Abstract
Environmental temperature remarkably impacts on metabolic homeostasis, raising a serious concern about the optimum housing temperature for translational studies. Recent studies suggested that mice should be housed slightly below their thermoneutral temperature (26°C). On the other hand, the external temperature, also known as a zeitgeber, can reset the circadian rhythm. However, whether housing temperature affects the circadian oscillators of the liver remains unknown. Therefore, we have compared the effect of two housing temperatures, namely 21°C (conventional; TC) and 26°C (thermoneutral; TN), on the circadian rhythms in mice. We found that the rhythmicity of food intake showed an advanced phase at TC, while the activity was more robust at TN, with a prolonged period onset. The serum levels of norepinephrine were remarkably induced at TC, but failed to oscillate rhythmically at both temperatures. Likewise, circulating glucose levels were increased but were non-rhythmic under TC. Both total cholesterol and triglycerides levels were induced at TN, but showed an advanced phase under TC. Additionally, the expression of hepatic metabolic genes and clock genes remained rhythmic at both temperatures, with the exception of G6Pase, Fasn, Cpt1a and Cry2, at TN. Nevertheless, the liver histology examination did not show any significant changes in response to housing temperature. Although the non-consistent trends of phase changes in each temperature, our results suggest a non-reductant role of temperature in mouse internal rhythmicity resetting. Thus, the temperature-controlled internal circadian synchronization within organs should be taken into consideration when optimizing housing temperature for mice.
Introduction
Temperature is a critical environmental parameter that profoundly impacts the physiological homeostasis of animals and humans. To date, mice are the most widely used animal model to mimic human diseases and to understand basic biological processes. Scientists have devoted great efforts in the optimization of mouse husbandry conditions to achieve the optimum translational outcomes, yet, optimal housing temperature remains a debate (Karp 2012, Maloney et al. 2014, Fischer et al. 2018, Reitman 2018). Historically and conventionally, mice are housed around 20–21°C, a range primarily chosen for human comfort (Karp 2012). However, it is now recognized that such a temperature leads to a mild cold stress in mice, as evidenced by a significant increase in their heart rates, mean arterial blood pressures and energy expenditure (Overton 2010, Karp 2012). Then, researchers postulated that mice should be housed at 30°C (Overton 2010, Gordon 2012), which falls within their thermoneutral zone, or the temperature of metabolic homeostasis (Gordon 1993). Such a recommendation was recently challenged, pointing out that 30°C was too hot and that mice do not tend to live at these temperatures when active (Keijer et al. 2019a). Based on a more accurate measurement of the basal metabolic rate (BMR), Keijer et al. suggested a new range of 25.5–27.6°C as an optimum housing temperature (Keijer et al. 2019a,b), which is slightly below thermoneutrality.
Among all the physiological processes affected by environmental temperature, the circadian clock might be the most representative, which plays an important role in the adaptation to external cues. Daily cycles of light and temperature are the most reliable timing signals for living organisms on Earth. Consequently, according to these signals, the endogenous circadian rhythms within organisms are entrained to the solar day (Pittendrich 1960, Refinetti 2010, Fonken et al. 2013). While the central suprachiasmatic nucleus (SCN), within the hypothalamus, is considered as the dominant circadian pacemaker to drive systemic rhythms, peripheral tissues have the capacity to generate specific oscillations in a cell-autonomous manner in response to external stimuli (Welsh et al. 1995, 2004, Balsalobre et al. 1998, Lowrey & Takahashi 2004, Nagoshi et al. 2004, Yoo et al. 2004). For instance, it has been reported that external temperature cycles can evoke rhythmic clock gene expression in Rat-1 fibroblasts (Izumo et al. 2003) and primary glial cells in vitro (Prolo et al. 2005), as well as the peripheral clocks within the organism (Lahiri et al. 2005, Machado et al. 2018). Therefore, temperature changes could be a potent zeitgeber to reset the circadian clock of mammalian peripheral tissues.
At the molecular level, the clock machinery involves a series of genes forming a transcriptional autoregulatory feedback loop. Of these important clock oscillators, Clock and Bmal1, two basic helix-loop-helix transcription factors, activate the transcription of period (Per) and cryptochrome (Cry) genes. In turn, Per and Cry proteins inhibit their own expression through the repression of Clock/Bmal1 activity, forming the critical feedback loop within the clock circuitry (Harmer et al. 2001, Reppert & Weaver 2001). The orphan nuclear receptors of the ROR and Rev-erb families are also implicated in the control of circadian clock function (Preitner et al. 2002). Given the importance of clock machinery, almost all the central and peripheral organs/tissues, such as the liver, possess an intact and functional clock system. It has been demonstrated that the key genes involved in hepatic principal functions (metabolic pathways, energy homeostasis, food processing and detoxification) exhibit robust rhythmic expression (Bass & Takahashi 2010, Ferrell & Chiang 2015, Reinke & Asher 2016). Thus, impairments of the clock alignments induced by unhealthy living styles would abolish normal hepatic rhythmicity, leading to metabolic diseases (Ferrell & Chiang 2015).
Hepatic metabolism responds sensitively to external changes. For instance, thermoneutral housing (30°C) exacerbates nonalcoholic fatty liver disease in mice, associated with an elevated expression of genes involved in lipid metabolism pathways and fatty acid oxidation (Giles et al. 2017). Although the sensitivity of hepatic metabolism to temperature is well known, the role of circadian genes in the hepatic adaptation to temperature remains unknown. Therefore, in the present study, mice were subjected respectively to TC and the new suggested TN, to determine whether the different housing temperature would affect the 24-h variation of circulating metabolites and the circadian expression of metabolic/clock genes in the liver. The aim of this study was to enhance the understanding of the impact of housing temperature on clock machinery and metabolism in the mouse liver, facilitating further optimization of housing conditions for translational research in a chronobiological perspective.
Materials and methods
Animals and housing conditions
All animal procedures were conducted according to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996) and were approved by the committee of the Ethics regulations set by the Laboratory Animal Care Committee at China Pharmaceutical University (permit number: SYXK-2018-0019).
Six-week-old male C57BL/6J mice were purchased from the Model Animal Research Center of Nanjing University (Nanjing, Jiangsu, China). Mice were acclimated in the standard environment (26°C, 50–60% humidity, 12 h light:12 h darkness cycle) for 1 week and fed a normal diet ad libitum (1 g of food contains 0.18 g of proteins, 0.04 g of crude fat and 0.59 g of carbohydrates, equivalent to 3.44 kcal/g of food). The light was on at 08:00 h (Zeitgeber 0, ZT0) and off at 20:00 h (ZT12). Two housing temperatures were set: 26 ± 1°C (TN) and 21 ± 1°C (TC). Seventy-two mice were randomly divided into these two temperature groups, where each group contained six sub-groups; randomly corresponding to six time-points during a day (ZT1, ZT5, ZT9, ZT13, ZT17 and ZT21). Mice were subjected to these two housing temperatures for 12 weeks. A thermometer was placed in each cage to monitor the authenticity of the temperature in situ and the temperature in each cage was measured daily. The body weight, food intake and water drinking were monitored weekly.
