MCB
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Other Versions of this Article:
MCB.02155-06v1
27/13/4876    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lim, C.
Right arrow Articles by Choe, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lim, C.
Right arrow Articles by Choe, J.

 Previous Article  |  Next Article 

Molecular and Cellular Biology, July 2007, p. 4876-4890, Vol. 27, No. 13
0270-7306/07/$08.00+0     doi:10.1128/MCB.02155-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Functional Role of CREB-Binding Protein in the Circadian Clock System of Drosophila melanogaster{triangledown}

Chunghun Lim, Jongbin Lee, Changtaek Choi, Juwon Kim, Eunjin Doh, and Joonho Choe*

Department of Biological Sciences, Korea Advanced Institute of Science and Technology, Daejeon 305-701, South Korea

Received 17 November 2006/ Returned for modification 8 January 2007/ Accepted 9 April 2007


   ABSTRACT
Top
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Rhythmic histone acetylation underlies the oscillating expression of clock genes in the mammalian circadian clock system. Cellular factors that contain histone acetyltransferase and histone deacetylase activity have been implicated in these processes by direct interactions with clock genes, but their functional relevance remains to be assessed by use of appropriate animal models. Here, using transgenic fly models, we show that CREB-binding protein (CBP) participates in the transcriptional regulation of the Drosophila CLOCK/CYCLE (dCLK/CYC) heterodimer. CBP knockdown in pigment dispersing factor-expressing cells lengthens the period of adult locomotor rhythm with the prolonged expression of period and timeless genes, while CBP overexpression in timeless-expressing cells causes arrhythmic circadian behaviors with the impaired expression of these dCLK/CYC-induced clock genes. In contrast to the mammalian circadian clock system, CBP overexpression attenuates the transcriptional activity of the dCLK/CYC heterodimer in cultured cells, possibly by targeting the PER-ARNT-SIM domain of dCLK. Our data suggest that the Drosophila circadian clock system has evolved a distinct mechanism to tightly regulate the robust transcriptional potency of the dCLK/CYC heterodimer.


   INTRODUCTION
Top
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The circadian clock is an evolutionarily conserved system that perceives environmental time cues, synchronizes the organism's inherent clock with external time, and exhibits the circadian physiology of organisms (e.g., diurnal or nocturnal locomotor activities) (14, 40, 52, 59). At the cellular level of circadian clock systems in animals, pacemaker cells in a small subset of brain neurons display a robust oscillation of clock gene products and dominate the circadian behaviors of the organism by governing the peripheral clock systems. These core clock cells correspond to ventral lateral neurons (LNvs) and the suprachiasmatic nucleus in Drosophila and mouse circadian clock systems, respectively (21, 23, 58). Drosophila LNvs express the neuropeptide pigment dispersing factor (pdf) gene, which is implicated in the synchronization of clock cells (33, 41, 42, 44). At the molecular level, some core clock genes are periodically expressed, and their rhythmic expression is maintained under free-running conditions (i.e., exclusion of external time cues). This molecular clock system involves several transcription factors, protein kinases, phosphatases, and proteosomal pathway components, which together mediate the transcriptional regulation of clock transcripts and control the posttranslational localization, quantity, and quality of clock proteins (19, 45).

Published data suggest that two interlocking feedback loops maintain the oscillating expression of core clock genes in Drosophila melanogaster and mouse. In Drosophila, a heterodimer of the dClock (dClk) and cycle (cyc) gene products activates the transcription of the period (per), timeless (tim), vrille (vri), and Par domain protein 1{varepsilon} (Pdp1{varepsilon}) genes during subjective night by binding to E-box sequences within their promoters (2, 6, 11, 12, 48). The PER/TIM heterodimer or the PER monomer then translocates to the nucleus and inhibits the transcriptional activity of the dCLK/CYC heterodimer, subsequently repressing the transcription of the per and tim genes (8, 12, 29, 39, 47, 49). In the second feedback loop, VRI and PDP1{varepsilon} bind to a site within the dClk promoter, repressing and activating the transcription of dClk gene, respectively (11, 17).

As far as the mammalian circadian clock system is concerned, it has been shown that the oscillating expression of the mPer and mCry genes is based on the rhythmic histone acetylation of their promoter regions (15). This acetylation may involve the intrinsic histone acetyltransferase activity of CLOCK protein (13) and/ or transcriptional coactivators such as CREB-binding protein (CBP), p300, and p300/CBP-associated factor (p/CAF), since they have been shown to augment the transcriptional activity of the CLOCK/BMAL1 heterodimer (10, 56). In addition, it was recently reported that mCRY1 might attenuate the transcriptional activity of the CLOCK/BMAL1 heterodimer through interactions with the components of a corepressor complex that contains histone deacetylase activity (38). However, no appropriate animal model was available for investigating the functional role of either histone acetyltransferases or histone deacetylases in a circadian clock system. In this study, we adopted transgenic fly models in which Drosophila CBP expression can be up- or downregulated in a tissue-specific manner via a GAL4/upstream activation sequence (UAS) system (7) and characterized their circadian behaviors as well as the molecular clocks in their pacemaker neurons. In contrast to the mammalian circadian clock system, both our in vivo and our in vitro data indicate that CBP may function as a negative regulator of the dCLK/CYC heterodimer, a Drosophila homolog of the mammalian CLOCK/BMAL1 heterodimer.


   MATERIALS AND METHODS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Plasmids. Total RNA from adult fly heads was isolated using the TRIzol reagent (Invitrogen) and reverse transcribed using Moloney murine leukemia virus (M-MuLV) reverse transcriptase according to the manufacturer's instructions (Roche). Pdp1{varepsilon} and cyc cDNAs were amplified by PCR using gene-specific primer sets, inserted into pAc5/V5-His (Invitrogen) for V5- and His-tagged expression in Schneider 2 (S2) cells, and confirmed by sequencing. The dClk, per, and mouse PKA catalytic subunit cDNAs were similarly cloned into pAc5/V5-His. The dClk cDNA was also inserted into pAc/FLAG, a modified version of pAc5/V5-His, to express N-terminally FLAG-tagged dCLK protein in S2 cells. Mammalian CBP cDNA with a C-terminal stop codon (32) was inserted into pAc5/V5-His and therefore could not express either the C-terminal V5 tag or the His tag. The per-luc, tim-luc, dClk-luc, and ATFx3-luc constructs have been described previously (11, 12, 31). The cDNAs corresponding to the deletion mutants of dCLK were amplified with the appropriate primer sets and inserted into pGEX 4T-1 (Amersham Biosciences) for expression of a glutathione S-transferase (GST) fusion protein in bacteria.

