- Email: [email protected]
Gender differences in the circadian rhythms of rhesus monkeys
Physiology & Behavior 101 (2010) 595–600
Contents lists available at ScienceDirect
Physiology & Behavior j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p h b
Gender differences in the circadian rhythms of rhesus monkeys Laura K. Barger, Tana M. Hoban-Higgins, Charles A. Fuller ⁎ Department of Neurobiology, Physiology & Behavior, University of California, Davis, Davis, CA 95616, USA
a r t i c l e
i n f o
Article history: Received 4 December 2009 Received in revised form 24 May 2010 Accepted 7 June 2010 Keywords: Body temperature Heart rate Activity Biotelemetry
a b s t r a c t Studies investigating gender differences in human circadian rhythms report equivocal results. In addition, many of these studies have been limited to examination of one circadian variable. This study examined gender differences in circadian rhythms of multiple physiological variables of rhesus monkeys under highly controlled conditions. Under general anesthesia, eight female and seven male rhesus were implanted with a biotelemetry transmitter to measure body temperature (Tb) and heart rate. An external accelerometer was used to measure physical activity. The Psychomotor Task System (PTS) provided environmental enrichment and delivered a pelletized diet and water was available ad libitum. Data were collected continuously under LD 16:8 for a minimum of 31 days. Mean, phase and amplitude of each rhythm were calculated and compared between genders. Although there were no significant differences between genders in mean or amplitude, circadian rhythms in females were significantly delayed compared to males in all variables (p range 0.001 to 0.030). The consistent pattern of delay suggests that a fundamental gender difference may be present in the circadian timing system. Mechanisms underlying this difference require further exploration. © 2010 Elsevier Inc. All rights reserved.
1. Introduction Circadian rhythms are endogenous oscillations in an organism's physiology and behavior with a period of about 24 h. These rhythms provide temporal organization for almost all behavioral, physiological and biochemical variables. The fundamental adaptive advantage of this temporal organization is that it allows for predictive, rather than entirely reactive, homeostatic regulation of function. For example, prior to waking, body temperature rises, heart rate and blood pressure increase, and plasma cortisol rises in anticipation of increased energetic demands. Disruption of circadian rhythms may lead to sleep–wake disturbances, decrements in cognitive performance and gastrointestinal problems [1–8]. Long term consequences include mental disorders, severely compromised physical and cognitive function and, most seriously, a possibly lowered life expectancy [2,5–9]. Circadian alterations include changes in period and phase relationships and a decrease in rhythm amplitude. One particularly notable example of a circadian rhythm-related affective disorder is winter depression or seasonal affective disorder (SAD) [10,11]. SAD is characterized by depression, lethargy, loss of libido, hypersomnia, weight gain, carbohydrate cravings, anxiety ⁎ Corresponding author. Department of Neurobiology, Physiology & Behavior, University of California, One Shields Avenue, Davis, CA 95616-8519, USA. Tel.: + 1 530 752 2979; fax: + 1 530 752 5851. E-mail address: [email protected] (C.A. Fuller). 0031-9384/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.physbeh.2010.06.002
and an inability to concentrate and focus [11–13]. SAD and other psychological illnesses, which have a greater prevalence in women, are predominantly associated with circadian delays [14,15]. In contrast, symptoms of endogenous depression include factors indicative of an advanced circadian phase, including early morning awakening and early onset of REM sleep [5]. Thus far, studies evaluating gender differences in human circadian rhythms have produced conflicting results. Some studies report females delayed relative to males [16], some report females are relatively advanced [17–19], and others find no difference between genders [20–22]. A limitation of these studies is that most have examined only one circadian variable. Additionally, many of these studies have not controlled the subject's LD cycle, one of the most important environmental signals for the circadian system [17,19,21]. As an alternative, the use of non-human primates in these types of studies provides several advantages. The rhesus monkey, being a diurnal Old World Primate with close phylogenetic ties to humans, provides a very useful model for biomedical research. Additional similarities to humans are that rhesus monkeys display consolidated, nocturnal sleep [23] and female rhesus have menstrual cycles. An added benefit of the rhesus monkey is that its larger size allows for the recording of multiple rhythms from every subject. The use of the non-human primate biomedical model alleviates the overwhelming operational demands inherent in long-duration human studies, allows for more tightly controlled environmental conditions and eliminates many social factors that could confound human investigations. This study examined gender differences in circadian rhythms of
596
L.K. Barger et al. / Physiology & Behavior 101 (2010) 595–600
multiple physiological variables of rhesus monkeys (Macaca mulatta) under strictly controlled conditions.
2. Methods 2.1. Subjects Eight female (5.5 ± 0.6 years; 6.5 ± 0.9 kg; mean ± SD) and seven male (6.05 ± 0.04 years; 8.5 ± 1.3 kg; mean ± SD) rhesus monkeys were trained to use the Psychomotor Test System (PTS), which provided nutrition and behavioral enrichment throughout the study. Data collection was segregated by gender; females were studied for 35 days and males for 31 days. Although our previous study did not show significant differences in the circadian rhythms of rhesus monkeys between follicular and luteal phases [24], we designed this study to encompass one complete menstrual cycle. All animal care met the NIH Guidelines for Animal Care and Use. All procedures were approved by the University of California Davis Institutional Animal Care and Use Committee.
2.2. Environmental conditions This study was performed at the California National Primate Research Center (CNPRC) in a room accommodating eight animals. Cages (24 × 26 × 32 inches) were hung on double-decker racks that kept the animals visually isolated from one another. Animals had auditory and olfactory contact. Each cage was illuminated by one 12inch, 8 Watt cool white fluorescent bulb that was mounted outside of the animal's cage producing approximately 400 lux in the front of the cage to approximately 90 lux in the back. Overhead lighting was off except for the time when husbandry tasks were performed (approximately one hour per day). The lighting schedule consisted of 16 h of light followed by 8 h of darkness (LD 16:8; L on 0600) to more closely mimic human conditions and was computer controlled. Room temperature was continuously monitored using a YSI thermister probe (Yellow Springs Instrument Company, Yellow Springs, OH) and recorded at 10 minute intervals. Mean room temperature during data collection was maintained at 22 ± 1 °C.
2.3. Psychomotor Test System Animals were fed and environmental enrichment was provided through the Psychomotor Test System (PTS) developed at the Language Research Center at Georgia State University. The PTS is a computer-based video task system that presents a selection of 5 tasks. The animal manipulates a joystick to control a cursor to select and complete a series of trials for each task. Successful completion of a video task was rewarded with a 190-mg pellet of a nutritionally complete diet (P.J. Noyes, Lancaster, NH). The PTS computer screen was illuminated during the 16 h of light. An example of a PTS task is Matching to Sample in which the subject must indicate which of two figures is identical to one presented previously. Animals were fully trained to use PTS to meet their daily nutritional requirements prior to serving as study subjects.
2.4. Husbandry Husbandry was performed on a non-24 hour basis, so as not to introduce circadian time cues. Due to the physiological disturbance associated with daily husbandry tasks, all data collected during the time investigators were in the room plus one-half hour following were not used for analysis.