Food rhythm
Food intake and water drinking were recorded every 30 min using an animal metabolic monitoring cage system (TSE-Phenomaster system, China Ltd) and data were analyzed with the TSE system software (TSE-Phenomaster system, China Ltd).
Locomotor activity analysis
Twelve mice were randomly assigned for TN or TC group and housed individually in a cage equipped with running wheels to detect their locomotor activity. They were fed with standard food and water ad libitum. After 2 weeks of accommodations to a 12 h light:12 h darkness cycle, the mice were switched into the constant darkness (DD) for 10 weeks. Wheel revolutions were recorded and analyzed using the ClockLab analytical software (Actimetrics, Wilmette, IL, USA). The circadian period and periodogram amplitude of the activity rhythm in the LD and DD cycles were calculated using the chi-squared periodogram (Sokolove & Bushell 1978). The amplitude of daily activity during the LD and DD cycles was determined via fast Fourier transformation analysis.
Blood and liver collection
To minimize the impact of handling and the retro-orbital bleed, we collected blood every 4 h from separate cohorts of mice for each time-point. For instance, for the ZT1 sub-group, blood samples were collected in non-heparinized tubes through retro-orbital bleeding at ZT1 and then stored at 4°C. After the blood collection, mice were directly sacrificed via cervical dislocation. Their livers were removed, washed with PBS, frozen in liquid nitrogen and stored at −80°C before use.
Serological analysis
The blood samples were centrifuged at 448 g for 10 min at 4°C to obtain the serum. The serum level of norepinephrine (NE), total cholesterol (TC) and triglycerides (TG) (Nanjing Jiancheng Institute of Biotechnology, Nanjing, China), as well as non-esterified fatty acids (NEFA) (Fujifilm Wako Pure Chemical Industries, Osaka, Japan) and glucose (Shanghai Rongsheng Biotechnology, Shanghai, China) were determined with commercial kits, following the manufacturer’s instructions.
Liver metabolite measurements
For the liver lipid measurements, 100 mg of frozen liver samples were homogenized with 900 µL ethyl alcohol on iced bath. The homogenate was centrifuged at 336 g for 10 min. The supernatant was used to determine the hepatic contents of TC, TG and NEFA using commercial kits, according to the manufacturer’s instructions. The protein content was quantified using a BCA kit (Jiancheng Institute of Biotechnology, Nanjing, China).
RNA extraction and RT-qPCR
Total RNA was extracted from the livers using TRIzol reagent (Thermo Fisher, Life Technologies) and quantified by measuring the absorbance at 260 nm. The RNA quality was determined by measuring the 260/280 ratio (>1.8). Thereafter, cDNA was synthesized by a PrimeScript 1st-strand cDNA Synthesis Kit (TaKaRa) according to the manufacturer’s instructions. RT-qPCR was performed with a StepOnePlus Real-Time PCR Detection System (Applied Biosystems) using a SYBR Premix Ex Taq kit (TaKaRa) for the relative mRNA expression levels of the genes. A total of 10 µL of PCR reaction mix contained 2 µL of cDNA, 5 µL of qPCR SYBR mix (TaKaRa), 0.3 µL of primer and 2.7 µL of double-distilled water. The primer sequences were listed in Supplementary Table 1 (see section on supplementary materials given at the end of this article). 36B4 was used as the housekeeping gene to normalize the gene expression levels.
Histology
For the hematoxylin and eosin (H&E) staining, the frozen samples were fixed with 4% (v/v) paraformaldehyde solution for 24-h in situ and processed for paraffin embedding. The liver blocks were cut into 4 μm transverse sections for staining. Another slice of 3 μm of a fresh frozen liver was stained with oil red O (ORO) to observe the lipid accumulation, by using a light microscopy at ×400 magnification (Nikon Eclipse Ts2R; Nikon).
Identification of the rhythmicity
JTK cycle was performed to determine the rhythmicity of the experimental measures during the course of the day. Cosinor analysis was assessed to obtain the key chronobiological parameters of MESOR (midline statistic of rhythm, a rhythm-adjusted mean), amplitude, and acrophase (peak time) of the rhythm. Each data set was fitted to a cosine regression model, f(t) = M + Acos(2πt/τ + ϕ) + e(t), in cosinor2 package of R software. We defined the variables as M (MESOR), A: amplitude (a measure of half the extent of predictable variation within a cycle), ϕ: acrophase (a measure of the time of overall high values recurring in each cycle), τ: period (duration of one cycle, here is 24 h), and e(t) is the error term. Meanwhile, Cosinor2 package was used to compare the cosinor parameters of two groups according to a previous study (Cornelissen 2014). Statistical significance was set at P < 0.05.
Statistical analysis
Repeated-measure ANOVA was used to determine the statistical significance of the weight-gain, food intake and water drinking measurements. Two-factor ANOVA was performed to determine the statistical significance of the main effects of temperature, time and temperature × time on serological analysis and mRNA quantifications. Data were presented as mean ± s.e. (s.e.m.). The data sets on figures were analyzed with Student’s t-test and group differences were considered statistically significant at P < 0.05.
Results
Housing temperature affects the circadian patterns of the mouse behavior
Although the increasing body weight over the 12-weeks of study (Fig. 1A), no significant difference was observed on the body weight-gain between the temperature groups (F = 0.66, P = 0.42), which suggested that the housing temperature did not affect the body weight. The calorie intake was significantly induced in the TC group (average = 23.19 ± 1.5 kcal/body weight vs 19.06 ± 1.86 kcal/body weight). The food intake showed a circadian rhythmicity over a 24 h, but with a phase delay under the TN temperature compared to the TC (Fig. 1B and Tables 1, 4). In contrast, the housing temperature did not affect the water drinking (Fig. 1D and Table 1). These data suggest that the temperature is a critical Zeitgeber for the food intake rhythmicity.
The housing temperatures alter the food intake in mice. (A) Weekly measure of the body weight-gain. (B) Weekly measure of calorie intake. (C) Rhythmicity of the food intake. (D) Weekly measure of the water drinking. Data are presented as mean ± s.e.m. (n = 36 for each temperature group of the body weight-gain, food intake and water intake; n = 6 for the food intake rhythmicity), *P < 0.05, **P < 0.005 vs TC.
Citation: Journal of Endocrinology 247, 2; 10.1530/JOE-20-0100
RM ANOVA analysis of the mouse weight-gain, the food intake and the water drinking.
| Name | F | P |
|---|---|---|
| Weight-gain | 0.66 | 0.42 |
| Food intake | 50.25 | 0.00 |
| Water drinking | 0.026 | 0.875 |
n = 36 for each temperature group. P = 0.00 indicates that P < 0.004.
