Skip to main content
NCBI home page
Endocrinology logoLink to Endocrinology
. 2010 Mar 9;151(5):2287–2296. doi: 10.1210/en.2009-1252

Episodes of Prolactin Gene Expression in GH3 Cells Are Dependent on Selective Promoter Binding of Multiple Circadian Elements

Sudeep Bose 1, Fredric R Boockfor 1
PMCID: PMC2869263  PMID: 20215567

Abstract

Prolactin (PRL) gene expression in mammotropes occurs in pulses, but the mechanism(s) underlying this dynamic process remains obscure. Recent findings from our laboratory of an E-box in the rat PRL promoter (E-box133) that can interact with the circadian factors, circadian locomoter output cycles kaput (CLOCK) and brain and muscle aryl hydrocarbon receptor nuclear translocator-like protein (BMAL)-1, and was necessary for pulse activity raised the intriguing possibility that the circadian system may be central to this oscillatory process. In this study, we used serum-shocked GH3 cells, established previously to synchronize PRL pulses between cells in culture, to reveal that pulses of PRL mRNA are linked temporally to the expression of bmal1, cry1, per1, and per3 mRNA in these cells. Moreover, we found that each of these circadian factors binds to the rat PRL promoter by chromatin immunoprecipitation analysis. Using EMSA analysis, we observed that two sites present in the proximal promoter region, E-box133 and E-box10, bind circadian factors differentially (E-box133 interacted with BMAL1, cryptochrome-1, period (PER)-1, and PER3 but not PER2 and E-box10 bound BMAL1, cryptochrome-1, PER2, PER3 but not PER1). More importantly, down-regulation of any factor binding E-box133 significantly reduced PRL mRNA levels during pulse periods. Our results demonstrate clearly that certain circadian elements binding to the E-box133 site are required for episodes of PRL mRNA expression in serum-shocked GH3 cultures. Moreover, our findings of binding-related differences between functionally distinct E-boxes demonstrate not only that E-boxes can bind different components but suggest that the number and type of circadian elements that bind to an E-box is central in dictating its function.

Prolactin-gene expression pulse activity is dependent on circadian factor input.

Prolactin (PRL) is important for a number of physiological processes. Its release is very complex, occurring in a highly coordinated fashion with distinct rises and falls that are regulated with respect to timing and the endocrine status of the animals (1). Investigations by several groups reveal that this high degree of regulation of PRL secretion is governed by a variety of factors released mainly from the hypothalamus that dictate the synthesis and release of PRL from the pituitary (1,2). From this type of evidence, it is generally accepted that the impact of exogenous factors on the pituitary is central to the timed release of PRL.

A growing body of evidence suggests that, in addition to these exogenous factors, some of this timing control occurs within mammotropes themselves (3,4). It was clear from earlier studies in our laboratory that prolactin gene expression (GE) occurred in distinct pulses in individual living mammotropes obtained from rat pituitaries (5). Although the reason for this pulsatile activity was unclear, more recent experiments demonstrated that PRL pulse activity was a promoter-driven phenomenon, depending on a specific promoter binding region termed an E-box, a site usually associated with the circadian system (6). In fact, mutation analysis revealed that only a specific E-box, E-box133, was required for PRL GE pulses and not a site in close proximity, E-box10. The interrelationship of the circadian system with PRL GE pulse activity was also suggested by studies of McFerran et al. (3), who found that exposure of GH3 cultures to high concentrations of serum (a treatment previously found to stimulate oscillations of circadian transcription factors) resulted in the synchronization of individual cells with respect to PRL GE oscillatory events. From these observations, it appeared not only that the control of PRL GE pulses was very precise, but also that part of the underlying mechanism involved the circadian system.

From previous work with other cell and tissue systems, it was clear that circadian timing depended on the interaction of transcription factors encoded by genes named clock genes that bound these E-box elements (7,8,9). Interestingly, recent work using Caenorhabditis elegans revealed that the affinity of specific circadian elements may differ from E-box to E-box within the same promoter (10). This, when viewed in light of our findings of the importance of E-box133 in PRL GE pulse activity, raised the possibility that the elaboration of a PRL pulse may depend on the association of a particular group of circadian factors with this specific E-box site. To address this in the following study, we used the serum-shocked GH3 model described above to identify whether the E-box133 site binds a defined group of circadian transcription factors and, if so, whether each of these factors is required to induce a burst of PRL mRNA synthesis.

Materials and Methods

Cell culture and synchronization of GH3 cells by serum shock treatment

GH3 cells (CCL-82.1; American Type Culture Collection, Manassas, VA) were maintained in high glucose and DMEM, supplemented with antibiotics/antimycotics (Ab/Am; 1% penicillin/streptomycin and 1% Fungizone), and 10% fetal bovine serum (FBS). Unless indicated otherwise, all tissue culture supplies were obtained from Invitrogen (Carlsbad, CA). The cells were grown in a 5% CO2 environment at 37 C in a water-saturated atmosphere. For experimentation, cells (0.25 × 106) were seeded in 35-mm plates in DMEM containing 10% FBS overnight. The next day cultures were serum shocked according to the protocol developed by Balsalobre et al. (11) and adapted by McFerran et al. (3) for use with GH3 cells. Briefly, seeded GH3 cells were treated with serum-shock medium (DMEM supplemented with 50% horse serum) for 2 h followed by incubation for 4–84 h in serum-free DMEM containing Ab/Am.

RNA isolation and real-time RT-PCR of PRL and circadian genes

Total RNA was extracted from serum-shocked cultures every 4 h using a RNA purification kit (5-PRIME, Inc., Gaithersburg, MD). First-strand cDNA synthesis was performed using a qScript cDNA synthesis kit (Quanta Biosciences, Inc., Gaithersburg, MD) according to the manufacturer’s protocol. Then real-time RT-PCR was performed on cDNA generated using Multi-Scribe reverse transcriptase from the Sensi-Mix Plus SYBR and fluorescein kit PCR master mix (Quantace, Inc., Taunton, MA). The primer pairs for rat PRL and different circadian genes (Table 1) were designed from Universal Probe Library (Roche Applied Science, Indianapolis, IN). Oligos were synthesized, desalted (Fisher-Oligos; Fisher Scientific Corp., Pittsburgh, PA), and amplification reactions were performed in an iCycler machine (Bio-Rad Laboratories, Inc., Hercules, CA) using a three-step thermal cycling method, which consisted of a 3-min step at 95 C followed by 40 cycles at 95 C for 15 sec, 58 C for 1 min, and 72 C for 30 sec. The fold changes in the levels of PRL or circadian components were determined by the comparative cycle (Ct) threshold method (12) with cyclophilin as internal control. As a negative control, reactions without cDNA template were also performed. All reactions were carried out three times in triplicate. Pulses were identified using the PC-PULSAR program (13). Peaks were identified as individual subseries raised above baseline with magnitude of at least G(n) by using Pulsar G (1) criteria [G (1) = 100, G (2) = 2.6, G (3) = 1.9, G (4) = 1.5, G (5) = 1.2].