2.5. Telemetry Each animal was implanted with a battery operated T4E biotelemetry unit (Konigsberg Instruments, Pasadena, CA). The implant consisted of a small temperature transmitter, two leads and one antenna. Each implant was cold sterilized by ethylene oxide (Mercy General Hospital, Sacramento, CA) and surgeries were performed at the CNPRC. The animals, fasted overnight prior to surgery, were premedicated with an intramuscular injection of ketamine (10 mg/kg). After endotracheal intubation, a surgical plane of anesthesia was initiated and maintained by inhalation using isoflurane. The transmitter was implanted within the muscular wall on the left side of the abdomen (except for one male implanted on the right side). ECG leads and antenna were passed subcutaneously and secured to the muscle in the left, right and center of the chest. Postoperative analgesics were administered, and antibiotics were given as required. Animals were allowed at least two weeks to recover from the surgery before data collection. Each implant had its own frequency-modulated (FM) signal and was decoded by a TI20-C Temperature/Biopotential Demodulator (Konigsberg Instruments, Pasadena, CA). Reception of this signal allowed continuous monitoring of body temperature (Tb) electrocardiogram (ECG) leads provided continuous ECG recording, which was converted to heart rate (HR) by an R wave detector. An in-cage antenna system maximized signal reception and all data, sampled at 1-second intervals, were saved to a computer hard disk. In-house software filtered and translated Tb and HR telemetry files into a form that could be analyzed for circadian rhythmicity. 2.6. Activity Each animal was fitted with a mesh jacket that had a dorsal pocket to hold a bi-axial activity monitor (Individual Monitoring Systems, Baltimore, MD). The activity monitor has been used previously in a study of non-human primates [25]. Activity counts, transferred from the monitor to a computer through an interface, were summed in two-minute bins. 2.7. Feeding Pellet delivery by each PTS was recorded by an in-house software program written at Georgia State University. It was assumed that all dispensed pellets were consumed by the animal. This was confirmed by observation during husbandry. The caloric intake of each subject was calculated daily to confirm that adequate nutrition was provided through PTS. To examine the circadian feeding rhythm, pellet counts were summed by hour with totals assigned to the start of the hour block. Water was available ad libitum. 2.8. Circadian and statistical analysis All results are presented as mean and standard deviation. Circadian analysis was completed using in-house software (EPL, Davis, CA). The phase-fitting software (least-squares harmonic regression analysis) uses a Fourier-based algorithm to calculate a best-fit sine wave to each 24 hour block of data [26]. The time at which the sine wave peaks is deemed the acrophase. Mean, acrophase, and amplitude were calculated for each variable. Since all variables did not present a perfect sine wave fit, the timing of mean crossing of the average daily waveform was used to confirm phase. Statview 5.0 (SAS Institute, Inc. Cary, NC) was used for statistical analysis. Analysis of variance (ANOVA) and post-hoc tests were used to analyze gender differences in circadian variables. An alpha of 0.05 was considered significant.
L.K. Barger et al. / Physiology & Behavior 101 (2010) 595–600
3. Results 3.1. Body temperature Tb has a very stable circadian rhythm, and is a powerful variable when studying rhythmicity [27]. Using non-human primates as subjects enabled us to utilize implantable telemetric devices and to record data continuously for over one month. Recording for such a long period revealed an accurate average rhythm. An example of the raw Tb telemetry output from a male and female rhesus is presented in Fig. 1. Average daily waveforms for a five day window are presented in Fig. 2. Retroperitoneal mean Tb was 36.50 °C [SD 0.53] in males. Female mean Tb [36.96 °C; SD 0.31] while higher than male mean Tb, was not statistically different (p = 0.056). To determine the timing of the rhythms, a sine wave was fitted to each day's data. The peak of the fitted curve was deemed the acrophase. Mean Tb acrophase was 13.74 h [SD 0.36] in males and 14.57 h [SD 0.43] in females. Acrophase of the body temperature rhythm was significantly delayed (p = 0.001) in the females compared to the males. Mean acrophases for all variables are shown in Fig. 3. Amplitude was the most variable circadian Tb characteristic between animals. Males had a mean Tb amplitude of 0.63 °C [SD 0.21]. Female Tb amplitude [0.75 °C; SD 0.11] was not statistically different from male Tb amplitude. (Mean and amplitude data for all physiological variables are presented in Table 1.)
597
Both genders showed a similar daily waveform in Tb. Female Tb during the light period was significantly higher (p = 0.031) than male Tb during the light period. There was no difference in the mean Tb between genders during the dark period. Fig. 2a illustrates that females had a significantly sharper slope in the decline of Tb in the evening (p = 0.012). 3.2. Heart rate Circadian rhythms of heart rate were more variable than those of Tb; some animals showed a robust circadian HR rhythm, whereas others had a relatively flat rhythm. HR raw data was noisier (Fig. 1) than Tb. This was expected as HR can be rapidly influenced by external factors. Representative raw HR data are presented in Fig. 1 and average 24-hour waveforms in Fig. 2. Mean HR was similar between genders (males = 114.3 beats/min [SD 18.8]; females = 120.4 beats/ min [SD 9.0]). Mean HR acrophase for males was 13.14 h [SD 0.59], 0.6 h earlier than male Tb acrophase. Mean female HR acrophase was 14.61 h [SD 1.24], which was approximately the same as Tb acrophase. Female HR acrophase was significantly delayed (p = 0.013) compared to male acrophase. HR amplitude differed widely between subjects, ranging from as low as b10 beats/min to as high as 28 beats/min. HR amplitude did not differ significantly between genders (males = 14.6 beats/min [SD 7.5]; females = 12.1 beats/min [SD 3.8]).
Fig. 1. One week of raw physiological data from a representative female (left) and male (right) rhesus monkey. Presented are, from top to bottom, body temperature (Tb °C), heart rate (beats/min), activity (counts/10-min), and feeding (pellets/h). With the exception of feeding, all data are presented at 10-minute intervals or in 10 minute bins. A three-point moving average was applied to filter the Tb and HR data.
598
L.K. Barger et al. / Physiology & Behavior 101 (2010) 595–600
Fig. 2. Average daily waveforms (± SD) of body temperature (Tb °C), heart rate (bpm), activity (counts/10-min), and feeding (pellets/h) for a group of female (black line) and a group of male (gray line) rhesus monkeys. N/group was 7–8, with the exception of male activity (see Results for details).
The daily average rhythm in heart rate was very similar in males and females (Fig. 2). The increase in HR at lights on demonstrated the masking effects light can have on the rhythm of heart rate. There was no significant difference between genders in HR during either the light period or the dark period. 3.3. Activity One male removed his activity monitor and another monitor malfunctioned leaving activity (ACT) data available for 5 males. The
overall mean activity level for males was 20.6 [SD 7.4] counts/10-min. The average ACT acrophase in males [12.46 h; SD 0.88] occurred 0.68 h earlier than the HR acrophase and 1.28 h earlier than that of Tb. Mean male ACT amplitude was 18.1 counts/10-min [SD 10.5]. One female removed her activity monitor; therefore activity data were available for seven females. Female mean ACT was 25.0 counts/ 10-min [SD 7.2], which was not significantly different from male mean ACT. Timing of the female ACT rhythm, with an average acrophase of 13.90 h [SD 1.03], was advanced approximately 0.7 h ahead of both the Tb and HR rhythms. The timing of the ACT rhythm was significantly delayed in females (p = 0.030) compared to males. Female mean amplitude, at 18.5 counts/10-min [SD 6.9], was similar to that of the males' average. Representative raw ACT data are presented in Fig. 1. Daily waveforms of activity are shown in Fig. 2.
Table 1 Mean, amplitude and phase for multiple circadian rhythms in male and female rhesus monkeys. Units for mean and amplitude are given in the first column. There were no significant differences between genders in mean or amplitude. Phase is in hours. Female circadian rhythms were significantly delayed compared to males in all variables. p Values are given in the right column. Variable
Gender
Mean
Amplitude
Phase (h)
p (Phase)
Tb (°C)
Male Female Male Female Male Female Male Female
36.50 36.96 114.3 120.4 20.6 25.0
0.63 0.75 14.6 12.1 18.1 18.5
13.74 14.57 13.14 14.61 12.46 13.90 12.51 13.84
0.001
HR (bpm) Fig. 3. Mean acrophase (± SD) of the rhythms of body temperature, heart rate, activity, and feeding for female and male rhesus monkeys. Rhythms are significantly delayed (see Table 1 for p values) in females (black squares) compared to males (gray diamonds) in all variables.
