Caloric restriction and melatonin substitution: Effects on murine circadian parameters

Brain Research 1048 (2005) 146 – 152 www.elsevier.com/locate/brainres

Research report

Caloric restriction and melatonin substitution: Effects on murine circadian parameters David Resuehr1, James Olcese* Institute for Hormone and Fertility Research, Centre for Innovative Medicine, University of Hamburg, Falkenried 88, 20251 Hamburg, Germany Accepted 22 April 2005 Available online 23 May 2005

Abstract Aging effects have been reported in endocrine, metabolic and behavioral circadian rhythms. The effects of age on the circadian system have been investigated primarily in rats and hamsters and only seldom in mice. Our aim was to assess the effects of two common ‘‘antiaging’’ treatments, namely caloric restriction (CR) and melatonin substitution, on the circadian system of mice. Animals were subjected to phase delays of the light – dark cycle and constant darkness (DD). The most pronounced change in the murine circadian system was the length of the endogenous period, s, which increased with age regardless of treatment. CR had diverse effects e.g., enabling a more rapid phase shift response while concomitantly leading to a fragmented circadian phenotype with considerable activity during the rest (light) phase. Melatonin enforced the adaptation to the light/dark cycle, thus facilitating a rapid reentrainment to phase delayed lighting conditions. Interestingly, the melatonin-substituted animals displayed an increase in locomotor activity under constant darkness and in 50% of all cases a biphasic (split) activity pattern. These results contribute to the phenotypic evaluation of two very different approaches to intervene in the age-related degeneration of the mammalian circadian system. As both CR and melatonin have negative and positive effects on the behavioral expression of clock function (i.e., fragmentation of rhythms vs. faster reentrainment), their usefulness in managing age-related circadian disorders may be limited. D 2005 Elsevier B.V. All rights reserved. Theme: Neuronal basis of behavior Topic: Biological rhythms and sleep Keywords: Melatonin; Caloric restriction; Circadian period s; SCN; Locomotor activity; DD; LD

1. Introduction The synchronization of cellular, physiological and/or behavioral rhythms to a cyclically changing environment is adaptive, ubiquitous and probably evolved early in the history of life. One of the most prominent rhythms among animals is the circadian day/night activity rhythm [34].

* Corresponding author. Current address: Department of Biomedical Sciences, Florida State University College of Medicine, Tallahassee, FL 32306, USA. Fax: +1 850 644 5781. E-mail address: [email protected] (J. Olcese). URL: http://www.med.fsu.edu (J. Olcese). 1 The data represent a portion of the doctoral dissertation of D.R. submitted to the University of Hamburg, Department of Biology. 0006-8993/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2005.04.063

In the aging animal, the circadian system undergoes many changes at multiple levels. Adaptation to altered daylengths takes longer, activity rhythms are more fragmented and the endogenous circadian period (s) changes as compared to young animals [21,29]. Disruptions in the circadian system lead to a reduced life expectancy in many animal models [11,15,16]. Melatonin has diverse biological functions, one of which is to serve as an output signal from the circadian clock [26]. In many animal species, it also controls more slowly changing processes, such as fertility and fur growth (e.g., in hamsters, sheep, [12,37]). The synthesis and release of pineal melatonin are greatly reduced in the aging individual when compared to youthful secretory levels [25]. Melatonin substitution in aging animals increases the maximum life

D. Resuehr, J. Olcese / Brain Research 1048 (2005) 146 – 152

span [31] and can reduce the severity of other senescenceaccompanied changes [17,33]. Another approach toward reducing age-related physical degeneration is caloric restriction (CR) which has been shown in many model systems to delay the onset and severity of age-related diseases [41,42]. However, the effect of CR on the circadian system has not been investigated at the behavioral level in terms of free-running periods or in terms of reentrainment efficiencies after phase shifting nor in terms of a comparison to melatonin effects on these parameters [8,9]. In the present study, mice were maintained for 10 months (beginning at 5 months of age) under one of three treatment groups: CR, melatonin substitution or a controlled normocaloric diet. During this time entrainment to light:dark (LD) cycles, as well as free-running periods under constant darkness (DD) and phase shifting efficiency were assessed in all groups.