Drosophila stocks. All flies were reared with standard cornmeal-yeast-agar medium at 25°C under light-dark (LD; 12 h of light and 12 h of darkness) cycles. GMR-GAL4, UAS-green fluorescent protein (GFP), and UAS-GFPRNAi lines were obtained from the Bloomington Drosophila Stock Center. EP element insertion lines including EP1179 and EP1149, GAL4 driver lines including pdf-GAL4, tim-GAL4-62, tim-GAL4-86, Mz520-GAL4, Mai179-GAL4, and C929-GAL4, and GAL80 repressor lines including pdf-GAL80 and cry-GAL80 have been described previously (18, 23, 36, 44, 53). pdf-GAL4 is expressed in both large and small LNvs (44). Two tim-GAL4 lines are expressed in all known clock neurons including both large and small LNvs, dorsal LNs (LNds), and three types of dorsal neurons (DNs) (23). Mz520-GAL4 is expressed similarly to pdf-GAL4 (18). Mai179-GAL4 is expressed in a subset of LNvs and LNds (18, 51). C929-GAL4 is expressed in large LNvs as well as in 100 peptidergic noncircadian neurons of the adult fly brain (18, 55). For the construction of transgenic flies in which CBP expression is downregulated via a GAL4/UAS system, a double-stranded RNA (dsRNA) construct for CBP was designed according to the genomic cDNA hybrid method (22). The DNA fragment from bp 6721 to 7454 of CBP cDNA and genomic DNA including the cDNA with internal and adjacent 3' introns were amplified by PCR from pDF378 (16) and fly genomic DNA, respectively. The fragments were ligated together into the pUAST vector, and the transgenic construct was injected with pUCHsp{pi}{Delta}2-3 into w1118 embryos, from which several germ line transformants were established. All experiments were performed using three independent lines containing the UAS-CBPRNAi construct on the third chromosome, which gave consistent results. Data from a representative line are shown.

Behavioral analysis. The locomotor activities of individual male flies were measured using Drosophila activity monitors (Trikinetics). Monitoring conditions included LD cycles for 2 to 4 days, followed by constant-dark (DD) cycles for 4 to 7 days. Data were analyzed using ClockLab analysis software (Actimetrics). Rhythmic flies were defined as described previously (61), except that the significance level of the {chi}2 periodogram was set at an {alpha} value of 0.05. Data were pooled from more than three independent experiments. The average locomotor activity profile for each genotype was analyzed using Microsoft Excel. Locomotor activity per half-hour bin was normalized by the average locomotor activity per day, and the relative locomotor activities per half-hour bin for individual flies were averaged for each genotype. For eclosion rhythm analysis, third-instar larvae or early pupae, reared under LD cycles, were individually placed in glass tubes. Their circadian rhythms were further entrained to LD cycles for 2 days and then subjected to DD cycles. Under DD conditions, the eclosion time was determined using Drosophila activity monitors and analyzed as described previously (35).

In situ hybridization. In situ hybridization using larval brains was performed as described previously (57) with minor modifications. RNA probes for in situ hybridization were transcribed in vitro using MAXIscript (Ambion) in the presence of digoxigenin (DIG)-11-UTP (Roche), precipitated with ethanol, dissolved in 1x hybridization buffer (5x SSC [1x SSC is 0.15 M NaCl plus 0.015 M sodium citrate], pH 5.0, 0.1% Tween 20, 50 µg/ml heparin, 50 µg/ml sonicated salmon sperm DNA, and 50% formamide), and stored at –70°C until use. The hybridizing region of the tim gene has been described previously (50). The probe for the Pdp1{varepsilon} gene encompasses the mRNA region from nucleotide +565 to +765 relative to the transcription start site, while the probe for the pdf gene encompasses the full-length cDNA. After hybridization, DIG-labeled probes were detected colorimetrically using an alkaline phosphatase-conjugated anti-DIG antibody and a nitroblue tetrazolium-5-bromo-4-chloro-3-indolyl phosphate stock solution (Roche).

Immunofluorescence assay (IFA) and image analysis. Adult flies were fixed in phosphate-buffered saline (PBS) containing 4% formaldehyde for 2.5 h, and their brains were subsequently dissected in PBS. Larval brains were dissected in PBS and fixed in PBS containing 4% formaldehyde for 1 h. After permeabilization with PBS containing 0.1% Triton X-100 (PBS-T), brain samples were blocked in PBS-T containing 2% bovine serum albumin and incubated overnight with primary antibodies in blocking solution at 4°C. After repeated washes, samples were incubated with secondary antibodies in PBS-T for 2 h, washed extensively, and mounted using Vectashield (Vector Laboratories). S2 cells were grown on coverslips and transfected using the calcium precipitation method. At approximately 40 h after transfection, cells were fixed in PBS containing 3.7% formaldehyde for 30 min and permeabilized with PBS containing 0.2% Triton X-100 at 4°C for 25 min. Blocking and antibody incubation were performed similarly with brain samples, except that the primary antibody was incubated at room temperature for 2 h. The primary antibodies used were a guinea pig anti-CBP antibody (30), a rat anti-PDF antibody (41), rabbit anti-PER, anti-dCLK, and anti-PDF antibodies (see below), rabbit anti-acetyl histone H3 and H4 antibodies (Upstate Biotechnology), and a mouse anti-V5 antibody (Invitrogen). The secondary antibodies used were rhodamine-conjugated anti-guinea pig, anti-rabbit, anti-mouse, and anti-rat antibodies and a fluorescein isothiocyanate (FITC)-conjugated anti-rabbit antibody (Jackson ImmunoResearch Laboratories). Samples were examined with a BX51 fluorescence microscope (Olympus) and a Pascal confocal laser scanning microscope (Carl Zeiss). For the quantitative analysis, the intensity of PER staining signals in each group of clock cells was quantified using ImageJ software. After subtraction of the background intensity, the total intensity value for each group of clock cells was averaged from the values of eight brains at each circadian time.

