Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Oct 1.
Published in final edited form as: Aging Cell. 2012 Jul 12;11(5):794–800. doi: 10.1111/j.1474-9726.2012.00846.x

Deviation of innate circadian period from 24 hours reduces longevity in mice

Sergiy Libert 1, Michael S Bonkowski 1, Kelli Pointer 1,2, Scott D Pletcher 3, Leonard Guarente 1,*
PMCID: PMC3526942  NIHMSID: NIHMS386160  PMID: 22702406

Summary

The variation of individual lifespans, even in highly inbred cohorts of animals and under strictly controlled environmental conditions, is substantial and not well understood. This variation in part could be due to epigenetic variation, which later affects the animal’s physiology and ultimately longevity. Identification of the physiological properties that impact health and lifespan is crucial for longevity research and the development of anti-aging therapies. Here we measured individual circadian and metabolic characteristics in a cohort of inbred F1 hybrid mice and correlated these parameters to their lifespans. We found that mice with innate circadian periods close to 24 hours (revealed during 30 days of housing in total darkness) enjoyed nearly 20% longer lifespans than their littermates, which had shorter or longer innate circadian periods. These findings show that maintenance of a 24 hour intrinsic circadian period is a positive predictor of longevity. Our data suggest that circadian period may be used to predict individual longevity and that processes that control innate circadian period affect aging.

Introduction

Longevity is genetically limited. Soon after reaching sexual maturity the organism starts aging, which manifests itself in fitness decline, increased susceptibility to numerous diseases, and ultimately death. However, some individuals age at different rates compared to others. Even in the cohorts of highly inbred animals, which are housed in identical and tightly controlled environments, the variability of lifespans can be on the order of 250% (Sanchez-Blanco and Kim, 2011). One of the explanations for such variability could be an innate instability of the epigenetic information set during the development of the organism (Skora and Spradling, 2010). Epigenetic variation that occurs early in development might manifest itself as differences in the specific characteristics of individuals, which would influence the rate of aging and lifespan. Other possibilities also exist; these include de novo mutations, or intrinsic stochasticity of the developmental processes.

A number of aging theories exist that link the rate of organismal aging to the damage caused by reactive oxygen species, instability of the genetic programs after the development is completed, or general genome instability (both nuclear and mitochondrial) (Prinzinger, 2005). The circadian clock plays an important role in the above processes. The circadian clock mechanism is well studied (Bass and Takahashi, 2010) and it coordinates behavioral and biochemical processes with day/night cycles. Almost all cells of the organism sustain their own innate clock systems, which interact with environmental cues to optimize metabolism temporally.

Deviation of the innate circadian period from that of the environment can lead to a decrease in fitness in diverse organisms, ranging from cyanobacteria (Ouyang et al., 1998) to mice (Wyse and Coogan, 2010). While investigating free-running period in green alga Euglena gracilis and fruit flies Drosophila melanogaster (Bruce and Pittendrigh, 1956; Pittendrigh et al., 1958) it was suggested that a match of the innate circadian period to the environmental photoperiod might be beneficial. Indeed, artificial alteration of light cycles shortens the lifespan of Drosophila melanogaster, and 24-hour cycle housing conditions yields the longest lifespan (Pittendrigh and Minis, 1972). Additionally, genetic mutations that disrupt the circadian clock tend to shorten longevity and decrease fitness of fruit flies (Klarsfeld and Rouyer, 1998). In mammals, genetic disruption of the innate clock genes, such as CLOCK or BMAL1, has been associated with numerous diseases of aging such as cataracts (Dubrovsky et al., 2010), cancer (Zhu et al., 2009), psychiatric disorders (Rosenwasser, 2010), and metabolic syndrome (Sadacca et al., 2011).

Here we use an unbiased approach to investigate the impact of innate circadian rhythm on longevity in a homogeneous population of mice. Our findings suggest that strategies to match innate circadian rhythms to the diurnal cycle may provide benefits for humans, including extended longevity.

Results

To investigate the impact of the individual metabolic characteristics on longevity we chose to use F1 hybrid mice (CB6F1, purchased from NIA, NIH at 1 year of age), which are derived by crossing two inbredmouse strains BALB/cBy and C57BL/6. Even though these F1 hybrid mice are genetically identical, hybrid vigor alleviates strain specific diseases, which are often observed in inbred strains (Turturro et al., 1999). Moreover, these F1 hybrid mice are known to exhibit a range of innate circadian periods (Siepka and Takahashi, 2005), possibly due to epigenetic variation.

