The Circadian Binding of CLOCK Protein to the Promoter of C/ebpα Gene in Mouse Cells

C/EBPα plays important roles in metabolism as well as in the maintenance of energy homeostasis. Here we describe loss of the circadian oscillation of C/ebpα expression in liver of Clock mutant mice. Reporter assays indicate Clock and Bmal significantly induced C/ebpα gene expression whereas Cry suppressed. Real time reporter assays showed that two mutated E-boxes disrupted C/ebpα promoter dependent-oscillation. Chromatin immunoprecipitation suggests Clock can bind to two E-boxes in the C/ebpα promoter with a circadian manner in vivo. Thus, C/ebpα gene transcription is under circadian control of a core clock component, Clock. The data suggests that circadian disturbances may affect metabolic abnormalities through the C/ebpα pathway in liver.


Introduction
Many organisms display physiological and behavioral rhythms of entrainment to a 24-h cycle of light and darkness. The master clock located in the suprachiasmatic nucleus of mammals controls most physical and physiological circadian rhythmicity [1,2] and the generation of circadian rhythms depends on the concerted coexpression of specific clock genes. Clock was the first clock gene identified in vertebrates [3]. CLOCK binds DNA and activates transcription after dimerization with BMAL1 [1,2] by driving the rhythmic transcription of other clock and circadian clockcontrolled genes through an E-box (CACGTG). Because the Clock allele of Clock mutant mice is truncated and causes a deletion of 51 amino acids, the mutation presumably would not significantly affect N-terminal basic helix-loop-helix and Per-Arnt-Sim domains, leaving CLOCK dimerization and DNA binding intact. The mutant CLOCK protein can still form heterodimers with BMAL1 that binds to DNA, but these heterodimers are deficient during transactivation [2]. The mammalian circadian clock is an intracellular, transcriptional-translational mechanism comprising the same molecular components in the suprachiasmatic nucleus and in peripheral cells. Because this endogenous timekeeper interacts with countless biological systems, circadian disruption has significant effects on health; for example, susceptibility to obesity, diabetes and related metabolic syndromes and various types of cancer is increased among long-term shift workers [4].
The founder of the family of related leucine zipper transcription factors is CCAAT enhancer binding protein alpha (C/EBPa), which plays an important role in numerous cellular processes including proliferation, differentiation, apoptosis, metabolic con-trol and other specific functions [5][6][7][8]. This gene is expressed at high levels in the liver and it is critical for the establishment and maintenance of energy homeostasis in neonates [9,10]. C/ebpa gene knockout mice die because they cannot accumulate glycogen in the liver and develop hypoglycemia [10].
Hundreds of tissue-specific circadian genes that regulate an impressive diversity of biological processes have been identified using DNA microarray technology [11][12][13][14], and we found that C/ ebpa is one of circadian expressing gene candidates that have been screened from Database of Circadian Gene Expression (http:// expression.gnf.org/cgi-bin/circadian/index.cgi) and Database for Systems Biology (http://sirius.cdb.riken.jp/). Some studies have shown that mutations or deletions of Clock and Bmal1 genes result in not only circadian disturbances but also disrupted glucose and lipid metabolism [15][16][17][18]. In fact, recent reports indicate that C/ EBPa directly controls Per2 and Rev-Erba, which are core members of clock genes for the circadian clock [19,20]. However, little is known about the molecular mechanism of circadian oscillation of C/ebpa gene expression.
We identified E-boxes in the C/ebpa promoter region investigated the relationship between C/ebpa and clock genes. Here, we show that the core clock gene product CLOCK regulates the circadian expression of C/ebpa gene in mouse cells.

Ethics Statement
All animal experiments, care and handling proceeded under the approval of our institutional Animal Care and Use Committee (Permission Number 2009-020).

Animals
Male Jcl:ICR (Clea Japan Inc., Tokyo, Japan) and homozygous Clock mutant mice on a Jcl:ICR background [21] aged 7-10 weeks were maintained under a 12 h light/12 h dark cycle (lights on at 0:00 and lights off at 12:00) for at least two weeks before experimentation. The mice were fasted overnight and sequentially sacrificed for some experiments.