2. Materials and methods 2.1. Animals Mice of the hybrid strain B6C3F1 were obtained at the age of 6 weeks (N = 80) (Charles River, Wilmington, MA). This strain has often been used in studies of aging [7,18]. In contrast to many other laboratory mouse strains [13], the B6C3F1 strain (derived from a C57BL/6  C3H cross) is melatonin-proficient (Olcese et al., unpublished). Young control mice (N = 20) were kept for longitudinal assessment of locomotor activity during the course of the study. Additionally, beginning at 5 months of age, another 60 mice were weighed and sorted into three groups of 20 animals each: controls (CT), melatonin-substituted and CR. All groups of mice had the same average initial body weight. Body weight was then measured weekly for the first 14 weeks and every 2 weeks thereafter to assess the impact of CR. Animals were housed individually in Plexiglas cages with overhead infrared sensors to record locomotor activity (see below for details). Ambient humidity was 68%, and room temperature was 20 T 2 -C. At the beginning of the experiment, the light cycle was set to 06:00 h lights on and 18:00 h lights off. For subsequent experimental purposes, the light cycle was occasionally modified as described below. Periods of constant darkness were imposed for a minimum of 3 weeks to determine endogenous period s. 2.2. Feeding Specially formulated mouse chow was purchased from Teklad (Madison, Wisconsin). Control and melatoninsubstituted animals were fed standard chow (AIN-93M adult maintenance precision pellets). CR animals were fed an especially prepared, vitamin- and mineral-enhanced restriction diet (based on AIN-93M) to assure proper

147

nutrition in the face of caloric restriction. This diet has been employed frequently in studies on the effects of caloric restriction [5,18,42]. At the beginning of the experiment, feeding was ad libitum. The amount of food consumed on a daily basis was assessed by offering each individual a preweighed amount and measuring the uneaten amount remaining on the following day. The average amount consumed was 4.3 g/day which was defined as 100%. Thereafter, food was provided three times a week (a commonly used feeding strategy [6]). On Monday and Wednesday, a 2-day portion was given, while on Friday the mice were fed a 3-day amount of chow. In the case of the CR mice, initial food amounts were reduced biweekly by 20% until a level of 60% was achieved which was then maintained over the course of the study (65.8 kcal/week). The melatonin mice received the same amount of food as the control mice, whereas the former received melatonin in their drinking water at a concentration of 5 Ag/ml. Melatonin (Sigma, St. Louis, MO) stock solutions were dissolved in 96% ethanol which were then diluted 1:1000 in water. The CR and CT animals also received alcohol in their drinking water (0.1%) as a vehicle control. The drinking bottles were painted black to protect melatonin from photodegradation. 2.3. Locomotor activity recordings To record individual locomotor activity, passive infrared detectors (Conrad-Elektronik, Art.-Nr.192236-3F) were mounted on top of all cages (N = 20 per group). The exact timing of the light –dark cycle in the facilities was recorded with a photodiode (Conrad-Elektronik, Art.-Nr. 606863-3F). Locomotor activity and ambient lighting signals were measured as TTL-Signals (electrical currents between 0 V and 5 V) and relayed to a personal computer (IBM PC 286, 500 MB HDD, 32 MB RAM) via a digital I/O-Card (Conrad-Elektronik, Model Nr. PIO 24/48 II). The absence of activity ‘‘0’’ was defined as 0– 0.8 V, whereas 2.4 – 5 V was defined as ‘‘1’’—or activity. A refractory period of 7 s after each signal was established in the computer program, during which time no further signals could be recorded. Signals were recorded in 6 min bins per recording interval, i.e., 240 intervals per day. The recording and analysis of locomotor activity and the determination of endogenous period s (using a chi square periodogram) were performed with a Quickbasic computer program written by Dr. Haiko Dernbach (Tiermedizinische Hochschule Hannover, Germany). Reentrainment analyses after phase shifts were conducted by calculating the amount of activity in the 40 6 min bins immediately following the onset of darkness before and after shifting. A threshold level was determined by forming the average activity under normal LD conditions i.e., before the shift (N = 20 animals per tested group). Attainment of this amount of nocturnal activity after shifting was considered as the threshold for reentrainment.

