| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
APPETITE, OBESITY, DIGESTION, AND METABOLISM
1Faculty of Pharmaceutical Sciences, Josai International University, Togane, Chiba; and 2Faculty of Pharmaceutical Sciences, Josai University, Sakado, Saitama, Japan
Submitted 2 November 2005 ; accepted in final form 24 December 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
O2), and respiratory quotient (RQ) in mice were monitored continuously throughout 12:12-h light-dark cycles before, during, and after time-restricted feeding (RF). Mice fed ad libitum showed robust daily rhythms in both parameters: high during the dark phase and low during the light phase. The daily profile of energy metabolism in mice under daytime-only feeding was reversed at the beginning of the first fasting night. A few days after daytime-only feeding began, RF also reversed the circadian core body temperature rhythm. Moreover, RF for 6 consecutive days shifted the phases of circadian expression patterns of clock genes in liver significantly by 8–10 h. When mice were fed a high-fat (HF) diet ad libitum, the daily rhythm of RQ dampened day by day and disappeared on the sixth day of RF, whereas O2 showed a robust daily rhythm. Mice fed HF only in the daytime had reversed O2 and RQ rhythms. Similarly, mice fed HF only in the daytime significantly phase shifted the clock gene expression in liver, whereas ad libitum feeding with HF had no significant effect on the expression phases of liver clock genes. These results suggested that O2 is a sensitive indicator of entrainment in the mouse liver. Moreover, physiologically, it can be determined without any surgery or constraint. On the basis of these results, we hypothesize that a change in the daily O2 rhythm, independent of the energy source, might drive phase shifts of circadian oscillators in peripheral tissues, at least in the liver. circadian rhythm; metabolic rate; respiratory quotient; fat; core body temperature
Time of food access is a time cue (20, 32) that entrains peripheral oscillators, called food-entrainable oscillators (FEO). When rodents were restricted to eating only during light-rest periods, they exhibited food-anticipatory activity (FAA) before daily food presentation (6, 19). The restricted feeding (RF) of mice for several days shifted the phases of the circadian rhythm of clock gene expression in peripheral tissues even under cyclic light-dark conditions without affecting the phases of central oscillators in the SCN (5, 12, 33, 36). These lines of evidence indicate that food metabolites or signals (neuronal and/or humoral) elicited by food processing would be the principal timing cues for peripheral oscillators. However, little is known about the mechanism by which biological oscillators are entrained to time cues by feeding or how they are integrated into a unified behavior and physiology, such as FAA.
Energy metabolism, defined as oxygen consumption rate (O2) and respiratory quotient (RQ), exhibits circadian rhythms in human and rodents (23, 26). The liver functions as a peripheral integrator of nutrient availability and the energy needs of organisms (17). A major task of circadian oscillators in liver cells may be to anticipate and adapt to the physiological conditions required for food processing. Restricting mice to daytime-only feeding phase shifted core body temperature (Tb) and clock gene expression in peripheral organs such as the liver (5, 15). Energy metabolism is involved in cellular redox levels, such as the NAD+-to-NADH ratio, which may affect the expression of clock genes. To date, however, no information has been reported regarding the effects of RF on the rhythmicity of energy metabolism.
In this study, we examined the effects of RF on energy metabolism, Tb, and clock gene expression in peripheral oscillators in mice to reveal the relationship between physiological indicators and endogenous clocks in the process of food entrainment. Moreover, because we observed that both food deprivation and high-fat feeding lowered RQ but had distinct effects on O2, we also examined whether scheduled feeding, including restriction and combinations of food components, may differentiate O2 from RQ in energy metabolism under the conditions that cause phase shifting of peripheral clock genes.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Two types of powder chow (high-carbohydrate diet, HC; or high-fat diet, HF) were prepared to analyze the effects of food components on daily rhythm. The energy components of HC consisted of 60% carbohydrate and 12.7% fat, consistent with MF, whereas HF consisted of 20% carbohydrate and 45% fat. In both diets, 27.3% of the calories were protein. Each diet was supplemented with a vitamin mix and a mineral mix. The physiological fuel values for HC and HF were 357 and 358 kcal/g body wt, respectively. Throughout the experiments, there was no significant difference between the HC and HF groups in the volume of food consumed, which was determined on the basis of the weight of food remaining after each feeding.
Scheduled feeding. During RF, mice were fed either at night (NT) or during the day (DT). Mice in the DT group were allowed access to food for 9 h, from ZT2 to ZT11, and mice in the NT group had access to food for 15 h, from ZT11 to ZT2. Throughout the experiments, there were no significant differences in the volumes of food consumed between the NT and DT groups. Some mice were also fed a combination diet, either HF/HC or HC/HF. The HF/HC combination was given to mice that had been fed HF during ZT2–ZT11 and HC during ZT11–ZT2, and the HC/HF combination was given to mice that had been fed in the reverse order.