Isolation of mRNA and RT-PCR
Total RNA was isolated from NIH 3T3 cells or liver tissue using RNAiso (TAKARA Bio Inc., Shiga, Japan) and then reversetranscribed using the PrimeScript RT reagent kit (TAKARA Bio) according to the manufacturer's protocol. 3 animals were used per each time. The cDNA levels of genes of interest were measured by real time quantitative PCR using a LightCycler (Roche Applied Science) with SYBR Premix Ex Taq (TAKARA Bio). The amount of mRNA was corrected relative to that of b-actin for liver tissue or Large Ribosome Protein (Rplpo) for NIH 3T3 cells. The maximal value for wild type mice is expressed as 100% and other values are expressed means 6SEM (n = 3).

Cell Culture
NIH3T3 cells were incubated in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and a mixture of penicillin and streptomycin at 37uC under a humidified 5% CO 2 atmosphere.

Transient Luciferase Assays
The upstream region of the C/ebpa transcription start site containing two (21386 to +113 bp) E-boxes were cloned into the pGL3 Basic vector (Promega). Mouse CLOCK, BMAL1, and CRY1 expression plasmids were provided by Dr. T. Todo [22]. Constructs (500 ng) were co-transfected with 1 ng of pRL-CMV (Promega) as the internal control into NIH3T3 cells (24-well plates) using HilyMax (DOJINDO Laboratories, Kumamoto, Japan) according to the manufacturer's protocols. Luciferase activities were measured using a dual luciferase reporter assay system (Promega) and a Luminometer model TD-20/20 (Turner Designs, Sunnyvale, CA). The transcriptional activities were normalized relative to Renilla luciferase activities.

Real time Luciferase Assays
The upstream region of the C/ebpa transcription start site containing two E-boxes (21386 to +113 bp) was cloned into SV40-dLuc harboring the SV-40 promoter and a rapid degradation domain modified from mouse ornithine decarboxylase at the C-terminal end of firefly luciferase [13]. The Per2 promoter regions (2798 to +331 relative to the cap site) were cloned into pGL3-dLuc [23], and then 2 mg of reporter plasmids were transfected into NIH3T3 cells (35-mm collagen type I-coated dishes) using HilyMax (DOJINDO Laboratories, Kumamoto, Japan). The cells were stimulated with 100 nM dexamethasone (Sigma-Aldrich) for 2 h in serum-free Dulbecco's modified Eagle's medium (D-MEM) and then the medium was replaced with fresh Dulbecco's MEM containing 100 mM luciferin (Wako Pure Chemical Industries), 25 mM HEPES (pH 7.2), and 10% fetal bovine serum. Bioluminescence was measured and integrated for 1 min at intervals of 10 min using the Photon Detection UNIT LM-2400 (Hamamatsu Photonics, Hamamatsu, Japan). To  compare the phase and amplitude of C/ebpa -WT-dLuc and each mutant, the data were detrended by subtracting an average of 12 h from the data.

Chromatin Immunoprecipitation (ChIP) Assays
Immunoprecipitation was adapted from the reported procedure [24,25]. NIH 3T3 cells were stimulated with dexamethasone and incubated for 10 min at room temperature. Formaldehyde was added directly to cell culture media at a final concentration of 1% at 0, 6, 12, 18, 24 or 30 h thereafter. Cross-linking was stopped by adding 125 mM glycine. The plates were rinsed with cold PBS and then NIH 3T3 cells harvested by scraping in 0.4 mL of ice cold 10 mM PMSF in PBS were sedimented by centrifugation. Binding between CLOCK and E-boxes was detected using the Simple ChIP Enzymatic Chromatin IP Kit (Cell Signaling Technology, Danvers, MA, USA) with anti-CLOCK antibody Real time reporter assay. NIH3T3 cells transfected with indicated mutant constructs were incubated with dexamethasone (100 nM) and then bioluminescence was measured. Plasmid pGL3-dLuc-Cebp-WT, wild type 1.4 kb upstream region of C/ebpa gene;, pGL3-dLuc-Cebp-Mut I, mutant E1 site; pGL3-dLuc-Cebp-Mut II, mutant E2 site; pGL3-dLuc-Cebp-Mut I-II, mutant E1 and E2 sites; mPer2-dLuc, Per2 promoter region (2798 to +331 relative to the cap site) cloned into pGL3-dLuc [23]. Each experiment was repeated 3-5 times and representative data were used for these graphs.