All animals were singly-housed in a climate-controlled environment for the duration of their lifespan. To measure the activity and circadian rhythm of the mice, we equipped each cage with a running wheel that contained an electronic transmitter to record revolutions. After habituation, all the animals established a typical circadian activity pattern, where they mostly rested during the 12-hour light period and were active during the 12 dark hours. Typical actograms of their running-wheel activity plotted versus time are presented on Figures 1A-F (each panel represents one typical animal). After 20 days of entraining, the light cycle was changed to 24-hour darkness for 30 days. During this testing phase, animals could not synchronize their activity with the external 12:12 light/dark illumination and their rest-activity patterns relied solely on their internal clock. In this cohort, we calculated each individual’s innate period (Τ) using the Lomb-Scargle periodogram (LSP) procedure (Van Dongen et al., 1999), and we identified individuals with a near 24-hour circadian period (typical actograms are presented in Fig. 1A-B), individuals with innate circadian period shorter than 24 hours (typical actograms are presented in Fig. 1C-D), and individuals with innate circadian period longer than 24 hours (typical actograms are presented in Fig. 1E-F). The distribution of Τ in this cohort of mice is rather broad and has a bell-like shape with its peak close to 24 hours (Figure 1G). After 30 days of circadian activity measurements, the light regime was returned to 12 hours light / 12 hours dark, running wheels were removed from cages, and lifespans of mice were monitored.

Figure 1.

Figure 1

Open in a new tab

Mice with innate circadian rhythm close to 24 hours have reduced mortality.

(A-F) Typical actograms of animal’s activity are shown (each panel represents one animal). Running activity of each animal was binned in 10 minute intervals and plotted against the time of the day (one line represents 24 hours) for 50 consecutive days. Red bars indicate continuous running activity for at least 20 minutes with an average speed greater than 2 feet per minute.

A-B) Typical actograms of animals with innate circadian period of 24 hours.

C-D) Typical actograms of animals with innate circadian period shorter than 24 hours.

E-F) Typical actograms of animals with innate circadian period longer than 24 hours.

G) The histogram of the distribution of the innate circadian period for the tested cohort of mice is shown. The distribution has symmetrical bell-like shape.

H) Deviation of the innate circadian period (Τ) by more than 7 minutes negatively impacts longevity of animals. Absolute deviation of the innate period (ΔΤ) was assigned to each animal. Animals were grouped into two categories, those with ΔΤ less than particular cutoff value and those with ΔΤ greater than that value. Statistical significance (using Wald test) of survival difference between these two groups of animals is plotted against ΔΤ cutoff values ranging from 1 to 24 minutes. ΔΤ of 7 minutes yields the maximum difference in survivorship.

I) Animals with innate circadian rhythm close to 24 hours (+/− 7 minutes, red survivorship curve, N=24) enjoy longer lifespan (p=0.0047) than animals with circadian rhythm significantly longer or shorter than 24 hours (greater deviation than 7 minutes, green curve, N=52, see also Supplementary Figure 1 and Supplementary Table 1).

J) Longevity of animals with longer versus shorter circadian rhythm does not differ. Survival curves for two groups are shown. Nred=33 (animals with circadian periods greater than or equal to 24 hours), Ngreen=43 (animals with circadian periods less than 24 hours), p=0.737.

Ample data from simpler organisms suggested that deviation of the circadian cycle from 24 hours might be detrimental for health and longevity. We decided to test this hypothesis. For each individual in our study, we computed a ΔΤ value, which is the deviation of that individual’s innate circadian period from 24 hours. We then asked whether animals with larger absolute deviations experienced significantly reduced lifespan compared to animals with deviations closer to zero. We did this by splitting our single experimental cohort into two sub-cohorts based on a specific cutoff value of |ΔΤ|. The first sub-cohort would be comprised of individuals whose innate circadian cycle deviated slightly from 24 hours (|ΔΤ|<cutoff value), and the second sub-cohort was comprised of individuals whose innate circadian cycle exhibited greater deviation from 24 hours (|ΔΤ|>cutoff value). Rather than choose our cutoff value a priori, we systematically assessed the difference in longevity between these two sub-cohorts using a range of different cutoff values, where Cox Regression and Wald statistics were applied to quantify the statistical significance of the longevity difference (Figure 1H). We found that a cutoff value of 7 minutes resulted in the two cohorts with the largest difference in lifespan and greatest Wald statistic. Indeed, the survivorship curves from these two sub-cohorts are strikingly divergent and reveal a statistically significant effect of tau deviation (P=0.0047, Figure 1H, 1I, and Supp. Table 1). Note that dividing the animals into two groups with shorter or longer than 24-hour periods showed no significant differences in longevities between groups (Fig. 1J and Supp. Table 2). To test the relationship between |ΔΤ| and survivorship in greater detail we treated |ΔΤ| as a continuous variable and performed Cox proportional hazard model analysis (Supplementary Figure 1). We found that ΔΤ3 was a significant predictor ofmortality risk (p=0.00054), while the linear and quadratic coefficients were not significant. These relationships were essentially unchanged following the inclusion of other physiological traits measured in this study as additional covariates in multi-covariate Cox hazard model (Supp. Figure 1).