In general, under LD, although the onset activity of the mice was slightly delayed at TN; it was unchanged in both temperatures under DD. In contrast, the offset activity at both light conditions was advanced (Fig. 2A and B), slightly shifting their phases (Fig. 2C). Yet, the housing temperature did not alter the period (Table 2). The periodogram analysis indicated that the mouse behavioral rhythms at TN was more robust (Fig. 2D), associated with a significant amplification of their amplitude activity (Fig. 2E). Besides, higher ratios of rho counts (resting) to alpha counts (active) were observed in both light conditions (Table 3) at TC. Our results demonstrated a disruption of the mouse physiological behavior at TC, and suggested a tendency of energy saving.
The housing temperatures alter the wheel-running activity of the mice. (A) Double plotted locomotor activities of the mice. The mice were subjected to the LD cycle for 15 days and then transferred to the DD cycle for 30 days. A 24-h light:12 h darkness cycle recording is represented by the white box (as the day) and black box (as the night). A 24-h DD recording is represented by black and grey boxes. (B) The onset and offset activity in the LD (left) and DD (right) cycles. (C) Graphic representation of the phase under different temperatures and light/dark or dark/dark exposures. (D) Periodogram amplitude of the wheel-running activity rhythms of the mice. (E) The ratio activities per hour, relative to the total daily activities. Left panel: under the LD cycle, the right panel: under the DD cycle. Data are presented as mean ± s.e.m. (n = 6 for each temperature group), **P < 0.005 vs TC.
Citation: Journal of Endocrinology 247, 2; 10.1530/JOE-20-0100
Average periods.
| ZT | CT | ||
|---|---|---|---|
| 21°C | 26°C | 21°C | 26°C |
| 24.02 ± 0.04 h | 24.01 ± 0.06 h | 23.81 ± 0.13 h | 23.89 ± 0.10 h |
n = 6 for each temperature group.
Average counts of the wheel revolutions per circadian cycle.
| Temperature | Parameters | LD | DD | ||||
|---|---|---|---|---|---|---|---|
| Alpha count | Rho count | Total | Alpha count | Rho count | Total | ||
| 21°C | Average activity | 2540.64 | 6490.74 | 9031.38 | 5787.02 | 9665.97 | 15452.96 |
| Activity/day | 169.38 | 432.72 | 602.1 | 192.9 | 322.2 | 515.1 | |
| 26°C | Average activity | 4046.89 | 18497.37 | 22544.25 | 3903.19 | 12008.08 | 15911.27 |
| Activity/day | 337.24 | 1541.45 | 1878.69 | 130.11 | 400.27 | 530.37 | |
n = 6 for each temperature group. LD = 15 days, DD = 30 days.
Housing temperature affects the expression and the circadian rhythmicity of serum metabolites
Since the lower temperature induces stress in mice, we firstly measured the serum levels of NE, which is a major transmitter of sympathetic nervous system for the thermoregulatory effects (Kozyreva et al. 2015). As expected, a significant elevated concentration was noticed in the TC group (Fig. 3A), although its failure to oscillate under both temperatures (P = 1, Table 4).
The housing temperatures alter the circulating levels of metabolites in mice. (A) Norepinephrine. (B) Circulating glucose levels. (C) Serum levels of TC. (D) Serum levels of TG. (E) Serum levels of NEFA. Data are presented as mean ± s.e.m., n = 6 per time-point. *P < 0.05, **P < 0.005 vs TC.
Citation: Journal of Endocrinology 247, 2; 10.1530/JOE-20-0100
JTK cycle and Cosinor analysis.
| Name | JTK cycle | Cosinor | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Rhythmicity | Amplitude | MESOR | Acrophase | True | ||||||||||||
| 21°C | 26°C | 21°C | 26°C | 21°C vs 26°C | 21°C | 26°C | 21°C vs 26°C | 21°C | 26°C | 21°C vs 26°C | Peak time | |||||
| P | P | P | P | F | P | P | P | F | P | P | P | F | P | 21°C | 26°C | |
| F.I | 0.00 | 0.00 | 0.15 | 0.11 | 33.09 | 0.00 | 0.09 | 0.11 | 0.10 | 0.75 | −0.12 | −6.11 | 142.53 | 0.00 | 16 | 15 |
| NE | 1.00 | 1.00 | 2.82 | 3.19 | 0.01 | 0.93 | 63.48 | 51.74 | 65.85 | 0.00 | −2.06 | −5.28 | 0.77 | 0.40 | 23 | 12 |
| Glucose | 1.00 | 0.00 | 0.54 | 1.18 | 1.13 | 0.31 | 9.15 | 8.11 | 25.57 | 0.00 | −0.67 | −5.47 | 14.62 | 0.00 | 18 | 12 |
| TC | 0.04 | 0.03 | 0.25 | 0.27 | 0.00 | 0.96 | 3.76 | 4.59 | 162.38 | 0.00 | −5.10 | −0.57 | 28.31 | 0.00 | 11 | 18 |
| TG | 0.00 | 0.00 | 0.27 | 0.64 | 2.95 | 0.12 | 1.32 | 1.57 | 12.64 | 0.01 | −1.02 | −1.13 | 31.66 | 0.00 | 19 | 20 |
| NEFA | 1.00 | 0.09 | 0.08 | 0.15 | 0.22 | 0.65 | 1.36 | 1.28 | 3.72 | 0.08 | −1.93 | −1.75 | 4.64 | 0.06 | 16 | 15 |
| Pepck | 0.00 | 0.00 | 1.34 | 0.72 | 33.57 | 0.00 | 0.66 | 0.53 | 0.22 | 0.65 | −3.79 | −4.16 | 24.16 | 0.00 | 6 | 7 |
| G6Pase | 0.00 | 0.50 | 1.48 | 0.89 | 111.30 | 0.00 | 0.80 | 0.06 | 4.18 | 0.07 | −4.59 | −5.54 | 32.01 | 0.00 | 9 | 13 |
| Pgc1a | 0.03 | 0.02 | 1.29 | 1.15 | 4.04 | 0.07 | 0.22 | 0.21 | 0.00 | 0.97 | −3.09 | −2.21 | 8.91 | 0.01 | 3 | 0 |
| Glut 2 | 0.00 | 0.00 | 2.85 | 1.58 | 382.59 | 0.00 | 1.58 | 0.89 | 6.49 | 0.03 | −3.28 | −1.82 | 122.89 | 0.00 | 4 | 22 |
| Gck | 0.00 | 0.00 | 1.37 | 0.85 | 223.36 | 0.00 | 0.72 | 0.59 | 0.76 | 0.40 | −3.39 | −4.33 | 111.85 | 0.00 | 4 | 8 |
| Gys2 | 0.00 | 0.00 | 2.26 | 2.53 | 17.50 | 0.00 | 1.27 | 2.08 | 9.57 | 0.01 | −2.95 | −2.00 | 182.69 | 0.00 | 3 | 23 |
| Ppp1r3b | 0.01 | 0.00 | 0.78 | 0.57 | 38.80 | 0.00 | 0.34 | 0.25 | 0.53 | 0.48 | −3.99 | −4.99 | 0.53 | 0.48 | 7 | 11 |