Table 1.

The primer pairs for rat PRL and different circadian genes were designed from Universal Probe Library (Roche Applied Science)

Gene name Primers Primer sequences (5′–3′)
PRL Sense GACCGGTTCTGTCATGTCG
Antisense ACCTGGTTCATCATCACTAATCAC
Cyclophilin Sense TTTTCGCCGCTTGCTGCAGAC
Antisense CACCCTGGCACATGAATCCTGGA
BMAL1 Sense ATTCCAGGGGGAACCAGA
Antisense GAAGGTGATGACCCTCTTATCCT
CRY1 Sense CGACCATGATGAGAAGTACGG
Antisense AGGACAAGCCATCGGTATCA
PER1 Sense GTGGGCTTGACACCTCTTCT
Antisense TGCTTTAGATCGGCAGTGGT
PER2 Sense CATCTGCCACCTCAGACTCA
Antisense CTGGTGTGACTTGTATCACTGCT
PER3 Sense TGGCCACAGCATCAGTACA
Antisense TACACTGCTGGCACTGCTTC
Open in a new tab

Desalted oligos were purchased from Operon Biotechnology (Fisher-Oligos, Fisher), and amplification reactions were performed by a three-step thermal cycling method using Bio-Rad iCycler as described in Materials and Methods. 

Chromatin immunoprecipitation (ChIP) assays

ChIP assays were performed according to our previous protocol using the ChIP-IT Express kit (Active Motif, Carlsbad, CA) with a few modifications (14). Briefly, the chromatin in GH3 cells (10–20 × 106 cells) was cross-linked by exposure to a 1.0% formaldehyde solution for 5 min at room temperature and then sheared using a VirSonic sonicator (VirTis, Gardiner, NY) at an output setting of 15% for 30 cycles of 15 sec each. The resulting chromatin solution containing fragment lengths of 150–200 bp was incubated with 8 μg of goat polyclonal antibodies against brain and muscle aryl hydrocarbon receptor nuclear translocator-like protein (BMAL)-1 (sc-8550x), cryptochrome (CRY)-1 (sc-5953), period (PER)-1 (sc-7724x), PER2 (sc-7728x), PER3 (sc-8912x), or DREAM-antigoat IgG (sc-9309) (all from Santa Cruz Biotechnology, Santa Cruz, CA) in 200 μl final reaction volume overnight at 4 C. Immunoprecipitation was then performed using protein G magnetic beads (Active Motif) and the DNA eluted according to kit instructions. The cross-linking was reversed at 94 C overnight. As input control, an aliquot of sheared chromatin was used and the samples were treated with proteinase K for 1 h, purified using a QIAquick PCR purification kit (QIAGEN, Valencia, CA). Finally, the samples were treated with proteinase K for 1 h and subjected to PCR amplification of rat PRL promoter region using the primer set 5′-ATCCTTCCTTTCTGGCCACT (forward) and 5′-TCTTCCCCTCCCAATCATCT (reverse) (Integrated DNA Technologies, Inc., Coralville, IA). The reaction mixtures were subjected to 2% agarose gel electrophoresis and visualized with ethidium-bromide staining.

Nuclear protein extract and EMSAs

The interaction of GH3 nuclear proteins with the DNA fragment containing the E-box133 or E-box10 binding site was analyzed as described previously (14) by using a fluorescent-based, nonradioactive EMSA procedure. Nuclear protein extracts from GH3 cells were obtained using the nuclear extract-PER nuclear and cytoplasmic extraction reagents supplemented with two times the final concentration of the Halt protease inhibitor cocktail kit (Pierce, Rockford, IL). Protein concentrations of the nuclear extracts were determined using the Micro BCA protein assay reagent kit (Pierce). DNA-protein binding reactions consisted of incubation of 8–10 μg of the nuclear extract-PER nuclear extracts mixed with a reaction buffer composed of 10 mm Tris-HCl (pH 7.5), 50 mm KCl, 1 mm dithiothreitol, 50 ng/μl poly (deoxyinosine-deoxycytosine) (GE Healthcare, Piscataway, NJ), 7.5% glycerol, and 0.05% Nonidet P-40 for 10 min at room temperature.

This was followed by addition of 150 nm of DNA duplexes consisting of an annealed pairs of 5′-cy5-labeled oligonucleotides for the E-box133 site (forward, 5′-AATAA AATACCATTTGATGT-3′ and reverse, 5′-ACATCAAATGGTATTTTATT-3′) for the E-box10 site (forward, 5′-cy5/AGTCAA TGTCTGCAGATGAGAA-3′ and reverse, 5′-CY5/TCAGTTACAGACGTCTACTCTT-3′; both from Integrated DNA Technologies) or the stimulating protein-1 (SP1) site (forward, 5′-ATTCGATCGGGGCGGGGCGAGC-3′ and reverse, 5′-GCTCGCCCCGCCCCGATCAAT-3′; Promega Corp., Madison, WI) to the reaction mixture and incubating for 30 min at room temperature. For competition experiments, a 15- to 60-fold molar excess of unlabeled duplex DNA fragments corresponding to the E-box133, E-box10, or SP1 site were added to the reaction mixture before addition of the labeled oligonucleotides. For band shift assays, we added 4 μg of IgG affinity-purified goat (13) polyclonal antibody raised against BMAL1 (sc-8550x) (6,15), CRY1 (sc-5953) (16), PER1 (sc-7724x) (17), PER2 (sc-12429x) (18), PER3 (sc-12020x), or a control IgG (DREAM, sc-9309; Santa Cruz Biotechnology) to the binding reaction mixture and incubated for 40 min at room temperature before the addition of labeled oligos. Additional antibodies for these factors BMAL1 (sc-48790x), PER1 (sc-25362x) (19), PER2 (sc-7729x), PER3 (sc-8912x), or control IgG, DREAM (sc-9142), were also tested to confirm our EMSA results. These reaction mixtures were then subjected to nondenaturing 6% PAGE (Invitrogen) at 4 C with 0.25× solution of 90 mm Tris-borate, 2 mm EDTA (pH 8.3). The gels were run for 150 min at 120 V. Wet fluorescent gels were scanned using the Typhoon 9400 phosphoimager system (GE Healthcare) with a setting of 600–700 V and a scanning resolution of 200 μm.

