Dietary proanthocyanidins modulate BMAL1 acetylation, Nampt expression and NAD levels in rat liver

Metabolism follows circadian rhythms, which are driven by peripheral clocks. Clock genes in the liver are entrained by daytime meals and food components. Proanthocyanidins (PAs), the most abundant flavonoids in the human diet, modulate lipid and glucose metabolism. The aim of this study was to determine whether PAs could adjust the clock system in the liver. Male Wistar rats were orally gavaged with 250 mg grape seed proanthocyanidin extract (GSPE)/kg body weight at zeitgeber time (ZT) 0 (light turned on), at ZT12 (light turned off), or before a 6 hour jet-lag and sacrificed at different times. The 24 hour rhythm of clock-core and clock-controlled gene expression indicated that nicotinamide phosphoribosyltransferase (Nampt) was the most sensitive gene to GSPE. However, Nampt was repressed or overexpressed after GSPE administration at ZT0 or ZT12, respectively. NAD levels, which are controlled by Nampt and also exhibit circadian rhythm, decreased or increased according to Nampt expression. Moreover, the ratio of acetylated Bmal1, that directly drives Nampt expression, only increased when GSPE was administered at ZT12. Therefore, GSPE modulated the clock system in the liver, suggesting that PAs can regulate lipid and glucose metabolism by adjusting the circadian rhythm in the liver.


Results
Acute administration of GSPE only modulated the expression pattern of clock-core and clock-controlled genes in the liver when was administered at ZT12. The capacity of GSPE to modify the molecular clock in the liver was evaluated by measuring the 24 h mRNA oscillation of Clock and Bmal1 (clock core genes), Per2 (a component of the negative loop of the circadian clock), Rorα and Rev-erbα (nuclear receptors whose expression is regulated by CLOCK:BMAL1 and which act as an activator or repressor, respectively, of Bmal1 gene expression), Nampt (a metabolic gene whose expression is directly regulated by CLOCK:BMAL1) and HmgCoAR (a metabolic gene that has circadian rhythm expression but that is not directly controlled by CLOCK:BMAL1).
Rats were administered GSPE at ZT0 and sacrificed at ZT0, ZT0.5, ZT1, ZT3, ZT6, ZT12 or ZT24. Nonetheless, to achieve a better visualization of the changes induced by GSPE administered at ZT0, we drew the figures (Fig. 1) with a 24 h curve for the control group by assembling the expression values of the control group administered at ZT0 and ZT12. Overall, the ANOVA test indicated that GSPE administered at ZT0 (light turned on) did not significantly affect the mRNA rhythm of any gene studied in the liver (Fig. 1A-G).
Rats are nocturnal animals and eat mainly at night. Therefore, we next studied whether PAs can modulate the molecular clock in the liver when GSPE is administered at night, when the liver actively manages the ingested nutrients. GSPE was administered when the light was turned off (ZT12), and rats were sacrificed at ZT12, ZT13, ZT15 or ZT18. Thus, the expression of clock-core and clock-controlled genes in the liver was determined at these four time points. Nonetheless, to achieve a better visualization of the changes induced by GSPE administered at ZT12, we drew the figures (Fig. 2) with a 24 h curve for the control group by assembling the expression values of the control group from both this experiment and the former experiment.
GSPE, administered at ZT12, induced slight effects on the mRNA levels of clock-core genes Bmal1 ( Fig. 2A) and Clock (Fig. 2B) as well as on the mRNA levels of the clock-controlled genes Rorα (Fig. 2C) and Rev-erbα (Fig. 2D). Nonetheless, the mRNA levels of Per2 (Fig. 2E) and Nampt (Fig. 2F), two clock-controlled genes, were significantly affected by GSPE. Moreover, the mRNA levels of HmgCoAR (Fig. 2G), a gene with circadian rhythm but not directly controlled by the clock-core genes, were also significantly increased by GSPE treatment.