| Ppp1r3c | 0.00 | 0.00 | 0.54 | 0.53 | 0.02 | 0.89 | 0.39 | 0.34 | 0.61 | 0.45 | −6.05 | −0.24 | 0.61 | 0.45 | 15 | 16 |
| Srebp1c | 0.00 | 0.02 | 1.08 | 0.49 | 418.25 | 0.00 | 0.83 | 0.16 | 15.78 | 0.00 | −2.47 | −1.11 | 120.31 | 0.00 | 1 | 20 |
| Fasn | 0.00 | 0.23 | 0.97 | 0.98 | 0.08 | 0.78 | 0.51 | 0.24 | 8.40 | 0.02 | −2.80 | −1.25 | 295.11 | 0.00 | 2 | 20 |
| Ppara | 0.00 | 0.00 | 1.71 | 1.17 | 138.66 | 0.00 | 0.73 | 0.33 | 6.79 | 0.03 | −3.05 | −3.12 | 42.17 | 0.00 | 3 | 3 |
| Cpt1a | 0.00 | 0.22 | 1.88 | 1.55 | 58.40 | 0.00 | 0.65 | 0.44 | 1.97 | 0.19 | −2.82 | −2.94 | 57.51 | 0.00 | 2 | 3 |
| Bmal1 | 0.00 | 0.00 | 0.48 | 0.46 | 0.59 | 0.46 | 0.53 | 0.52 | 0.02 | 0.89 | −5.56 | −5.93 | 142.30 | 0.00 | 13 | 14 |
| Clock | 0.00 | 0.00 | 0.64 | 0.73 | 4.19 | 0.07 | 0.33 | 0.33 | 0.00 | 0.96 | −5.75 | −5.86 | 92.93 | 0.00 | 13 | 14 |
| Per1 | 0.00 | 0.00 | 3.05 | 2.32 | 23.80 | 0.00 | 3.07 | 1.93 | 4.81 | 0.05 | −3.04 | −3.41 | 139.98 | 0.00 | 3 | 5 |
| Per2 | 0.00 | 0.00 | 1.88 | 1.28 | 104.05 | 0.00 | 1.54 | 1.16 | 3.02 | 0.11 | −3.21 | −3.63 | 138.72 | 0.00 | 4 | 5 |
| Cry1 | 0.00 | 0.00 | 1.24 | 2.87 | 475.53 | 0.00 | 0.95 | 3.56 | 119.65 | 0.00 | −5.08 | −5.04 | 306.18 | 0.00 | 11 | 11 |
| Cry2 | 0.00 | 0.15 | 1.91 | 1.93 | 0.51 | 0.49 | 0.98 | 1.51 | 27.19 | 0.00 | −2.67 | −1.58 | 382.37 | 0.00 | 2 | 22 |
| Rev-Erba | 0.00 | 0.00 | 6.80 | 6.68 | 0.74 | 0.41 | 9.92 | 9.23 | 1.54 | 0.24 | −1.70 | −1.72 | 1028.10 | 0.00 | 22 | 22 |
| DBP | 0.00 | 0.00 | 98.25 | 81.92 | 9.76 | 0.01 | 151.89 | 125.31 | 1.51 | 0.25 | −2.31 | −2.42 | 166.35 | 0.00 | 0 | 1 |
| Liver TC | 0.87 | 0.05 | 0.27 | 0.26 | 0.38 | 0.55 | 0.04 | 0.06 | 0.18 | 0.68 | −1.70 | −4.00 | 312.60 | 0.00 | 22 | 7 |
| Liver TG | 1.00 | 0.00 | 0.50 | 0.44 | 1.87 | 0.20 | 0.05 | 0.11 | 0.94 | 0.36 | −4.69 | −3.65 | 388.82 | 0.00 | 9 | 5 |
The acrophase is measured in degrees (in relation to a reference time set to 0°, with 360° equated to the period); the peak time is measured in ZT (Zeitgebers), ZT0 refers to the time of 08:00 h 21°C vs 26°C is the statistical difference between the groups, calculated by the regression model with R software. n = 36 for each temperature group. P = 0.00 indicates that P < 0.004.
F.I, food intake.
Likewise, we measured the concentration of the circulating metabolites, which exhibit dramatic oscillations in energy demands and nutrients supply throughout 24 h. The glucose concentration was significantly induced at the day time under TC (Fig. 3B), while both serum concentrations of TC and TG were significantly increased under TN (Fig. 3C and D). The circulating levels of NEFA remained similar between the two temperature groups and, together, failed to show a rhythmic pattern (P = 1, Fig. 3E). Besides, the level of glucose was non-rhythmic under 21°C. On the other hand, the serum levels of TC and TG of the TC group were significantly rhythmic and presented an advanced phase (Amplitude = 0.25; MESOR = 3.76, P < 0.05; Acrophase = −5.10, P < 0.05; Peak time = ZT11, and Amplitude = 0.27; MESOR = 1.32, P = 0.01; Acrophase = −1.02, P < 0.05; Peak time = ZT19, respectively). More importantly, two-factor ANOVA analysis showed that the temperature alone sufficiently altered their contents, suggesting the role of temperature on the circulating metabolites and their rhythmicity. The detailed parameters analyses were reported in Table 4 and Supplementary Table 2.
Housing temperature affects the circadian rhythmicity of the hepatic metabolic genes
Given that the temperature influences the circadian rhythmicity of the circulating metabolites, we also analyzed the diurnal oscillation of the key metabolic gene expression of the liver. First, the expression levels of the genes involved in the gluconeogenesis and glycolysis were downregulated in the TN group (Fig. 4A and B). A delayed phase occurred for the phosphoenolpyruvate carboxykinase (Pepck) (Table 4), while glucose 6-phosphatase (G6Pase) lost its diurnal oscillation (P = 0.5) within this group. The circadian rhythmicity of the master regulator, peroxisome proliferator-activated receptor gamma coactivator 1α (Pgc1a) expressed a statistically significant phase advance (peak time = 83.22 at 26°C vs 116.65 at 21°C). Additionally, the key regulators of glycolysis remained rhythmic in both temperature groups (P < 0.005). The circadian expression of glucose transporter 2 (Glut2) and glucokinase (Gck) showed a phase delay in the TN group (peak time = ZT22 and ZT8, respectively) (Table 4). On the other hand, the hepatic expression of the key genes for glycogenesis, such as the glycogen synthase (Gys2), was significantly induced in these mice. In contrast, the mRNA expression levels of protein phosphatase 1 regulatory subunit 3C (Ppp1r3c) and subunit 3B (Ppp1rc3b) were decreased (Fig. 4C and Table 4). However, the overall mRNA expression of key glycogenesis genes remained rhythmic in both temperature groups (P < 0.05), with a remarkable phase delay at 26°C (Table 4).
Housing temperatures affect the circadian rhythmicity of the hepatic metabolic genes. RT-qPCR analysis of mRNA expression levels of key metabolic genes involved in (A) gluconeogenesis, (B) glycolysis, (C) glycogenesis (D) fatty acid synthesis, and (E) fatty acid oxidation. Data are presented as mean ± s.e.m., n = 6 per time-point. *P < 0.05, **P < 0.005 vs TC.