For the analysis of mRNA levels of clock genes, mice were killed by hemorrhage and their livers were removed after saline reflux was performed at 4-h intervals for 24 h on the seventh day of scheduled feeding. The livers were immediately frozen using liquid nitrogen and kept at –80°C until use.
Energy metabolism.
In vivo indirect open circuit calorimetry was performed in metabolic chambers at a controlled ambient temperature (24 ± 2°C). A constant air flow (0.6 L/min) was drawn through the chamber and monitored using a metabolic analyzer (O2/CO2 Analyzer MM202R; Muromachi Kikai, Tokyo, Japan). To calculate O2, the CO2 production rate (CO2) and the RQ (ratio of CO2 to O2) gas concentrations were monitored at the inlets and outlets of the sealed chambers. Throughout these experiments, the mice had access to water ad libitum, and food availability was controlled according to the experimental protocols.
The mice in each group were fed a diet appropriate to their feeding conditions. In the case of RF of MF-fed mice as shown in Fig. 1, mice (n = 4) were individually transferred to each chamber so that their metabolic indicators could be monitored before, during, and after RF. Two of the mice were placed in an NT group, and the other two were placed in a DT group. Another group of mice (n = 5) were transferred to the metabolic chambers, and their metabolic indicators were monitored for 24 h on the sixth day after the beginning of each feeding schedule. The average values of O2 and RQ during ZT16–ZT20 and ZT4–ZT12 were calculated.
|
Locomotor activity. To assess the daily rhythm of locomotor activity, mice were housed individually in exclusive cages equipped with a running wheel. Wheel revolutions were counted in 1-min bins, and wheel-running activity was expressed by constructing double-plotted figures using CompACT AMS version 3 (Muromachi Kikai, Tokyo, Japan). After mice (n = 12) fed MF ad libitum exhibited stable onset patterns under 12:12-h light-dark cycle conditions, two groups, NT (n = 6) and DT (n = 6), were subjected to RF for 6 days. Subsequently, all mice were returned to a diet of MF ad libitum.
RNA isolation and semiquantitation of mRNA levels by RT-PCR. For the analysis of mRNA levels of clock genes, mice (n = 3–5) were killed by hemorrhage and their livers were removed after saline reflux at 4-h intervals for 24 h on the seventh day after each scheduled feeding or 1 day after ad libitum feeding for 7 days. The livers were immediately frozen in liquid nitrogen and kept at –80°C until use.
Total RNA was extracted from the livers using a commercially available kit (RNeasy; Qiagen, Valencia, CA), and the remaining DNA was completely removed by RNase-free DNase treatment. Total RNA (50 ng) was reverse-transcribed with poly(dT)12–18 as a first-strand primer using RT-PCR beads (Amersham Biosciences, Uppsala, Sweden) according to the manufacturer's instructions. After the addition of the gene-specific primers, target genes were amplified by performing PCR using a thermal cycler (GeneAmp PCR System 9700; Applied Biosystems, Foster City, CA). The following PCR primers were used: mPer1, 5'-AGCCAGATTGGTGGAGGTTAC-3' (forward), 5'-GCCAGAGTCTTATTGGAGCAGT-3' (reverse); mPer2, 5'-GTGAAGGCTAATGAGGAGTACTACCA-3' (forward), 5'-CAGCAAACATATCCGCGTTAT-3' (reverse); and mDbp, 5'-GGAACTGAAGCCTCAACCAATC-3' (forward), 5'-CTCCGGCTCCAGTACTTCTCAT-3' (reverse). PCR was executed under the following conditions: PCR amplification for 30 cycles with denaturation at 95°C for 15 s, annealing at 57°C for 30 s, and extension at 72°C for 30 s. The PCR products were resolved by performing electrophoresis using 12% polyacrylamide gel and stained with 0.5 µg/ml ethidium bromide solution. Fluorescence intensity was quantified using an image analyzer (GeneGenius System, 100 V; Syngene, Frederick, MD).