Comparing the effects of GSPE administered at ZT12 (beginning of the night) or at ZT0 (beginning of the day), it is evident that the capacity of GSPE to modulated the clock system in the liver was dependent on the time of GSPE administration, being only effective at ZT12. Acute administration of GSPE modulated the peripheral clock in the liver of jet-lagged rats. The capacity of GSPE to modulate the peripheral clock was also evaluated in a situation where circadian rhythm was disrupted using rats subjected to a 6 hour jet-lag. Rats at ZT6 (middle of the light period) were administered GSPE and moved to ZT12 (light turned off) and they were sacrificed at ZT12, ZT13, ZT15 or ZT18. Thus, the expression of clock-core and clock-controlled genes in the liver was determined at these four time points.
In control animals, the jet lag induced a clear shift in the mRNA rhythmicity for all of the genes that were studied (Fig. 3) when the rhythms were compared with the 24 h control waves (built by assembling the expression values of the control groups, such as in the former experiments).
GSPE, administered at the beginning of jet lag, did not modulate Clock (Fig. 3B) or Per2 (Fig. 3E), whereas it significantly altered the expression rhythm of Rev-erbα (Fig. 3D), Bmal1 (Fig. 3A), Nampt Figure 1. Effects of an oral dose of a grape seed proanthocyanidin extract (GSPE) administered at Zeitgeber Time 0 on clock-core and clock-controlled genes in the liver. The capacity of GSPE to modify the peripheral clock was evaluated by measuring the oscillation of the levels of mRNA from the clock core genes (A) Bmal1 and (B) Clock as well as the CLOCK:BMAL1-controlled genes, (C) Rorα, (D) Rev-erbα, (E) Per2 and (F) Nampt. (G) The expression of HmgCoAR, a gene that has circadian rhythm expression but that is not directly controlled by CLOCK:BMAL1, was also evaluated. Rats were orally gavaged with tap water (control group) or 250 mg of GSPE /kg body weight dissolved in tap water at ZT0 (light turned on), and rats were sacrificed at ZT0, ZT0.5, ZT1, ZT3, ZT6, ZT12 or ZT24. mRNA levels were measured by real-time qRT-PCR and their expression was normalized to cyclophilin mRNA levels. Each graph shows the mean± s.e. for each data point (n = 3). For the control group, a 24 hour curve was constructed by assembling the expression values of the control group from this experiment and those from Fig. 2. T, significant effect of proanthocyanidins; t, significant effect of Zeitgeber Time; T*t, interaction between the two variables by twoway ANOVA.  The expression of HmgCoAR, a gene that has circadian rhythm expression but that is not directly controlled by CLOCK:BMAL1, was also evaluated. Rats were orally gavaged with tap water (control group) or 250 mg of GSPE /kg body weight dissolved in tap water at ZT12 (light turned off), and rats were sacrificed at ZT12, ZT13, ZT15 or ZT18. mRNA levels were measured by real-time qRT-PCR and their expression was normalized to cyclophilin mRNA levels. Each graph shows the mean± s.e. for each data point (n = 3). For the control group, a 24 hour curve was constructed by assembling the expression values of the control group from this experiment and those from Fig. 1. T, significant effect of proanthocyanidins; t, significant effect of Zeitgeber Time; T*t, interaction between the two variables by twoway ANOVA. The expression of HmgCoAR, a gene that has circadian rhythm expression but that is not directly controlled by CLOCK:BMAL1, was also evaluated. An oral dose of a grape seed proanthocyanidin extract (GSPE) administered to jet-lagged rats significantly alters the expression rhythms of Rev-erbα, Bmal1, Nampt and HmgCoAR. Rats were orally gavaged with tap water (control group) or 250 mg of GSPE /kg body weight dissolved in tap water at ZT6 (middle of the light period) and were moved to ZT12 (light turned off). Rats were sacrificed at ZT12, ZT13, ZT15 or ZT18. mRNA levels were measured by real-time qRT-PCR and their expression was normalized to cyclophilin mRNA levels. Each graph shows the mean± s.e. for each data point (n = 3). For the control group with no jet lag, a 24 hour curve was constructed by assembling the expression values of the control groups from Figs. 1 and 2. T, significant effect of proanthocyanidins; t, significant effect of Zeitgeber Time; T*t, interaction between the two variables by two-way ANOVA.