Citation: Journal of Endocrinology 247, 2; 10.1530/JOE-20-0100
The expression of sterol regulatory element-binding protein 1c (Srebp1c), a gene controlling fatty acid synthesis and the TG homeostasis, was highly induced at 21°C, accompanied by a delay oscillation phase at TN (Fig. 4D and Table 4). On the contrary, the fatty acid synthase (Fasn) mRNA expression variated considerably during a day and showed a non-rhythmic expression pattern at TN (Fig. 4D, P = 0.23). Under TC-housed conditions, the expression of genes involved in the fatty acid oxidation, such as the peroxisome proliferator-activated receptor α (Ppara), was highly induced (Fig. 4E). Its expression was statistically rhythmic under both temperature groups (P < 0.005), and peaked at the same time-point (Table 4). However, the TN group exerted an extreme impact on the expression of carnitine palmitoyl transferase 1 alpha (Cpt1a) throughout a day (Fig. 4E) and dampened its rhythmicity (P = 0.22, MESOR = 0.44 vs 0.65 at 21°C).
With the exception of Ppp1r3c and Fasn, two-factor ANOVA analysis showed a statistical difference between groups, implying that involvement of other factors existed in the resetting of their circadian oscillation. Nevertheless, the collective results suggested the role of temperature in the hepatic metabolic rhythmicity. Detailed analysis reflecting the impacts of temperature, time and their interaction was presented in Supplementary Table 2.
Housing temperature affects the circadian rhythmicity of hepatic clock genes
The mRNA quantitative analysis showed that the temperature affected the expression levels of Pers and Crys family. Notably, the expression levels of Pers family were decreased, whereas the Crys family was increased in the TN group (Fig. 5C, D, E and F). In general, under TN, both the circadian expressions of Per1 and Per2 were phase delayed, and peaked at the same time-point (Per1 Acrophase = −3.41, P < 0.005; Peak time = ZT5; and Per2 Acrophase = −3.63, P < 0.005; Peak time = ZT5). In contrast, the circadian expression of Cry1 showed a similar phase under both temperatures (Peak time = ZT11), while Cry2 exhibited non-rhythmic under 26°C (P = 0.15, Table 4). Furthermore, two-factor ANOVA analysis demonstrated that temperature was an independent factor for the Cry2 expression. Besides, the time, temperature and their interaction were critical determining factors for the circadian expression of clock genes (Supplementary Table 2).
Housing temperatures affect the circadian rhythmicity of the hepatic clock genes. RT-qPCR analysis of mRNA expression levels of key clock genes including (A) Bmal1, (B) Clock (C) Per1, (D) Per2, (E) Cry1, (F) Cry2, (G) Rev-erbα, and (H) Dbp. Data are presented as mean ± s.e.m., n = 6 per time-point. *P < 0.05, **P < 0.005 vs TC.
Citation: Journal of Endocrinology 247, 2; 10.1530/JOE-20-0100
Housing temperature affects the liver histology and lipid contents
Given the physiological outputs of the tested metabolic gene expression, H&E and ORO staining were therefore performed at six time-points. The hepatic phenotype was similar between the two temperature groups and no obvious lipid droplets were noticed (Fig. 6A and B). The concentration levels of liver TC showed an antiphase pattern in both groups (Fig. 6C), while the hepatic level of TG significantly peaked at ZT13 under 26°C (Fig. 6D). Nevertheless, both liver TC and TG lost their rhythmicity under 21°C (P = 0.87 and P = 1, respectively) (Table 4). Two-factor ANOVA showed that the temperature was a determining factor for the liver TG contents, while the liver TC contents were neither affected by the temperature nor by the time, but were affected by their interactions (Supplementary Table 2). These results suggest that the temperature is a potential Zeitgeber for the liver TG, but not for the liver TC.
The housing temperatures affect liver histology and lipid contents. (A) H&E staining of liver sections collected at different time-points. (B) ORO staining. (C) Liver contents of TC. (D) Liver contents of TG. Data are presented as mean ± s.e.m., n = 6 for each time-point for the staining. For the lipid quantification, n = 6 per time-point. *P < 0.05, **P < 0.005 vs TC.
Citation: Journal of Endocrinology 247, 2; 10.1530/JOE-20-0100
Discussion
Environmental temperature is a critical parameter for a translational study for mice; as an increasing research demonstrates its effects on mouse physiological homeostasis. On the other hand, the circadian rhythms orchestrate several physiological and behavioral processes, thus allowing an adjusted anticipation and adaptation to the external challenges. Additionally, a previous research indicated that the chronic non-circadian environmental challenges could alter the circadian metabolism in a tissue-specific manner (Guan et al. 2018). The mouse metabolism is controlled by the internal circadian clock system, thus exhibiting a strong diurnal fluctuation (Bass & Takahashi 2010). However, the evaluation of the housing temperature’s actions on the circadian rhythmicity of the liver remains unexplored. Here, we described how the conventional TC and latest suggested TN (Keijer et al. 2019a,b) differentially regulated the mouse behavioral activity and energy metabolism, as well as the circadian rhythmicity of hepatic gene expression. Based on our findings, we concluded that the external environmental temperature entrains the rhythmicity of the food intake, serum metabolites, as well as the hepatic oscillations of metabolic and clock gene expression in mice.
As the general phenotypes, our results were in consistent with previous findings (Overton 2010, McKie et al. 2019) where mice at TN ate less compared to mice at TC, hence, their body weight remained identical. The changes in the rhythmicity of food intake at TC suggested that the feeding rhythmicity was temperature-induced, since the food consumption is largely governed by a cold-induced enhancement of food intake.
Interestingly, a recent paper from Raun et al. (2020) showed that the thermoneutral housing temperature blunted the mouse exercise-training and the metabolic fluctuations. In contrast, our result displayed an increasing exercise rate under the thermoneutral temperature, which was associated with a reduction of the metabolic fluctuation. We hypothesized that such a discrepancy might be caused by the rodent sexual differences within the two studies. In their research, Raun et al. have used female mice, whereas we used male mice. Effectively, it is known that female mice are more active compared to male mice (Konhilas et al. 2015). However, male mice are more susceptible to develop heat-induced hyperthermia compared to female mice (Chen & Yu 2018). Therefore, female mice are more active under lower temperatures to preserve the body temperature homeostasis. Besides, aligning with prior study (DeVallance et al. 2017), we suggested that the thermal stress (here, lower housing temperature) negatively impacts the mouse activity. Such a thermal stress hypothesis was verified by the increase of the NE under the thermoregulatory effects (Kozyreva et al. 2015). Additionally, Barnett and Dickson stipulated an ontogenetical adaptation of the mice by shivering and by the reduction of activity when exposed to a temperature below the thermoneutral zone (Barnett & Dickson 1989) to minimalize their energy expenditure (Hankenson et al. 2018, Serin & Acar 2019). Such a stipulation was in consistent with both ours and prior findings (McKie et al. 2019).
On the other hand, the variations in the mouse activity can be reset by the temperature considering that the locomotor activities were different at night (active state). Of course, we recognized that the circadian clock can be reset by more than one cues, such as light, food and temperature (Xie et al. 2019). Hence, the phenotypes we observed here might be the final outcomes of the systemic effects of activity, sleep, feeding, metabolic state, together with the effects of housing temperature.
