhibited anticipatory behavior during food deprivation. The mean onset of the bout of activity during food deprivation was 1025 h ? 17 tin, with an average bout length of 207 ? 20 min. In contrast, in the control group of young animals that were not previously exposed to FR, none showed an anticipatory bout of activity prior to 1300 h (Fig. 2d). Aged animals had disrupted circadian activity patterns prior to FR (Fig. 1A). Mean peak power spectrum values within the a circadian range were significantly different from mean values for the young, SCN-intact group (young animals 0.019t 0.0002; aged animals 0.004 k 0.0005;.- p < 0.05). Three of the . eleven aged animals showed no significant peaks in the circadian range (periods between 22-27 h). During FR, 10 of the 11 aged animals showed anticipatory activity (Fig. IB ), with a mean onset time of 1002 h t 7 rein, and a bout length of 290 min i 20 min. In addition, 8 of the 11 animals exhibited a consolidated nocturnal bout of activity during food restriction. Two of these animals had not exhibited clear nocturnal bouts prior to FR. Eight of the eleven animals displayed an anticipatory bout during food deprivation (Fig. 2b), with a mean onset of 1013 h k 10 tin, and a bout length of 305 min ~ 33 min. The mean lemzths of the anticiuatorv bouts of the aged animals did not differ sig~ificantly from tke o~er groups. Of he 3 animals that did not entrain, 1 showed anticipatory activity during FR, but not during deprivation. The other 2 animals showed anticipatory bouts during deprivation with durations of less than 100 rnin, but these rats were neither the oldest nor the ones with the most severely disrupted rhythms during baseline data collection. The SCN-lesioned animrds had severely disrupted activity rhythms. Mean peak power spectrum value (mean t SEM) within the circadian range was 0.001 ~ 0.0002 (SCN vs. young intact, p < 0.001). All SCN-lesioned animals showed an anticipatory activity bout during food restriction (Fig. IB ), with a mean time of onset of 1027 h t 6 rein, and a mean bout length of 313 f 17 min. One of the SCN-lesioned rats failed to show

F f a r

A a1 e

d f

12 f

s

to 1

t p

D h

a b 1

f d

g

f

d

a

e f r

e e

f i

1 a g

t y

a y

a

d G

r

S y p

p p

e

c f r F b f

1000

1208

WALCOTT AND TATE

an anticipatory bout during deprivation; for the other 5 animals, the average time of onset of the bout of anticipatory activity was 1012 h f 10 rein, with a mean duration of 274 min t 18 min (Fig. 2c). Histological analysis confirmed complete lesions in all 6 animals (data not shown). DISCUSSION

All young animals showed entrainment to FR. The activity bout anticipatory to food presentation during the restricted feeding schedule is an indication of response to the food presentation, but not necessarily indicative of entrainment to it. However, the bout of activity that occurs during food deprivation, with a phase similar to the time of previous food presentation, does indicate that entraittment occurred. The onset of anticipatory activity in the young animals shifled from 0938 t 12 mitt during FR, to 1025 t 17 min during deprivation; however, the mean length of the bout did not change. The onset times that the young animals exhibited during deprivation more closely approximate the onset times of the aged and SCN-lesioned groups. It is not clear why this shift occurred. SCN-lesiotted rats successfully entrained to the restricted feeding schedule. Therefore, the restrictive feeding schedule employed in this study was sufficientto entrain an animal in the absence of theprirnary circadian oscillator, as has been previously reported. A food-restriction schedule within the circadian range elicited anticipatory behavior in aged rats that had previously displayed abnormal rhythms on light-dark cycles. The capacity of aged rats to entrain to food restriction, although unable to show regular entrainment to light-dark cues, indicates that food can be an entraining cue, even when light entrainment appears to be weak. Several of the aged animals also exhibited a consolidated bout of nocturnal activity during FR that had not been present during baseline data collection. Thus, when entrained to FR, aged ani-

reals show anticipatory activity prior to food presentation and an a consolidated bout of nocturnal activity. A portion of the ~oup of young rats displayed a loss of nocturnal activity from day 8 of food restriction, in addition to an extended daytime bout of activity often continuing beyond the time of food presentation. This suggests the possibility that the daytime activity bout induced by a resticted feeding schedule is actually a shifting of nighttime activity, specifically of nocturnal activity previously associated with eating. The remaining nocturnal activity may be light-entmined activity that continues to be expressed nocturnally in the presence of FR. The shift in activity in the young animals during FR does not appear to be the evocation of additional extmneous activity, because overall mean activity levels do not differ from baseline data collection to FR. It maybe that the expression of an intact food-entrainable oscillator gates food-related activity to a nocturnal time during ad lib feeding. During FR, this activity becomes phasesltitled to the day hours. In summary, we report here that aged rats that display compromised activity rhythms in the presence of light/dark cycles cart entrain to restricted feeding schedules. The results suggest that there may be a sequence to aging, such that the light-entrainable rhythms become impaired while food entrainable rhythms may still be intact. An intact food-entrainable oscillator may be one mechanism to restore some components of rhythmicity to otherwise severely dysrhythmic animals. ACKNOWLEDGEMENTS

a i C s R

w J

t

J

W

M w

a

A

c

a

r

ag

f

w

t m

m w

A

F

A

REFERENCES B

R

1 B r C S C

F r

b

s B

B

R

1 T

F

N

B C r

P

m 1

d 4 m

E P

B

F

a

r

l B

P

A

e 1

m

c

c

1

c 1 o

b

r

a

r

1 H

N

h

H

n p

r 1

3

u

H 2 V r

v

e

p

r

n 1 h d f

d

f

P

n

f l

r

e

G

a

rat. B

s

R

1 r P

R f B

r 3

H f

c

R w v 1 M r

a h r

d

l

r M

s

2

a A

N D r

f

e

c B

C

N

1 W

b

A

P r

p

l 1

c F

a

d w s

r 1 B f a S

s N

B B

r

C

p r

i

P

M b

P

B

e l

B

Tc R

y

i

S f

c

e

a

E P r 2

B

N

s

1 s

Ab 1

1

4 S d S f B S r B P r

o

7 E

r

E

G C

I

M

r

f

c

C

1

R o S f 3 S

P

l

H c

f

r A

c

r

1 M r B P

1 r

e

b

1

3

t 1 M

Y A

r

a

R P B C o

H as

2 E d E l B H m t h H o I a 2 M t l M i

T o

c

w

t B S

a 4

y

A s

l

2 n

e

1 S

s 2 M R G

various

a 1

r 1 A V

w

E s

c l c

S H

e

r F o1

5 e i i