148

D. Resuehr, J. Olcese / Brain Research 1048 (2005) 146 – 152

Statistical data analysis was performed on a Dell Dimension 2400\ PC using GraphPad Prism\ and MS Excel\. 3. Results

Fig. 1. Total average activity of animals from different treatment groups at the age of 14 months (N = 10 per group) under LD (12 h light: 12 h dark) or DD (constant dark) conditions. Night LD refers to nocturnal activity under LD, whereas day LD refers to daytime activity. Note the higher total activity of melatonin mice under DD (P < 0.05) and the elevated daytime activity of CR mice under LD (P < 0.05). The data represented here are derived from the electronic files displayed in actogram form in Fig. 2.

The untreated CT group had an average body weight of 29.1 T 0.4 g, the melatonin-substituted animals had average body weight of 29.4 T 0.4 g and the CR animals had an average body weight of 28.6 T 0.3 g at the beginning of the experiment. At the end of the experiment (40th week), the control animals had an average body weight of 41.0 T 0.9 g and the melatonin-substituted animals had an average body weight of 35.0 T 0.9 g. The weight difference at the end of the experiment between the CT and melatonin groups was statistically significant after a two sided t test (P < 0.001). The body weight of the CR animals decreased to 20.5 T 0.3 g in the 10th week and increased thereafter to 24.1 T 0.4 g in the 40th week (P < 0.001). The activity in all groups of mice in the dark phase was higher than during the light phase (Fig. 1, P < 0.05). However, the nocturnal activity of the CR animals was a little lower ( 11%, P < 0.05), while the daytime activity was higher (+43%, P < 0.05) when compared to CT or

Fig. 2. Representative actograms of mouse locomotor activity at the ages of 7 (A – C) and 11 months (D – F). The light cycle was 06:00 – 18:00 h followed by DD (*) and subsequent reentrainment (triangle) (A – C). (A) Seven-month-old control animal. (B) Seven-month-old CR animal. (C) Seven-month-old melatonin-treated animal displaying biphasic activity. In panels (D – F), the light period was 10:00 – 22:00 h followed by a 4 h phase delay to 14:00 – 02:00 h (arrow Y). The DD in all actograms began at the * and lasted until the reentrainment to the pre-DD conditions (triangle). (D) Control animal aged 11 months. (E) CR animal aged 11 months. (F) Melatonin animal aged 11 months.

D. Resuehr, J. Olcese / Brain Research 1048 (2005) 146 – 152

melatonin animals. The total activity increased (+26.3%, P < 0.05) during DD conditions in the melatonin-treated animals (Fig. 1). There was no significant inter-group difference in total activity between the examined groups of the same age (Fig. 1, LD 24 h). Older control mice (21 months) displayed significantly less activity under all lighting conditions (data not shown). CT animals displayed a robust circadian rhythmicity under LD conditions and a free-running period <24 h in DD (Figs. 2A and D). During LD conditions, but even more so during DD conditions, there was a tendency for the main bout of activity to split slightly into two separate bouts. The major trait of the CR group was a strongly fragmented activity rhythm with significant activity during the dark/ light transition (Figs. 2B and E upper portion of picture, above * and Y). As was the case with control animals, CR mice individuals often displayed two activity peaks, which