Statistical analysis. Data are means ± SD. The significance of differences between two groups was determined using the unpaired t-test. The circadian rhythms of clock gene expression levels were analyzed using the single-cosinor model (22) to calculate the peak time (i.e., acrophase) in each group (n = 3–5). The significance of differences among more than two groups was determined using non-repeated-measures ANOVA with Dunnett's test.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
O2 and RQ values; that is, the values were high during the dark phase and low during the light phase. During RF, NT mice showed almost the same daily energy metabolism profile as mice fed MF ad libitum, with a decrease in RQ levels during the light phase due to the lack of a food supply (Fig. 1A). On the other hand, DT mice exhibited distinct daily rhythms in both O2 and RQ, and the phase was almost reversed beginning with the first fasting night (Fig. 1B). The reversed daily rhythm continued during RF. RQ levels during the dark phase in DT mice were decreased because they were in the light phase of NT mice. In both groups, refeeding ad libitum after RF restored the daily patterns of energy metabolism, although it took a few days for DT mice to return to a pattern resembling that before RF. Circadian rhythm of Tb. Figure 2 shows continuously monitored Tb in mice subjected to 12:12-h light-dark conditions before, during, and after RF. Mice fed MF ad libitum also exhibited a robust circadian rhythm of Tb, with the peak occurring during the dark phase and the nadir occurring during the light phase. RF induced the reversion of phases in the Tb rhythm in DT mice, whereas NT mice showed a rhythm similar to that of mice fed ad libitum. However, compared with energy metabolism, Tb rhythm took a much longer period to achieve a new phase after the onset of DT. During RF, the Tb amplitude of DT mice was greater than that of NT mice (Fig. 2B). In DT mice, on the first day after the end of RF, Tb remained low during the dark phase. The next day Tb returned to a temperature profile similar to that before RF.
|
|
|
|
O2 had a robust rhythm even after 6 days of HF feeding. Restriction only in the light phase with HF feeding reversed the daily rhythms of O2 and RQ, as HC feeding did, although HF feeding decreased RQ compared with HC feeding for both NT and DT. Moreover, mice were intentionally fed a combination of HC and HF. The HC-HF combination did not reverse the day-night variation of O2 or RQ in mice fed the HF-HC combination diet, although whole RQ levels in the HC-HF mice decreased as a result of ingestion of HF during the dark active period.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
O2 and RQ exhibited robust daily rhythms under ad libitum feeding conditions and adapted immediately to the restriction of food presentation periods. Because nocturnal rodents consume 
80% of their energy source during the dark phase when they are fed ad libitum, the daily rhythm of energy metabolism was almost parallel with the activity rhythm. Under RF with feeding during the light phase only, the daily profile of energy metabolism reversed immediately. These results indicate O2 and RQ sensitively reflect energy use in the body.
An increase of lipid use as an energy source lowers RQ, whereas an increase in glucose use raises it (2). Theoretical RQ ranges from 0.7 to 1.0. The reduction of RQ by HF feeding for several consecutive days resulted in dampened RQ rhythmicity without affecting the levels or rhythmicity of O2. Fasting also induces a metabolic change from glucose to lipid as the energy source used (11). Our results indicate that fasting decreased O2 and RQ, whereas feeding with either MF or HC raised O2 and RQ and HF feeding increased only O2. Therefore, O2 is a sensitive indicator of the fasting-to-fed cycle of mice, independent of energy source.
In general, Tb and locomotor activity increase and decrease approximately in parallel (24). It is widely thought that Tb equilibrium results from heat production and heat loss (14). Heat production consists of nonshivering thermogenesis, such as locomotor activity and energy metabolism, and shivering thermogenesis. Our results indicate that the reversal of Tb rhythm followed the reversal of energy metabolism by DT, whereas the daily locomotor rhythm remained unchanged throughout RF. Moreover, the reversed rhythm of Tb as well as energy metabolism persisted for at least two cycles after RF was interrupted in DT mice. Therefore, it is not likely that these phenomena were masking effects caused simply by the reversed feeding protocol. The results of Tb and behavior rhythms induced by RF are consistent with those reported in a study by Damiola et al. (5), in which peripheral molecular clocks in mice uncoupled from the master pacemaker in the SCN when feeding time was restricted.
We observed FAA 1–2 h before food presentation only in DT mice. Davidson et al. (7) reported evident FAA even when food was presented to mice at ZT16 under 12:12-h light-dark cycle conditions. In our study, food was presented to NT mice for 15 h from 1 h before the beginning of the dark phase until 2 h after the start of the light phase, and these mice showed the daily onset of behavioral activity once the lights were turned off. The absence of FAA in our NT mice may be due to the duration of food access, because when food access exceeded 12 h, the amount of FAA diminished (31).
We also demonstrated the phase shifts of clock gene expression in liver by 8–10 h in response to daytime feeding during ZT2–ZT11, regardless of food components, compared with those in mice fed ad libitum or at night. The acrophases of mPer1 in liver under stable light-dark cycle conditions with ad libitum or NT feeding were ZT18, which is consistent with previously reported data (5, 38). Stokkan et al. (33) revealed almost 12-h phase shifts of clock genes in liver after an 8-h (ZT3–ZT11) restriction from food access during the light inactive phase. We consider our results to be similar to those of previous works (5, 33) in spite of the shorter duration of phase shifts of clock genes.