These data show that mice with innate periods close to 24-hours enjoy greater longevity, and that epigenetic modifications or other processes that influence innate circadian rhythm may also affect the rate of aging.

It was of interest to determine whether other physiological parameters were impacted by the deviation of innate circadian period from 24 hours (ΔΤ). We performed this analysis in two ways. First, we used regression analysis to examine the linear relationship between ΔΤ and the value of interest, for example weight (Figure 2A). This analysis revealed that heavier individuals tended to have longer innate circadian periods, or greater ΔΤ (R=0.27 for linear fit, p=0.018). Indeed, the sub-cohort consisting of individuals with ΔΤ<0 (i.e.,. Τ<24 hours) has a higher average weight that the sub-cohort that consists of individuals with ΔΤ>0 (i.e., Τ>24 hours, Figure 2B, p=0.0004). A summary of the regression and binning analyses for all of the parameters measured in this study reveals significant association between Τ and weight as well as between Τ and metabolic rate. The association between Τ and physical activity was modestly significant, while we found no association with either fasted glucose levels or response to a glucose tolerance test (Figure 2C).

Figure 2.

Figure 2

Open in a new tab

Innate circadian rhythm correlates with certain physical parameters of individual animals.

A) Higher weight of animals correlates with longer innate circadian period (Τ). Scatter plot for individual animals is shown. Weight of animals is plotted against deviation of innate circadian rhythm from 24 hours (ΔΤ). Linear regression (R=0.27, p=0.018) reveals weak positive correlation between weight and Τ.

B) Animals with innate circadian period Τ greater than 24 hours on average weigh 8% more than animals with Τ less than 24 hours (p=0.0004).

C) The table presents the summary of correlations between individual’s weight, physical activity, glucose levels, metabolic rate, and glucose handling with their innate circadian period. Weight has positive correlation Τ (also see A-B), physical activity and metabolic rate have negative correlation with Τ, and glucose handling parameters do not show any correlation.

Second, we investigated the impact of other individual characteristics, which were previously reported to be associated with longevity, at the population level. Alterations in insulin signaling and glucose homeostasis have been associated with longevity in numerous species, including mammals (Bartke, 2011). We measured fasted glucose levels of individual animals as well as their response to glucose injection (glucose tolerance test) and found no correlation between longevity and fasted glucose levels (Figure 3A), or between longevity and the ability to handle glucose in a glucose tolerance test (Figure 3B). These factors also had little effect on organism lifespan, as assessed by Cox proportional hazard analysis (Supplementary Figure 2).

Figure 3.

Figure 3

Open in a new tab

Voluntary physical activity correlates with metabolic rate.

A) Fasted glucose levels do not correlate with longevity. We separated all the animals into three groups, according to their levels of blood glucose: high (125-190 mg/dL), medium (110-125 mg/dL), and low (60-109 mg/dL). Average longevity for these three groups is shown (+/− SEM). See also Supplementary Figure 2 for statistical analysis via Cox proportional hazard model.

B) Glucose tolerance does not correlate with longevity. We separated all the animals into three groups, according to their ability to absorb glucose in glucose tolerance test (GTT) as measured by the calculated area under the curve. Groups of animals with high area under the GTT curve (1150-1600 mmol/L/120 min), medium (1000-1150 mmol/L/120 min), and low (500-1000 mmol/L/120 min) were identified. Average longevity for these three groups is shown (+/− SEM). There are no statistically significant differences via t-test for all the possible pair-wise combinations or ANOVA (p=0.62). See also Supplementary Figure 2 for statistical analysis via Cox proportional hazard model.