Cell culture, transfection, and semiquantitative RT-PCR. Drosophila S2 cells were maintained in Shields and Sang M3 insect medium (Sigma) supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin (Invitrogen). Cells were transiently transfected using the standard calcium precipitation method. For semiquantitative reverse transcription-PCR (RT-PCR), total RNA from transfected cells was isolated using the TRIzol reagent (Invitrogen). After the removal of contaminating genomic DNA by DNase I digestion, RNA was reverse transcribed with M-MuLV reverse transcriptase and oligo(dT) primers according to the manufacturer's instructions (Promega). The cDNA for each clock gene was amplified by PCR with a gene-specific primer set under nonsaturating conditions. PCR products were resolved by 1% agarose gel electrophoresis, stained with ethidium bromide, and photographed after exposure to UV light.

ChIP assay. Chromatin immunoprecipitation (ChIP) assays were performed according to the manufacturer's instructions (Upstate Biotechnology) with minor modifications. S2 cells in 6-well plates were cotransfected with the expression vectors for V5-tagged dCLK and/or CBP. At approximately 40 h after transfection, cells were cross-linked with 1% formaldehyde at 25°C for 10 min. Cross-linked chromatin was sheared by sonication and immunoprecipitated with a rabbit anti-acetyl H3 antibody (Upstate Biotechnology) or a guinea pig anti-dCLK antibody (24). Immunoprecipitated DNA was analyzed by semiquantitative PCR with primers specific for each clock gene. The amplified region of the vri gene promoter was from –1.9 kb to –1.64 kb relative to the transcription start site, which includes two canonical E-box sequences. The amplified region of the Pdp1{varepsilon} gene promoter was from –1.05 kb to –0.75 kb, which includes two canonical E-box sequences. The amplified region of the dClk gene promoter was from –0.2 kb to +0.18 kb, which includes no canonical E-box sequence. Representative data from two independent experiments are shown.

Antibody production, immunoprecipitation, and Western blotting. The C-terminal 544 amino acids of the PER protein, the N-terminal 388 amino acids of the PDP1{varepsilon} protein, and the C-terminal 287 amino acids of the dCLK protein fused to GST were expressed in bacteria, purified using glutathione-Sepharose 4B beads (Amersham Biosciences), and used for the immunization of rabbits by subcutaneous injection. For the priming injection, the proteins were mixed with complete Freund's adjuvant (Sigma). After 1 month, the rabbits were injected four times, at 1-week intervals, with proteins in incomplete Freund's adjuvant (Sigma). Peptides corresponding to the C-terminal 20 amino acids of TIM (26) or PDF protein were synthesized (Anygen, Korea) and similarly used to immunize guinea pigs or rabbits, respectively. Polyclonal antibodies were affinity purified from rabbit antisera, dialyzed, and stored at –70°C. For immunoprecipitation from cultured cell extracts, transfected cells were lysed in a lysis buffer (25 mM Tris-Cl, pH 7.5, 300 mM NaCl, 10% glycerol, 1 mM EDTA, 1 mM dithiothreitol, 0.5% Nonidet P-40, and 1 mM phenylmethylsulfonyl fluoride) at 4°C for 20 min. After clarification by centrifugation, soluble extracts were diluted to adjust salt concentration (final concentration, 100 mM NaCl) and then incubated with 2 µl of guinea pig anti-CBP serum (30) or control guinea pig serum at 4°C for 1.5 h. For immunoprecipitation from fly head extracts, approximately 50 µl of adult fly heads was homogenized in the same lysis buffer and incubated at 4°C for 20 min. After two clarifications by centrifugation, soluble extracts were similarly diluted and then incubated with 3 µl of guinea pig anti-dCLK serum (24) or control guinea pig serum at 4°C for 1.5 h. After the addition of preequilibrated protein A Sepharose (Amersham Biosciences), extracts were further incubated at 4°C for 1.5 h. The beads were then washed three times in the same buffer. Bound proteins were eluted, resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, transferred to a Hybond-C membrane (Amersham Biosciences), and analyzed by Western blotting. Proteins were detected using rabbit anti-CBP (43), anti-dCLK, anti-PER, and anti-PDP1{varepsilon} sera, a mouse anti-V5 antibody, and guinea pig anti-TIM serum with ECL Plus reagents (Amersham Biosciences).

In vitro binding assay. GST fusion proteins were incubated with [35S]methionine-labeled proteins synthesized by the TNT T7 coupled transcription-translation reticulocyte lysate system according to the manufacturer's instructions (Promega). After a 30-min incubation at room temperature in a binding buffer (25 mM Tris-Cl, pH 7.5, 150 mM NaCl, 10% glycerol, 1 mM EDTA, 1 mM dithiothreitol, 0.5% Nonidet P-40, and 1 mM phenylmethylsulfonyl fluoride), glutathione-Sepharose 4B beads were added and incubated further for 30 min at room temperature. The beads were washed four times in the same buffer. Bound proteins were eluted, resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and subjected to autoradiography. For competition experiments, 2 to 5 µg of maltose-binding protein (MBP) or the cysteine-histidine-rich 3 (C/H3) region of CBP fused to MBP was preincubated with GST-dCLK protein for 10 min at room temperature before the addition of in vitro-translated CYC protein.


   RESULTS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Construction of CBP mutant flies via a GAL4/UAS system. To investigate the possible role of CBP in the Drosophila circadian clock system, we first established transgenic fly models in which CBP expression is up- or downregulated. It was previously reported that two EP lines, EP1179 and EP1149, contain an EP element insertion in the promoter of the CBP gene and can be used for CBP overexpression by a GAL4 driver (36). Meanwhile, we found that the EP1149 line contains an SLIH mutation in the per gene (reference 20 and data not shown), and this line was excluded from further study. Flies hemizygous for a CBP null mutation have been shown to display embryonic lethality (1), making it difficult to study their circadian behavior. To knock down CBP expression in a cell type-specific manner, we made a transgenic fly in which dsRNA for CBP was expressed using a GAL4/UAS system. Accordingly, the EP1179 and UAS-CBPRNAi lines were crossed with circadian clock-related GAL4 drivers, and their progeny was characterized.

