4’-demethylnobiletin-rich fermented Citrus reticulata (ponkan) attenuated the disturbance in clock gene expression and locomotor activity rhythms caused by high-fat diet feeding

ABSTRACT

Demethylnobiletins are nobiletin metabolites, with 4′-demethylnobiletin being the most abundant in the demethylated compounds upon feeding with nobiletin. Previous studies have repeatedly shown that demethylnobiletin has higher physiological activity than nobiletin; however, no studies have yet suggested that 4′-demethylnobiletin could affect biological rhythms. Therefore, in the present study, we analyzed the effects of fermented Citrus reticulata, containing a large amount of 4′-demethylnobiletin, on the locomotor activity level, locomotor activity rhythms under normal light-dark and constant-dark conditions, and peripheral clock rhythms. We demonstrated that fermented Citrus reticulata not only modified the changes in the locomotor activity rhythms caused by high-fat diet feeding in mice but also inhibited the phase shift of the peripheral clock rhythms caused by high-fat diet feeding. Taken together, the results of this study suggest that fermented Citrus reticulata could potentially be a new functional food that could help to maintain normal biological rhythms.

KEYWORDS:

• Chrono-nutrition
• functional food
• high-fat diet
• locomotor activity
• peripheral clocks

Introduction

The circadian clock system is an inherent mechanism in almost all species. In mammals, this system is composed of several clock genes, such as period 1/2 (Per1/2), cryptochrome 1/2 (Cry1/2), brain and muscle arnt-like protein 1 (Bmal1), and circadian locomotor output cycles kaput (Clock) (Buhr and Takahashi 2013). These clock genes generate a rhythmic cycle of approximately 24 h by creating a transcriptional and translational negative feedback loop (Reppert and Weaver 2002). This system has central and peripheral clocks (Hastings 1997; Weaver 1998). The central clock, which dominates in the suprachiasmatic nucleus (SCN) in the hypothalamus, is entrained by the environmental light-dark cycle (Nelson and Zucker 1981). Subsequently, the central clock synchronizes the peripheral clocks, located in almost all peripheral organs, through neural signals, hormonal signals, and other pathways (Albrecht 2012; Bass and Takahashi 2010). Peripheral clocks can be organized not only by the central clock, but also by environmental stimuli, such as feeding, temperature, and stress (Hara et al. 2001; Ohnishi et al. 2014; Tahara et al. 2015). The central and peripheral clocks play an important role in modulating physiological functions, metabolism, body temperature, and sleep-wake behavior (Gachon et al. 2004). Accordingly, we need to control not only the timing of light exposure but also the timing of environmental stimulation in order to maintain healthy physiological functions and a well-balanced circadian clock.

Circadian clocks, especially the peripheral clocks, are affected not only by meal timing but also by some nutrients. Nobiletin (Nob) is a nutrient that affects clock gene expression rhythms. Many previous studies have indicated that Nob prolongs the period length and shifts the phases of PER2 expression rhythm in mouse embryonic fibroblasts (MEFs) (Shinozaki et al. 2017), and that Nob prevents a decrease in locomotor activity levels caused by high-fat diet (HFD) feeding in mice (He et al. 2016). Previous studies have suggested that these effects are mediated through the inhibition of adenosine receptors, the increase in intracellular cyclic adenosine monophosphate (cAMP) by non-selective phosphodiesterase inhibition, and other signaling pathways, such as the retinoic acid receptor-related orphan receptor (ROR) nuclear receptor-related and extracellular signal-regulated kinase (ERK)/cAMP response element-binding protein (CREB) pathways (Kawahata et al. 2013; He et al. 2016). Taken together, multiple pieces of evidence suggest that some nutrients affect circadian clocks by themselves.