C) Body weight of the animals does not strongly correlate with their lifespan. Median body weight of animals in our cohort was 44 grams (body weight was measured at 1 year of age after overnight fasting). Survival curves for two halves are shown. Open circles represent lighter half of the cohort (body weight <44 g, N=38), and filled triangles represent heavier half of the cohort (body weight >44 g, N=38). There is no statistically significant difference in survival between these two groups (p =0.441).

D) Animals with low voluntary physical activity tend to have shorter lifespans. We separated all the animals into three groups, according to their free-running activity: highly active individuals travelled from 1.25 to 3.25 kilometers per day, individuals with moderate activity from 500 to 1250 meters per day, and low activity from 0 to 500 meters per day. Animals with higher activity had on average longer lifespans (p=0.041). However, Cox proportional hazard analysis did not reveal linear relation between physical activity and longevity (see also Supplementary Figure 2).

E) Animals with higher metabolic rate (as measured by calories consumed per gram of body weight per day) tend to have longer lifespans (p=0.030). We separated all the animals into three groups, according to their metabolic rate. Individuals with high metabolic rate (those that consumed from 0.47 to 0.6 kilocalories per gram of body weight per day), moderate metabolic rate (0.42-0.47), and low metabolic rate (0.37-0.42). Average longevity for these three groups is shown (+/− SEM). Animals with high metabolic rate had on average greater longevity then moderate or low metabolic rate animals. However, Cox proportional hazard analysis did not reveal a linear relation between metabolic rate and longevity (see also Supplementary Figure 2).

F) There is a strong correlation between voluntary physical activity and metabolic rate. Individuals with higher relative metabolic rate also tend to be the individuals with higher voluntary physical activity (p=2.9*10−4). Scatter plot for these two values is shown. Each point represents an individual animal. This correlation shows that longevity increases shown on panels (D) and (E) are linked.

Across different species, animals with larger body size tend to live longer (Speakman, 2005), perhaps because they have smaller surface to volume ratio, which can support lower metabolic rates. There are a number of outliers for this observation, such as bats, which despite extremely high metabolic rate, enjoy lengthy lifespan. Paradoxically, within a species smaller individuals tend to have enhanced longevity, perhaps due to decreased levels of IGF and growth hormone (Bartke, 2000; Harper et al., 2006). We analyzed the effect of body weight on lifespan of individual animals in our cohort, and found no significant difference (Fig. 3C), although a trend for increased longevity of lighter animals was evident. Specifically, the average longevity of the lightest 38 animals (~50th percentile) was not statistically different from the heavier half of the cohort (p=0.439). Cox proportional hazard analysis did not support a significant impact of weight on longevity (Supplementary Figure 2).

It is generally agreed that exercise improves health and has the ability to extend longevity. Rats that had the ability to exercise enjoyed extended longevity when compared to their sedentary littermates (Holloszy and Schechtman, 1991). Recently, it has been shown that physical exercise is correlated withreduced mortality and extended longevity in humans (Wen et al., 2011). However, despite numerous improvements of health characteristics, exercise did not extend lifespan in mice (Samorajski et al., 1985). While we found a trend in the association between voluntarily physical activity and longevity (individuals with high voluntary physical activity (1.25-3.25 km/day) benefited from longer lifespans than their less active (0-0.5 km/day) siblings, Figure 3D), Cox proportional hazard model did not support a statistically significant impact of physical activity on longevity (Supplementary Figure 2).

Likewise, whereas there was only a trend of correlation between higher metabolic rates and increased longevity (Figure 3E), there was also a strong correlation (Figure 3F, R=0.42, p=2.9*10−4) between the ability of the animal to exercise and their metabolic rate, indicating that high metabolic rates and heightened ability to exercise are not independent parameters.

Discussion

In summary, we investigated the longevity variation within a genetically homogeneous cohort of mice and found that individual lifespans correlate with animal’s ability to maintain a circadian period close to 24 hours. This observation adds to our knowledge base relating the circadian clock to the rate of aging in mammals, and extends observations previously made in simpler organisms. Our findings also show trends for correlations between physical activity, metabolic rates and longevity, but these are not nearly as strong as that between longevity and innate circadian period.