Statistical Analysis
All data are expressed as means 6 SEM. Differences in expression levels and peak times were statistically evaluated using Student's t-test for single comparisons and one-way ANOVA with post-hoc Student's t-test for multiple comparisons.

Oscillatory Expression of C/ebpa mRNA in Mouse Liver
Since C/ebpa is essential for liver development, we isolated C/ ebpa mRNA from livers of wild-type and Clock mutant mouse to determine the relationship between the circadian clock and temporal C/ebpa gene expression using RT-PCR (Fig. 1). Diurnal expression of C/ebpa mRNA was robust in the liver of wild-type mice, but rhythmic expression was significantly dampened in those from Clock mutant mice. The C/ebpa mRNA levels obviously peaked at Zeitgeber time (ZT) 8 in the wild type liver.

CLOCK Up-regulates C/ebpa Expression in vitro
Analysis of the mouse genome revealed two perfect E-box motifs within 1.4 kb of the 59-flanking region of the C/ebpa gene ( Fig. 2A), suggesting that clock gene products may control C/ebpa. We therefore analyzed the functions of these E-box elements using a genomic DNA sequence of this region fused to the luciferase reporter plasmid (Fig. 2B). The promoter reporter assay showed that the transcriptional activity of a 1.4 kb fragment containing the two E-boxes was increased 3-fold under CLOCK and BMAL1 overexpression in NIH 3T3 cells (Fig. 2C). This increase was suppressed by co-expression with CRY1, which is the negative component of CLOCK-BMAL1-dependent transcriptional activation [1]. The findings suggest that a feedback loop of the molecular circadian clock may regulate the upstream region of C/ ebpa containing these two E-boxes. Horizontal bars indicate amplified regions (E-box unrelated sequence, from 22239 to 22063; E1 site, from 21240 to 21128; E2 site, from 2295 to 2137.) NIH 3T3 cells were rapidly sampled and processed every 6 hours after dexamethasone stimulation. Chromatin immunoprecipitation was performed using antibodies against CLOCK, and enrichment of E-boxes containing sequence of the promoter region of C/ebpa were measured using quantitative PCR. Data were normalized to a control condition (Normal Rabbit IgG). The ChIP assay was repeated two times. Each time point comes from triplicate experiments. Data are presented as mean 6 SEM (n = 3, *: p,0.0001). (B) Temporal mRNA expression of endogenous C/ebpa in NIH 3T3.Total RNA was extracted every 6 hours after dexamethasone stimulation and mRNA transcripts were quantitated relative to Rplpo transcript expression (housekeeping gene) by using quantitative RT-PCR. Each time point comes from triplicate experiments. Data are presented as mean 6 SEM (n = 3, *: p,0.05). doi:10.1371/journal.pone.0058221.g004 Figure 5. Links between C/ebpa and Clock gene expressions. An interactive feedback loop between C/ebpa and the core clock genes feedback loop is indicated. The core clock genes feedback loop is enclosed in an oval. Transcriptional activators CLOCK and BMAL1 drive Per2 and C/ebpa gene expression. Per and CRY proteins inhibit CLOCK/ BMAL1 activity. Thick line denotes regulation based on present findings. Dotted arrow comes from other papers [20,21]. doi:10.1371/journal.pone.0058221.g005 Upstream E1 and E2 Sites Regulate the Circadian Expression of C/ebpa Gene We analyzed the roles of the two E-boxes in circadian expression of the C/ebpa gene using real time reporter assays with NIH3T3 cells containing functional circadian clock components. The transcriptional activity of the construct containing two Eboxes ( Fig. 3A; pGL3-dLuc-Cebp-WT) exhibited circadian oscillation ( Fig. 3B; Left upper). When E1 or E2 site was mutated ( Fig. 3A; pGL3-dLuc-Cebp-Mut I, pGL3-dLuc-Cebp-Mut II), circadianoscillations of the reporter activity was still observed although with abnormal phase when E1 or E2 site was mutated ( Fig. 3B; Upper right and Left lower). In contrast, mutations in both of Eboxes ( Fig. 3A; pGL3-dLuc-Cebp-Mut I-II) disrupted the circadian oscillation of the reporter activity ( Fig. 3B; Lower right). These findings show that both of the E1 and E2 sites are involved in circadian expression of the C/ebpa reporter gene in NIH3T3 cells.