ACT (counts/2 min) FD (pellets/h)
0.013 0.030 0.001
L.K. Barger et al. / Physiology & Behavior 101 (2010) 595–600
3.4. Feeding Female daily mean food consumption of 1039.21 pellets/day [SD 40.86] was significantly (p = 0.006) less than the male daily mean of 1309.24 pellets/day [SD 82.82]. However, this difference was negated when food consumption is considered on a per mass basis; females consumed 161.03 pellets/kg [SD 8.19] body weight and males 155.08 pellets/kg [SD 4.65] body weight. PTS was not available during the hours of dark and consequently no pellets were consumed between 2200 and 0600. This necessary restriction to control lighting conditions (i.e., no computer monitor illumination during the 8 h of darkness) affected the curve to which the sine wave was fitted and admittedly may have been different if animals fed throughout the night. However, although the feeding rhythm was artificially influenced by PTS availability being limited to the light hours, because both the male and female studies were conducted under this same condition, the timing of the feeding rhythms was comparable. In this variable as well, female acrophase [13.84 h; SD 0.76] was significantly (p = 0.001) delayed compared to male acrophase [12.51 h; SD 0.53]. 4. Discussion To our knowledge, this was the first comprehensive study examining gender differences in circadian rhythms using the rhesus monkey biomedical model. We found that the timing of multiple physiological rhythms was delayed in females compared to males. The literature on gender differences in circadian rhythms of human subjects is inconsistent. Conflicting results could be due to lack of environmental or social controls, studies of too short duration to obtain good average data, or other confounding factors such as the effects of aging. In humans, the influence of the menstrual cycle on circadian rhythms may also account for some of the conflicting literature, although no circadian differences were found between phases of the menstrual cycle in a recent rhesus study [24]. Additionally, the length of data collection in this study allowed each female rhesus to undergo a complete menstrual cycle negating any influence of the follicular or luteal phases. In this study, female mean Tb was higher than male Tb, although this difference only reached significance in the light period. This finding is consistent with Moe et al. who found higher peak temperatures in elderly females [18] and Winget et al. who reported that female rectal temperature exceed that of males at all time points [20]. Campbell also reported higher mean temperatures for female subjects [28]. The issue of gender differences in the timing of the Tb rhythm in humans is most inconsistent. There are reports of females having an advanced phase [17–19,29], a delayed phase [16], and a phase no different than males [20]. This study showed a significant phase delay in the female Tb rhythm relative to male subjects. The sharp decline in evening body temperature of female rhesus in this study was similar to that reported in human females [30]. Whereas this study revealed no gender differences in Tb amplitude, a larger amplitude in Tb has been reported for human males [17,20]. In contrast, Campbell reported a larger amplitude in Tb for females in subjects with a mean age of 49.1 years [28]. Other studies have reported changes in amplitude of this rhythm as a result of aging [31,32]. Gender may be a covariate in the magnitude of age-related amplitude changes, with males showing a greater amplitude decrease with age [29,30]. Circadian differences in heart rate may also be age related. For example, while mean HR is “generally higher” in females [20], Stein and colleagues reported no gender differences in HR of subjects aged 64–76 years, but in subjects aged 26–42 years, men had a lower mean HR [33]. Another study of human subjects, mean age of 26 years, showed no gender differences in mean HR [34]. Finally, Ryan and
599
colleagues found no gender or age differences in HR [35]. The monkeys in this study showed no differences in mean HR. As was the case for Tb, HR acrophase was significantly delayed in female monkeys compared to males. Young adult human females (ages 25–35) also showed a later acrophase than comparably aged males, however, this gender difference did not reach significance [20]. The highly variable HR amplitude did not differ between male and female monkeys in this study. Similarly, Wang and colleagues reported no gender difference in HR amplitude in young adults [34]. The high degree of variability within and between monkeys in HR was also noted in a study using pigtailed macaques (Macaca nemestrina; [36]). There was no gender difference in the amount of “spontaneous activity” in humans measured by microwave radars and reported as percent of time subject was moving [37] or in activity measured by energy expenditure [38]. No gender differences in the amount of physical activity were observed in a previous study using rhesus monkeys [39]. Similar to this study, Tapp and Natelson reported activity acrophase occurring prior to the acrophase of Tb in rhesus monkeys [40]. The advanced phase of ACT in relation to the HR and TB suggests that these rhythms do not merely parallel each other (e.g., HR does not simply reflect activity). Rhythms normally maintain a stable phase relationship within an individual; disturbance of this normal relationship, such as is seen in jet-lag, leads to physiological dysfunction [27]. In contrast to the delayed activity phase in female compared to male rhesus, human females performed habitual evening events significantly earlier than males as reported on daily logs [41]. This observation could potentially be due to social pressures rather than endogenous circadian regulation, again, underscoring the usefulness of the rhesus model for performing experiments under controlled conditions [41]. The timing of all rhythms studied in this experiment was significantly delayed in females compared to males (Fig. 3). Delays ranged from as little as 0.83 h in Tb, to as much as 1.47 h in HR. This consistent pattern in all variables suggests a fundamental difference between genders in the circadian timing system (CTS). Serotonin and estrogen have both been implicated as having possible roles in this difference [42]. Lepage and Steiner propose that females have a more responsive serotonergic system that causes differences in the sleep/ wake and circadian cycles [43]. Moreover, estrogen affects the function of the serotonergic system [44]. Separately or together, estrogen or serotonin may mediate inputs to the CTS. Of course, there may be many other interacting mechanisms contributing to these observed gender differences. Answering this question will require further studies. 4.1. Conclusions Consistent with what has been reported in the human literature, there were no reliable differences between genders in circadian mean or amplitude. There was, however, a consistent phase delay in the female rhesus in all physiological variables measured. Determining the mechanisms underlying this difference requires further exploration. Future study of rhythms and their gender differences could clarify the role of rhythm disruption in diseases such as SAD, sleeping and eating disorders and other psychological illnesses that have a greater preponderance in females [14,15]. Acknowledgements The authors would like to thank Peter Takeuchi and the staff at the California National Primate Research Center for their invaluable assistance during this study. Dr. Laura Barger is now affiliated with the Division of Sleep Medicine at Brigham and Women's Hospital and Harvard Medical School, Boston, MA 02115. Grants This study was supported by a grant from the National Aeronautics and Space Administration (NNJ04HF44G). Dr. Barger was supported