Why do animals with innate periods close to 24 hours have longer lifespans? An intriguing possibility is that synchrony between the intrinsic circadian period and the externally cued light cycle is advantageous. Animals with intrinsic periods significantly different from 24 hours must reset their clocks daily in accord with the light cycle. This resetting itself may impose a metabolic stress that increases the rate of aging. An interesting prediction of this model is that placing individuals in an all-dark environment for the duration of their lives would normalize the differences in lifespans.

Another interesting question is whether animals with a circadian period longer than 24 hours and those with circadian period shorter than 24 hours have shortened lifespans for similar reasons. Our data suggest that reasons for shorter longevities are different for these two groups of animals. A number of physiological parameters, such as body weight, appear to be different between these two groups of animals. For instance animals with Τ>24 hours weigh on average more than animals with Τ<24 hours (Figure 2A-B). One can imagine that high body weight in animals with long periods might be deleterious to health and longevity due to the obesity-induced inflammation. Conversely, low body weight in animals with short periods might be a signal of a health problem, or require a very high metabolic rate to sustain normal temperature thereby taxing the organism with high ROS production.

It will be intriguing to determine whether any of our findings in these F1 hybrid mice apply to humans, which might suggest that strategies to match innate and extrinsic circadian periods would slow aging in at least some people. In this regard, it is interesting to note that melatonin, a drug that resets circadian rhythm (often used to alleviate jet lag), is also shown to increase longevity in simpler organisms (Bonilla et al., 2002) and mice (Anisimov et al., 2001). This is one of many possibilities of how adjustment of innate circadian rhythm could be beneficial for improving health and longevity.

Experimental Procedures

Animals

All procedures were performed according to guidelines and under supervision of Committee for Animal Care (CAC) of Massachusetts Institute of Technology.

Only males were used in this study. Seventy six CB6F1 (F1 progeny of BALB/cBy and C57BL/6 cross) one-year-old males were purchased from NIA, NIH (http://www.nia.nih.gov/research/dab/aged-rodent-colonies-handbook/available-strains). Animals were individually housed for the duration of life.

Diet

Ad libitum standard mouse chow (AIN-93G standard diet) and drinking water were provided for the animals, except one day before Glucose Tolerance Test (described below) and one day before 1 year-old body weight measurement (described in main text).

Running Wheel Activity

At 14 months of age low profile running wheels (Med Associates, Inc., St.Albans, VT) were introduced into each cage of singly housed mice. Each wheel was equipped with magnetic rotation counter which were electronically monitored by a computer using PCI Input/output board (PCI-DIO-120, Access I/O products, Inc., San Diego, CA). Activity of mice was monitored for 20 days on 12 hour dark: 12 hour light regimes, after which activity of mice was monitored for 30 days at 24 hour dark. During dark phase of the experiment, double black-out doors were used, so that veterinary personal could enter the mouse housing facilities without introducing ambient light. All mouse handling (during cage changes) was done using dim (less than 15 lux as measured by handheld light flux meter, VWR) deep red light (rodents have dichromatic vision and do not see red light, (Szel et al., 1992)). All the procedures that required humans to enter mouse facility were scheduled at random times of the day, so that no pattern could be established by experimental animals. After 50 days, running wheels were removed from the cages and lighting was returned to 12:12 schedule.

Statistical analysis

Free running period (tau, or Τ) was estimated using Lomb-Scargle periodogram (LSP) procedure (Van Dongen et al., 1999) using script written by Dr. Refinetti (ver. 2.7, http://www.circadian.org/main.html). Statistical analyses were conducted in R (version 2.13.1) using the ‘survival’ package.

Glucose Tolerance Test

Glucose Tolerance Test (GTT) was performed as described (Andrikopoulos et al., 2008) at 14 months of age. Briefly, after overnight fast basal glucose was determined by using a glucometer (Lifescan; Jonson & Jonson, New Brunswick, NJ) in blood obtained by removing the tip of the tail. Glucose (Sigma) was injected i.p. at 2 mg/kg of body weight. Blood was subsequently sampled at 15, 30, 60, 90 and 120 minutes thereafter for glucose measurements.

Relative metabolic rate

Relative metabolic rate is a value calculated from individual’s food consumption and weight. It reflects amount of calories consumed per gram of weight per unit of time, when food is provided ad libitum.

Longevity

Animals in this longevity study were checked daily for health and survival and were minimally handled for cage changes and measurements described in this study. Animals that appeared to be near death (listless, unable to walk, and cold to the touch) or had large bleeding tumors or neoplastic growth approaching 10% of body weight were euthanized, and the date of euthanasia was considered the date of death.