Treatment with small interfering RNA (siRNA)

GH3 cells were seeded first in DMEM supplemented with 10% FBS but no Ab/Am and cultured overnight. The next day, the cells were transfected with 100 nm of siRNA for bmal1 (sc-77369), cry1 (sc-108035), per3 (sc-108003), or scrambled siRNA (sc-44232) (all from Santa Cruz) or per1 (129816A09) or per2 (129103A10) (from Invitrogen) using Lipofectamine-2000 (Invitrogen) according to the manufacture’s protocol and as reported by others (20). After 6 h of transfection, cells were incubated overnight in serum-free medium and serum shocked the next day as described above. Cells were harvested and total RNA was prepared 70 h after serum shock. PRL mRNA levels were measured by real-time RT-PCR.

Cell extraction and Western analysis

siRNA-treated and untreated (control) cells were harvested by trypsin-EDTA treatment, washed with ice-cold PBS, and centrifuged at 100 × g for 5 min at 4 C. Cells were placed into lysis buffer (Roche Applied Bioscience) containing 1 mm dithiothreitol and 0.1% protease inhibitor cocktail (Sigma-Aldrich Corp., St. Louis, MO). Lysates were incubated for 15 min on ice followed by centrifugation at 12,000 × g for 20 min at 4 C. Total protein recovered in the supernatant was estimated by Micro BCA protein assay (Pierce) and stored in aliquots at −80 C. For Western analysis, 40 μg of protein sample was mixed with 5× Laemmli sample buffer and boiled for 5 min followed by separation on a 4–12% gradient NUPAGE gel (Invitrogen). Proteins were transferred from the gel to a nitrocellulose membrane followed by blocking for 2 h at room temperature in 5% nonfat dried milk in buffer of Tris-buffered saline and Tween-20. Blots were then probed with primary antibody (anti-BMAL1, anti-CRY1, anti-PER1, anti-PER2, and anti-PER3; all from Santa Cruz Biotechnology) at 1:500–1000 dilutions. After washing, the membranes were incubated with antigoat antibody conjugated to alkaline-phosphatase (dilution 1:5000; Santa Cruz) and the signal visualized using CDP-Star chemiluminescence reagents (PerkinElmer, Boston, MA). As a control, blots were stripped and reprobed with mouse anti-α-tubulin (catalog no. T6199) primary antibody (1:5000; Sigma-Aldrich). The quantification of immunoblots was performed densitometrically using Image J gel analysis software (developed by the National Institutes of Health and available at http://rsb.info.nih.gov/ij). With this program, the protein signal of a target band was determined and normalized by comparison with the housekeeping protein from same sample.

Statistical analysis

Statistical differences in gel-shift band intensities for different circadian components and mRNA levels after siRNA treatment were assessed by one-way ANOVAs followed by the Newman-Keuls multiple comparison test. Differences were considered statistically significant at P < 0.05, unless noted otherwise.

Results

In our experiments, we used serum-shocked GH3 cells to explore the process underlying PRL GE pulse elaboration. As mentioned above, McFerran et al. (3) demonstrated that exposure to high concentrations of serum resulted in synchronization of individual GH3 cells with respect to PRL GE pulse activity. In a manner similar to this study, we exposed GH3 cells to 50% horse serum for 2 h followed by culture in serum-free medium for 84 h. During this period, total RNA was extracted from cells every 4 h followed by measurement of PRL mRNA by real-time RT-PCR. As shown in a representative example in Fig. 1, we found that PRL mRNA occurred in two distinct pulses, one from 44–56 h and another from 60–80 h. The PRL mRNA peak amplitudes were 8.06 ± 1.05- and 6.65 ± 0.87-fold for the first and second pulses, respectively, compared with basal expression (values represent the mean ± se for four separate experiments). Earlier studies from our laboratory raised the possibility that a relationship might exist between circadian clock components and PRL oscillatory behavior (6). To address this, we investigated first whether bmal1 may be expressed rhythmically with temporal patterns related to PRL pulse activity. Interestingly, we observed that bmal1 mRNA was expressed in a periodic fashion with two peaks occurring at 48 and 68 h (Fig. 1). Our analysis revealed increases of 3.00 ± 0.49-fold at 48 h and 3.70 ± 0.24-fold at 68 h when compared with basal gene expression. Surprisingly, a comparison of bmal1 and PRL mRNA expression revealed a close association between the two profiles with pulses for each gene occurring at approximately the same time.

Figure 1.

Figure 1

Open in a new tab

Pulse profile of PRL and bmal1 mRNA levels in GH3 cultures after serum-shock treatment. Cells in culture were exposed to medium containing 50% horse serum for 2 h. The medium was removed and replaced with serum-free medium. Cells were extracted for RNA from 4 to 84 h after treatment and mRNA levels for PRL and bmal1 were measured by real-time RT-PCR. Relative mRNA abundance is determined by comparing each value obtained with those measured at 4 h. The solid-lined profile depicts PRL mRNA levels and the dashed line represents bmal1 mRNA levels. The values represent the mean ± se of at least three replicates, and the profiles are representative of four separate experiments for PRL and three for bmal1. Each asterisk represents the peak value of a pulse of PRL mRNA. Note the differences in the y-axis for relative values of mRNA expression.

The next experiments were conducted to determine whether other circadian factors were expressed rhythmically and, if so, whether some type of temporal relationship could be identified between expression of these components and PRL. We found that three of the four factors tested, cry1, per1, and per3 mRNA, were expressed in pulses (Fig. 2). Cry1 exhibited mRNA peak expression at 16, 48, and 72 h (Fig. 2A) with mRNA levels elevated by 2.15 ± 0.25-fold at 16 h, 3.05 ± 0.25-fold at 48 h, and 5.35 ± 1.54-fold at 68 h over basal levels. The timing of the latter two pulses was similar to those found for PRL mRNA. Per1 was found to exhibit three mRNA pulses peaking at 24, 48, and 68 h (Fig. 2B). In this case, Per1 mRNA levels were increased by 4.04 ± 0.56-fold at 24 h, 4.80 ± 1.97-fold at 48 h, and 6.50 ± 0.97-fold at 64 h compared with basal gene expression. Again, the last two pulses had timing similar to that for the PRL mRNA pulses. Analysis of Per3 mRNA changes after serum shock revealed only one mRNA peak occurring at 72 h (Fig. 2C). The mRNA level was increased by 2.70 ± 0.88-fold over basal levels and occurred at a time similar to the latter PRL mRNA pulse. Our testing of a fourth circadian factor, Per2 mRNA, revealed no pulse activity after serum shock (Fig. 2D). (Values listed for the pulses above each represent the mean ± se of three separate experiments.)

Figure 2.