Using an IFA, we first confirmed GAL4 driver-specific CBP overexpression and knockdown in these transgenic flies. As shown in Fig. 1A, CBP was overexpressed in tim-expressing cells of the EP1179 line by tim-GAL4. Endogenous CBP was barely identifiable in tim-expressing cells of control flies, making it difficult to directly assay the knockdown by CBP dsRNA in these clock cells. However, we found that a nuclear staining signal with the anti-CBP antibody was barely detectable in the ring gland by expression of CBP dsRNA, suggesting that endogenous CBP expression was reduced in the transgenic fly. In addition, using the abnormal eye phenotype induced by CBP overexpression and knockdown (27, 34), we genetically validated that the CBP dsRNA used in our study specifically reduces CBP expression. An eye-specific expression of CBP dsRNA but not CG17100 dsRNA by GMR-GAL4 rescued the smooth-eye phenotype induced by CBP overexpression, while CBP overexpression but not PDP1{varepsilon} overexpression rescued the abnormal eye phenotype by CBP knockdown (Fig. 1B; also data not shown). Taken together, we verified two transgenic fly models in which cell type-specific CBP expression could be up- or downregulated by a GAL4 driver, and we subsequently analyzed their respective circadian behaviors.


Figure 1
View larger version (58K):
[in this window]
[in a new window]

  		FIG. 1. Construction of transgenic flies in which CBP expression is either up- or downregulated via a GAL4/UAS system. (A) Clock cell-specific overexpression and knockdown of CBP by a tim-GAL4 driver. Larval brains were immunostained with a guinea pig anti-CBP antibody and a rhodamine-conjugated anti-guinea pig antibody (red), while tim-expressing cells were visualized by GFP expression (green). Confocal images of one hemisphere (upper panels) and the ring gland (lower panels) in each larval brain are shown. The genotypes of transgenic flies are given on the left. l-LN, larval lateral neuron; mGFP, myristylated GFP. (B) Eye-specific overexpression and knockdown of CBP by a GMR-GAL4 driver. Eye-specific expression of CBP dsRNA results in an abnormal phenotype, which is rescued by coexpression of CBP. CBP overexpression by a GMR-GAL4 driver also causes a smooth-eye phenotype. Images from light microscopy (LM) (upper panel) and scanning electron microscopy (SEM) (lower panel) are shown. The genotypes of transgenic flies are given at the top.

 
Circadian behaviors of CBP mutant flies. For behavioral analysis, we crossed several GAL4 drivers with the UAS-CBPRNAi and EP1179 lines, and their progeny was monitored for adult locomotor activity under both LD and DD cycles. Expression from a single copy of the UAS-CBPRNAi transgene by circadian clock-related GAL4 drivers had no apparent effect on circadian behavior (Table 1). Of note, tim-GAL4-mediated expression of CBP dsRNA decreased longevity, making it impossible to estimate the rhythmicity and period of adult locomotor activity under free-running conditions. However, CBP knockdown from two copies of the UAS-CBPRNAi transgene by LNv-specific GAL4 drivers (e.g., pdf-GAL4, Mz520-GAL4, and Mai179-GAL4) lengthened the period of adult locomotor activity (Fig. 2A). Because pdf-GAL4-mediated expression of GFP dsRNA did not significantly affect the adult locomotor rhythm, it seems unlikely that the altered circadian behavior by CBP knockdown would be an artifact arising from the nonspecific overexpression of dsRNAs in the pacemaker cells. In contrast, CBP overexpression in tim-expressing cells dramatically reduced the rhythmicity of adult locomotor activity under DD cycles (Fig. 2B). Although transgenic flies in which CBP was overexpressed by Mai179-GAL4 and C929-GAL4 showed more arrhythmic behavior than did control flies (Table 2), the averaged actograms of these transgenic flies were comparable to those of control flies (data not shown). When tim-GAL4-mediated expression of CBP was suppressed in pdf-expressing cells by the pdf-GAL80 transgene (53), the locomotor rhythm was partially rescued. However, the suppression of tim-GAL4-mediated CBP overexpression in cryptochrome (cry)-expressing cells by the cry-GAL80 transgene dramatically recovered rhythmicity. A profile analysis of locomotor activity revealed that CBP overexpression by tim-GAL4 resulted in a dampened evening peak of locomotor activity during LD cycles, which was also rescued by the cry-GAL80 transgene (data not shown). CBP overexpression in tim-expressing cells also disrupted eclosion rhythm, an additional circadian behavior of Drosophila, suggesting that CBP overexpression may generally disturb the Drosophila circadian clock system (Fig. 2C).


View this table:
[in this window]
[in a new window]

  		TABLE 1. Locomotor activity analysis of transgenic flies in which CBP knockdown is driven by specific GAL4 driversa

 

Figure 2
View larger version (43K):
[in this window]
[in a new window]

  		FIG. 2. Circadian behaviors of CBP mutant flies. (A) Actogram analysis of transgenic flies in which CBP expression is downregulated by LNv-specific GAL4 drivers. The locomotor activities of flies with the same genotype were averaged and double plotted. The genotypes and numbers of transgenic flies used in the analysis are given at the top. The rhythmicity and the period (with the standard error of the mean) of adult locomotor activity are also shown. (B) Actogram analysis of transgenic flies in which CBP is overexpressed in tim-expressing cells. Where indicated, CBP overexpression by a tim-GAL4 driver is suppressed by either a pdf-GAL80 or a cry-GAL80 transgene. (C) Eclosion activity profile analysis of transgenic flies in which CBP is overexpressed in tim-expressing cells. After entrainment to LD cycles, the eclosion activities of individual pupae were monitored during subsequent DD cycles. The relative eclosion activity at each circadian time was calculated by normalizing to the total eclosion activity.

 

View this table:
[in this window]
[in a new window]

  		TABLE 2. Locomotor activity analysis of transgenic flies in which CBP overexpression is driven by specific GAL4 driversa

 
Rhythmic expression of clock genes is altered in clock neurons of CBP mutant flies. To dissect the molecular mechanisms underlying the circadian behaviors of CBP mutant flies, we analyzed clock gene expression in these transgenic fly models. Since a transgene under the control of the GAL4/UAS system would be heterogeneously expressed by a circadian clock-related GAL4 driver, we reasoned that clock gene expression in GAL4-expressing cells could be differentially affected by the transgene. More importantly, it has been shown that the altered expression of clock genes only in LNvs can significantly affect the period and rhythmicity of adult locomotor rhythm (18, 53). Because a quantitative analysis using head extracts would mask any subtle changes in clock gene expression in each group of pacemaker cells, we focused on clock gene expression in individual clock cells of CBP mutant flies.