Supplementary Material

Supp Fig S1-S2 & Table S1-S3

Acknowledgments

We are grateful to Dr. Veerle Rottiers for useful discussions. This work was supported by the fellowship from Leukemia and Lymphoma Society (5089-09) to S.L., fellowship from EMF/AFAR to M.B., and grants from the NIH and a gift from the Glenn Foundation for Medical Research to L.G.

Footnotes

Author Contributions

S.L., M.B., and L.G. - designed the experiments; S.L., M.B., and K.P. - performed the experiments; S.L., M.B., and S.P. analyzed the data; S.L., M.B., S.P., and L.G. wrote the manuscript.

References

  1. Andrikopoulos S, Blair AR, Deluca N, Fam BC, Proietto J. Evaluating the glucose tolerance test in mice. Am J Physiol Endocrinol Metab. 2008;295:E1323–1332. doi: 10.1152/ajpendo.90617.2008. [DOI] [PubMed] [Google Scholar]
  2. Anisimov VN, Zavarzina NY, Zabezhinski MA, Popovich IG, Zimina OA, Shtylick AV, Arutjunyan AV, Oparina TI, Prokopenko VM, Mikhalski AI, et al. Melatonin increases both life span and tumor incidence in female CBA mice. J Gerontol A Biol Sci Med Sci. 2001;56:B311–323. doi: 10.1093/gerona/56.7.b311. [DOI] [PubMed] [Google Scholar]
  3. Bartke A. Delayed aging in Ames dwarf mice. Relationships to endocrine function and body size. Results Probl Cell Differ. 2000;29:181–202. doi: 10.1007/978-3-540-48003-7_10. [DOI] [PubMed] [Google Scholar]
  4. Bartke A. Growth hormone, insulin and aging: the benefits of endocrine defects. Exp Gerontol. 2011;46:108–111. doi: 10.1016/j.exger.2010.08.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bass J, Takahashi JS. Circadian integration of metabolism and energetics. Science. 2010;330:1349–1354. doi: 10.1126/science.1195027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bonilla E, Medina-Leendertz S, Diaz S. Extension of life span and stress resistance of Drosophila melanogaster by long-term supplementation with melatonin. Exp Gerontol. 2002;37:629–638. doi: 10.1016/s0531-5565(01)00229-7. [DOI] [PubMed] [Google Scholar]
  7. Bruce VG, Pittendrigh CS. Temperature Independence in a Unicellular “Clock”. Proc Natl Acad Sci U S A. 1956;42:676–682. doi: 10.1073/pnas.42.9.676. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Dubrovsky YV, Samsa WE, Kondratov RV. Deficiency of circadian protein CLOCK reduces lifespan and increases age-related cataract development in mice. Aging (Albany NY) 2010;2:936–944. doi: 10.18632/aging.100241. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Harper JM, Durkee SJ, Dysko RC, Austad SN, Miller RA. Genetic modulation of hormone levels and life span in hybrids between laboratory and wild-derived mice. J Gerontol A Biol Sci Med Sci. 2006;61:1019–1029. doi: 10.1093/gerona/61.10.1019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Holloszy JO, Schechtman KB. Interaction between exercise and food restriction: effects on longevity of male rats. J Appl Physiol. 1991;70:1529–1535. doi: 10.1152/jappl.1991.70.4.1529. [DOI] [PubMed] [Google Scholar]
  11. Klarsfeld A, Rouyer F. Effects of circadian mutations and LD periodicity on the life span of Drosophila melanogaster. J Biol Rhythms. 1998;13:471–478. doi: 10.1177/074873098129000309. [DOI] [PubMed] [Google Scholar]