149

drifted apart under DD conditions (Fig. 2E). The main characteristic of the melatonin-substituted animals was a very robust entrainment to the LD cycle (Fig. 2F) and a rapid adjustment to changes in LD phase. About 50% of the animals treated with melatonin displayed a very strong biphasic behavior (Fig. 2C) after onset of DD. A splitting of the activity bouts with this severity was only observed in the melatonin-treated group. The effects of a 4 h phase delay (depicted in Figs. 2D, E, F at the Y) were evaluated statistically to determine if CR or melatonin treatment modulated the time necessary to resynchronize to the novel lighting conditions (Fig. 3). Untreated animals (Fig. 3A: CT) reached the pre phase shift level of nocturnal activity (the criterion for reentrainment) by the third night after the phase shift. CR animals (Fig. 3B) achieved resynchronization by the second night after the phase delay. The locomotor activity of CR continued to rise from the second to fourth night before dropping to normal levels (data not shown). Melatonin animals (Fig. 3C) also resynchronized within two nights, comparable to the CR animals. The endogenous free-running period (s) under DD was shorter than 24 h in all mice of the strain B6C3F1. In all treatment groups, we observed a significant increase of s with age (Fig. 4). Neither melatonin nor CR reversed or attenuated the lengthening of s, although the circadian clock tended to run a little slower in CR mice compared to CT and melatonin (two sided t test, N = 10, P < 0.05).

4. Discussion

Fig. 3. Reentrainment after a 4 h phase delay in CT (A), CR (B) and melatonin (C) mice. The letter B on the abscissa of each graph indicates the average nocturnal activity in each group before the phase shift. The letters A1 to A5 indicate the average nocturnal activity on the first five consecutive days after applying the phase shift. Untreated CT animals reached their average amount of nocturnal activity by the third night after the phase shift (*P < 0.05). CR animals achieved resynchronization by the second night after the phase delay (*P < 0.05), melatonin animals were also resynchronized by the second night (*P < 0.05).

The starting weight of the animals in all groups was similar (ca. 29 g). During the experimental period, however, the CR animals significantly lost weight until the 10th week, at which time body weight stabilized or even increased slightly by the end of the experiment. This weight loss was anticipated and is comparable to that which has been reported previously [41,42]. It should perhaps be noted that the CR regime established in the present study was somewhat less severe than that employed by others [7,18]. Interestingly, there was a significant decrease in body weight between the melatonin-treated and CT animals. In female SHR mice [1], rats [31] and hamsters [4], a reduction in body mass due to melatonin has been reported, but not in male mice of our strain. One logical explanation for this melatonin effect might be the increase of locomotor activity (Figs. 1, 2), leading to increased calorie consumption. However, as increased activity was only observed under DD conditions, this is unlikely to be the cause. In earlier studies in which young female BALB/c (melatonin deficient) mice were substituted with melatonin at a concentration of 10 Ag/ ml drinking water, no significant difference in body weight between control- and melatonin-substituted animals was seen [27,28]. However, a life-extending effect of melatonin was noted [28]. The amount of food consumed by both

150

D. Resuehr, J. Olcese / Brain Research 1048 (2005) 146 – 152

Fig. 4. The endogenous free-running period (s) of mice in DD. In all groups, s showed a significant increase with age. Neither melatonin nor CR treatment reversed or attenuated this change. By the age of 14 months, the period is significantly longer in CR and melatonin animals as compared to controls (two sided t test, N = 10, P < 0.05). At 8 and 14 months of age, the CR animals display the longest s as compared to both CT and melatonin animals.

melatonin-treated and CT animals did not appear to differ; therefore, food intake does not appear to be a likely explanation for the weight difference. Recent studies reveal that the B6C3F1 strain is rather long lived with a median life span of 30.7 months, which can be increased up to 35.4 months by a 40% caloric restriction [7], such as used here. In the present study, we only assessed animals up to an age of 14 months. Nevertheless, it has been shown that even short term CR or CR beginning in the middle of the murine life span leads to a significant change in aging-related physiological parameters, toward a more juvenile character [5,18]. Furthermore, oral administration of melatonin beginning later in life leads to prominent changes in mouse physiology, even at levels less than half the concentration of those used in our study [23]. Our results indicate a significant effect of melatonin substitution, and even more dramatically, CR on the locomotor activity of B6C3F1 mice. The endogenous freerunning period s (under DD conditions) and the pattern of activity under LD conditions clearly lengthened over the course of 14 months in control animals. However, neither CR nor melatonin treatments prevented this trend (Fig. 4). Valentinuzzi et al. [38] observed a lengthening of s in aging mice (C57BL/6, 6 –22 months), an effect comparable to our results. However, the reported effects of aging on the endogenous free-running circadian period are not always consistent, even in the same strain of mice. For example, Possidente and coworkers [30] reported a lengthening of s with age in C57BL/6 mice, while Teena and Wax [40] observed no changes in this strain. This heterogeneity of