Figure 2

Open in a new tab

RNA profiles of various circadian elements in GH3 cultures after 50% serum shock: cry1 (A), per1 (B), per3 (C), and per2 (D). Each point represents the mean ± se of three replicates and is representative of three separate experiments. As before, the values presented are expressed relative to the first time period determined. An asterisk represents the a peak point within pulses of mRNA.

In our next studies, we investigated whether an interaction occurred between the PRL promoter and some or all of these circadian factors by performing ChIP analyses on GH3 cells using antibodies against BMAL1, PER1, PER2, PER3, and CRY1 for precipitation. The presence of the PRL promoter in conjunction with the precipitated factors was determined by PCR amplification, using primers specific for the 117-bp segment of the proximal portion of the rat PRL promoter. As shown in Fig. 3, a PCR product (117 bp) was observed in the sample precipitated with the anti-BMAL1 antibody (lane 1), anti-PER1 (lane 2), anti-PER2 (lane 3), anti-PER3 (lane 4), and CRY1 antibody (lane 5) when compared with goat IgG control (lane 6). From these results, it appears that each of these circadian factors tested associated with some region of the PRL promoter in these cells.

Figure 3.

Figure 3

Open in a new tab

The in vivo interaction of circadian factors with the rat PRL promoter by ChIP analysis. GH3 cells were cross-linked with formaldehyde, sheared using sonication, and immunoprecipitated. The chromatin cross-linking was reversed and then subjected to PCR using primers designed to amplify a 117-bp segment of the PRL proximal promoter region. Products immunoprecipitated with BMAL1 antibody (lane 1), anti-PER1 (lane 2), anti-PER2 (lane 3), anti-PER3 (lane 4), and anti-CRY1 (lane 5) are shown. Lane 6 represents antigoat IgG control (negative control), whereas total DNA input (lane 7) represents positive control subjected to PCR. These results are representative of those obtained in three separate experiments.

As we know from previous observations from our laboratory (6), there are two E-box elements in the initial portions of the proximal PRL promoter (Fig. 4), one of which (E-box133) was identified to be functionally active for PRL GE pulse activity and the other (E-box10) was not. Interestingly, the E-box133 site also appeared to complex with CLOCK and BMAL1. To test whether, in addition to BMAL1, other circadian components also interact with this E box-133 site, we performed EMSA analysis. In this study, 5′-cy5-labeled DNA probe (containing the E-box133 site) was incubated with nuclear proteins prepared from GH3 cells and separated by 6% DNA retardation gel electrophoresis. This resulted in the formation of multiple, DNA-protein complexes (Fig. 5A, lane 1).

Figure 4.

Figure 4

Open in a new tab

Diagram of a portion of the PRL promoter region containing Ebox-133 and Ebox-10. Ebox-133 and Ebox-10 binding sites are denoted by small black boxes on the line representing the initial proximal 331 bp of the rat PRL promoter. These boxes are expanded to show the CATTTG and CAGATG nucleotide sequences for Ebox-133 and Ebox10, respectively.

Figure 5.

Figure 5

Open in a new tab

EMSA analysis of circadian factors binding to the E-box133 and E-box10 sites of the rat PRL proximal promoter region. EMSAs were performed with nuclear protein extracts from GH3 cells and cy5-labeled oligonucleotides containing DNA probes with either the E-box133 (1) or the E-box10 binding site (2) (see Materials and Methods). A, Competition gel-shift assays were performed to test the specificity of binding of nuclear proteins with the probe. Incubation of the labeled E-box133 binding site probe with GH3 cell nuclear extract alone (lane 1), in the presence of a 15-fold (lane 2) or a 60-fold (lane 3) molar excess of unlabeled specific E-box133 oligonucleotide DNA duplex, or a 15- or 60-fold molar excess of nonspecific DNA probe, SP1 (lanes 4 and 5) are shown. Protein-DNA interactions are denoted by the appearance of two specific bands (C1 and C2). B, EMSAs were performed with labeled E-box133 and antigoat DREAM antibody (control) (lane 2) or BMAL1 antibody (lane 3) as described in Materials and Methods. C, EMSA analysis using labeled E-box133 and antibodies to DREAM (lane 2, control), CRY1 (lane 3), PER1 (lane 4), and PER3 (lane 5) antibodies. D, Band shift patterns after use of labeled E-box133 and DREAM (lane 2, control) and PER2 (lane 3) antibody. Analyses were also performed using labeled E-box10 with a 15-(lane 2) and 30-fold (lane 3) of unlabeled SP1 oligos or a 15- (lane 4) and 30-fold (lane 5) molar excess of unlabeled E-box10 (E). Finally, band patterns after incubation of cy5-labeled E-box10 binding site with nuclear extracts and antiserum for DREAM (lane 2, control), BMAL1 (lane 3), CRY1 (lane 4), PER1 (lane 5), PER2 (lane 6), and PER3 (lane 7) are shown (F). These results are representative of those obtained in three separate experiments.

To test the specificity of these complexes, gel-shift competition assays were performed. We found that complexes C1 and C2 were virtually eliminated when the reaction mixture was incubated with 15× (lane 2) and 60× (lane 3) unlabeled E-box133 DNA. Additionally, these complexes were unaffected when unrelated SP1 DNA oligos were used (lanes 4 and 5), establishing that these two complexes are specific for the E-box133 site. The other bands that appeared and were altered by treatment with unrelated oligos were deemed nonspecific and were not considered for further study. Then, we tested whether some of or all of the circadian factors found above to associate with the PRL promoter, also bound to the E-box133 site by using EMSA band-shift analysis with antibodies directed against these circadian elements. In these assays, the presence of a certain circadian factor in a binding complex was indicated by a shift in the amount or position of the complexes formed after use of an antibody to that factor in the binding reaction. As before, incubation of E-box133 DNA with GH3 nuclear extract yielded two strong binding complexes, C1 and C2 (Fig. 5B, lane 1). Addition of a nonspecific antibody, α-DREAM, had little impact on C1 or C2. However, our use of BMAL1 antibody in the binding reaction markedly reduced the amount and size of the binding complexes C1 and C2 by 57.52 ± 3.30 and 77.40 ± 3.50%, respectively (lane 3), indicating that BMAL1 was present in the complex. Similarly with the other factors tested, we observed reductions in the intensities of C1 and C2 with CRY1 antibody by 71.00 ± 3.70 and 78 ± 6.32% (lane 3), PER1 antibody by 63.41 ± 5.00 and 65.20 ± 7.75% (lane 4) and PER3 antibody by 54.23 ± 7.65 and 66.50 ± 1.12%, respectively (lane 5), compared with DREAM antibody control (Fig. 5C, lane 2). Interestingly, PER2 antibody did not show any detectable changes for either complex (Fig. 5D, lane 3). (The percentage values presented above and subsequently for EMSA each represent the mean ± se of three independent experiments.)