Since the initial phase of clock gene expression would be daily synchronized by the light during LD cycles, we thought that the effect of CBP knockdown on circadian gene transcription could be directly evaluated under these conditions. Therefore, we first examined the rhythmic expression of the tim transcript in larval LNs during LD cycles. An in situ hybridization assay revealed that the tim transcript in larval LNs expressing CBP dsRNA persisted throughout later zeitgeber (ZT; lights on at ZT time zero and lights off at ZT12 during LD cycles) times, suggesting that the rhythmic expression of the tim transcript may be prolonged by CBP knockdown (Fig. 3A). Also, we examined and quantified the rhythmic PER expression in clock cells of adult brains at different circadian times during DD cycles. We concentrated on PER expression in LNvs and LNds, because these clock cells function as the predominant pacemakers of the Drosophila circadian clock system (18, 53). As shown in Fig. 3B, CBP knockdown by pdf-GAL4 delayed and sustained the phase of cycling PER expression in PDF-positive small LNvs, while it constitutively elevated PER expression in large LNvs. Of note, the initial phase of rhythmic PER expression in PDF-negative small LNvs and LNds of mutant flies was similar to that of control flies, but its phase was extended during the time that PER expression in PDF-positive small LNvs of mutant flies displayed its peak level.


Figure 3
View larger version (75K):
[in this window]
[in a new window]

  		FIG. 3. Rhythmic expression of clock genes is prolonged by CBP knockdown. (A) Oscillating expression of a tim transcript is detectable in larval LNs expressing CBP dsRNA throughout later ZT times during LD cycles. Male flies homozygous for both the pdf-GAL4 and the UAS-CBPRNAi transgene were sterile (data not shown). For the simplicity of the assay, female flies homozygous for both the pdf-GAL4 and the UAS-CBPRNAi transgene were crossed with male flies homozygous for the UAS-CBPRNAi transgene, and their progeny was subsequently analyzed. Of note, these transgenic flies also display a long period rhythm of locomotor activity (see Table 1). Third-instar larvae were dissected at the indicated times during the LD cycle. In situ hybridization was performed using a antisense probe for the tim gene. The genotypes of transgenic flies are given on the left. Four representative images of larval LNs at each ZT are shown. (B) Oscillating expression of PER protein is delayed and elevated by CBP knockdown. Adult flies were entrained to LD cycles, transferred to DD conditions, and fixed at the indicated times during the first and second DD cycles. Dissected brains were immunostained with a rabbit anti-PER antibody and an FITC-conjugated anti-rabbit antibody (green) and with a rat anti-PDF antibody and a rhodamine-conjugated anti-rat antibody (red). Representative confocal images of the anterior brains are shown (upper panels). The genotypes of transgenic flies are given at the top. PDF-negative small LNvs are indicated by white arrows. The intensity of PER staining signals in each group of clock cells was quantified using ImageJ software and averaged from the values for eight brains at each circadian time (lower panel; graph). Error bars, standard errors of the means. l-LNv, large LNv; s-LNv, small LNv.

 
In contrast to CBP knockdown, the rhythmic expression of tim and Pdp1{varepsilon} transcripts in larval LNs of CBP-overexpressing flies was barely detectable during both LD and DD cycles (Fig. 4A). We also performed an IFA to monitor the rhythmic expression of clock proteins in adult brains of CBP-overexpressing flies. The PER protein in adult brains was not observed at ZT8 (data not shown), but it was strongly expressed in a different subset of clock cells at ZT0 during LD cycles (Fig. 4B). CBP overexpression by two copies of pdf-GAL4 slightly reduced PER expression only in pdf-expressing cells, whereas CBP overexpression by a single copy of tim-GAL4 abolished PER expression in all LNvs. In addition, the number and intensity of PER protein signals in LNds and DNs were decreased by tim-GAL4-mediated CBP overexpression (Fig. 4B; also data not shown). In the rescue experiment, the pdf-GAL80 transgene rescued PER expression only in pdf-expressing cells, while the cry-GAL80 transgene recovered it in all clock cells of CBP-overexpressing flies. Similar results were obtained during DD cycles, except that PER expression in large LNvs was significantly weakened (data not shown). Meanwhile, we found that the expression level of PDF protein was also reduced in LNvs of CBP-overexpressing flies, making their neural projections undetectable by immunostaining with an anti-PDF antibody.


Figure 4
View larger version (72K):
[in this window]
[in a new window]

  		FIG. 4. Rhythmic expression of clock genes is abolished by CBP overexpression. (A) Oscillating expression of tim (upper panel) and Pdp1{varepsilon} (lower panel) transcripts is not detectable in larval LNs of CBP-overexpressing flies. Third-instar larvae were dissected at the indicated times during the LD cycle or the first DD cycle. In situ hybridization was performed using antisense probes for the tim and Pdp1{varepsilon} genes. Of note, some DNs of CBP-overexpressing flies also did not display oscillating expression of the tim transcript. It was difficult to accurately identify DN1s and DN2s among several DNs expressing the Pdp1{varepsilon} transcript. Only one hemisphere of each larval brain is shown. l-LN, larval LN; DN1 and DN2, DNs that express the tim transcript in phase and out of phase, respectively, with l-LN. (B) The expression level of PER protein is reduced by CBP overexpression. Adult flies were fixed at ZT0 during LD cycles. Dissected brains were immunostained with a rabbit anti-PER antibody and an FITC-conjugated anti-rabbit antibody (green) and with a rat anti-PDF antibody and a rhodamine-conjugated anti-rat antibody (red). Representative confocal images of the anterior brains are shown. White arrows indicate PDF-negative small LNvs. l-LNv, large LNv; s-LNv, small LNv.