( Fig. 3F) and HmgCoAR (Fig. 3G) when compared with the jet lag control group. Remarkably, Nampt and HmgCoAR were again two of the genes most sensitive to GSPE, as in when GSPE was administered at ZT12. Thus, next we focused in these two genes.

Acute administration of GSPE had opposite effects on both Nampt expression and NAD levels in the liver at different treatment times.
Given that Nampt gene was susceptible to GSPE in the liver and that this protein is the rate-limiting enzyme of the NAD salvage pathway, we next focused specifically on the oscillations of Nampt RNAm and protein levels as well as NAD concentration during the first 6 hours after GSPE administration at ZT0 and ZT12. These analyses were performed using the livers of rats from the previous experiments.
Remarkably, GSPE induced opposite effects when it was administered at ZT0 or ZT12 (Fig. 4). Nampt protein and mRNA levels were decreased three hours after GSPE administration at ZT0 (Fig. 4A), while Nampt mRNA and protein levels were significantly elevated at three and six hours, respectively, after GSPE administration at ZT12 (Fig. 4B). These modifications in Nampt expression agreed with the alterations in NAD levels that were induced by GSPE in each situation: NAD levels were significantly decreased six hours after GSPE administration at ZT0, whereas they were significantly increased six hours after GSPE administration at ZT12.
Altogether, these results implicate Nampt and NAD modulation as key factors in GSPE activity in the liver and confirm Nampt as a target of GSPE in the liver Acute administration of GSPE at ZT12 increased the ratio of acetylated Bmal1 in rat liver. Nampt is a direct target gene of CLOCK:BMAL1. However, GSPE did not induce a strong modification in the mRNA rhythms of Clock or Bmal1 that could explain the observed alteration in Nampt expression. However, the transcriptional activity of CLOCK:BMAL1 is dependent on Bmal1 acetylation. Thus, we focused further on the ratio of Bmal1 acetylation by measuring Bmal1 mRNA, protein and acetylated protein during the first six hours after GSPE administration at ZT0 or ZT12. Bmal1 mRNA, protein and acetylated protein were analyzed in the livers of rats from the previous experiments.
GSPE administered at either time increased mRNA and protein levels of Bmal1 after 1 hour of treatment (Fig. 5A,B; ZT1 and ZT13, respectively). Nonetheless, the ratio of acetylated Bmal1 was only increased when GSPE was administered at night (ZT12); these are the same circumstances under which Nampt was overexpressed.
Acute administration of GSPE repressed the relative expression of HmgcoAR in rat liver. Finally, the mRNA and protein levels of HmgcoAR, a metabolic gene not controlled by the clock molecular machinery, were determined according to the three experimental designs (day, night and jet lag) during only the first six hours after GSPE treatment (ZT1, ZT3 and ZT6 or ZT13, ZT15 and ZT18). HmgcoAR mRNA and protein were analyzed in the livers of rats from the previous experiments.
HmgcoAR expression was significantly repressed at ZT1 (mRNA) and ZT3 (protein) after GSPE treatment was administered at ZT0 (Fig. 6A). While mRNA levels were increased at ZT15 after GSPE treatment was administered at ZT12, the protein levels did not reflect that fact (Fig. 6B). Finally, in the jet-lagged rats, HmgcoAR protein and mRNA levels were decreased at ZT18 (Fig. 6C).

Discussion
While light is the major synchronizer of the central clock in the suprachiasmatic nucleus (SCN) in the hypothalamus, many other external cues such as temperature, social events or meal timing 10 can entrain circadian rhythms in other cerebral regions or peripheral tissues. This phenomenon is especially the case in the liver, which is the most important metabolic organ due to its involvement in glucose 6 and lipid 5 metabolism, among other crucial physiological functions 23 . Even specific components in foods could also be important synchronizers, such as dietary fat 24 or phenolic compounds like resveratrol [25][26][27][28] . Therefore, the aim of this work was to determine the capacity of an acute dose of GSPE to act as a signal to modulate the molecular clock in the liver.