To determine whether this combination of factors binds to another E-box site or whether it was specific for the E-box133, EMSA analysis was performed using oligos for E-box10. As before, we established the specificity of DNA-protein interaction by gel-shift competition assays. Again, we found the formation of multiple complexes when cy5-labeled E-box10 oligos were incubated with GH3 cell nuclear extract (Fig. 5E, lane 1). Although incubation with nonspecific DNA (SP1) did not greatly impact the formation of C1 and C2 (lanes 2 and 3), we found that when incubating with unlabeled E-box10, a 15-fold molar excess decreased and a 30-fold excess abolished the formation of C1 and C2 (lanes 4 and 5). To identify the circadian proteins that associate with the E-box10 site, EMSA was performed in the presence of specific antibodies. As before, an example of these results (Fig. 5F), shows the formation of two major complexes, C1 and C2, when nuclear extract was incubated with E-box10 oligos (lane 1). The presence of BMAL1 in the complex was indicated by a reduction in the intensity of C2 by 33.20 ± 3.86% with BMAL1 antibody (lane 3). In addition, we found C1 and C2 were reduced by 53.2 ± 1.86 and 48.3 ± 3.12%, respectively, with CRY1 antibody (lane 4) compared with antigoat DREAM antibody control (lane 2). Interestingly, we observed no detectable change in the formation of C1 or C2 with PER1 antibody (lane 5). Moreover, we found that PER2 Ab reduced the binding of both C1 and C2 by 42.05 ± 6.58 and 45.05 ± 2.56% (lane 6), respectively, whereas PER3 antibody decreased C1 intensity by 76.05 ± 4.59% and abolished the formation of the C2 complex (lane 7). From these results, it appears that the combination of circadian elements that binds to E-box10 is not the same as that that binds to E-box133.

To test the requirement of these factors in stimulating PRL GE pulses, we determined whether reductions in expression of these circadian components may have any impact on PRL mRNA expression levels. Toward this goal, cultures of GH3 cells were first treated with 100 nm of siRNA for bmal1, cry1, per1, per2, or per3 followed by serum-shock treatment. Total RNA was extracted and real-time RT-PCR was performed for PRL at 72 h after shock, a time associated with pulse activity for most of the circadian factors. We found that knockdown of bmal1 expression by siRNA significantly reduced the PRL mRNA levels compared with that for cells treated with scrambled siRNA (Fig. 6A). As shown in Fig. 6B, PRL mRNA levels were reduced markedly after treatment with per1 siRNA. Interestingly, there was no significant change in PRL mRNA found after treatment with per2 siRNA. However, use of per3 and cry1 siRNA both decreased the levels of PRL mRNA. To determine whether siRNA influenced earlier pulses as well, we tested siRNA-treated cultures at 48 h after serum shock. We found decreases in PRL mRNA that were virtually identical with those above after siRNA treatment for circadian elements pulsing at 48 h. (Analysis revealed a decrease of 70 ± 10.6, 70 ± 5.4, and 46 ± 8.7% in PRL mRNA after treatment with siRNA for bmal1, per1, and cry1, respectively, when compared with scrambled control; n = 3 separate experiments.) As before, treatment with siRNA for per2 had no affect on PRL mRNA levels at 48 h. To confirm the knockdown efficiency of siRNA on these genes, Western blot analyses were done and examples of the results are presented (Fig. 6C). Our analysis reveals a decrease in BMAL1, CRY1, PER1, PER2, and PER3 protein by 67.09 ± 3.89, 68.23 ± 5.48, 65.00 ± 2.56, 78.93 ± 5.79, and 80.07 ± 4.03%, respectively, after treatment with the corresponding siRNA (these results represent the mean ± se, n = 3 independent experiments).

Figure 6.

Figure 6

Open in a new tab

Circadian factor expression after siRNA treatment. To determine the impact of specific circadian elements on PRL gene expression, GH3 cells transfected with siRNAs against specific circadian factors were serum shocked, and the RNA extracted and analyzed for PRL mRNA levels by real-time RT-PCR. A, Cultures treated with siRNA against bmal1 and compared with those treated with scrambled siRNA. Relative PRL expression was determined by real-time RT-PCR performed on samples 70 h after serum shock and expressed relative to values obtained from cells exposed to all reagents in the transfection procedure except for siRNA. B, Other cultures were transfected with scrambled siRNA or siRNA against per1, per2, per3, and cry1, serum shocked, and the cells extracted at 70 h after treatment for PRL mRNA analysis. Again the values presented are expressed relative to that found with scrambled siRNA. The asterisks denotes significant difference, P < 0.01, when compared with scrambled siRNA control. To confirm the knockdown efficiency of these genes with siRNA, Western blot analysis was done and an example of the results are presented (C). The top row of each blot contains protein immunoreactive for a specific circadian factors and the bottom row contains protein immunoreactive for α-tubulin. These results are representative of those obtained from three separate experiments.

Discussion

A growing body of evidence indicates that expression of several output genes are under the regulation of circadian elements (21,22). Our previous findings that PRL GE pulse expression requires the presence of a functional E-box binding site (E-box133) (6) indicated that PRL oscillatory behavior is linked mechanistically to the process of circadian oscillation. This, when coupled with observations that many of the genes identified as fundamental key regulators of circadian oscillation are expressed in GH3 cells, reinforced this possibility and prompted us to attempt to identify circadian factors that may be involved in this process. In this study, the use of serum-shocked GH3 cells, established previously to synchronize pulses in PRL cells, enabled us to not only identify circadian elements that appear to be associated with the timing of PRL pulse activity but also test whether they play a role in PRL GE. In our analysis of these factors, we found that the E-box133 site binds a specific array of circadian components. Interestingly, this group of factors was different from those that bound another site, E-box10, shown previously to be unrelated to PRL GE pulse activity (6). Finally, the functional importance of these E-box133 binding factors was revealed by the observation that reduction of any of these elements greatly reduced PRL message levels during pulses, suggesting that circadian input, especially these components, are critical for PRL pulse activity.