600
L.K. Barger et al. / Physiology & Behavior 101 (2010) 595–600
by NASA's Graduate Student Research Program (NFT-51417) and a Zonta International Amelia Earhart Fellowship. References [1] Colquhoun P. Rhythms in performance. In: Aschoff J, editor. Handbook of behavioral neurobiology: biological rhythms. New York: Plenum Press; 1981. p. 333–49. [2] Bunney WE, Bunney BG. Molecular clock genes in man and lower animals: possible implications for circadian abnormalities in depression. Neuropsychopharmacology 2000;22(4):335–45. [3] Czeisler CA, Richardson GS, Coleman RM, Zimmerman JC, Moore-Ede MC, Dement WC, et al. Chronotherapy: resetting the circadian clocks of patients with delayed sleep phase insomnia. Sleep 1981;4:1–21. [4] Fredrickson PA, Richardson JW, Esther MS, Lin SC. Sleep disorders in psychiatric practice. Mayo Clin Proc 1990;65:861–8. [5] Halaris A. Chronobiology and psychiatric disorders. New York: Elsevier; 1987. [6] Lewy AJ, Sack RL, Singer CM. Assessment and treatments of chronobiologic disorders using plasma melatonin levels and bright light exposure: the clock-gate model and the phase response curve. Psychopharmacol Bull 1984;20:561–5. [7] Van Cauter E, Turek FW. Depression: a disorder of time-keeping? Perspect Biol Med 1986;29:511–9. [8] Winget CM, DeRoshia CW, Markley CL, Holley DC. A review of human physiological and performance changes associated with desynchronosis of biological rhythms. Aviat Space Environ Med 1984;55:1085–96. [9] Davidson AJ, Sellix MT, Daniel J, Yamazaki S, Menaker M, Block GD. Chronic jet-lag increases mortality in aged mice. Curr Biol 2006;16(21):R914–6. [10] Lewy AJ, Kern HA, Rosenthal NE, Wehr TA. Bright artificial light treatment of a manic-depressive patient with a seasonal mood cycle. Am J Psychiatry 1982;139 (21):1496–8. [11] Rosenwasser AM, Wirz-Justice A. Depression: clinical and experimental models. In: Redfern PM, Lemmer B, editors. Physiology and pharmacology of biological rhythms. Berlin: Springer; 1996. p. 457–86. [12] Lewy AJ, Sack RL, Miller LS, Hoban TM. Antidepressant and circadian phaseshifting effects of light. Science 1987;235:352–4. [13] Lewy AJ, Sack RL. Light therapy and psychiatry. Proc Soc Exp Biol Med 1986;183: 11–8. [14] Ferrari E, Fraschini F, Brambilla F. Hormonal circadian rhythms in eating disorders. Biol Psychiatry 1990;27(9):1007–20. [15] Gorman J. Gender differences in depression and response to psychotropic medication. Gend Med 2006;3(2):93–109. [16] Mellette H, Hutt B, Askovitz S, Norvarth S. Diurnal variation in body temperatures. J Appl Physiol 1951;3:665–75. [17] Baker F, Waner J, Vieira E, Taylor S, Driver H, Mitchell D. Sleep and 24 hour body temperatures: a comparison in young men, naturally cycling women and women taking hormonal contraceptives. J Physiol 2001;530(Part 3):565–74. [18] Moe K, Prinz M, Vitiello M, Marks A, Larsen L. Healthy elderly women and men have different entrained circadian temperature rhythms. J Am Geriatr Soc 1991;39 (4):383–7. [19] Lee K. Circadian temperature rhythms in relation to menstrual cycle phase. J Biol Rhythms 1988;3(3):255–63. [20] Winget C, DeRoshia C, Vernikos-Danellis J, Rosenblatt W, Hetherington N. Comparison of circadian rhythms in male and female humans. Waking Sleeping 1977;1:359–63. [21] Jean-Louise G, Mendlowicz MV, Von Gizycki H, Zizi F, Nunes J. Assessment of physical activity and sleep by actigraphy: examination of gender differences. J Womens Health Gender Based Med 1999;8(8):1113–7. [22] Kerkhof GA. Inter-individual differences in the human circadian system: a review. Biol Psychol 1985;20(2):83–112.
[23] Hsieh K-C, Robinson EL, Fuller CA. Sleep architecture in unrestrained rhesus monkeys (Macaca mulatta) synchronized to 24-hour light–dark cycles. Sleep 2008;31(9):1239–50. [24] Barger L, Hoban Higgins T, Fuller C. Assessment of circadian rhythms throughout the menstrual cycle of female rhesus monkeys. Am J Primatol 2007;69:1–10. [25] Golub MP, Takeuchi PT, Hoban-Higgins TM. Nutrition and delayed activity offset in adolescent rhesus monkeys. In: Carskadon M, editor. Adolescent sleep: biological, psychological and social influences. England: Cambridge University Press; 2002. p. 50–68. [26] Halberg F, Tong YL, Johnson EA. Circadian system phase: an aspect of temporal morphology; procedures and illustrative examples. In: von Mayersbach H, editor. The cellular aspects of biorhythms. Berlin: Springer-Verlag; 1967. p. 20–48. [27] Moore-Ede M, Sulzman F, Fuller C. The clocks that time us. Cambridge, MA: Harvard University Press; 1982. [28] Campbell S. Effects of timed bright light exposure on shift-work adaptation in middle-aged subjects. Sleep 1995;18(6):408–16. [29] Prinz P, Moe K, Vitiello M, Larsen L, Lee K. Cosinor temperature measures in healthy entrained men and women: relationships with age, gender and sleep timing and quality. Sleep Res 1993;22:630. [30] Campbell S, Gillin J, Kripke D, Erikson P, Clopton P. Gender differences in the circadian temperature rhythms of healthy elderly subjects: relationships to sleep quality. Sleep 1989;12(6):529–36. [31] Vitiello M, Smallwood R, Avery D, Pascualy R, Martin D, Prinz P. Circadian temperature rhythms in young adult and aged men. Neurobiol Aging 1986;7(2): 97–100. [32] Weitzman E, Moline M, Czeisler C, Zimmerman J. Chronobiology of aging: temperature, sleep–wake rhythms and entrainment. Neurobiol Aging 1982;3(4): 299–309. [33] Stein P, Kleiger R, Rottman J. Differing effects of age on heart rate variability in men and women. J Cardiol 1997;80(3):302–5. [34] Wang Z, Wang L, Zhang L, Liu Q, Xue Z, Cornelissen G, et al. Circadian relations among cardiovascular variables of young adults. Chronobiologia 1992;19(3–4): 111–20. [35] Ryan S, Goldberg A, Pincus S, Mietus J, Lipsitz L. Gender- and age-related differences in heart rate dynamics: are women more complex than men? J Am Coll Cardiol 1994;24(7):1700–7. [36] Reite M, Short R. Circadian rhythms in monkeys: variability and behavioral correlations. Physiol Behav 1981;27(4):663–71. [37] Ferraro R, Lillioja S, Fontvielle A, Rising R, Bogardus C, Ravussin E. Lower sedentary metabolic rate in women compared with men. J Clin Invest 1992;90(3):780–4. [38] Morio B, Beaufrere B, Montaurier C, Verdier E, Ritz P, Fellmann N, et al. Gender differences in energy expenditure during activities and in daily energy expenditure of elderly people. Am J Physiol 1997;273:E321–7. [39] Ramsey J, Laatsch J, Kemnitz J. Age and gender differences in body composition, energy expenditure, and glucoregulation of adult rhesus monkeys. J Med Primatol 2000;29:11–9. [40] Tapp W, Natelson B. Circadian rhythms and patterns of performance before and after simulated jet lag. Am J Physiol 1989;257:R796–803. [41] Monk T, Petrie S, Hayes A, Kupfer D. Regularity of daily life in relation to personality, age, gender, sleep quality and circadian rhythms. J Sleep Res 1994;3: 196–205. [42] Leibenluft E. Do gonadal steroids regulate circadian rhythms in humans? J Affect Disord 1993;29(2–3):175–81. [43] Lepage P, Steiner M. Gender and serotonergic dysregulation: implications for the late luteal phase dysphoric disorder. In: Cassano GB, editor. Serotonin-related psychiatric syndromes: clinical and therapeutic links. London: Soc. Med. Servic.; 1991. [44] Biegon A, Fischette T, Mitchell D. Serotonin receptor modulation by estrogen in discrete brain nuclei. Neuroendocrinology 1982;35:287–91.