  12. Ouyang Y, Andersson CR, Kondo T, Golden SS, Johnson CH. Resonating circadian clocks enhance fitness in cyanobacteria. Proc Natl Acad Sci U S A. 1998;95:8660–8664. doi: 10.1073/pnas.95.15.8660. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Pittendrigh C, Bruce V, Kaus P. On the Significance of Transients in Daily Rhythms. Proc Natl Acad Sci U S A. 1958;44:965–973. doi: 10.1073/pnas.44.9.965. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Pittendrigh CS, Minis DH. Circadian systems: longevity as a function of circadian resonance in Drosophila melanogaster. Proc Natl Acad Sci U S A. 1972;69:1537–1539. doi: 10.1073/pnas.69.6.1537. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Prinzinger R. Programmed ageing: the theory of maximal metabolic scope. How does the biological clock tick? EMBO Rep. 2005;6(Spec No):S14–19. doi: 10.1038/sj.embor.7400425. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Rosenwasser AM. Circadian clock genes: non-circadian roles in sleep, addiction, and psychiatric disorders? Neurosci Biobehav Rev. 2010;34:1249–1255. doi: 10.1016/j.neubiorev.2010.03.004. [DOI] [PubMed] [Google Scholar]
  17. Sadacca LA, Lamia KA, deLemos AS, Blum B, Weitz CJ. An intrinsic circadian clock of the pancreas is required for normal insulin release and glucose homeostasis in mice. Diabetologia. 2011;54:120–124. doi: 10.1007/s00125-010-1920-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Samorajski T, Delaney C, Durham L, Ordy JM, Johnson JA, Dunlap WP. Effect of exercise on longevity, body weight, locomotor performance, and passive-avoidance memory of C57BL/6J mice. Neurobiol Aging. 1985;6:17–24. doi: 10.1016/0197-4580(85)90066-1. [DOI] [PubMed] [Google Scholar]
  19. Sanchez-Blanco A, Kim SK. Variable pathogenicity determines individual lifespan in Caenorhabditis elegans. PLoS Genet. 2011;7:e1002047. doi: 10.1371/journal.pgen.1002047. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Siepka SM, Takahashi JS. Methods to record circadian rhythm wheel running activity in mice. Methods Enzymol. 2005;393:230–239. doi: 10.1016/S0076-6879(05)93008-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Skora AD, Spradling AC. Epigenetic stability increases extensively during Drosophila follicle stem cell differentiation. Proc Natl Acad Sci U S A. 2010;107:7389–7394. doi: 10.1073/pnas.1003180107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Speakman JR. Body size, energy metabolism and lifespan. J Exp Biol. 2005;208:1717–1730. doi: 10.1242/jeb.01556. [DOI] [PubMed] [Google Scholar]
  23. Szel A, Rohlich P, Caffe AR, Juliusson B, Aguirre G, Van Veen T. Unique topographic separation of two spectral classes of cones in the mouse retina. J Comp Neurol. 1992;325:327–342. doi: 10.1002/cne.903250302. [DOI] [PubMed] [Google Scholar]
  24. Turturro A, Witt WW, Lewis S, Hass BS, Lipman RD, Hart RW. Growth curves and survival characteristics of the animals used in the Biomarkers of Aging Program. J Gerontol A Biol Sci Med Sci. 1999;54:B492–501. doi: 10.1093/gerona/54.11.b492. [DOI] [PubMed] [Google Scholar]
  25. Van Dongen HP, Olofsen E, VanHartevelt JH, Kruyt EW. A procedure of multiple period searching in unequally spaced time-series with the Lomb-Scargle method. Biol Rhythm Res. 1999;30:149–177. doi: 10.1076/brhm.30.2.149.1424. [DOI] [PubMed] [Google Scholar]
  26. Wen CP, Wai JP, Tsai MK, Yang YC, Cheng TY, Lee MC, Chan HT, Tsao CK, Tsai SP, Wu X. Minimum amount of physical activity for reduced mortality and extended life expectancy: a prospective cohort study. Lancet. 2011;378:1244–1253. doi: 10.1016/S0140-6736(11)60749-6. [DOI] [PubMed] [Google Scholar]
  27. Wyse CA, Coogan AN. Impact of aging on diurnal expression patterns of CLOCK and BMAL1 in the mouse brain. Brain Res. 2010;1337:21–31. doi: 10.1016/j.brainres.2010.03.113. [DOI] [PubMed] [Google Scholar]
  28. Zhu Y, Stevens RG, Hoffman AE, Fitzgerald LM, Kwon EM, Ostrander EA, Davis S, Zheng T, Stanford JL. Testing the circadian gene hypothesis in prostate cancer: a population-based case-control study. Cancer Res. 2009;69:9315–9322. doi: 10.1158/0008-5472.CAN-09-0648. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp Fig S1-S2 & Table S1-S3
NIHMS386160-supplement-Supp_Fig_S1-S2___Table_S1-S3.pdf (321.3KB, pdf)

RESOURCES