To accomplish this outcome, three different experimental approaches were performed to determine whether PAs can modulate the liver clock: the administration of GSPE at ZT0, at the beginning of the light phase; at ZT12, at the beginning of the dusk phase; and to rats with 6 h of jet lag. The data clearly show that the power of PAs to modulate the circadian rhythm of clock-core and clock-controlled genes in the liver depends on the time of their administration.
Interestingly, Nampt and NAD emerge as molecular targets of PAs in the liver. Nampt is the rate-limiting enzyme in NAD biosynthesis through its salvage pathway 29 . NAD plays a major role as a coenzyme in numerous oxidation-reduction reactions 30 and is required in a number of important signaling pathways in mammalian cells, including poly-ADP-ribosylation in DNA repair 31 , mono-ADP-ribosylation in both the immune response and G protein-coupled signaling 32 , and the synthesis of cyclic ADP-ribose and nicotinate adenine dinucleotide phosphate (NAADP) in intracellular calcium signaling 33 . Furthermore, NAD activates several NAD(+ )-dependent deacetylases (SIRT), such as SIRT1 and SIRT3, thus controlling the activity of many cellular proteins by cycling them between their acetylated and deacetylated forms. Specifically, the circadian oscillations of NAD levels have been shown to modulate mitochondrial respiration by controlling the activity of SIRT3, thus generating rhythms in the acetylation and activity , and rats were sacrificed at ZT1, ZT3 or ZT6 and ZT13, ZT15 or ZT18, respectively. Nampt RNAm levels were measured by real-time qRT-PCR and its expression was normalized to cyclophilin mRNA levels. Nampt protein was analyzed by Western blot and was normalized to β -actin. Un-cropped blots/gels are presented in Supplementary Figure 1. NAD quantification was performed using an ELISA kit following the manufacturer's instructions. Each value is the mean ± s.e. of the same 3 animals for all the parameters quantified. White bars, control group; colored bars, GSPE-treated groups. *Statically significant differences found by independent Student T-test (p < 0.05) between the control group and GSPE-treated group for each ZT.
Scientific RepoRts | 5:10954 | DOi: 10.1038/srep10954 Figure 5. An oral dose of a grape seed proanthocyanidin extract (GSPE) administered at Zeitgeber Time 12 or to jet-lagged rats increases the ratio of acetylated Bmal1 in rat liver. Rats were orally gavaged with tap water (control group) or 250 mg of GSPE /kg body weight dissolved in tap water both at (A) ZT0 (light turned on), (B) ZT12 (light turned off) or (C) ZT6 (middle of the light period), in the case of the jet-lagged rats, which were moved to ZT12 (light turned off) after GSPE administration. Rats were sacrificed at ZT1, ZT3 or ZT6 when GSPE was administered at ZT0 and ZT13, ZT15 or ZT18 when GSPE was administered at ZT12 or to jet-lagged rats. Bmal1 RNAm levels were measured by real-time qRT-PCR and its expression was normalized to cyclophilin mRNA levels. Acetylated and total BMAL1 proteins was analyzed by Western blot and was normalized to β -actin. Acetylated and total BMAL1 protein samples were then divided to obtain the acetylated Bmal1/total Bmal1 protein ratio, shown as a percentage. Un-cropped blots/gels are presented in Supplementary Figure 1. Each value is the mean± s.e. of the same 3 animals for all the parameters quantified. White bars, control group; colored bars, GSPE-treated groups. *Statically significant differences found by independent Student T-test (p < 0.05) between the control group and GSPEtreated group for each ZT.