 
The impaired expression of clock genes raised the possibility that these clock cells would be absent or abnormally developed in CBP-overexpressing flies. An in situ hybridization assay revealed that LNs were present in larval brains of CBP-overexpressing flies but that the pdf gene transcript was weakly expressed in the clock cells, as was observed for Clk and cyc mutant flies (6, 41) (Fig. 5A). In contrast, dCLK protein displayed comparable expression levels in larval brains of control and CBP-overexpressing flies (Fig. 5B). The presence of LNvs in adult brains of CBP-overexpressing flies could be also visualized either by coexpression with GFP or by detecting the elevated histone acetylation state with an anti-acetyl histone H4 antibody (34) (Fig. 5C and D). Taken together, these data suggest that the abnormal circadian behaviors of CBP mutant flies may arise from alteration of dCLK/CYC-dependent clock gene expression in clock cells but not from absence of the pacemaker cells or reduced expression of dCLK protein.


Figure 5
View larger version (73K):
[in this window]
[in a new window]

  		FIG. 5. Pacemaker cells are normally developed in larval and adult brains of CBP-overexpressing flies. (A) The expression level of the pdf transcript is reduced by CBP overexpression. In situ hybridization using an antisense probe for the pdf gene was performed for larval brains. While the pdf transcript was detected comparably in the ventral ganglion cells of control and CBP-overexpressing flies (upper panel, black arrows), larval LNs expressing the pdf transcript were barely identifiable in CBP-overexpressing flies under the same conditions (upper panel, white rectangles). However, weak expression of the pdf gene in larval LNs of CBP-overexpressing flies was observed during a longer detection time. The genotypes of transgenic flies are given at the top. Images outlined in the upper panel are shown at a higher magnification in the lower panel. (B) The expression level of dCLK protein is not altered by CBP overexpression. Dissected larval brains were immunostained with a rabbit anti-dCLK antibody and a rhodamine-conjugated anti-rabbit antibody (red), while tim-expressing cells were visualized by GFP expression (green). Under our experimental conditions, we observed dCLK expression in larval LNs, DN1s, and DN2s. Also, weak signals of anti-dCLK antibody staining were detected in some tim-negative cells. l-LN, larval lateral neuron. l-LN, DN1, and DN2 are as defined for Fig. 4. (C) Reduced expression of PDF protein is not due to an absence of LNvs in CBP-overexpressing flies. Dissected adult brains were immunostained with a rabbit anti-PDF antibody and a rhodamine-conjugated anti-rabbit antibody (red), while tim-expressing cells were visualized by GFP expression (green). (D) The acetylation state of histone H4 protein is elevated by CBP overexpression. Dissected adult brains were immunostained with a rabbit anti-acetyl histone H4 antibody and a rhodamine-conjugated anti-rabbit antibody (red), while tim-expressing cells were visualized by GFP expression (green). l-LNv, large LNv; s-LNv, small LNv.

 
The transcriptional activity of dCLK/CYC is inhibited by CBP overexpression. Our in vivo analysis of clock gene expression suggested that CBP may function as a negative regulator of the dCLK/CYC heterodimer. To validate this hypothesis, we adopted an in vitro cell culture system comprised of the S2 cell line derived from Drosophila embryos. It was previously shown that S2 cells endogenously express the cyc gene, and therefore the ectopic expression of dCLK is sufficient for the activation of dCLK/CYC-dependent transcription from both reporter plasmids and endogenous promoters (12, 37). Because CBP knockdown in S2 cells retarded cell growth and significantly inhibited the transient expression of dCLK under our experimental conditions (data not shown), it was difficult to directly evaluate the effect of CBP knockdown on the transcriptional activity of the dCLK/CYC heterodimer. As shown in Fig. 6A, CBP overexpression suppressed dCLK-activated transcription from reporter plasmids containing per and tim gene promoters in a dose-dependent manner. Under similar conditions, CBP and protein kinase A (PKA) synergically activated transcription from a reporter plasmid containing tandem CREB/ATF-binding sites (Fig. 6B, ATFx3-luc). We could also observe moderate activation of per-luc by CBP and PKA. In view of the fact that CBP functions as a transcriptional coactivator of the classic CREB/PKA pathway (28), these data are consistent with the previous observations that per-luc reporter expression is reduced in dCREB2 mutant flies and that there are three putative CREB-binding sites within 4 kb to 1.2 kb upstream of the transcription start site for the per gene (5). The dCLK-induced expression of endogenous clock genes in S2 cells was also repressed by CBP overexpression at their transcript and protein levels (Fig. 6C and D), further confirming our in vivo observations. Taken together, these data suggest that CBP overexpression may specifically inhibit dCLK/CYC-dependent transcription.


Figure 6
View larger version (38K):
[in this window]
[in a new window]

  		FIG. 6. The transcriptional activity of dCLK/CYC is inhibited by CBP overexpression. (A) CBP overexpression inhibits the transcriptional activation of the per and tim reporter plasmids by dCLK expression. S2 cells in 6-well plates were cotransfected with a reporter plasmid (1 µg), a V5-tagged dCLK expression vector (10 ng for per-luc and 1 ng for tim-luc), and increasing amounts of a CBP expression vector (0.1 µg and 1 µg). Extracts from transfected cells were prepared at approximately 40 h after transfection, and luciferase assays were performed. The cell extracts were also used in Western blotting to confirm that CBP overexpression does not affect the expression level of V5-tagged dCLK protein (data not shown). Levels of activation were calculated by normalizing values to the luciferase activity in the presence of the reporter plasmid, which was set to 1. Results are averages from three independent experiments. Error bars, standard deviations. (B) CBP and PKA synergically activate transcription from ATF/CREB-responsive reporter plasmids. S2 cells in 6-well plates were cotransfected with a reporter plasmid (1 µg), a V5-tagged PKA expression vector (100 ng), and increasing amounts of a CBP expression vector (0.1 µg and 1 µg). Luciferase assays, Western blotting, and calculation of increases in activation were performed as for panel A. (C) CBP overexpression inhibits the transcriptional activation of endogenous clock gene promoters by dCLK expression. S2 cells in 6-well plates were cotransfected with a V5-tagged dCLK expression vector (10 ng) and/or increasing amounts of a CBP expression vector (0.25 µg and 1 µg). Total RNA was prepared at approximately 40 h after transfection, and cDNAs were synthesized using M-MuLV reverse transcriptase. Semiquantitative PCR was performed using the clock gene-specific primers shown on the left. (D) CBP overexpression inhibits the dCLK-induced expression of endogenous clock proteins. S2 cells in 6-well plates were cotransfected with a V5-tagged dCLK expression vector (+, 10 ng; ++, 50 ng) and/or increasing amounts of a CBP expression vector (0.25 µg and 1 µg). Total extracts from transfected cells were prepared at approximately 40 h after transfection, resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and immunoblotted with the clock gene-specific antibodies shown on the left.