Contents lists available at ScienceDirect
Physiology & Behavior j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p h b
Gender differences in the circadian rhythms of rhesus monkeys Laura K. Barger, Tana M. Hoban-Higgins, Charles A. Fuller ⁎ Department of Neurobiology, Physiology & Behavior, University of California, Davis, Davis, CA 95616, USA
a r t i c l e
i n f o
Article history: Received 4 December 2009 Received in revised form 24 May 2010 Accepted 7 June 2010 Keywords: Body temperature Heart rate Activity Biotelemetry
a b s t r a c t Studies investigating gender differences in human circadian rhythms report equivocal results. In addition, many of these studies have been limited to examination of one circadian variable. This study examined gender differences in circadian rhythms of multiple physiological variables of rhesus monkeys under highly controlled conditions. Under general anesthesia, eight female and seven male rhesus were implanted with a biotelemetry transmitter to measure body temperature (Tb) and heart rate. An external accelerometer was used to measure physical activity. The Psychomotor Task System (PTS) provided environmental enrichment and delivered a pelletized diet and water was available ad libitum. Data were collected continuously under LD 16:8 for a minimum of 31 days. Mean, phase and amplitude of each rhythm were calculated and compared between genders. Although there were no significant differences between genders in mean or amplitude, circadian rhythms in females were significantly delayed compared to males in all variables (p range 0.001 to 0.030). The consistent pattern of delay suggests that a fundamental gender difference may be present in the circadian timing system. Mechanisms underlying this difference require further exploration. © 2010 Elsevier Inc. All rights reserved.
1. Introduction Circadian rhythms are endogenous oscillations in an organism's physiology and behavior with a period of about 24 h. These rhythms provide temporal organization for almost all behavioral, physiological and biochemical variables. The fundamental adaptive advantage of this temporal organization is that it allows for predictive, rather than entirely reactive, homeostatic regulation of function. For example, prior to waking, body temperature rises, heart rate and blood pressure increase, and plasma cortisol rises in anticipation of increased energetic demands. Disruption of circadian rhythms may lead to sleep–wake disturbances, decrements in cognitive performance and gastrointestinal problems [1–8]. Long term consequences include mental disorders, severely compromised physical and cognitive function and, most seriously, a possibly lowered life expectancy [2,5–9]. Circadian alterations include changes in period and phase relationships and a decrease in rhythm amplitude. One particularly notable example of a circadian rhythm-related affective disorder is winter depression or seasonal affective disorder (SAD) [10,11]. SAD is characterized by depression, lethargy, loss of libido, hypersomnia, weight gain, carbohydrate cravings, anxiety ⁎ Corresponding author. Department of Neurobiology, Physiology & Behavior, University of California, One Shields Avenue, Davis, CA 95616-8519, USA. Tel.: + 1 530 752 2979; fax: + 1 530 752 5851. E-mail address: [email protected] (C.A. Fuller). 0031-9384/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.physbeh.2010.06.002
and an inability to concentrate and focus [11–13]. SAD and other psychological illnesses, which have a greater prevalence in women, are predominantly associated with circadian delays [14,15]. In contrast, symptoms of endogenous depression include factors indicative of an advanced circadian phase, including early morning awakening and early onset of REM sleep [5]. Thus far, studies evaluating gender differences in human circadian rhythms have produced conflicting results. Some studies report females delayed relative to males [16], some report females are relatively advanced [17–19], and others find no difference between genders [20–22]. A limitation of these studies is that most have examined only one circadian variable. Additionally, many of these studies have not controlled the subject's LD cycle, one of the most important environmental signals for the circadian system [17,19,21]. As an alternative, the use of non-human primates in these types of studies provides several advantages. The rhesus monkey, being a diurnal Old World Primate with close phylogenetic ties to humans, provides a very useful model for biomedical research. Additional similarities to humans are that rhesus monkeys display consolidated, nocturnal sleep [23] and female rhesus have menstrual cycles. An added benefit of the rhesus monkey is that its larger size allows for the recording of multiple rhythms from every subject. The use of the non-human primate biomedical model alleviates the overwhelming operational demands inherent in long-duration human studies, allows for more tightly controlled environmental conditions and eliminates many social factors that could confound human investigations. This study examined gender differences in circadian rhythms of
596
L.K. Barger et al. / Physiology & Behavior 101 (2010) 595–600
multiple physiological variables of rhesus monkeys (Macaca mulatta) under strictly controlled conditions.
2. Methods 2.1. Subjects Eight female (5.5 ± 0.6 years; 6.5 ± 0.9 kg; mean ± SD) and seven male (6.05 ± 0.04 years; 8.5 ± 1.3 kg; mean ± SD) rhesus monkeys were trained to use the Psychomotor Test System (PTS), which provided nutrition and behavioral enrichment throughout the study. Data collection was segregated by gender; females were studied for 35 days and males for 31 days. Although our previous study did not show significant differences in the circadian rhythms of rhesus monkeys between follicular and luteal phases [24], we designed this study to encompass one complete menstrual cycle. All animal care met the NIH Guidelines for Animal Care and Use. All procedures were approved by the University of California Davis Institutional Animal Care and Use Committee.
2.2. Environmental conditions This study was performed at the California National Primate Research Center (CNPRC) in a room accommodating eight animals. Cages (24 × 26 × 32 inches) were hung on double-decker racks that kept the animals visually isolated from one another. Animals had auditory and olfactory contact. Each cage was illuminated by one 12inch, 8 Watt cool white fluorescent bulb that was mounted outside of the animal's cage producing approximately 400 lux in the front of the cage to approximately 90 lux in the back. Overhead lighting was off except for the time when husbandry tasks were performed (approximately one hour per day). The lighting schedule consisted of 16 h of light followed by 8 h of darkness (LD 16:8; L on 0600) to more closely mimic human conditions and was computer controlled. Room temperature was continuously monitored using a YSI thermister probe (Yellow Springs Instrument Company, Yellow Springs, OH) and recorded at 10 minute intervals. Mean room temperature during data collection was maintained at 22 ± 1 °C.
2.3. Psychomotor Test System Animals were fed and environmental enrichment was provided through the Psychomotor Test System (PTS) developed at the Language Research Center at Georgia State University. The PTS is a computer-based video task system that presents a selection of 5 tasks. The animal manipulates a joystick to control a cursor to select and complete a series of trials for each task. Successful completion of a video task was rewarded with a 190-mg pellet of a nutritionally complete diet (P.J. Noyes, Lancaster, NH). The PTS computer screen was illuminated during the 16 h of light. An example of a PTS task is Matching to Sample in which the subject must indicate which of two figures is identical to one presented previously. Animals were fully trained to use PTS to meet their daily nutritional requirements prior to serving as study subjects.
2.4. Husbandry Husbandry was performed on a non-24 hour basis, so as not to introduce circadian time cues. Due to the physiological disturbance associated with daily husbandry tasks, all data collected during the time investigators were in the room plus one-half hour following were not used for analysis.