Only a few studies have shown that Nob can negatively impact health. Almost all Nob is transformed into demethylnobiletin, especially 4′-demethylnobiletin (4DMNob) (Wu et al. 2015a, Yasuda et al. 2003). Indeed, in rat colonic mucosa, 4DMNob is the most abundant demethylated metabolite upon Nob feeding (Wu et al. 2015b). Moreover, many studies have demonstrated that in comparison with Nob, demethylnobiletin has a more relevant contribution to various processes, such as anti-inflammatory and anti-cancer mechanisms, and carbohydrate or lipid metabolism (Lo et al. 2010; Wu et al. 2018, Wu et al. 2015a). Taken together, these studies suggest that fermented Citrus reticulata (FCR), a functional food containing increased amounts of 4DMNob, would have stronger effects on certain physiological functions than Nob.

To the best of our knowledge, no studies to date have demonstrated the impact of 4DMNob and FCR on circadian clocks. Shinozaki et al. suggested that not only does Nob affect PER2 expression rhythm but also that the effects of flavonoids with similar chemical structures do not always similarly affect the circadian clocks (Kawahata et al. 2013; He et al. 2016; Shinozaki et al. 2017). Therefore, 4DMNob and FCR might not always affect the circadian clock in the same manner as Nob. In the current study, we investigated the effect of 4DMNob and FCR on circadian clocks using both in vitro and in vivo experimental approaches.

Materials and methods

Nob, 4DMNob, and FCR

For the in vitro experiments, 4DMNob was dissolved in 0.1% dimethyl sulfoxide (DMSO; Wako Pure Chemical Industries, Ltd., Japan). For the in vivo experiments, FCR, containing a large amount of 4DMNob and little Nob, was blended with HFD (D12451M; RESEARCH DIETS Inc., Japan) (Kawahata et al. 2017). FCR preparation has been described previously (Kawahata et al. 2017). 4DMNob and FCR were provided by Fuji Sangyo Co., Ltd. (Japan).

Cell culture

MEFs were derived from PERIOD2::LUCIFERASE (PER2::LUC) knock-in mice (Yoo et al. 2004). MEFs were cultured in Dulbecco’s modified Eagle’s medium (D-MEM, Wako Pure Chemical Industries, Ltd.) supplemented with 10% FBS (Bio West, USA) with heat-inactivation and 1% penicillin/streptomycin (Wako Pure Chemical Industries, Ltd.), and were passaged twice a week. MEFs were incubated at 37°C under a 5% CO₂ atmosphere.

PER2::LUC MEF bioluminescence measurement

The rhythmic expression of PER2::LUC in MEFs was measured using a real-time LUCIFERASE assay. MEFs were passaged at a concentration of 6 × 10⁴ cells onto 35-mm dishes (AGC Techno Glass Co. Ltd., Japan). The following day, MEFs were stimulated with 100 nM dexamethasone (Sigma-Aldrich Corp, USA) for 2 h to synchronize the clock gene expression rhythm between cells, before being replaced with D-MEM as mentioned above, containing 0.1 mM D-luciferin sodium salt (Invitrogen, USA). MEFs were incubated at 37°C, and bioluminescence was monitored for 1 min at each 10-min interval with a dish-type luminometer (LumiCycle 32, Actimetrics, USA). For chronic treatment, MEFs were treated with the vehicle (0.1% DMSO) or each nutrient after dexamethasone stimulation, and bioluminescence was monitored for 6 days.

Circadian rhythm evaluation in MEFs

Original data were smoothed by an adjusting-averaging method using 2-h running means, as previously described (Hayasaka et al. 2007; Ohta et al. 2008). Then, the data set was de-trended by subtracting the 24-h running average from the raw data using R software (http://www.r-project.org/). Bioluminescence was measured for six consecutive days. The first, second, and third peaks were defined as peaks 1, 2, and 3, respectively (Figure 1A). The amplitude was calculated based on peak 1 bioluminescence. During the MEF bioluminescence measurements, the period length was calculated as half the interval between peaks 1 and 3 of the waveform.