 
CBP physically interacts with dCLK. Recently, it was reported that dCLK localized to nuclear spots in transiently expressed S2 cells (25). We also found the nuclear spots of dCLOCK in a subset of transfected cells highly expressing dCLK (data not shown), but they became more obvious in the presence of CBP (Fig. 7A, upper panel). For the present, it is difficult to evaluate their physiological significance. Since CYC overexpression weakened the punctate localization of dCLK in the presence of CBP (Fig. 8B), one possibility would be that the nuclear spots serve as a subnuclear reservoir for CYC-free dCLK. Interestingly, CBP colocalized with dCLK protein but not with other clock proteins in the nuclear spots of transfected cells (Fig. 7A, lower panel), suggesting that dCLK may associate with CBP in transiently expressed cells. As expected, dCLK protein was coimmunoprecipitated with an anti-CBP antibody, but not with a control antibody, from transfected cell extracts (Fig. 7B). At two different circadian times, we could also observe a physical interaction between CBP and dCLK in adult head extracts of wild-type but not ClkJrk flies (Fig. 7C), indicating that the reduced expression level and/or aberrant conformation of the dCLK mutant protein would diminish its association with CBP. To map the interaction domain between CBP and dCLK, an in vitro binding assay was performed using wild-type dCLK protein and deletion mutants. As shown in Fig. 7D, the PER-ARNT-SIM (PAS) A domain of dCLK commonly interacted with in vitro-translated PER, CYC, and CBP proteins. In a reciprocal in vitro binding assay, we found that dCLK protein associates with the evolutionarily conserved C/H3 region of CBP (data not shown), which was sufficient for binding to the PAS A domain of dCLK.


Figure 7
View larger version (39K):
[in this window]
[in a new window]

  		FIG. 7. CBP physically interacts with dCLK. (A) dCLK colocalizes with CBP in the punctate nuclear spots of transfected cells. S2 cells grown on coverslips were cotransfected with expression vectors for V5-tagged clock proteins and/or a CBP expression vector. An IFA was performed at approximately 40 h after transfection. (Upper panel) V5-tagged dCLK was immunostained with a mouse anti-V5 antibody (rhodamine; red), while acetylated histone H3 proteins were immunostained with a rabbit anti-acetyl H3 antibody (FITC; green). (Lower panel) V5-tagged proteins were immunostained with a mouse anti-V5 antibody (rhodamine; red), while CBP was immunostained with a rabbit anti-CBP antibody (FITC; green). Images were obtained using confocal laser scanning microscopy. Differential interference contrast images are overlaid with merged confocal images. (B) dCLK protein associates with endogenous CBP in transiently transfected S2 cells. S2 cells were transfected with a V5-tagged dCLK expression vector or a blank vector. At approximately 40 h after transfection, soluble extracts from transfected cells were immunoprecipitated with a guinea pig anti-CBP serum or a control serum. Coimmunoprecipitated proteins were immunoblotted with mouse anti-V5 and rabbit anti-CBP antibodies. INPUT, approximately 4% of soluble extracts used for immunoprecipitation. (C) CBP is coimmunoprecipitated with dCLK from fly head extracts. Adult flies were entrained to LD cycles, transferred to DD conditions, and harvested at the indicated times during the first DD cycle. Soluble extracts from adult fly heads were prepared and immunoprecipitated with a guinea pig anti-dCLK antibody. Bound proteins were immunoblotted with rabbit anti-CBP and anti-dCLK antibodies. The genotypes of adult flies are given at the top. Wild-type dCLK and mutant dCLKJrk proteins are indicated by single and double asterisks, respectively. INPUT, approximately 5% of soluble extracts used for immunoprecipitation. (D) The PAS A domain of dCLK is targeted by CBP. 35S-labeled in vitro-translated proteins were incubated with dCLK and with its deletion mutants fused to GST protein. Proteins pulled down by glutathione-Sepharose 4B were subjected to autoradiography. A schematic diagram of dCLK deletion mutants used in this assay is shown on the left. bHLH, basic helix-loop-helix; polyQ, glutamine repeat; KIX, CREB-binding domain; INPUT, 10% of in vitro-translated protein used in the binding assay.

 

Figure 8
View larger version (42K):
[in this window]
[in a new window]

  		FIG. 8. CBP disrupts the association of dCLK with CYC. (A) The C/H3 region of CBP is sufficient for the competition with CYC for binding to dCLK protein. GST or GST-dCLK was preincubated with increasing amounts of MBP or the C/H3 region of CBP fused to MBP (2 µg and 10 µg) and was then further incubated with 35S-labeled in vitro-translated CYC. After pulldown with glutathione-Sepharose 4B, bound proteins were subjected to autoradiography. INPUT, 20% of in vitro-translated CYC protein included in the binding assay. (B) CBP overexpression inhibits the nuclear translocation of CYC by dCLK in transiently expressed S2 cells. S2 cells grown on coverslips were cotransfected with a V5-tagged CYC expression vector (50 ng), a FLAG-tagged dCLK expression vector (50 ng), and/or a CBP expression vector (1 µg). IFA was performed at approximately 40 h after transfection. V5-tagged CYC was immunostained with a mouse anti-V5 antibody (rhodamine; red), while FLAG-tagged dCLK was immunostained with a rabbit anti-dCLK antibody (FITC; green). Transfected cells were examined using confocal laser scanning microscopy. Representative results are shown in the upper panel, and differential interference contrast images are overlaid with merged confocal images. In addition, the subcellular localization of V5-tagged CYC protein in the absence or presence of dCLK and CBP was scored for each cell (n = 460 to 520) and quantitatively analyzed. The percentages of cells in which CYC protein is evenly expressed in both the nucleus and cytoplasm, is expressed more strongly in the nucleus than in the cytoplasm, and is expressed exclusively in the nucleus are indicated by red, yellow, and green bars, respectively, in the lower panel. (C) CBP overexpression inhibits the association of dCLK with endogenous clock gene promoters. S2 cells in 6-well plates were cotransfected with a V5-tagged dCLK expression vector (10 ng) and/or a CBP expression vector (1 µg). Cross-linked chromatin was prepared at approximately 40 h after transfection, immunoprecipitated with an anti-acetyl histone H3 (upper panel) or an anti-dCLK (lower panel) antibody, and amplified by PCR with each clock gene-specific primer shown on the left. PCR products were resolved by 6% acrylamide gel electrophoresis and visualized by ethidium bromide staining.