2.5. Telemetry Each animal was implanted with a battery operated T4E biotelemetry unit (Konigsberg Instruments, Pasadena, CA). The implant consisted of a small temperature transmitter, two leads and one antenna. Each implant was cold sterilized by ethylene oxide (Mercy General Hospital, Sacramento, CA) and surgeries were performed at the CNPRC. The animals, fasted overnight prior to surgery, were premedicated with an intramuscular injection of ketamine (10 mg/kg). After endotracheal intubation, a surgical plane of anesthesia was initiated and maintained by inhalation using isoflurane. The transmitter was implanted within the muscular wall on the left side of the abdomen (except for one male implanted on the right side). ECG leads and antenna were passed subcutaneously and secured to the muscle in the left, right and center of the chest. Postoperative analgesics were administered, and antibiotics were given as required. Animals were allowed at least two weeks to recover from the surgery before data collection. Each implant had its own frequency-modulated (FM) signal and was decoded by a TI20-C Temperature/Biopotential Demodulator (Konigsberg Instruments, Pasadena, CA). Reception of this signal allowed continuous monitoring of body temperature (Tb) electrocardiogram (ECG) leads provided continuous ECG recording, which was converted to heart rate (HR) by an R wave detector. An in-cage antenna system maximized signal reception and all data, sampled at 1-second intervals, were saved to a computer hard disk. In-house software filtered and translated Tb and HR telemetry files into a form that could be analyzed for circadian rhythmicity. 2.6. Activity Each animal was fitted with a mesh jacket that had a dorsal pocket to hold a bi-axial activity monitor (Individual Monitoring Systems, Baltimore, MD). The activity monitor has been used previously in a study of non-human primates [25]. Activity counts, transferred from the monitor to a computer through an interface, were summed in two-minute bins. 2.7. Feeding Pellet delivery by each PTS was recorded by an in-house software program written at Georgia State University. It was assumed that all dispensed pellets were consumed by the animal. This was confirmed by observation during husbandry. The caloric intake of each subject was calculated daily to confirm that adequate nutrition was provided through PTS. To examine the circadian feeding rhythm, pellet counts were summed by hour with totals assigned to the start of the hour block. Water was available ad libitum. 2.8. Circadian and statistical analysis All results are presented as mean and standard deviation. Circadian analysis was completed using in-house software (EPL, Davis, CA). The phase-fitting software (least-squares harmonic regression analysis) uses a Fourier-based algorithm to calculate a best-fit sine wave to each 24 hour block of data [26]. The time at which the sine wave peaks is deemed the acrophase. Mean, acrophase, and amplitude were calculated for each variable. Since all variables did not present a perfect sine wave fit, the timing of mean crossing of the average daily waveform was used to confirm phase. Statview 5.0 (SAS Institute, Inc. Cary, NC) was used for statistical analysis. Analysis of variance (ANOVA) and post-hoc tests were used to analyze gender differences in circadian variables. An alpha of 0.05 was considered significant.
L.K. Barger et al. / Physiology & Behavior 101 (2010) 595–600
3. Results 3.1. Body temperature Tb has a very stable circadian rhythm, and is a powerful variable when studying rhythmicity [27]. Using non-human primates as subjects enabled us to utilize implantable telemetric devices and to record data continuously for over one month. Recording for such a long period revealed an accurate average rhythm. An example of the raw Tb telemetry output from a male and female rhesus is presented in Fig. 1. Average daily waveforms for a five day window are presented in Fig. 2. Retroperitoneal mean Tb was 36.50 °C [SD 0.53] in males. Female mean Tb [36.96 °C; SD 0.31] while higher than male mean Tb, was not statistically different (p = 0.056). To determine the timing of the rhythms, a sine wave was fitted to each day's data. The peak of the fitted curve was deemed the acrophase. Mean Tb acrophase was 13.74 h [SD 0.36] in males and 14.57 h [SD 0.43] in females. Acrophase of the body temperature rhythm was significantly delayed (p = 0.001) in the females compared to the males. Mean acrophases for all variables are shown in Fig. 3. Amplitude was the most variable circadian Tb characteristic between animals. Males had a mean Tb amplitude of 0.63 °C [SD 0.21]. Female Tb amplitude [0.75 °C; SD 0.11] was not statistically different from male Tb amplitude. (Mean and amplitude data for all physiological variables are presented in Table 1.)
597
Both genders showed a similar daily waveform in Tb. Female Tb during the light period was significantly higher (p = 0.031) than male Tb during the light period. There was no difference in the mean Tb between genders during the dark period. Fig. 2a illustrates that females had a significantly sharper slope in the decline of Tb in the evening (p = 0.012). 3.2. Heart rate Circadian rhythms of heart rate were more variable than those of Tb; some animals showed a robust circadian HR rhythm, whereas others had a relatively flat rhythm. HR raw data was noisier (Fig. 1) than Tb. This was expected as HR can be rapidly influenced by external factors. Representative raw HR data are presented in Fig. 1 and average 24-hour waveforms in Fig. 2. Mean HR was similar between genders (males = 114.3 beats/min [SD 18.8]; females = 120.4 beats/ min [SD 9.0]). Mean HR acrophase for males was 13.14 h [SD 0.59], 0.6 h earlier than male Tb acrophase. Mean female HR acrophase was 14.61 h [SD 1.24], which was approximately the same as Tb acrophase. Female HR acrophase was significantly delayed (p = 0.013) compared to male acrophase. HR amplitude differed widely between subjects, ranging from as low as b10 beats/min to as high as 28 beats/min. HR amplitude did not differ significantly between genders (males = 14.6 beats/min [SD 7.5]; females = 12.1 beats/min [SD 3.8]).
Fig. 1. One week of raw physiological data from a representative female (left) and male (right) rhesus monkey. Presented are, from top to bottom, body temperature (Tb °C), heart rate (beats/min), activity (counts/10-min), and feeding (pellets/h). With the exception of feeding, all data are presented at 10-minute intervals or in 10 minute bins. A three-point moving average was applied to filter the Tb and HR data.
598
L.K. Barger et al. / Physiology & Behavior 101 (2010) 595–600
Fig. 2. Average daily waveforms (± SD) of body temperature (Tb °C), heart rate (bpm), activity (counts/10-min), and feeding (pellets/h) for a group of female (black line) and a group of male (gray line) rhesus monkeys. N/group was 7–8, with the exception of male activity (see Results for details).
The daily average rhythm in heart rate was very similar in males and females (Fig. 2). The increase in HR at lights on demonstrated the masking effects light can have on the rhythm of heart rate. There was no significant difference between genders in HR during either the light period or the dark period. 3.3. Activity One male removed his activity monitor and another monitor malfunctioned leaving activity (ACT) data available for 5 males. The
overall mean activity level for males was 20.6 [SD 7.4] counts/10-min. The average ACT acrophase in males [12.46 h; SD 0.88] occurred 0.68 h earlier than the HR acrophase and 1.28 h earlier than that of Tb. Mean male ACT amplitude was 18.1 counts/10-min [SD 10.5]. One female removed her activity monitor; therefore activity data were available for seven females. Female mean ACT was 25.0 counts/ 10-min [SD 7.2], which was not significantly different from male mean ACT. Timing of the female ACT rhythm, with an average acrophase of 13.90 h [SD 1.03], was advanced approximately 0.7 h ahead of both the Tb and HR rhythms. The timing of the ACT rhythm was significantly delayed in females (p = 0.030) compared to males. Female mean amplitude, at 18.5 counts/10-min [SD 6.9], was similar to that of the males' average. Representative raw ACT data are presented in Fig. 1. Daily waveforms of activity are shown in Fig. 2.
Table 1 Mean, amplitude and phase for multiple circadian rhythms in male and female rhesus monkeys. Units for mean and amplitude are given in the first column. There were no significant differences between genders in mean or amplitude. Phase is in hours. Female circadian rhythms were significantly delayed compared to males in all variables. p Values are given in the right column. Variable
Gender
Mean
Amplitude
Phase (h)
p (Phase)
Tb (°C)
Male Female Male Female Male Female Male Female
36.50 36.96 114.3 120.4 20.6 25.0
0.63 0.75 14.6 12.1 18.1 18.5
13.74 14.57 13.14 14.61 12.46 13.90 12.51 13.84
0.001
HR (bpm) Fig. 3. Mean acrophase (± SD) of the rhythms of body temperature, heart rate, activity, and feeding for female and male rhesus monkeys. Rhythms are significantly delayed (see Table 1 for p values) in females (black squares) compared to males (gray diamonds) in all variables.