results may be explained by the use of different illumination protocols before or after the measurement of s or may be due to varying starting ages of the test animals. The agerelated increase in circadian period of our B6C3F1 mice is comparable to previously described changes in s as a function of age in rats [43]. Thus, a change in s seems to be a common trait of the aging circadian system, consistent with reports that the endogenous clock in the suprachiasmatic nuclei (SCN) itself undergoes changes during senescence [42]. Under LD conditions CR mice exhibited an increased daytime and a decreased nighttime locomotor activity compared to the melatonin and CT mice (Figs. 1 and 2B). This appears to reflect a fragmented circadian rhythmicity and the animals’ tendency to search for food even during the daytime hours [9]. Restrictive feeding can be expected to result in a certain amount of mild stress and an increased excitability as is seen in semi-starved rats and anorexia nervosa patients [14]. In fact, we observed that even small disturbances in the animal housing area were sufficient to cause the entire CR group to become very active. Reduced caloric availability is known to increase spontaneous daytime wheel-running activity in rodents, which has been suspected to be foraging activity [24]. It is likely that hypoleptinemia could be the cause for this phenomenon because the exogenous application of leptin abolishes the anorexia-induced increase of activity in semi-starved rats [10,14]. In order to minimize this masking effect of hunger, feeding times in our experiments were randomly varied and limited to the dark phase. The decreased nighttime activity in CR mice is probably a compensatory increased nocturnal

D. Resuehr, J. Olcese / Brain Research 1048 (2005) 146 – 152

rest phase due to excessive daytime activity. The total locomotor activity of all three treatment groups under LD conditions is however not significantly different. In contrast, under DD conditions, many of the melatonin-treated animals displayed an increase in total activity (Figs. 1, 2C), in many cases due to rhythm splitting (see discussion below). The time required to reentrain to a shifted LD after DD has been reported to increase with age in mice [38]. In the present study, we observed that the melatonin-treated and CR animals adjusted to a 4 h phase delay in the lighting conditions faster than control mice (Figs. 2A – C and 3A – C). However, the CR animals were unusual in that 3 –5 days after the phase delay they showed a remarkable excess of activity (Fig. 3B). This overcompensation is likely to be artificial and may be related to the increased excitability of the CR mice subsequent to refeeding after a 3-day (weekend) interval. Melatonin’s effects on the circadian system are well known. In several species, melatonin induces phase shifts and entrainment of the circadian clock [32]. Melatonin treatment can also entrain the circadian system of humans [36]. In the present study, melatonin facilitated adjustment to a 4 h phase delay, such that mice were resynchronized within 48 h as opposed to 72 h for the control animals. Similar positive effects of melatonin on synchronization to new lighting conditions have been reported previously in hamsters [39]. Quite unexpectedly, this mouse strain frequently developed a ‘‘split’’ free-running rhythm under DD. Splitting is commonly seen in nocturnal rodents when they are maintained under LL conditions [21]. It is defined as the spontaneous separation of the main activity bout into two smaller bouts with different free-running periods. Pronounced splitting of the free-running locomotor activity rhythm was seldom seen in either CT or CR animals (Figs. 2A, B, D, E); however, it was common in the melatonintreated animals, especially under DD conditions. Only unsplit animals were used for the determination and graphical illustration of s (Fig. 4). As depicted in Fig. 2C, the rhythm splitting in melatonin-treated animals was often substantial, resulting in two free-running rhythms with different periods (23 vs. 23.6 h, calculated by v 2 periodogram analysis). Only 20% of the CT animals displayed biphasic free-running activity under DD, while the incidence of splitting in the melatonin animals was 50%. In view of its known effects on the SCN clock [19, 20], it seems conceivable that pharmacological levels of melatonin, self-administered in a regular nocturnal manner to B6C3F1 mice, may paradoxically facilitate the uncoupling of dual SCN oscillators under Zeitgeber-free conditions (i.e., DD), which would manifest itself as rhythm splitting. Further studies at the molecular level are needed to better understand this unusual behavior. It would be informative to determine whether the SCN expression of the MT1 and/or MT2 melatonin receptor is vital for the splitting phenom-