Nagai et al. (21) reported that SCN ablation flattened the daily rhythms of energy metabolism (i.e., O2 and RQ) in rats. However, homozygous Clock-mutant mice showed a significant difference in O2 between day and night (35). These findings suggest that Clock-based oscillators are not essential to the daily rhythm of energy metabolism but that energy metabolism may be based on the daily feeding activity rhythm driven by endogenous clock components that the SCN integrates. Our data indicate that DT not only shifted but almost reversed the phases of the circadian rhythms of clock genes in mouse liver, with concomitant phase shifts of daily O2 and Tb rhythms, but not with locomotor activity rhythm. Moreover, Damiola et al. (5) revealed that the circadian rhythm of clock genes in peripheral organs phase shifted independently of the SCN by RF. Brown et al. (3) reported that Tb oscillation was capable of sustaining peripheral clock gene rhythmicity in vitro but that the physiological Tb rhythm was incapable of establishing circadian gene expression de novo. Altogether, these findings led us to speculate that adaptation to environmental food availability may induce an immediate change in O2 rhythmicity. In addition, subsequent alteration of FEOs, such as liver clock genes, Tb, and FAA, may result in desynchronization with major locomotor activity in a process driven by the master pacemaker in the SCN.
Switching the feeding time in rats to the light phase activates the central nervous system (CNS) (39). In rodents, activation of the CNS increases O2 (16, 25). Terazono (34) reported that the activation of sympathetic nerves through adrenergic neurons plays important roles in the CNS's regulation of clock gene expression in the liver. Escobar et al. (10) reported that several humoral factors changed during the 2 h before the anticipated onset of mealtime during RF. For example, free fatty acids and ketone bodies are high in concert with a decrease in triglycerides. Therefore, it is likely that humoral signals as well as neuronal signals induced by RF may be involved in the alteration of peripheral circadian oscillators.
Redox states are one of the most likely factors in the modification of clock gene expression. Rutter et al. (27) revealed that, at least in vitro, the DNA binding of the Clock-Bmal1 or NPAS2-Bmal1 heterodimers, which constitute the positive limbs of the central circadian feedback loop, is highly sensitive to the proportion of NAD cofactors. Schibler et al. (28) speculated that redox states are involved in food-induced phase shifts of circadian oscillators and suggested the necessity of studying the relationship between cellular redox states and circadian timing through in vivo experiments. In the present study, we clearly have shown a relationship between peripheral oscillators and O2, which is closely related to the cellular ratio of reduced to oxidized NAD cofactors under RF conditions in vivo. Merrow and Roenneberg (18) further hypothesized the metabolic feedback regulation of the molecular clock by redox states in both its positive and negative aspects, because the negative limb of clock components, cryptochromes, may bind a flavine cofactor, which is also regulated by redox states. Immediately before RF rats were allowed access to food, an increase of the NAD+-to-NADH ratio in hepatic cytoplasm and mitochondria was observed (8), and oxidative phosphorylation and NADH shuttling activity (1) were both high. These lines of evidence and considerations strongly support the hypothesis that the metabolic state of cells may influence molecular oscillators in the liver and other peripheral organs directly through changes in redox potential.
In conclusion, the present study indicates that O2 is a sensitive indicator of peripheral oscillators in the liver without any surgery or constraint. Moreover, we hypothesize that the daily rhythm of O2, independent of energy source, might drive the phase shift of circadian clock gene expression in peripheral organs, at least in the liver.
|
GRANTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. S. Bray and M. E. Young Diurnal variations in myocardial metabolism Cardiovasc Res, July 15, 2008; 79(2): 228 - 237. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. C. Russell, S. D. Proctor, S. E. Kelly, and D. N. Brindley Pair feeding-mediated changes in metabolism: stress response and pathophysiology in insulin-resistant, atherosclerosis-prone JCR:LA-cp rats Am J Physiol Endocrinol Metab, June 1, 2008; 294(6): E1078 - E1087. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
, ad libitum; 
, night (NT); 
, day (DT). Circadian rhythmicity was analyzed using single-cosinor model to estimate acrophases. Estimated acrophases of each clock gene expression are summarized in 
with dotted line, intentional feeding as described in detail in text. Circadian rhythmicity was analyzed using single-cosinor model to estimate acrophases. Estimated acrophases of the expression of each clock gene are summarized in