CLOCK Binds to E-boxes in C/ebpa Control Region in Living Cells
We analyzed in vivo binding of CLOCK for the putative DNA sequence using chromatin immunoprecipitation (ChIP) assays in NIH 3T3 cells. To confirm the validity of this anti-CLOCK antibody for ChIP in 3T3 cells, we evaluated CLOCK binding to-E box element in Dbp (D site albumin promoter binding protein) as a positive control (Fig. S1). We found that CLOCK bound to both fragments of the C/ebpa 59-flanking region that contain E-boxes (Fig. 4A). CLOCK binding to two E-boxes was detected in a circadian manner and peaked at 18 hour after dexamethasone treatment. In contrast, E-box unrelated sequence in upstream region has consistently detected very low binding with CLOCK (Negative control). We also determined temporal expression of endogenous C/ebpa by using RT-PCR and confirmed its circadian oscillation (Fig. 4B). These findings indicate that CLOCK protein bind to two E-boxes of the C/ebpa promoter in NIH 3T3 cells at the chromatin level and suggest that CLOCK controls the circadian expression of C/ebpa gene through these two E-boxes even in living cells.

Discussion
Here, we discovered that the circadian expression of C/ebpa gene was regulated by two E-boxes in the upstream region of this gene through the core clock protein, CLOCK. Considering with transient luciferase assays, real-time reporter assays in vitro, and ChIP assays and loss of circadian mRNA expression in Clock mutant mouse liver, the data indicates that the molecular circadian feedback loop affects C/ebpa gene circadian expression through two E-boxes. C/ebpa protein is thought to play important roles in numerous cellular processes including cell proliferation, differentiation and apoptosis [5][6][7][8]. But, C/ebpa gene is expressed at high levels in the liver and recent reports show that it is critical for the establishment and maintenance of energy homeostasis in mammals [9][10]. Furthermore, the core clock proteins are reported to affect energy homeostasis in many species [29].
Genome-wide screening of BMAL1 and CLOCK targets has confirmed that carbohydrate and lipid metabolism comprises the major output of the circadian clock in the mouse liver [14,30,31].
Duplicate E-boxes were reported to be important for the circadian rhythmic mRNA expression of clock-controlled genes [30,32]. We also found two E-boxes are required to generate the distinct circadian oscillation of Gys2 (Glycogen synthase 2) gene expression in mouse liver [25]. These papers suggest that two Eboxes system are important for the circadian expression of clockcontrolled genes in peripheral tissues. In this paper, ChIP assay and real time reporter assay indicated that the binding peak of CLOCK to E-boxes preceded the peak expression of endogenous C/ebpa in NIH 3T3 cell. This suggests that CLOCK binding in vivo might occur earlier than transcription. Furthermore, CLOCK binds to Dbp promoter in a similar fashion ( Figure S1), [27]. This also suggests that CLOCK binds to C/ebpa promoter in a circadian manner.
Our findings suggest that C/ebpa is one of clock-controlled genes regulated by the circadian negative feedback (Oval in Figure 5). Interestingly, others have reported that the Per2 promoter region has several potential C/ebpa binding sites to activate Per2 transcription [19,20] (Dotted line in Figure 5). Furthermore, previous papers indicate that Per2 and C/ebpa have the same effects for tumor suppression events [19,20]. Thus, we propose an interactive feedback loop between the negative feedback loop and C/ebpa rhythmic expression ( Figure 5). Recently, the circadian expression of the gene encoding glycogen synthase was found to be damped in livers of C/ebpa conditional-knockout mice [33]. These data support that glucose metabolism might be involved in an interactive feedback loop in livers.
Recent studies have revealed an association between the circadian clock disruption and metabolic syndrome [28,[33][34][35]. The C/ebpa gene has divergent roles and functions, especially in glucose metabolism [10,35,36]. These findings also suggest close links between the circadian clock-controlled gene, C/ebpa and metabolic activity. We propose that disrupted clock genes expression may cause metabolic syndrome through deregulation of C/ebpa. Figure S1 ChIP assay for CLOCK binding to Dbp promoter. Binding activity between CLOCK and E box element in Dbp (D site albumin promoter binding protein) promoter sequence was evaluated to confirm the validity of our ChIP assay. Experiment was performed as described in the materials and methods. Each time point comes from triplicate experiments. Data are presented as mean 6 SEM (n = 3, *: p = 0.01).