It has been shown that PRL GE occurs in pulses in individual pituitary cells, but the mechanism(s) underlying this dynamic process, especially the role of the circadian system, is not well understood. Our initial observations of pulsatile gene expression in pituitary cells from lactating rats were generated by microscopic observation of individual cells after microinjection with PRL promoter driven luciferase plasmid (5,6). Although this powerful tool enabled us to address aspects of cell function that were unable to be approached in another manner, we were limited in scope because of the technical difficulty and laboriousness of the studies. To more effectively study the mechanisms underlying this pulse activity, we used another approach. Recently McFerran et al. (3) demonstrated that exposure to medium containing high concentrations of horse serum causes a functional synchronization of individual GH3 cells, giving rise to coordinated oscillations of PRL-promoter-directed transcription. This extended earlier work by Balsalobre et al. (11), who observed circadian oscillations in clock genes following the identical serum-shock protocol. A treatment regimen that could synchronize PRL pulse activity between individual cells and induce circadian gene oscillations appeared to be an excellent model system to determine whether there was an the interaction between these two processes. Moreover, because these PRL episodes were specifically timed and cell components could be obtained at any time during this process, the use of this model afforded us the opportunity to expand our studies on the relationship between circadian elements and PRL gene expression with other molecular and biochemical tools. Using this approach, we observed that PRL mRNA occurred in two episodes consistent with the findings of others (3). Subsequent analysis of circadian expression after this treatment revealed that four of five factors tested were expressed rhythmically. This was not unexpected in light of observations that several circadian factors were reported to be expressed in a pulsatile fashion in other cell systems (23,24,25,26). More importantly, a comparison of these patterns revealed that one circadian factor (BMAL1) was expressed with a pattern that was virtually identical temporally to that of PRL mRNA and that three of four other factors analyzed had at least one pulse occurring at the same time as a PRL pulse. These findings, although correlative, strengthened the possibility that circadian elements were in some manner associated with the timing of PRL pulse activity.

It would stand to reason that a process that dictated the timing of PRL GE should involve the PRL promoter and that the functional association of circadian elements with PRL expression is likely to be mediated by this regulatory region. To address this, we performed ChIP assays using standard conditions and found that a 117-bp segment in the proximal PRL promoter was amplified when fragmented DNA was precipitated with antibodies to each circadian protein tested. ChIP analysis is routinely used to confirm the in vivo association of transcription factors with the promoter of the target gene (22,27). Thus, our findings demonstrate clearly that one or more transcriptional complexes containing these circadian proteins are associated with the PRL promoter in living GH3 cells.

A central element in the control of circadian timing is the DNA binding site, termed the E-box. It is has been shown in a number of systems that this type of promoter site serves as the focus of the binding and action of various circadian elements (28). Probably the most well-studied of these is the binding of CLOCK and BMAL1, considered fundamental to a functioning E-box (27,29,30,31). This and the association of other circadian components with the BMAL1-CLOCK complex differs from system to system and is thought to be part of the regulatory mechanism(s) underlying timing control (32,33,34,35). As mentioned above, previous observations from our laboratory identified two E-box elements in the proximal PRL promoter, one of which (E-box133) was identified by mutation analysis to be required for pulse activity and the other (E-box10) was not (6).

Our findings in this study of the binding of CRY1, PER1 and PER3, and BMAL1 but not PER2 to the E-box133 clearly demonstrate that a specific array of circadian elements associate with this active PRL promoter site. The requirement of this specific combination of factors to function is suggested by our observations that a different group of circadian elements binds to E-box10 (BMAL1, CRY1, PER2, and PER3 but not PER1), a site not associated with pulse activity. The importance of this group of E-box binding factors to PRL expression is further strengthened by our data revealing that each factor associated with the E-box133, when reduced (via siRNA treatment), resulted in a decrease of PRL mRNA expression. In contrast, a reduction in PER2 (associated with E-box10 and not E-box133) does not affect pulse-associated PRL mRNA levels. The idea of differential regulation from E-box to E-box for a specific gene has been observed in a few cases. For example, in C. elegans, it was found that within the arg-1 gene, one E-box appeared to be critical for general expression in muscle tissues, whereas two other E-boxes were found to be important for expression in distinct muscle types (10). Although very little information is available on the functional equivalency from E-box to E-box in mammalian systems, it is clear from recent studies with the nocturin gene promoter that certain E-boxes can bind higher levels of CLOCK and BMAL1 than others (27).

The reason for differential E-box activity in the same promoter is obscure. One intriguing study by Munoz et al. (36) reported that certain functional aspects of E-boxes in the arginine vasopressin gene promoter were related to surrounding base sequences. Their results revealed that the binding efficiency of certain E-boxes can be altered by changing adjacent base elements. In light of this in the PRL promoter, the ability of E-box133 and not E-box10 to contribute to pulse activity may reside in adjacent DNA binding sites that may influence the combination of factors that bind. Indeed, this part of the PRL promoter contains a number of binding sites that are associated with regulatory control factors such as TRH, dopamine, and insulin (37). Such sites may bring modulatory proteins in close proximity to the E-box and influence function. This could result in several potential changes such as an increase in activity via stabilization of a complex, a decrease in activity by disruption of the complex, or perhaps a coordination of E-box activity with other control elements of the promoter. The possibility that the E-box functions in concert with other aspects of the promoter is consistent with the concept of an enhanceosome proposed by Merika and Thanos (38), suggesting that promoter function depends on a complex of interacting factors. In fact, the integration of an E-box with other promoter elements may explain how rhythmic or timed function, residing primarily in the E-box apparatus, can be integrated with other aspects of gene function.

In summary, our study reveals that several circadian elements complex with specific E-box regions of the PRL promoter. Moreover, it appears that the number and type of factors that bind differ from one E-box region to another and that those that bind to the E-box133 site are necessary for PRL gene expression. These findings, when combined with our previous observations demonstrating the necessity of the E-box133 for pulse activity, suggest that a specific array of circadian elements acting through this E-box is central to the elaboration of a PRL GE pulse. Although intriguing, our results are based on analysis of a limited number of factors. In light of the wide variety of circadian elements have been identified, it would be reasonable to suggest that the full array of factors that impinge on a specific E-box may be more extensive and that the impact of these circadian elements on gene expression may be fundamental to the proper functioning of a number of cells and tissues. Circadian oscillations have been shown in a wide variety of tissues such as skin, bone marrow, retina, liver, heart, and kidney (39,40,41,42,43,44). Also, it appears that these factors can contribute to the expression of many types of genes. For example, Chong et al. (45) demonstrated that clock gene products play a vital role in activation of human plasminogen activator inhibitor-1 activity in myocardial infarction. Most recently it has been shown that BMAL1 is necessary for rhythmic expression of cell cycle inhibitor p21WAF1/C1P1 in mouse hepatocytes (46). Thus, the idea of an array of circadian elements acting on a promoter to influence the expression of a particular gene, especially its timing, may not only reveal an important aspect of PRL gene regulation but may also represent a fundamental process by which timing events can be integrated with other aspects of gene control in various cells and tissues.