151

enon that we observed in the B6C3F1 mice. This could, for example, be assessed by backcrossing this strain with melatonin receptor knockout mice [35]. In conclusion, the present study has compared two anti-aging treatments, which are likely to influence the endogenous circadian system. Melatonin’s use as a chronobiotic has been well explored in rodents and humans [2,3]. Timed restricted feeding under the conditions of caloric restriction [22] is a potent synchronizer of peripheral and central oscillators. In the present report, we deliberately alternated the feeding times and fed mice in the dark phase to avoid offering a light coupled ‘‘Zeitgeber’’ i.e., feeding in the light phase which could lead to animals associating light with food. Nevertheless, on a behavioral level, we could detect a direct effect of caloric restriction on the circadian clock (in terms of locomotor activity), rendering it more sensitive to changed light cycles. A comparable effect was seen in animals administered melatonin. The mechanisms through which melatonin and caloric restriction exert their action on the circadian clock however are probably completely different, that is, melatonin is likely to work through Gprotein-coupled receptors [35], while caloric restriction is thought to operate via other pathways, such as the orexigenic system and leptin [10,14].

Acknowledgments We would like to thank Dr. Haiko Dernbach for assistance in setting up the locomotor activity electronics and for his generosity in donating the software for the activity recordings. We are also grateful to Jenny Behrens and Helena Fischer for assistance in maintaining the animals. This project was financed by the LeidenbergerMu¨ller Foundation and the Graduiertenkolleg 336 (DFG).

References [1] V.N. Anisimov, I.N. Alimova, D.A. Baturin, I.G. Popovich, M.A. Zabezhinski, S.V. Rosenfeld, K.G. Manton, A.V. Semenchenko, A.I. Yashin, Dose-dependent effect of melatonin on life span and spontaneous tumor incidence in female SHR mice, Exp. Gerontol. 38 (2003) 449 – 461. [2] J. Arendt, D.J. Skene, Melatonin as a chronobiotic, Sleep Med. Rev. 9 (2005) 25 – 39. [3] J. Arendt, S. Deacon, J. English, S. Hampton, L. Morgan, Melatonin and adjustment to phase shift, J. Sleep Res. 4 (1995) 74 – 79. [4] S.D. Bilbo, R.J. Nelson, Melatonin regulates energy balance and attenuates fever in Siberian hamsters, Endocrinology 143 (2002) 2527 – 2533. [5] S.X. Cao, J.M. Dhahbi, P.L. Mote, S.R. Spindler, Genomic profiling of short- and long-term caloric restriction effects in the liver of aging mice, Proc. Natl. Acad. Sci. U. S. A. 98 (2001) 10630 – 10635. [6] J.M. Dhahbi, P.L. Mote, J. Wingo, J.B. Tillman, R.L. Walford, S.R. Spindler, Calories and aging alter gene expression for gluconeogenic, glycolytic, and nitrogen-metabolizing enzymes, Am. J. Physiol. 277 (1999) E352 – E360.