Scientific RepoRts | 5:10954 | DOi: 10.1038/srep10954 Figure 6. An oral dose of a grape seed proanthocyanidin extract (GSPE), administered at Zeitgeber Time 0, 12, or to jet-lagged rats, modulates HmgcoAR expression in rat liver. Rats were orally gavaged with tap water (control group) or 250 mg of GSPE / kg body weight dissolved in tap water at (A) ZT0 (light turned on), (B) ZT12 (light turned off) or (C) ZT6 (middle of the light period), in the case of the jet-lagged rats, which were moved to ZT12 (light turned off) after GSPE administration. Rats were sacrificed at ZT1, ZT3 or ZT6 when GSPE was administered at ZT0 and ZT13, ZT15 or ZT18 when GSPE was administered at ZT12 or to jet-lagged rats. HmgcoAR RNAm levels were measured by real-time qRT-PCR and its expression was normalized to cyclophilin mRNA levels. HmgcoAR protein was analyzed by Western blot and was normalized to β -actin. Un-cropped blots/gels are presented in Supplementary Figure 1. White bars, control group; colored bars, GSPE-treated groups. *Statically significant differences found by independent Student T-test (p < 0.05), between the control group and GSPE-treated group for each ZT. of oxidative enzymes that synchronize mitochondrial oxidative functions across the daily cycles of fasting and feeding 34 . Therefore, NAD is a key molecule in the synchronization of liver metabolism, and the modulation of its levels in the liver by PAs is an attractive candidate for the explanation of some of the metabolic effects of PAs. GSPE modulated Nampt and NAD levels in opposite ways 6 h after its administration during the day versus at night, reducing or increasing their levels, respectively, suggesting that Nampt expression and NAD levels peaked at night. Therefore, in this sense, PAs could act as an element of adaptation in the liver, improving the energetic profile of rats and increasing mitochondrial function and oxidation at night, when rats are active. In keeping with the idea of an adaptation mechanism, as these animals are resting during the light phase, PA activity could be acting as an energy saver through the decreased levels of NAD after PA administration at ZT0.
NAD concentration oscillates in a circadian manner due to the circadian expression of Nampt, which in turn is mediated by the CLOCK:BMAL1 heterodimer 35 . The rhythm of Clock and Bmal1 expression was not altered by GSPE administration, either during the day or at night ( Figs. 1 and 2). However, when Bmal1 expression was studied during the first six hours of GSPE administration (Fig. 5), both mRNA and protein levels were always increased one hour after PA consumption, regardless of whether GSPE was administered diurnally, at night, or even under jet lag conditions, suggesting a robust relationship between Bmal1 and PAs. However, to be active, BMAL1 should be acetylated by CLOCK, which is its own partner. Therefore, the ratio of acetylated to total BMAL1 protein provides direct information about the transactivation activity of BMAL1:CLOCK 35 . We found that GSPE significantly increased the ratio of BMAL1 that was acetylated at ZT13, whereas this effect was not observed at ZT1. This differential pattern of BMAL1 acetylation, which depends on the time of GSPE administration, could explain the overexpression of NAMPT and therefore the peak in NAD levels in the liver only when GSPE was administered at night. Therefore, it is globally supposed in this work that NAD levels peaked 6 hours after GSPE consumption at ZT12 (thereby at ZT18) as a consequence of an increased BMAL1 acetylation ratio at ZT13 that, in turn, increased Nampt mRNA and protein levels at ZT15-ZT18.
PAs modulate lipid metabolism in the liver 20 . Therefore, we also analyzed the expression of HmgcoAR, the key enzyme in the cholesterol biosynthetic pathway, which has circadian rhythm but is not directly controlled by the clock-core genes 36 . GSPE had a dual effect on HmgcoAR expression depending on the time of its administration. GSPE repressed both HmgcoAR mRNA and protein levels at ZT0, whereas protein levels at ZT12 were not affected. Therefore, as in the case of BMAL1 acetylation, Nampt expression and NAD levels, the time of PA administration conditions the circadian regulation outputs.