Acknowledgments

The authors thank Dr. Gilles M. Leclerc, and Dr. Waleed Twaal for technical assistance, and Dr. Scott Argaves and Dr. E. K. Spicer for allowing access to real-time PCR and sonication equipment.

Footnotes

This work was supported by National Institutes of Health Grant DK 073270 (to F.R.B.).

Disclosure Summary: The authors have nothing to disclose.

First Published Online March 9, 2010

Abbreviations: Ab/Am, Antibiotics/antimycotics; BMAL, brain and muscle aryl hydrocarbon receptor nuclear translocator-like protein; ChIP, chromatin immunoprecipitation; CRY, cryptochrome; FBS, fetal bovine serum; GE, gene expression; PRL, prolactin; PER, period; siRNA, small interfering RNA; SP1, stimulating protein-1.

References

  1. Freeman ME, Kanyicska B, Lerant A, Nagy G 2000 Prolactin: structure, function, and regulation of secretion. Physiol Rev 80:1523–1631 [DOI] [PubMed] [Google Scholar]
  2. Leong DA, Frawley LS, Neill JD 1983 Neuroendocrine control of prolactin secretion. Annu Rev Physiol 45:109–127 [DOI] [PubMed] [Google Scholar]
  3. McFerran DW, Stirland JA, Norris AJ, Khan RA, Takasuka N, Seymour ZC, Gill MS, Robertson WR, Loudon AS, Davis JR, White MR 2001 Persistent synchronized oscillations in prolactin gene promoter activity in living pituitary cells. Endocrinology 142:3255–3260 [DOI] [PubMed] [Google Scholar]
  4. Takasuka N, White MR, Wood CD, Robertson WR, Davis JR 1998 Dynamic changes in prolactin promoter activation in individual living lactotrophic cells. Endocrinology 139:1361–1368 [DOI] [PubMed] [Google Scholar]
  5. Shorte SL, Leclerc GM, Vazquez-Martinez R, Leaumont DC, Faught WJ, Frawley LS, Boockfor FR 2002 PRL gene expression in individual living mammotropes displays distinct functional pulses that oscillate in a noncircadian temporal pattern. Endocrinology 143:1126–1133 [DOI] [PubMed] [Google Scholar]
  6. Leclerc GM, Boockfor FR 2005 Pulses of prolactin promoter activity depend on a noncanonical E-box that can bind the circadian proteins CLOCK and BMAL1. Endocrinology 146:2782–2790 [DOI] [PubMed] [Google Scholar]
  7. Okamura H, Yamaguchi S, Yagita K 2002 Molecular machinery of the circadian clock in mammals. Cell Tissue Res 309:47–56 [DOI] [PubMed] [Google Scholar]
  8. Reppert SM, Weaver DR 2002 Coordination of circadian timing in mammals. Nature 418:935–941 [DOI] [PubMed] [Google Scholar]
  9. Richter HG, Torres-Farfán C, Rojas-García PP, Campino C, Torrealba F, Serón-Ferré M 2004 The circadian timing system: making sense of day/night gene expression. Biol Res 37:11–28 [DOI] [PubMed] [Google Scholar]
  10. Zhao J, Wang P, Corsi AK 2007 The C. elegans Twist target gene, arg-1, is regulated by distinct E box promoter elements. Mech Dev 124:377–389 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Balsalobre A, Damiola F, Schibler U 1998 A serum shock induces circadian gene expression in mammalian tissue culture cells. Cell 93:929–937 [DOI] [PubMed] [Google Scholar]
  12. Fink L, Seeger W, Ermert L, Hänze J, Stahl U, Grimminger F, Kummer W, Bohle RM 1998 Real-time quantitative RT-PCR after laser-assisted cell picking. Nat Med 4:1329–1333 [DOI] [PubMed] [Google Scholar]
  13. Merriam GR, Wachter KW 1982 Algorithms for the study of episodic hormone secretion. Am J Physiol 243:E310–E318 [DOI] [PubMed] [Google Scholar]
  14. Leclerc GM, Bose SK, Boockfor FR 2008 Specific GATA-binding elements in the GnRH promoter are required for gene expression pulse activity: role of GATA-4 and GATA-5 in this intermittent process. Neuroendocrinology 88:1–16 [DOI] [PubMed] [Google Scholar]
  15. Oishi K, Shirai H, Ishida N 2005 CLOCK is involved in the circadian transactivation of peroxisome-proliferator-activated receptor α (PPARα) in mice. Biochem J 386:575–581 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Naruse Y, Oh-hashi K, Iijima N, Naruse M, Yoshioka H, Tanaka M 2004 Circadian and light-induced transcription of clock gene Per1 depends on histone acetylation and deacetylation. Mol Cell Biol 24:6278–6287 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Mansuy V, Risold PY, Glauser M, Fraichard A, Pralong FP 2009 Expression of the GABAA receptor associated protein Gec1 is circadian and dependent upon the cellular clock machinery in GnRH secreting GnV-3 cells. Mol Cell Endocrinol 307:68–76 [DOI] [PubMed] [Google Scholar]
  18. Chen ST, Choo KB, Hou MF, Yeh KT, Kuo SJ, Chang JG 2005 Deregulated expression of the PER1, PER2 and PER3 genes in breast cancers. Carcinogenesis 26:1241–1246 [DOI] [PubMed] [Google Scholar]
  19. Vieira E, Nilsson EC, Nerstedt A, Ormestad M, Long YC, Garcia-Roves PM, Zierath JR, Mahlapuu M 2008 Relationship between AMPK and the transcriptional balance of clock-related genes in skeletal muscle. Am J Physiol Endocrinol Metab 295:E1032–E1037 [DOI] [PubMed] [Google Scholar]
  20. Toda K, Yamamoto D, Fumoto M, Ikeshita N, Herningtyas EH, Iida K, Takahashi Y, Kaji H, Chihara K, Okimura Y 2008 Involvement of mPOU (Brn-5), a class VI POU protein, in the gene expression of Pit-1 as well as PRL. Mol Cell Endocrinol 280:20–29 [DOI] [PubMed] [Google Scholar]
  21. Lemos DR, Goodspeed L, Tonelli L, Antoch MP, Ojeda SR, Urbanski HF 2007 Evidence for circadian regulation of activating transcription factor 5 but not tyrosine hydroxylase by the chromaffin cell clock. Endocrinology 148:5811–5821 [DOI] [PubMed] [Google Scholar]
  22. Matsunaga N, Ikeda M, Takiguchi T, Koyanagi S, Ohdo S 2008 The molecular mechanism regulating 24-hour rhythm of CYP2E1 expression in the mouse liver. Hepatology 48:240–251 [DOI] [PubMed] [Google Scholar]