Figure 1. PER2::LUC expression rhythm effects on MEFs. (A) Representative de-trended data for cells treated with 0.1% DMSO as a vehicle or with 4DMNob (1–30 μM). (B) Effect of each treatment on the period length of the rhythmic PER2::LUC expression. (C) Effect of each treatment on PER2::LUC bioluminescence amplitude of peak 1. Data are presented as means ± SEM (n = 4). * p < 0.05, versus vehicle (one-way ANOVA with Tukey’s multiple comparison test). # p < 0.05, ## p < 0.01 (Kruskal–Wallis test with Dunn’s multiple comparison test).

Animals

Eight-week-old male ICR mice and heterozygous PER2::LUC mice (background: ICR mice) were housed in an animal room maintained at a temperature of 22 ± 2°C, a humidity level of 60 ± 5%, and under a 12-h light/dark cycle (LD condition; lights on from 08:00 to 20:00). Zeitgeber time 0 (ZT0) and 12 were defined as the lights-on and light-off times, respectively. PER2::LUC mice were bred in our laboratory and housed in an animal room as previously described (Tahara et al. 2012). They were provided with an EF mouse chow (Oriental Yeast Co. Ltd., Japan) – mouse food whose exact composition is unknown – and water was provided ad libitum before the experiments. During the experiments, mice were provided with HFD or AIN-93 M (Oriental yeast Co., Ltd., Japan) – mouse food with known exact composition – and water was provided ad libitum before the experiments. Experimental animal care was conducted with the permission of the Animal Welfare Committee of Waseda University (2018-A026 and 2019-A056).

Locomotor activity rhythm analysis

Locomotor activity was monitored using a SE-10 infrared radiation sensor (Akizuki Denshi Tsusho Co. Ltd., Japan) and analyzed using CLOCKLAB software (Actimetrics), as previously described (Haraguchi et al. 2018). Using the CLOCKLAB software, the locomotor activity level was analyzed in each minute while mice were kept under LD conditions, and the free-running period length, which is regarded as an endogenous rhythm, was calculated while mice were kept under a constant dark (DD) condition. In the locomotor activity rhythm under LD conditions, original data (1-min bins) were smoothed by an adjusting-averaging method with 2-h running means. The free-running period length under DD conditions was calculated using the chi-squared periodogram (Sokolove and Bushell 1978).

In vivo monitoring system protocol and data analysis

The experimental procedure and data analysis of the in vivo monitoring system were performed as described previously (Haraguchi et al. 2018). In short, PER2::LUC mice were anesthetized with isoflurane (Mylan Inc., Japan) using an in vivo imaging system (IVIS, PerkinElmer, Inc., USA). Mice were anesthetized before being injected subcutaneously on the back and near the neck with D-luciferin potassium salt (Promega, USA) at a concentration of 15 mg/kg body weight. Images were taken at 8 and 10 min post-injection in the dorsal-up position for the kidney and the ventral-up position for the liver and the submandibular gland (sub gla), respectively. The imaging exposure time was 1 min. Images from the same mice were taken at 6 consecutive time points per day at 4-h intervals. Mice were recovered quickly from isoflurane anesthesia, following the imaging session, and returned to their home cages. Bioluminescence emitted by each organ was calculated automatically using Living Image 3.2 software (PerkinElmer, Inc.). For each organ in each mouse, the average photon/min level from 6 consecutive time points was calculated as 100%, and the bioluminescent rhythm was expressed as a percentage. The acrophase of the bioluminescent rhythm was analyzed using the single cosinor procedure program (Acro.exe, 3.5) (Refinetti et al. 2007). These interventions did not affect peripheral clocks in a previous study (Tahara et al. 2012).