Further studies are needed in order to find out the actual compounds that can modulate the molecular clock in the liver. Studies on the bioavailability of proanthocyanidins are controversial 37 . Some studies indicate no absorption of oligomeric proanthocyanidins, thus attributing the biological effects of PA to their colonic metabolites such as phenolic acids and valerolactones 37 . However, others studies have detected oligomeric PA in plasma 38 and in several tissues 39 of rats. Moreover, GSPE can be metabolized by intestinal and hepatic enzymes, producing a large array of metabolites 39 .
Overall, these findings agree with the Xenohormesis Hypothesis, which proposes that heterotrophs are able to sense chemical cues, such as polyphenols, that are synthesized by plants in response to stress 40 . In fact, circadian rhythms allow the anticipation of environmental changes and adaptation to the time of day and food availability, which has been shown in this work through NAD, NAMPT and BMAL1 acetylation levels. Thus, PAs can advise animals about environmental conditions by modulating biological rhythms to obtain a better ability to adapt to changing conditions over the course of their lives. Despite this study has been performed with a pharmacological dose of PAs, animals could consume such a huge amount of PAs in a wildlife situation.
In conclusion, PAs modulate the molecular clock in the liver even though their effectiveness depends largely on the time of administration. Specifically, Bmal1 and Nampt, as well NAD, emerge as targets of GSPE in the liver.

Animals.
Administration of GSPE at ZT12. Twenty-two rats were singly caged in animal quarters at 22 °C with a 12 h light/dark cycle (light from 21:00 pm to 9:00 am). After three weeks of adaptation, the rats were orally gavaged with tap water (control group) or 250 mg of GSPE/kg body weight dissolved in tap water at ZT12 (9:00 am, light off). Rats were sacrificed by beheading at ZT12, ZT13, ZT15, and ZT18 (n = 3 for control and n = 3 for GSPE-treated groups).
Administration of GSPE to jet-lagged rats. Twenty-two rats were singly caged in animal quarters at 22 °C with a 12 h light/dark cycle (light from 15:00 pm to 03:00 am). After three weeks of adaptation, rats were orally gavaged with tap water (control group) or 250 mg of GSPE /kg body weight dissolved in tap water at ZT6 (9:00 am, middle of the light day) and immediately moved to a dusk room (ZT12), thus giving rats a jet lag of 6 hours. Rats were sacrificed by beheading at ZT12, ZT13, ZT15, and ZT18 (n = 3 for control and n = 3 for GSPE-treated groups).
For the three experiments, the liver was excised, frozen immediately in liquid nitrogen and stored at − 80 °C until RNA and protein extraction.
RNA extraction and cDNA synthesis. Total RNA from liver was extracted using TRIzol reagent and an RNeasy Mini Kit (Qiagen, 74106, Barcelona, Spain) according to manufacturer protocols. RNA was quantified by spectrophotometry (Nanodrop 1000 Spectrophotometer, Thermo Scientific, Madrid, Spain) at λ = 260 nm and tested for purity (by A260/280 ratio) and integrity (by denaturing gel electrophoresis). Complementary DNA was generated using the High-Capacity complementary DNA Reverse Transcription Kit from Applied Biosystems (4368814, Madrid, Spain). mRNA quantification by real-time qRT-PCR. A total of 10 ng of cDNA was subjected to quantitative RT-PCR amplification using SYBR Green PCR Master Mix from Bio-Rad (172-5200, Barcelona, Spain). The forward and reverse primers of the genes analyzed are shown in Table 1. Reactions were run on a quantitative real-time PCR system (CFX96 touch of Bio-Rad, Barcelona, Spain); the thermal profile settings were 50 °C for 2 min, 95 °C for 2 min, and then 40 cycles at 95 °C for 15 s and 60 °C for 2 min. Finally, statistical data were converted and normalized to the linear form by the 2 − CT (∆∆C T ) calculation 42 . The relative expression of the clock genes was normalized to cyclophilin mRNA levels.
Data and statistical analysis. The results are presented as the mean with the associated standard error (SE).The data were analyzed using a two-way ANOVA and Student t-test to determine the significant difference using SPSS statistical software (version 17.0 for Windows; SPSS, Inc.). P values < 0.05 were considered statistically significant.