Multiple studies demonstrated that the housing temperature affects the metabolic adaptations of the mice, leading to the unexpected metabolic diseases, such as nonalcoholic fatty liver disease (Giles et al. 2017, Hankenson et al. 2018, McKie et al. 2019). Specifically, according to Giles et al. (2017), the thermoneutral housing exacerbates nonalcoholic fatty liver disease in a sex-independent manner. It should be noted that the expression levels of lipid metabolic genes (Srebp1c, Fasn, Cpt1a, and Ppara) are reduced during the nonalcoholic Steatohepatitis (Nagaya et al. 2010). However, we have noticed an absence of hepatic lipid accumulation in the TN group. Therefore, we speculated that such an inconsistence between these results might probably due to the difference in the thermoneutral housing temperature selection (30°C in their settings vs 26°C in ours) and in the sacrifice time-point, since they neglected the influence of circadian clock. Nevertheless, the loss of Fasn rhythmicity at 26°C possibly contributed to the potential obesity phenotypes of mice when housed at the thermoneutral temperature.
Such a non-rhythmic expression might be due to the variation of gene expression during a 24-h period, which was mediated by nutritional status. As a key gene for de novo lipogenesis (DNL), Fasn expression is involved in the pathway for energy homeostasis, which is finely controlled by nutritional status and energy balance (Nguyen et al. 2008). Of note, the circadian expression of Cpt1a was similarly abolished in the mouse livers of TN group. Since Cpt1a is regulated by the Fasn product (Nguyen et al. 2008), its impaired oscillation pattern might be the consequence of the dampened rhythmicity of Fasn expression.
As mentioned previously, molecular clocks are reset by the external cues, while the peripheral oscillators might be adjusted by neural, humoral and food/temperature-induced signals (Mohawk et al. 2012, Chen et al. 2019). The circulating NE and epinephrine are potential signals from the master clock to entrain the peripheral organs (Bartness et al. 2001, Astiz et al. 2019). Effectively, aligning with the previous study (Terazono et al. 2003), Per1 expression was induced in the liver corresponding to NE stimulation. Given that NE is known to regulate the circadian rhythmicity of Cry2 in the pineal gland (Wongchitrat et al. 2009), we postulated that a similar regulation was existed within the liver.
In contrast, Per2 expression is regulated by multiple factors other than the Bmal1-Clock interaction, such as the glucocorticoids (Albrecht et al. 2007). It is known that the secretion of glucocorticoids is a common endocrine response to stress (Sapolsky et al. 2000). However, the thermal changes are generally associated with the secretion of glucocorticoids and are often perceived as a stressor in most animals (de Bruijn & Romero 2018). Importantly, a literature review showed an augmentation of glucocorticoids is released during an increasing temperature (de Bruijn & Romero 2018). Therefore, we suggested that under TN, the sustained elevation of glucocorticoids altered the rhythmicity of Per2 expression with a marked phase delay. Of note, Per2 is the central dogma linking the clock and metabolism in an interdependent fashion; it can regulate glycogen metabolism (Zani et al. 2013). Therefore, its phase shift potentially affected the gene expression involved in glycogenesis (mostly Ppp1r3c) at TN.
In our study, we noticed that hepatic G6Pase mRNA expression was reduced, accompanied by a dampened rhythmicity under the TN-housed temperature at ZT17. Since the Cry1 is a well-known clock gene that represses the hepatic gluconeogenesis (Jang et al. 2016), the temperature-induced Cry1 under 26°C may function as a potential clock mediator linking the housing temperature to the homeostasis of hepatic glucose metabolic rhythmicity.
The overall data suggest that housing temperature would affect the metabolic rhythmicity in mice, as well as the circadian expression of major genes involved in liver metabolism. Together with other studies (Hoevenaars et al. 2014, Cui et al. 2016, Tian et al. 2016, Giles et al. 2017, Hylander et al. 2017), our findings indicated that the metabolic diseases, or the apparent metabolic phenotypes, are induced by the interaction of temperature and nutritional signals. In this sense, it is of great interest to investigate the synergistic effects of temperature and high-fat diet feeding on the peripheral circadian clocks.
Conclusion
In conclusion, we demonstrated the non-reductant role of the temperature in the mouse internal rhythmicity resetting. Since the clock machinery controls almost all the mammalian physiological processes, the influence of circadian clock was overlooked in the past decades, which is actually critical for the evaluation of housing temperature. Ideally, mice should be housed in a temperature cycle, for example, cooler at night (active phase) and warmer in the day (rest phase). However, setting up this housing environment would be difficult due to the technique limitation. Future studies are thus required to precisely investigate the changes of circadian clock and consequent physiological events in rodents at various recommended housing temperatures.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/JOE-20-0100.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This work was supported by grants from the National Natural Science Foundation of China (31771298, 31800992), the Natural Science Foundation of Jiangsu Province (BK20180554), the Project of State Key Laboratory of Natural Medicines, China Pharmaceutical University (SKLNMZZ202005), the ‘Double First-Class’ University Project (CPU2018GY17), the Fundamental Research Funds for the Central Universities (2632018PY15), the Open Fund of State Key Laboratory of Pharmaceutical Biotechnology, Nanjing University, China (KF-GN-201901) and the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).
Author contribution statement
A R and C L designed research. A R and H L took care of the mice. A R performed research experiments. A R and S Z performed rhythmicity and statistical analysis. S C, X A and C L provided suggestions for analytic tools and critical input to the study. A R and C L analyzed the data. A R wrote the manuscript. All authors read and approved the final manuscript.