Data analysis

All values are expressed as the mean ± standard error of the mean (SEM). Statistical analyses were performed using the GraphPad Prism version 6.03 software (USA). We verified whether the data showed a normal or non-normal distribution and equal or biased variation, by assessing with the D’Agostino-Pearson normality test/one-sample t-test and F-value test/Bartlett’s test, respectively. The parametric analysis was performed using one-way analysis of variance (ANOVA) with Tukey’s multiple comparison test, and non-parametric analysis was conducted using the Mann Whitney test, t-test with false discovery rate (FDR) multiple testing correction, or Kruskal-Wallis test with Dunn’s multiple comparison test.

Results

4DMNob strongly affects the PER2::LUC expression rhythm in MEFs

To analyze the effects of 4DMNob on the PER2::LUC expression rhythm, MEFs were treated with the vehicle (0.1% DMSO) or 4DMNob (1–30 μM) (Figure 1A). Compared with the vehicle group, the 4DMNob (10–30 μM) group extended the PER2::LUC expression rhythm circadian period. There were significant differences between the vehicle group and the 4DMNob (20 or 30 μM) groups (Figure 1B). Moreover, these results suggest that 4DMNob has a stronger prolongation effect on the PER2::LUC expression rhythm period length compared with the prolongation effect of Nob in a previous study. The respective previous study showed similar PER2::LUC expression rhythm circadian period of the vehicle and the 30 μM Nob groups, and found that 100 μM of Nob significantly extended the PER2::LUC expression rhythm circadian period compared with that observed in the vehicle group (Shinozaki et al. 2017). The PER2::LUC expression rhythm amplitudes in the 4DMNob group were higher than those in the vehicle group. There were significant differences between the vehicle group and the 4DMNob (3 or 10 μM) groups (Figure 1C). These results indicate that 4DMNob has increasing and strong prolongation effects on the PER2::LUC expression rhythm and that this effect of 4DMNob is stronger than that of Nob. Therefore, in the following in vivo experiments, we addressed the effects of FCR, containing a large amount of 4DMNob, on biological rhythms, such as the locomotor activity rhythm and peripheral clock rhythms.

FCR attenuates locomotor activity rhythm disturbance caused by HFD feeding

To confirm whether FCR affects locomotor activity rhythms, mice were fed with HFD containing 0.257% FCR, which corresponded to 0.06% 4DMNob, and the locomotor activity levels were monitored under LD conditions. In this experiment, mice were acclimated to HFD feeding for the first week and then divided into HFD and HFD+FCR groups, in which they were fed HFD or 0.257% FCR-supplemented HFD, respectively. Mice were kept under each food condition for 1 week under LD conditions and then released under DD conditions for the next 4 weeks (Figure 2A). First, we analyzed locomotor activity levels under LD conditions. The total activity levels in the HFD+FCR group tended to be higher than those in the HFD group throughout the day (Figure 2C), especially at the beginning of the active period (Figure 2B), and the percentage of the activity levels during the active period in the HFD+FCR group tended to be lower than that in the HFD group (Figure 2D). In addition, we examined the effects of FCR on mice fed a normal diet (ND; AIN-93 M) under the same experimental schedule (Supplemental Figure S1). The ND+FCR group showed similar locomotor activity rhythms, total activity levels, and percentage of activity levels during the active period compared to those observed in the ND group (Supplemental Figure S1A-1D). The HFD+FCR group showed higher activity levels at the beginning of the active period than those observed in the HFD group (Supplemental Figure S1B), consistent with the results shown in Figure 2B. The total activity levels in the HFD+FCR group were similar to those in the HFD group, and the percentages of activity levels during the active period in the HFD+FCR group were higher than those in the HFD group (Supplemental Figure S1C and 1D). However, there were no significant differences between the HFD and HFD+FCR groups (Supplemental Figure S1C and 1D). Moreover, in the additional experimet, the tolal locomotor activity levels in the ND group were higher than those in the HFD group and the percentage of the activity levels during the active period in the ND group were higher than those in the HFD group (Supplemental Figure S1C and 1D). Taken together, our first experiment showed the preventive effects of FCR on the decrease in the total locomotor activity levels caused by HFD feeding (Figure 2); an additional experiment showed the preventive effects of FCR on the decrease in the percentage of the activity levels during the active period caused by HFD feeding (Supplemental Figure S1). In the first experiment, the percentage of activity levels during the active period in the HFD group was approximately 90%, and such high values may be unaffected by HFD feeding (Figure 2). These results indicated that FCR modified the locomotor activity under HFD feeding.