  23. Dorton AM 2000 The pituitary gland: embryology, physiology, and pathophysiology. Neonatal Netw 19:9–17 [DOI] [PubMed] [Google Scholar]
  24. Hirota T, Okano T, Kokame K, Shirotani-Ikejima H, Miyata T, Fukada Y 2002 Glucose down-regulates Per1 and Per2 mRNA levels and induces circadian gene expression in cultured Rat-1 fibroblasts. J Biol Chem 277:44244–44251 [DOI] [PubMed] [Google Scholar]
  25. Oishi K, Miyazaki K, Uchida D, Ohkura N, Wakabayashi M, Doi R, Matsuda J, Ishida N 2009 PERIOD2 is a circadian negative regulator of PAI-1 gene expression in mice. J Mol Cell Cardiol 46:545–552 [DOI] [PubMed] [Google Scholar]
  26. Yoo SH, Ko CH, Lowrey PL, Buhr ED, Song EJ, Chang S, Yoo OJ, Yamazaki S, Lee C, Takahashi JS 2005 A noncanonical E-box enhancer drives mouse Period2 circadian oscillations in vivo. Proc Natl Acad Sci USA 102:2608–2613 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Li R, Yue J, Zhang Y, Zhou L, Hao W, Yuan J, Qiang B, Ding JM, Peng X, Cao JM 2008 CLOCK/BMAL1 regulates human nocturin transcription through binding to the E-box of nocturin promoter. Mol Cell Biochem 317:169–177 [DOI] [PubMed] [Google Scholar]
  28. Gachon F, Nagoshi E, Brown SA, Ripperger J, Schibler U 2004 The mammalian circadian timing system: from gene expression to physiology. Chromosoma 113:103–112 [DOI] [PubMed] [Google Scholar]
  29. Lowrey PL, Takahashi JS 2004 Mammalian circadian biology: elucidating genome-wide levels of temporal organization. Annu Rev Genomics Hum Genet 5:407–441 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Muñoz E, Baler R 2003 The circadian E-box: when perfect is not good enough. Chronobiol Int 20:371–388 [DOI] [PubMed] [Google Scholar]
  31. Ripperger JA, Schibler U 2001 Circadian regulation of gene expression in animals. Curr Opin Cell Biol 13:357–362 [DOI] [PubMed] [Google Scholar]
  32. Bertolucci C, Cavallari N, Colognesi I, Aguzzi J, Chen Z, Caruso P, Foá A, Tosini G, Bernardi F, Pinotti M 2008 Evidence for an overlapping role of CLOCK and NPAS2 transcription factors in liver circadian oscillators. Mol Cell Biol 28:3070–3075 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Fujimoto K, Hamaguchi H, Hashiba T, Nakamura T, Kawamoto T, Sato F, Noshiro M, Bhawal UK, Suardita K, Kato Y 2007 Transcriptional repression by the basic helix-loop-helix protein Dec2: multiple mechanisms through E-box elements. Int J Mol Med 19:925–932 [PubMed] [Google Scholar]
  34. Nonaka H, Emoto N, Ikeda K, Fukuya H, Rohman MS, Raharjo SB, Yagita K, Okamura H, Yokoyama M 2001 Angiotensin II induces circadian gene expression of clock genes in cultured vascular smooth muscle cells. Circulation 104:1746–1748 [DOI] [PubMed] [Google Scholar]
  35. Sato F, Kawamoto T, Fujimoto K, Noshiro M, Honda KK, Honma S, Honma K, Kato Y 2004 Functional analysis of the basic helix-loop-helix transcription factor DEC1 in circadian regulation. Interaction with BMAL1. Eur J Biochem 271:4409–4419 [DOI] [PubMed] [Google Scholar]
  36. Muñoz E, Brewer M, Baler R 2006 Modulation of BMAL/CLOCK/E-Box complex activity by a CT-rich cis-acting element. Mol Cell Endocrinol 252:74–81 [DOI] [PubMed] [Google Scholar]
  37. Gourdji D, Laverrière JN 1994 The rat prolactin gene: a target for tissue-specific and hormone-dependent transcription factors. Mol Cell Endocrinol 100:133–142 [DOI] [PubMed] [Google Scholar]
  38. Merika M, Thanos D 2001 Enhanceosomes. Curr Opin Genet Dev 11:205–208 [DOI] [PubMed] [Google Scholar]
  39. Allaman-Pillet N, Roduit R, Oberson A, Abdelli S, Ruiz J, Beckmann JS, Schorderet DF, Bonny C 2004 Circadian regulation of islet genes involved in insulin production and secretion. Mol Cell Endocrinol 226:59–66 [DOI] [PubMed] [Google Scholar]
  40. Bjarnason GA, Jordan RC, Wood PA, Li Q, Lincoln DW, Sothern RB, Hrushesky WJ, Ben-David Y 2001 Circadian expression of clock genes in human oral mucosa and skin: association with specific cell-cycle phases. Am J Pathol 158:1793–1801 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Lopez-Molina L, Conquet F, Dubois-Dauphin M, Schibler U 1997 The DBP gene is expressed according to a circadian rhythm in the suprachiasmatic nucleus and influences circadian behavior. EMBO J 16:6762–6771 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Sakamoto K, Nagase T, Fukui H, Horikawa K, Okada T, Tanaka H, Sato K, Miyake Y, Ohara O, Kako K, Ishida N 1998 Multitissue circadian expression of rat period homolog (rPer2) mRNA is governed by the mammalian circadian clock, the suprachiasmatic nucleus in the brain. J Biol Chem 273:27039–27042 [DOI] [PubMed] [Google Scholar]
  43. Sun ZS, Albrecht U, Zhuchenko O, Bailey J, Eichele G, Lee CC 1997 RIGUI, a putative mammalian ortholog of the Drosophila period gene. Cell 90:1003–1011 [DOI] [PubMed] [Google Scholar]
  44. Zanello SB, Jackson DM, Holick MF 2000 Expression of the circadian clock genes clock and period1 in human skin. J Invest Dermatol 115:757–760 [DOI] [PubMed] [Google Scholar]
  45. Chong NW, Codd V, Chan D, Samani NJ 2006 Circadian clock genes cause activation of the human PAI-1 gene promoter with 4G/5G allelic preference. FEBS Lett 580:4469–4472 [DOI] [PubMed] [Google Scholar]
  46. Gréchez-Cassiau A, Rayet B, Guillaumond F, Teboul M, Delaunay F 2008 The circadian clock component BMAL1 is a critical regulator of p21WAF1/CIP1 expression and hepatocyte proliferation. J Biol Chem 283:4535–4542 [DOI] [PubMed] [Google Scholar]

Articles from Endocrinology are provided here courtesy of The Endocrine Society

RESOURCES