References
Albrecht U, Bordon A, Schmutz I & Ripperger J 2007 The multiple facets of Per2. Cold Spring Harbor Symposia on Quantitative Biology 72 . (https://doi.org/10.1101/sqb.2007.72.001)
Astiz M, Heyde I & Oster H 2019 Mechanisms of communication in the mammalian circadian timing system. International Journal of Molecular Sciences 20 343. (https://doi.org/10.3390/ijms20020343)
Balsalobre A, Damiola F & Schibler U 1998 A serum shock induces circadian gene expression in mammalian tissue culture cells. Cell 93 . (https://doi.org/10.1016/S0092-8674(0081199-X)
Barnett SA & Dickson RG 1989 Wild mice in the cold: some findings on adaptation. Biological Reviews of the Cambridge Philosophical Society 64 . (https://doi.org/10.1111/j.1469-185X.1989.tb00679.x)
Bartness TJ, Song CK & Demas GE 2001 SCN efferents to peripheral tissues: implications for biological rhythms. Journal of Biological Rhythms 16 . (https://doi.org/10.1177/074873040101600302)
Bass J & Takahashi JS 2010 Circadian integration of metabolism and energetics. Science 330 . (https://doi.org/10.1126/science.1195027)
Chen Y & Yu T 2018 Testosterone mediates hyperthermic response of mice to heat exposure. Life Sciences 214 . (https://doi.org/10.1016/j.lfs.2018.10.058)
Chen S, Feng M, Zhang S, Dong Z, Wang Y, Zhang W & Liu C 2019 Angptl8 mediates food-driven resetting of hepatic circadian clock in mice. Nature Communications 10 3518. (https://doi.org/10.1038/s41467-019-11513-1)
Cornelissen G 2014 Cosinor-based rhythmometry. Theoretical Biology and Medical Modelling 11 16. (https://doi.org/10.1186/1742-4682-11-16)
Cui X, Nguyen NLT, Zarebidaki E, Cao Q, Li F, Zha L, Bartness T, Shi H & Xue B 2016 Thermoneutrality decreases thermogenic program and promotes adiposity in high-fat diet-fed mice. Physiological Reports 4 e12799. (https://doi.org/10.14814/phy2.12799)
de Bruijn R & Romero LM 2018 The role of glucocorticoids in the vertebrate response to weather. General and Comparative Endocrinology 269 . (https://doi.org/10.1016/j.ygcen.2018.07.007)
DeVallance E, Riggs D, Jackson B, Parkulo T, Zaslau S, Chantler PD, Olfert IM & Bryner RW 2017 Effect of chronic stress on running wheel activity in mice. PLoS ONE 12 e0184829. (https://doi.org/10.1371/journal.pone.0184829)
Ferrell JM & Chiang JYL 2015 Circadian rhythms in liver metabolism and disease. Acta Pharmaceutica Sinica B 5 . (https://doi.org/10.1016/j.apsb.2015.01.003)
Fischer AW, Cannon B & Nedergaard J 2018 Optimal housing temperatures for mice to mimic the thermal environment of humans: an experimental study. Molecular Metabolism 7 . (https://doi.org/10.1016/j.molmet.2017.10.009)
Fonken LK, Aubrecht TG, Meléndez-Fernández OH, Weil ZM & Nelson RJ 2013 Dim light at night disrupts molecular circadian rhythms and increases body weight. Journal of Biological Rhythms 28 . (https://doi.org/10.1177/0748730413493862)
Giles DA, Moreno-fernandez ME, Stankiewicz TE, Graspeuntner S, Cappelletti M, Wu D, Mukherjee R, Chan CC, Lawson MJ, Klarquist J, et al. 2017 Thermoneutral housing exacerbates non-alcoholic fatty liver disease in mice and allows for sex-independent disease modeling. Nature Letters 23 . (https://doi.org/10.1038/nm.4346)
Gordon CJ 1993 Temperature Regulation in Laboratory Rodents. Cambridge, UK: Cambridge University Press. (https://doi.org/10.1017/CBO9780511565595)
Gordon CJ 2012 Thermal physiology of laboratory mice: defining thermoneutrality. Journal of Thermal Biology 37 . (https://doi.org/10.1016/j.jtherbio.2012.08.004)
Guan D, Xiong Y, Borck PC, Jang C, Doulias PT, Papazyan R, Fang B, Jiang C, Zhang Y, Briggs ER, et al. 2018 Diet-induced circadian enhancer remodeling synchronizes opposing hepatic lipid metabolic processes. Cell 174 831 .e12–842.e12. (https://doi.org/10.1016/j.cell.2018.06.031)
Hankenson FC, Marx JO, Gordon CJ & David JM 2018 Effects of rodent thermoregulation on animal models in the research environment. Comparative Medicine 68 . (https://doi.org/10.30802/AALAS-CM-18-000049)
Harmer SL, Panda S & Kay SA 2001 Molecular bases of circadian rhythms. Annual Review of Cell and Developmental Biology 17 . (https://doi.org/10.1146/annurev.cellbio.17.1.215)
Hoevenaars FPM, Bekkenkamp-grovenstein M, Janssen RJ, Heil SG, Bunschoten A, Hoek-Van Den Hil EF, Snaas-Alders S, Teerds K, van Schothorst EM, Keijer J, et al. 2014 Thermoneutrality results in prominent diet-induced body weight differences in C57BL/6J mice, not paralleled by diet-induced metabolic differences. Molecular Nutrition and Food Research 58 . (https://doi.org/10.1002/mnfr.201300285)
Hylander BL, Eng JW & Repasky EA 2017 The impact of housing temperature-induced chronic stress on preclinical mouse tumor models and therapeutic responses: an important role for the nervous system. Advances in Experimental Medicine and Biology 1036 . (https://doi.org/10.1007/978-3-319-67577-0_12)
Izumo M, Johnson CH & Yamazaki S 2003 Circadian gene expression in mammalian fibroblasts revealed by real-time luminescence reporting: temperature compensation and damping. PNAS 100 . (https://doi.org/10.1073/pnas.2536313100)
Jang H, Lee GY, Selby CP, Lee G, Jeon YG, Lee JH, Cheng KKY, Titchenell P, Birnbaum MJ, Xu A, et al. 2016 SREBP1c-CRY1 signalling represses hepatic glucose production by promoting FOXO1 degradation during refeeding. Nature Communications 7 12180. (https://doi.org/10.1038/ncomms12180)
Karp CL 2012 Unstressing intemperate models: how cold stress undermines mouse modeling. Journal of Experimental Medicine 209 . (https://doi.org/10.1084/jem.20120988)
Keijer J, Li M & Speakman JR 2019a What is the best housing temperature to translate mouse experiments to humans? Molecular Metabolism 25 . (https://doi.org/10.1016/j.molmet.2019.04.001)
Keijer J, Li M & Speakman JR 2019b To best mimic human thermal conditions, mice should be housed slightly below thermoneutrality. Molecular Metabolism 26 4. (https://doi.org/10.1016/j.molmet.2019.05.007)
Konhilas JP, Chen H, Luczak E, McKee LA, Regan J, Watson PA, Stauffer BL, Khalpey ZI, Mckinsey TA, Horn T, et al. 2015 Diet and sex modify exercise and cardiac adaptation in the mouse. American Journal of Physiology: Heart and Circulatory Physiology 308 H135–H145. (https://doi.org/10.1152/ajpheart.00532.2014)
Kozyreva TV, Meyta ES & Khramova GM 2015 Effect of the sympathetic nervous system co-transmitters ATP and norepinephrine on thermoregulatory response to cooling. Temperature 2 . (https://doi.org/10.1080/23328940.2014.1000705)