Figure 2. The effects of FCR on the locomotor activity levels and the free-running period length of the locomotor activity rhythm. (A) Representative double-plotted actograms for mice fed with HFD (the HFD group) or with HFD supplemented with 0.257% FCR (the HFD+FCR group). The horizontal lines indicate the local time and the vertical lines indicate the passing day. The gray shadow on the actograms indicates the dark period. (B) The locomotor activity rhythm under LD condition. We averaged the locomotor activity rhythm for 7 days. Open and closed bars indicate light and dark periods, respectively. (C) The average of total locomotor activity levels for 7 days in both groups under LD condition. (D) The average percentage of activity levels during the active period under LD condition. (E) The free-running period of locomotor activity rhythm was analyzed for all days during DD condition. Data are presented as means ± SEM (n = 5). * p < 0.05 vs the HFD group (t-test with FDR multiple testing correction). $ p < 0.05 (Mann Whitney test).

In the next experiment, we analyzed the free-running period length of the locomotor activity rhythm under DD conditions (Figure 2A). The free-running period length in the HFD+FCR group was significantly longer than that in the HFD group (Figure 2D). We also examined the effects of FCR on mice fed with ND under the same experimental schedule (Supplemental Figure S1). The ND+FCR group showed a free-running period length similar to that of the ND group (Supplemental Figure S1E). In the additional experiment, the HFD+FCR group showed a longer free-running period length than the HFD group (Supplemental Figure S1E), consistent with the results shown in Figure 2E, but there wes no significant difference by two-way ANOVA with Tukey’s multiple comparison test. On the other hand, the free-running period length in the HFD+FCR group was similar to that of the ND and ND+FCR groups. These results indicate that FCR has a preventive effect on the changing of the free-running period length caused by HFD feeding.

We monitored food intake and body weight in an additional experiment (Supplemental Figure S2). The ND+FCR group consumed more food than the ND group, and the HFD+FCR group consumed the amount of food similar to the HFD group (Supplemental Figure S2A). The ND+FCR group showed body weight similar to that of the ND group, and the HFD+FCR group showed body weight similar to that of the HFD group (Supplemental Figure S2B). Taken together, these results demonstrate that FCR would prevent the disturbance of locomotor activity rhythms caused by HFD feeding, regardless of the effects of FCR on food intake and body weight.

FCR attenuates peripheral clock impairment caused by HFD feeding

To analyze whether FCR affects the peripheral clocks, PER2::LUC mice were fed with 0.257% FCR-supplemented HFD under LD conditions, and the PER2::LUC expression rhythms in peripheral organs were monitored using IVIS (Figure 3). We included an ND group in this experiment, in which mice of the HFD and HFD+FCR groups were fed with AIN-93 M. After 2 weeks under each food condition, the peripheral clocks were monitored using IVIS. HFD feeding reduced the PER2::LUC expression rhythm amplitude compared with ND feeding, while FCR inhibited the HFD feeding-mediated amplitude decrease only in the kidneys. However, no significant differences were observed (Figure 3A-3H). Regarding the acrophase, the HFD group delayed the PER2::LUC expression rhythm phases in the peripheral organs, and there were significant differences between the ND and HFD groups in the kidney and sub gla. Interestingly, both the HFD+FCR and ND groups showed a similar acrophase in the peripheral organs, and there were significant differences in the kidney and sub gla between the HFD and HFD+FCR groups (Figure 3I-3K). Previous studies have suggested that HFD feeding reduces the amplitude and shifts the peripheral clock acrophase compared to ND feeding (Kohsaka et al. 2007; Pendergast et al. 2013). These results demonstrate that FCR attenuates the peripheral clock impairment caused by HFD feeding.