ACT (counts/2 min) FD (pellets/h)
0.013 0.030 0.001
L.K. Barger et al. / Physiology & Behavior 101 (2010) 595–600
3.4. Feeding Female daily mean food consumption of 1039.21 pellets/day [SD 40.86] was significantly (p = 0.006) less than the male daily mean of 1309.24 pellets/day [SD 82.82]. However, this difference was negated when food consumption is considered on a per mass basis; females consumed 161.03 pellets/kg [SD 8.19] body weight and males 155.08 pellets/kg [SD 4.65] body weight. PTS was not available during the hours of dark and consequently no pellets were consumed between 2200 and 0600. This necessary restriction to control lighting conditions (i.e., no computer monitor illumination during the 8 h of darkness) affected the curve to which the sine wave was fitted and admittedly may have been different if animals fed throughout the night. However, although the feeding rhythm was artificially influenced by PTS availability being limited to the light hours, because both the male and female studies were conducted under this same condition, the timing of the feeding rhythms was comparable. In this variable as well, female acrophase [13.84 h; SD 0.76] was significantly (p = 0.001) delayed compared to male acrophase [12.51 h; SD 0.53]. 4. Discussion To our knowledge, this was the first comprehensive study examining gender differences in circadian rhythms using the rhesus monkey biomedical model. We found that the timing of multiple physiological rhythms was delayed in females compared to males. The literature on gender differences in circadian rhythms of human subjects is inconsistent. Conflicting results could be due to lack of environmental or social controls, studies of too short duration to obtain good average data, or other confounding factors such as the effects of aging. In humans, the influence of the menstrual cycle on circadian rhythms may also account for some of the conflicting literature, although no circadian differences were found between phases of the menstrual cycle in a recent rhesus study [24]. Additionally, the length of data collection in this study allowed each female rhesus to undergo a complete menstrual cycle negating any influence of the follicular or luteal phases. In this study, female mean Tb was higher than male Tb, although this difference only reached significance in the light period. This finding is consistent with Moe et al. who found higher peak temperatures in elderly females [18] and Winget et al. who reported that female rectal temperature exceed that of males at all time points [20]. Campbell also reported higher mean temperatures for female subjects [28]. The issue of gender differences in the timing of the Tb rhythm in humans is most inconsistent. There are reports of females having an advanced phase [17–19,29], a delayed phase [16], and a phase no different than males [20]. This study showed a significant phase delay in the female Tb rhythm relative to male subjects. The sharp decline in evening body temperature of female rhesus in this study was similar to that reported in human females [30]. Whereas this study revealed no gender differences in Tb amplitude, a larger amplitude in Tb has been reported for human males [17,20]. In contrast, Campbell reported a larger amplitude in Tb for females in subjects with a mean age of 49.1 years [28]. Other studies have reported changes in amplitude of this rhythm as a result of aging [31,32]. Gender may be a covariate in the magnitude of age-related amplitude changes, with males showing a greater amplitude decrease with age [29,30]. Circadian differences in heart rate may also be age related. For example, while mean HR is “generally higher” in females [20], Stein and colleagues reported no gender differences in HR of subjects aged 64–76 years, but in subjects aged 26–42 years, men had a lower mean HR [33]. Another study of human subjects, mean age of 26 years, showed no gender differences in mean HR [34]. Finally, Ryan and
599
colleagues found no gender or age differences in HR [35]. The monkeys in this study showed no differences in mean HR. As was the case for Tb, HR acrophase was significantly delayed in female monkeys compared to males. Young adult human females (ages 25–35) also showed a later acrophase than comparably aged males, however, this gender difference did not reach significance [20]. The highly variable HR amplitude did not differ between male and female monkeys in this study. Similarly, Wang and colleagues reported no gender difference in HR amplitude in young adults [34]. The high degree of variability within and between monkeys in HR was also noted in a study using pigtailed macaques (Macaca nemestrina; [36]). There was no gender difference in the amount of “spontaneous activity” in humans measured by microwave radars and reported as percent of time subject was moving [37] or in activity measured by energy expenditure [38]. No gender differences in the amount of physical activity were observed in a previous study using rhesus monkeys [39]. Similar to this study, Tapp and Natelson reported activity acrophase occurring prior to the acrophase of Tb in rhesus monkeys [40]. The advanced phase of ACT in relation to the HR and TB suggests that these rhythms do not merely parallel each other (e.g., HR does not simply reflect activity). Rhythms normally maintain a stable phase relationship within an individual; disturbance of this normal relationship, such as is seen in jet-lag, leads to physiological dysfunction [27]. In contrast to the delayed activity phase in female compared to male rhesus, human females performed habitual evening events significantly earlier than males as reported on daily logs [41]. This observation could potentially be due to social pressures rather than endogenous circadian regulation, again, underscoring the usefulness of the rhesus model for performing experiments under controlled conditions [41]. The timing of all rhythms studied in this experiment was significantly delayed in females compared to males (Fig. 3). Delays ranged from as little as 0.83 h in Tb, to as much as 1.47 h in HR. This consistent pattern in all variables suggests a fundamental difference between genders in the circadian timing system (CTS). Serotonin and estrogen have both been implicated as having possible roles in this difference [42]. Lepage and Steiner propose that females have a more responsive serotonergic system that causes differences in the sleep/ wake and circadian cycles [43]. Moreover, estrogen affects the function of the serotonergic system [44]. Separately or together, estrogen or serotonin may mediate inputs to the CTS. Of course, there may be many other interacting mechanisms contributing to these observed gender differences. Answering this question will require further studies. 4.1. Conclusions Consistent with what has been reported in the human literature, there were no reliable differences between genders in circadian mean or amplitude. There was, however, a consistent phase delay in the female rhesus in all physiological variables measured. Determining the mechanisms underlying this difference requires further exploration. Future study of rhythms and their gender differences could clarify the role of rhythm disruption in diseases such as SAD, sleeping and eating disorders and other psychological illnesses that have a greater preponderance in females [14,15]. Acknowledgements The authors would like to thank Peter Takeuchi and the staff at the California National Primate Research Center for their invaluable assistance during this study. Dr. Laura Barger is now affiliated with the Division of Sleep Medicine at Brigham and Women's Hospital and Harvard Medical School, Boston, MA 02115. Grants This study was supported by a grant from the National Aeronautics and Space Administration (NNJ04HF44G). Dr. Barger was supported