152

D. Resuehr, J. Olcese / Brain Research 1048 (2005) 146 – 152

[7] J.M. Dhahbi, H.J. Kim, P.L. Mote, R.J. Beaver, S.R. Spindler, Temporal linkage between the phenotypic and genomic responses to caloric restriction, Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 5524 – 5529. [8] P.H. Duffy, R. Feuers, K.D. Nakamura, J. Leakey, R.W. Hart, Effect of chronic caloric restriction on the synchronization of various physiological measures in old female Fischer 344 rats, Chronobiol. Int. 7 (1990) 113 – 124. [9] P.H. Duffy, R.J. Feuers, R.W. Hart, Effect of chronic caloric restriction on the circadian regulation of physiological and behavioral variables in old male B6C3F1 mice, Chronobiol. Int. 7 (1990) 291 – 303. [10] C. Exner, J. Hebebrand, H. Remschmidt, C. Wewetzer, A. Ziegler, S. Herpertz, U. Schweiger, W.F. Blum, G. Preibisch, G. Heldmaier, M. Klingenspor, Leptin suppresses semi-starvation induced hyperactivity in rats: implications for anorexia nervosa, Mol. Psychiatry 5 (2000) 476 – 481. [11] E. Filipski, V.M. King, X. Li, T.G. Granda, M.C. Mormont, B. Claustrat, M.H. Hastings, F. Levi, Disruption of circadian coordination accelerates malignant growth in mice, Pathol. Biol. (Paris) 51 (2003) 216 – 219. [12] T. Gerlach, J.E. Aurich, Regulation of seasonal reproductive activity in the stallion, ram and hamster, Anim. Reprod. Sci. 58 (2000) 197 – 213. [13] M. Goto, I. Oshima, T. Tomita, S. Ebihara, Melatonin content of the pineal gland in different mouse strains, J. Pineal Res. 7 (1989) 195 – 204. [14] J. Hebebrand, C. Exner, K. Hebebrand, C. Holtkamp, R.C. Casper, H. Remschmidt, B. Herpertz-Dahlmann, M. Klingenspor, Hyperactivity in patients with anorexia nervosa and in semistarved rats: evidence for a pivotal role of hypoleptinemia, Physiol. Behav. 79 (2003) 25 – 37. [15] M.W. Hurd, M.R. Ralph, The significance of circadian organization for longevity in the golden hamster, J. Biol. Rhythms 13 (1998) 430 – 436. [16] A. Klarsfeld, F. Rouyer, Effects of circadian mutations and LD periodicity on the life span of Drosophila melanogaster, J. Biol. Rhythms 13 (1998) 471 – 478. [17] G.C. Koster-van Hoffen, M. Mirmiran, N.P. Bos, W. Witting, P. Delagrange, B. Guardiola-Lemaitre, Effects of a novel melatonin analog on circadian rhythms of body temperature and activity in young, middle-aged, and old rats, Neurobiol. Aging 14 (1993) 565 – 569. [18] C.K. Lee, D.B. Allison, J. Brand, R. Weindruch, T.A. Prolla, Transcriptional profiles associated with aging and middle age-onset caloric restriction in mouse hearts, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 14988 – 14993. [19] C. Liu, D.R. Weaver, X. Jin, L.P. Shearman, R.L. Pieschl, V.K. Gribkoff, S.M. Reppert, Molecular dissection of two distinct actions of melatonin on the suprachiasmatic circadian clock, Neuron 19 (1997) 91 – 102. [20] A.J. McArthur, A.E. Hunt, M.U. Gillette, Melatonin action and signal transduction in the rat suprachiasmatic circadian clock: activation of protein kinase C at dusk and dawn, Endocrinology 138 (1997) 627 – 634. [21] J.D. McAuley, J.P. Miller, E. Beck, Z.M. Nagy, K.C. Pang, Agerelated disruptions in circadian timing: evidence for ‘‘split’’ activity rhythms in the SAMP8, Neurobiol. Aging 23 (2002) 625 – 632. [22] J. Mendoza, C. Graff, H. Dardente, P. Pevet, E. Challet, Feeding cues alter clock gene oscillations and photic responses in the suprachiasmatic nuclei of mice exposed to a light/dark cycle, J. Neurosci. 25 (2005) 1514 – 1522. [23] Y. Okatani, A. Wakatsuki, R.J. Reiter, Y. Miyahara, Hepatic mitochondrial dysfunction in senescence-accelerated mice: correction

[24] [25]

[26]

[27]

[28]

[29]

[30]

[31]

[32] [33] [34] [35]

[36]

[37]

[38]

[39]

[40]