Figure 3. The effects of FCR on PER2::LUC expression rhythm in peripheral organs. Representative images (A and B), analyzed waveforms (C–E), amplitudes (F–H), and peak phases (I–K) of in vivo PER2::LUC bioluminescence. Open and closed bars indicate light and dark periods, respectively. Data are presented as the mean ± SEM (the ND and HFD+FCR groups, n = 6; the HFD group, n = 5). * p < 0.05, ** p < 0.01 (one-way ANOVA with Tukey’s multiple comparison test).

Discussion

Our present study demonstrated that 4DMNob increased the PER2::LUC expression rhythm amplitude and prolonged its period in MEFs. Moreover, we showed that FCR modified HFD feeding-induced locomotor activity rhythm disruption and prevented phase delay of the peripheral clock caused by HFD feeding. Overall, this study suggests that FCR can be a new functional food that contributes to the maintenance of a normal biological rhythm and that 4DMNob is a key FCR effector.

We revealed that 4DMNob increased the PER2::LUC expression rhythm amplitude and period length, and that FCR prevented the HFD feeding-induced disturbance of locomotor activity rhythms. A previous study with MEFs showed that Nob significantly increased the PER2::LUC expression rhythm amplitude and extended its period length compared with the vehicle (Shinozaki et al. 2017). Moreover, previous studies on mice have shown that ND with Nob prevents the locomotor activity level decrease induced by aging and that HFD with Nob prevents the same induced by HFD feeding (He et al. 2016; Nohara et al. 2019). These studies suggest that such effects can be obtained through the action of Nob as an ROR nuclear receptor agonist (He et al. 2016; Nohara et al. 2019). ROR is one of the clock genes and has been identified as a Bmal1 transcription activator, leading to the enhancement of Per and Cry transcription. The effects of Nob on locomotor activity levels could be induced by the existence of circadian clocks, as Nob had no protective effects on the locomotor activity level decrease caused by HFD feeding in Clock^Δ19/Δ19 mutant mice, which is regarded as a model of circadian clock disruption (He et al. 2016). Taken together, these results suggest that the effects of 4DMNob on the PER2 expression rhythm and the effects of FCR on locomotor activity rhythm may be mediated through the stimulation of the ROR nuclear receptor.

We also found that FCR modified the HFD feeding-induced locomotor activity level decrease, especially at the beginning of the active period. Previous studies have shown that Nob and 4DMNob stimulate ERK/CREB signaling to increase Cre-mediated transcription levels in neuronal cell lines and hippocampal neurons in primary cultures (Nagase et al. 2005; Al Rahim et al. 2009). Indeed, synRas mice, in which endogenous Ras activity continues to be regulated and ERK phosphorylation level is upregulated, were found to be more sensitive to light than wild-type mice, because synRas mice showed higher phosphorylation levels of CREB in the SCN induced by light pulses than wild-type mice, especially during the early subjective night period (Serchov et al. 2016). Moreover, CBP^KIX/KIX mice that do not show stimulus-induced activation of CREB, showed lower locomotor activity levels at the beginning of the active period (Chatterjee et al. 2020). Taken together, these results suggest that FCR modified the HFD feeding-induced locomotor activity level decrease at of the time LD switching via the activation of the ERK/CREB signaling.