Lahiri K, Vallone D, Gondi SB, Santoriello C, Dickmeis T & Foulkes NS 2005 Temperature regulates transcription in the zebrafish circadian clock. PLoS Biology 3 e351. (https://doi.org/10.1371/journal.pbio.0030351)
Lowrey PL & Takahashi JS 2004 Mammalian circadian biology: elucidating genome-wide levels of temporal organization. Annual Review of Genomics and Human Genetics 5 . (https://doi.org/10.1146/annurev.genom.5.061903.175925)
Machado FSM, Zhang Z, Su Y, de Goede P, Jansen R, Foppen E, Coimbra CC & Kalsbeek A 2018 Time-of-day effects on metabolic and clock-related adjustments to cold. Frontiers in Endocrinology 9 199. (https://doi.org/10.3389/fendo.2018.00199)
Maloney SK, Fuller A, Mitchell D, Gordon C & Michael Overton JM 2014 Translating animal model research: does it matter that our rodents are cold? Physiology 29 . (https://doi.org/10.1152/physiol.00029.2014)
McKie GL, Medak KD, Knuth CM, Shamshoum H, Townsend LK, Peppler WT & Wright DC 2019 Housing temperature affects the acute and chronic metabolic adaptations to exercise in mice. Journal of Physiology 597 . (https://doi.org/10.1113/JP278221)
Mohawk JA, Green CB & Takahashi JS 2012 Central and peripheral circadian clocks in mammals. Annual Review of Neuroscience 35 . (https://doi.org/10.1146/annurev-neuro-060909-153128)
Nagaya T, Tanaka N, Suzuki T, Sano K, Horiuchi A, Komatsu M, Nakajima T, Nishizawa T, Joshita S, Umemura T, et al. 2010 Down-regulation of SREBP-1c is associated with the development of burned-out NASH. Journal of Hepatology 53 . (https://doi.org/10.1016/j.jhep.2010.04.033)
Nagoshi E, Saini C, Bauer C, Laroche T, Naef F & Schibler U 2004 Circadian gene expression in individual fibroblasts: cell-autonomous and self-sustained oscillators pass time to daughter cells. Cell 119 . (https://doi.org/10.1016/j.cell.2004.11.015)
Nguyen P, Leray V, Diez M, Serisier S, Le Bloc’h J, Siliart B & Dumon H 2008 Liver lipid metabolism. Journal of Animal Physiology and Animal Nutrition 92 . (https://doi.org/10.1111/j.1439-0396.2007.00752.x)
Overton JM 2010 Phenotyping small animals as models for the human metabolic syndrome: thermoneutrality matters. International Journal of Obesity 34 (Supplement 2) S53–S58. (https://doi.org/10.1038/ijo.2010.240)
Pittendrich CS 1960 Circadian rhythms and the circadian organization of living systems. Cold Spring Harbor Symposia on Quantitative Biology 25 . (https://doi.org/10.1101/SQB.1960.025.01.015)
Preitner N, Damiola F, Lopez-Molina L, Zakany J, Duboule D, Albrecht U & Schibler U 2002 The orphan nuclear receptor REV-ERBα controls circadian transcription within the positive limb of the mammalian circadian oscillator. Cell 110 . (https://doi.org/10.1016/S0092-8674(0200825-5)
Prolo LM, Takahashi JS & Herzog ED 2005 Circadian rhythm generation and entrainment in astrocytes. Journal of Neuroscience 25 . (https://doi.org/10.1523/JNEUROSCI.4133-04.2005)
Raun SH, Henriquez-Olguín C, Karavaeva I, Ali M, Møller LLV, Kot W, Castro-Mejía JL, Nielsen DS, Gerhart-Hines Z, Richter EA, et al. 2020 Housing temperature influences exercise training adaptations in mice. Nature Communications 11 1560. (https://doi.org/10.1038/s41467-020-15311-y)
Refinetti R 2010 Entrainment of circadian rhythm by ambient temperature cycles in mice. Journal of Biological Rhythms 25 . (https://doi.org/10.1177/0748730410372074)
Reinke H & Asher G 2016 Circadian clock control of liver metabolic functions. Gastroenterology 150 . (https://doi.org/10.1053/j.gastro.2015.11.043)
Reitman ML 2018 Of mice and men – environmental temperature, body temperature, and treatment of obesity. FEBS Letters 592 . (https://doi.org/10.1002/1873-3468.13070)
Reppert SM & Weaver DR 2001 Molecular analysis of mammalian circadian rhythms. Annual Review of Physiology 63 . (https://doi.org/10.1146/annurev.physiol.63.1.647)
Sapolsky RM, Romero LM & Munck AU 2000 How do glucocorticoids influence stress responses? Integrating permissive, suppressive, stimulatory, and preparative actions. Endocrine Reviews 21 . (https://doi.org/10.1210/edrv.21.1.0389)
Serin Y & Acar N 2019 Effect of circadian rhythm on metabolic processes and the regulation of energy balance. Annals of Nutrition and Metabolism 74 . (https://doi.org/10.1159/000500071)
Sokolove PG & Bushell WN 1978 The chi square periodogram: its utility for analysis of circadian rhythms. Journal of Theoretical Biology 72 . (https://doi.org/10.1016/0022-5193(7890022-x)
Terazono H, Mutoh T, Yamaguchi S, Kobayashi M, Akiyama M, Udo R, Ohdo S, Okamura H & Shibata S 2003 Adrenergic regulation of clock gene expression in mouse liver. PNAS 100 . (https://doi.org/10.1073/pnas.0936797100)
Tian XY, Ganeshan K, Hong C, Nguyen KD, Qiu Y, Kim J, Tangirala RK, Tontonoz P & Chawla A 2016 Thermoneutral housing accelerates metabolic inflammation to potentiate atherosclerosis but not insulin resistance. Cell Metabolism 23 . (https://doi.org/10.1016/j.cmet.2015.10.003)
Welsh DK, Logothetis DE, Meister M & Reppert SM 1995 Individual neurons dissociated from rat suprachiasmatic nucleus express independently phased circadian firing rhythms. Neuron 14 . (https://doi.org/10.1016/0896-6273(9590214-7)
Welsh DK, Yoo SH, Liu AC, Takahashi JS & Kay SA 2004 Bioluminescence imaging of individual fibroblasts reveals persistent, independently phased circadian rhythms of clock gene expression. Current Biology 14 . (https://doi.org/10.1016/j.cub.2004.11.057)
Wongchitrat P, Felder-Schmittbuhl MP, Phansuwan-Pujito P, Pévet P & Simonneaux V 2009 Endogenous rhythmicity of Bmal1 and Rev-erb α in the hamster pineal gland is not driven by norepinephrine. European Journal of Neuroscience 29 . (https://doi.org/10.1111/j.1460-9568.2009.06742.x)
Xie Y, Tang Q, Chen G, Xie M, Yu S, Zhao J & Chen L 2019 New insights into the circadian rhythm and its related diseases. Frontiers in Physiology 10 682. (https://doi.org/10.3389/fphys.2019.00682)
Yoo SH, Yamazaki S, Lowrey PL, Shimomura K, Ko CH, Buhr ED, Siepka SM, Hong HK, Oh WJ, Yoo OJ, et al. 2004 PERIOD2::luciferase real-time reporting of circadian dynamics reveals persistent circadian oscillations in mouse peripheral tissues. PNAS 101 . (https://doi.org/10.1073/pnas.0308709101)
Zani F, Breasson L, Becattini B, Vukolic A, Montani JP, Albrecht U, Provenzani A, Ripperger JA & Solinas G 2013 PER2 promotes glucose storage to liver glycogen during feeding and acute fasting by inducing Gys2 PTG and GL expression. Molecular Metabolism 2 . (https://doi.org/10.1016/j.molmet.2013.06.006)