[41] [42] [43]

by long-term, orally administered physiological levels of melatonin, J. Pineal Res. 33 (2002) 127 – 133. J.M. Overton, T.D. Williams, Behavioral and physiologic responses to caloric restriction in mice, Physiol. Behav. 81 (2004) 749 – 754. S.F. Pang, F. Tang, P.L. Tang, Negative correlation of age and the levels of pineal melatonin, pineal N-acetylserotonin, and serum melatonin in male rats, J. Exp. Zool. 229 (1984) 41 – 47. S. Perreau-Lenz, A. Kalsbeek, J. Van Der Vliet, P. Pevet, R.M. Buijs, In vivo evidence for a controlled offset of melatonin synthesis at dawn by the suprachiasmatic nucleus in the rat, Neuroscience 130 (2005) 797 – 803. W. Pierpaoli, W. Regelson, Pineal control of aging: effect of melatonin and pineal grafting on aging mice, Proc. Natl. Acad. Sci. U. S. A. 91 (1994) 787 – 791. W. Pierpaoli, A. Dall’Ara, E. Pedrinis, W. Regelson, The pineal control of aging. The effects of melatonin and pineal grafting on the survival of older mice, Ann. N. Y. Acad. Sci. 621 (1991) 291 – 313. C.S. Pittendrigh, S. Daan, Circadian oscillations in rodents: a systematic increase of their frequency with age, Science 186 (1974) 548 – 550. B. Possidente, S. McEldowney, A. Pabon, Aging lengthens circadian period for wheel-running activity in C57BL mice, Physiol. Behav. 57 (1995) 575 – 579. B. Prunet-Marcassus, M. Desbazeille, A. Bros, K. Louche, P. Delagrange, P. Renard, L. Casteilla, L. Penicaud, Melatonin reduces body weight gain in Sprague Dawley rats with diet-induced obesity, Endocrinology 144 (2003) 5347 – 5352. J.R. Redman, Circadian entrainment and phase shifting in mammals with melatonin, J. Biol. Rhythms 12 (1997) 581 – 587. R.J. Reiter, Oxidative damage in the central nervous system: protection by melatonin, Prog. Neurobiol. 56 (1998) 359 – 384. S.M. Reppert, D.R. Weaver, Coordination of circadian timing in mammals, Nature 418 (2002) 935 – 941. S.M. Reppert, D.R. Weaver, T. Ebisawa, Cloning and characterization of a mammalian melatonin receptor that mediates reproductive and circadian responses, Neuron 13 (1994) 1177 – 1185. R.L. Sack, R.W. Brandes, A.R. Kendall, A.J. Lewy, Entrainment of free-running circadian rhythms by melatonin in blind people, N. Engl. J. Med. 343 (2000) 1070 – 1077. S.J. Schoeman, M.A. Botha, The effect of melatonin treatment on reproductive efficiency in an accelerated lambing system, J. S. Afr. Vet. Assoc. 66 (1995) 230 – 234. V.S. Valentinuzzi, K. Scarbrough, J.S. Takahashi, F.W. Turek, Effects of aging on the circadian rhythm of wheel-running activity in C57BL/6 mice, Am. J. Physiol. 273 (1997) R1957 – R1964. O. Van Reeth, L. Weibel, E. Olivares, S. Maccari, E. Mocaer, F.W. Turek, Melatonin or a melatonin agonist corrects age-related changes in circadian response to environmental stimulus, Am. J. Physiol., Regul. Integr. Comp. Physiol. 280 (2001) R1582 – R1591. T.M. Wax, Runwheel activity patterns of mature – young and senescent mice: the effect of constant lighting conditions, J. Gerontol. 30 (1975) 22 – 27. R. Weindruch, The retardation of aging by caloric restriction: studies in rodents and primates, Toxicol. Pathol. 24 (1996) 742 – 745. R. Weindruch, Caloric restriction and aging, Sci. Am. 274 (1996) 46 – 52. W. Witting, M. Mirmiran, N.P. Bos, D.F. Swaab, The effect of old age on the free-running period of circadian rhythms in rat, Chronobiol. Int. 11 (1994) 103 – 112.