600
L.K. Barger et al. / Physiology & Behavior 101 (2010) 595–600
by NASA's Graduate Student Research Program (NFT-51417) and a Zonta International Amelia Earhart Fellowship. References [1] Colquhoun P. Rhythms in performance. In: Aschoff J, editor. Handbook of behavioral neurobiology: biological rhythms. New York: Plenum Press; 1981. p. 333–49. [2] Bunney WE, Bunney BG. Molecular clock genes in man and lower animals: possible implications for circadian abnormalities in depression. Neuropsychopharmacology 2000;22(4):335–45. [3] Czeisler CA, Richardson GS, Coleman RM, Zimmerman JC, Moore-Ede MC, Dement WC, et al. Chronotherapy: resetting the circadian clocks of patients with delayed sleep phase insomnia. Sleep 1981;4:1–21. [4] Fredrickson PA, Richardson JW, Esther MS, Lin SC. Sleep disorders in psychiatric practice. Mayo Clin Proc 1990;65:861–8. [5] Halaris A. Chronobiology and psychiatric disorders. New York: Elsevier; 1987. [6] Lewy AJ, Sack RL, Singer CM. Assessment and treatments of chronobiologic disorders using plasma melatonin levels and bright light exposure: the clock-gate model and the phase response curve. Psychopharmacol Bull 1984;20:561–5. [7] Van Cauter E, Turek FW. Depression: a disorder of time-keeping? Perspect Biol Med 1986;29:511–9. [8] Winget CM, DeRoshia CW, Markley CL, Holley DC. A review of human physiological and performance changes associated with desynchronosis of biological rhythms. Aviat Space Environ Med 1984;55:1085–96. [9] Davidson AJ, Sellix MT, Daniel J, Yamazaki S, Menaker M, Block GD. Chronic jet-lag increases mortality in aged mice. Curr Biol 2006;16(21):R914–6. [10] Lewy AJ, Kern HA, Rosenthal NE, Wehr TA. Bright artificial light treatment of a manic-depressive patient with a seasonal mood cycle. Am J Psychiatry 1982;139 (21):1496–8. [11] Rosenwasser AM, Wirz-Justice A. Depression: clinical and experimental models. In: Redfern PM, Lemmer B, editors. Physiology and pharmacology of biological rhythms. Berlin: Springer; 1996. p. 457–86. [12] Lewy AJ, Sack RL, Miller LS, Hoban TM. Antidepressant and circadian phaseshifting effects of light. Science 1987;235:352–4. [13] Lewy AJ, Sack RL. Light therapy and psychiatry. Proc Soc Exp Biol Med 1986;183: 11–8. [14] Ferrari E, Fraschini F, Brambilla F. Hormonal circadian rhythms in eating disorders. Biol Psychiatry 1990;27(9):1007–20. [15] Gorman J. Gender differences in depression and response to psychotropic medication. Gend Med 2006;3(2):93–109. [16] Mellette H, Hutt B, Askovitz S, Norvarth S. Diurnal variation in body temperatures. J Appl Physiol 1951;3:665–75. [17] Baker F, Waner J, Vieira E, Taylor S, Driver H, Mitchell D. Sleep and 24 hour body temperatures: a comparison in young men, naturally cycling women and women taking hormonal contraceptives. J Physiol 2001;530(Part 3):565–74. [18] Moe K, Prinz M, Vitiello M, Marks A, Larsen L. Healthy elderly women and men have different entrained circadian temperature rhythms. J Am Geriatr Soc 1991;39 (4):383–7. [19] Lee K. Circadian temperature rhythms in relation to menstrual cycle phase. J Biol Rhythms 1988;3(3):255–63. [20] Winget C, DeRoshia C, Vernikos-Danellis J, Rosenblatt W, Hetherington N. Comparison of circadian rhythms in male and female humans. Waking Sleeping 1977;1:359–63. [21] Jean-Louise G, Mendlowicz MV, Von Gizycki H, Zizi F, Nunes J. Assessment of physical activity and sleep by actigraphy: examination of gender differences. J Womens Health Gender Based Med 1999;8(8):1113–7. [22] Kerkhof GA. Inter-individual differences in the human circadian system: a review. Biol Psychol 1985;20(2):83–112.
[23] Hsieh K-C, Robinson EL, Fuller CA. Sleep architecture in unrestrained rhesus monkeys (Macaca mulatta) synchronized to 24-hour light–dark cycles. Sleep 2008;31(9):1239–50. [24] Barger L, Hoban Higgins T, Fuller C. Assessment of circadian rhythms throughout the menstrual cycle of female rhesus monkeys. Am J Primatol 2007;69:1–10. [25] Golub MP, Takeuchi PT, Hoban-Higgins TM. Nutrition and delayed activity offset in adolescent rhesus monkeys. In: Carskadon M, editor. Adolescent sleep: biological, psychological and social influences. England: Cambridge University Press; 2002. p. 50–68. [26] Halberg F, Tong YL, Johnson EA. Circadian system phase: an aspect of temporal morphology; procedures and illustrative examples. In: von Mayersbach H, editor. The cellular aspects of biorhythms. Berlin: Springer-Verlag; 1967. p. 20–48. [27] Moore-Ede M, Sulzman F, Fuller C. The clocks that time us. Cambridge, MA: Harvard University Press; 1982. [28] Campbell S. Effects of timed bright light exposure on shift-work adaptation in middle-aged subjects. Sleep 1995;18(6):408–16. [29] Prinz P, Moe K, Vitiello M, Larsen L, Lee K. Cosinor temperature measures in healthy entrained men and women: relationships with age, gender and sleep timing and quality. Sleep Res 1993;22:630. [30] Campbell S, Gillin J, Kripke D, Erikson P, Clopton P. Gender differences in the circadian temperature rhythms of healthy elderly subjects: relationships to sleep quality. Sleep 1989;12(6):529–36. [31] Vitiello M, Smallwood R, Avery D, Pascualy R, Martin D, Prinz P. Circadian temperature rhythms in young adult and aged men. Neurobiol Aging 1986;7(2): 97–100. [32] Weitzman E, Moline M, Czeisler C, Zimmerman J. Chronobiology of aging: temperature, sleep–wake rhythms and entrainment. Neurobiol Aging 1982;3(4): 299–309. [33] Stein P, Kleiger R, Rottman J. Differing effects of age on heart rate variability in men and women. J Cardiol 1997;80(3):302–5. [34] Wang Z, Wang L, Zhang L, Liu Q, Xue Z, Cornelissen G, et al. Circadian relations among cardiovascular variables of young adults. Chronobiologia 1992;19(3–4): 111–20. [35] Ryan S, Goldberg A, Pincus S, Mietus J, Lipsitz L. Gender- and age-related differences in heart rate dynamics: are women more complex than men? J Am Coll Cardiol 1994;24(7):1700–7. [36] Reite M, Short R. Circadian rhythms in monkeys: variability and behavioral correlations. Physiol Behav 1981;27(4):663–71. [37] Ferraro R, Lillioja S, Fontvielle A, Rising R, Bogardus C, Ravussin E. Lower sedentary metabolic rate in women compared with men. J Clin Invest 1992;90(3):780–4. [38] Morio B, Beaufrere B, Montaurier C, Verdier E, Ritz P, Fellmann N, et al. Gender differences in energy expenditure during activities and in daily energy expenditure of elderly people. Am J Physiol 1997;273:E321–7. [39] Ramsey J, Laatsch J, Kemnitz J. Age and gender differences in body composition, energy expenditure, and glucoregulation of adult rhesus monkeys. J Med Primatol 2000;29:11–9. [40] Tapp W, Natelson B. Circadian rhythms and patterns of performance before and after simulated jet lag. Am J Physiol 1989;257:R796–803. [41] Monk T, Petrie S, Hayes A, Kupfer D. Regularity of daily life in relation to personality, age, gender, sleep quality and circadian rhythms. J Sleep Res 1994;3: 196–205. [42] Leibenluft E. Do gonadal steroids regulate circadian rhythms in humans? J Affect Disord 1993;29(2–3):175–81. [43] Lepage P, Steiner M. Gender and serotonergic dysregulation: implications for the late luteal phase dysphoric disorder. In: Cassano GB, editor. Serotonin-related psychiatric syndromes: clinical and therapeutic links. London: Soc. Med. Servic.; 1991. [44] Biegon A, Fischette T, Mitchell D. Serotonin receptor modulation by estrogen in discrete brain nuclei. Neuroendocrinology 1982;35:287–91.