In this study, we also showed that FCR prevented the HFD-induced impairment of circadian clocks coupled with low amplitude and abnormal phase. As mentioned above, the FCR effect may be explained by the 4DMNob effects on the ROR nuclear receptors and ERK/CREB signaling pathways. In addition to these effects, previous studies indicated that Nob and 4DMNob also have antioxidative effects. Nob reduces intracellular reactive oxygen species (ROS) levels in C2C12 myoblast cells treated with the oxidant tert-butyl hydroperoxide (Nohara et al. 2019), while 4DMNob induces anti-oxidative enzymes, such as heme oxygenase-1 and NADH quinone oxidoreductase 1, in RAW 264.7 macrophages treated with lipopolysaccharide (Wu et al. 2015a). HFD feeding induces oxidative stress, which prolongs the clock gene expression rhythm in PER2::LUC MEFs (Hsu and Yen 2007; Imamura et al. 2018); however, in this experiment, we did not observe any prolongation of the free-running period induced by HFD feeding. Therefore, the findings of the present and the related past studies taken together suggest that FCR may be a new potential functional food, which improves the HFD feeding-induced disruption of the circadian clock system through not only the ROR nuclear receptor and ERK/CREB signaling pathway agonists, but also through the enhancement of anti-oxidative effects.

We found that FCR modified the HFD-induced decrease in locomotor activity level and prevented the HFD-induced impairment of peripheral circadian clocks coupled with a phase delay. The circadian clock system closely interacts with the feeding rhythm and regulates it (Kalsbeek et al. 2001); feeding is the strongest entrainment factor for peripheral clocks (Tahara and Shibata 2016). Previous studies have demonstrated that male mice and ovariectomized mice fed an HFD for 1 week showed dispersed eating behavior, locomotor activity rhythm, and advanced phase of liver molecular rhythm (Pendergast et al. 2013; Palmisano et al. 2017). Other studies found that mice fed with HFD under a time-restricted feeding regimen showed increased activity toward the end of the active period and improved oscillations of circadian clock components in the liver compared with mice fed with HFD under an ad libitum regimen (Hatori et al. 2012). Taken together, these results suggest that the disruption of locomotor activity rhythms interacts with the phase shift of peripheral clocks. However, it is difficult to determine whether the phase shift of the peripheral clock by FCR supplementation is associated with activity rhythm changes or not.

In conclusion, our study revealed that FCR containing a substantial amount of 4DMNob may be a new functional food that maintains healthy biological rhythms, such as clock gene expression rhythm and locomotor activity rhythm. Indeed, FCR modified HFD feeding-induced locomotor activity rhythm disturbance and peripheral clock rhythm delay. However, FCR contains substances other than 4DMNob (Kawahata et al. 2017) which could also contribute to its effects; however, the analysis of the same is beyond the scope of this study. Regardless, FCR could still be expected to improve the biological rhythm disturbances caused by aging and those caused by lifestyle-related factors such as work shifts or jetlag.

Acknowledgments

This work was partially supported by a Grant-in-Aid for Early-Career Scientists (19K14018) from JSPS (A.H.), the Council for Science, Technology and Innovation, SIP, “Technologies for creating next-generation agriculture, forestry and fisheries” (funding agency: Bio-oriented Technology Research Advancement Institution, NARO) (S.S.), and a Grant-in-Aid for Scientific Research (S) (26220201) from the Ministry of Education, Culture, Sports, Science and Technology of Japan (S.S.).

Disclosure statement

Atsushi Haraguchi, Shuhei Sato, Tomohiro Yamazaki, Con Ryan, Masataka Sekiguchi, and Shigenobu Shibata: Declarations of interest: None. Shuichi Kusano is an employee of Fuji Sangyo Co., Ltd. The authors declare no other conflicts of interest associated with this manuscript.

Supplementary material

Supplemental data for this article can be accessed here

Additional information

Funding

This work was supported by the Council for Science, Technology and Innovation, SIP, “Technologies for creating next-generation agriculture, forestry and fisheries” (funding agency: Bio-oriented Technology Research Advancement Institution, NARO); a Grant-in-Aid for Scientific Research (S) from the Ministry of Education, Culture, Sports, Science and Technology of Japan [26220201]; Grant-in-Aid for Early-Career Scientists from JSPS [19K14018].