 
CBP disrupts the association of dCLK with CYC. Since both CBP and CYC target the PAS A domain of dCLK, we reasoned that they could compete with each other for association with dCLK. As shown in Fig. 8A, a molar excess of the C/H3 region of CBP fused to MBP, but not MBP itself, inhibited the heterodimerization of dCLK and CYC proteins in an in vitro binding assay. We also observed that the exclusive nuclear localization of CYC in the presence of dCLK (25) was suppressed by CBP overexpression (Fig. 8B), suggesting that CBP may disrupt the association of dCLK with CYC in transiently expressed S2 cells. Finally, a ChIP assay revealed that the association of dCLK with the vri and Pdp1{varepsilon} gene promoters was reduced by CBP overexpression (Fig. 8C, lower panel). Interestingly, the acetylation state of histone proteins in the endogenous promoter of the Pdp1{varepsilon} gene, but not the vri gene, was increased by dCLK expression (Fig. 8C, upper panel). Under our experimental conditions, the vri transcript was readily detectable even in the absence of dCLK, and its expression was moderately elevated by dCLK, while expression of the Pdp1{varepsilon} transcript, in contrast, was dramatically induced by dCLK (Fig. 6C). Therefore, it is possible that the striking activation of the Pdp1{varepsilon} gene promoter by dCLK would be accompanied by an increased acetylation state of histone proteins. Taken together, these data suggest that CBP may inhibit dCLK/CYC-induced transcription by disrupting the formation of the functional heterodimer capable of binding to target promoters and activating their transcription.


   DISCUSSION
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
To elucidate the functional role of Drosophila CBP in the circadian clock system, we generated transgenic fly models in which CBP expression can be up- or downregulated in clock cells using a GAL4/UAS system, and we characterized their circadian behaviors and molecular clocks. Our behavioral analyses revealed that CBP knockdown in pdf-expressing cells resulted in a long period rhythm of adult locomotor activity, with a concomitant prolongation and delay of the phase of rhythmic clock gene expression. Although the possibility that CBP knockdown may increase the stability of the tim transcript and/or PER protein cannot be excluded, our in vivo data support the hypothesis that dCLK/CYC-dependent clock gene expression may be derepressed by CBP knockdown. We propose that the extended phase of clock gene expression due to CBP knockdown during the subjective night of the LD cycle may be reset daily by the light but that it may subsequently delay the following phase of clock gene expression and adult locomotor rhythm during DD cycles. In addition, we found that the phase of rhythmic PER expression in PDF-negative small LNvs and LNds was similarly delayed at a specific circadian time by CBP knockdown in pdf-expressing cells, supporting the previous observations that (i) molecular clocks in PDF-negative small LNvs and LNds are in phase and (ii) PDF-positive LNvs control the phase of the circadian clock in LNds (46, 54).

CBP overexpression in tim-expressing cells causes arrhythmic circadian behaviors, but CBP overexpression in LNvs had no apparent effect. Since tim-GAL4 is believed to express more GAL4 proteins in LNvs than pdf-GAL4 (54), these results suggest two possibilities: (i) that CBP is not sufficiently overexpressed by LNv-specific GAL4 drivers to affect the locomotor rhythm and/or (ii) that clock cells other than LNvs may be responsible for the arrhythmic circadian behavior. In the rescue experiment, the suppression of tim-GAL4-mediated CBP expression by pdf-GAL80 partially rescued the arrhythmic behavior of mutant flies, while suppression by cry-GAL80 recovered a normal locomotor rhythm, indicating that both pdf- and cry-expressing clock cells contribute to the arrhythmic behavior due to CBP overexpression. In addition, the average profile of adult locomotor activity showed that CBP overexpression by tim-GAL4 dampens the evening peak of locomotor activity during LD cycles, which was rescued only by cry-GAL80. Accumulating data suggest that the morning and evening peaks of adult locomotor activity are governed by pdf-expressing LNvs (PDF+ cells) and by clock cells that express the cry gene but not the pdf gene (CRY+/PDF– cells), respectively (18, 53, 54). With regard to this model, our profile analyses support the hypothesis that the oscillator function of CRY+/PDF– cells is disrupted by CBP overexpression.

Molecular analysis of clock gene expression in CBP-overexpressing flies revealed that CBP overexpression abolishes dCLK/CYC-dependent clock gene expression. Although CBP overexpression by 2 copies of pdf-GAL4 reduced PER expression in pdf-expressing cells, CBP overexpression by a single copy of tim-GAL4 completely abolished it, suggesting that CBP may not be sufficiently overexpressed by pdf-GAL4 to affect the molecular clock as well as circadian behavior. However, we cannot exclude the possibility that aberrant clock gene expression in CRY+/PDF– cells due to CBP overexpression influences PER expression in PDF+ cells. We also showed that PER expression in PDF– small LNvs and LNds was recovered only by cry-GAL80, in agreement with a recent report that these clock cells may be responsible for the evening peak of locomotor rhythm (46). Of note, LNvs were more sensitive to the effect of CBP overexpression than LNds and DNs. We reason that the expression level of CBP by tim-GAL4 and/or the nature of the circadian clock system differs among clock cells, resulting in the differential effects of CBP overexpression. It is also possible that CBP may have different roles in the central and peripheral clock systems, as recently exemplified by the cry gene (9).

Flies homozygous for the pdf null mutation gradually lose the rhythmicity of adult locomotor activity under free-running conditions (44), in contrast to the locomotor activity of CBP-overexpressing flies, which becomes arrhythmic immediately after transfer to DD cycles. In addition, the rhythmicity of adult locomotor activity was not recovered when the pdf gene was ectopically coexpressed with CBP by tim-GAL4 (Table 2), suggesting that the reduced expression of the pdf gene due to CBP overexpression may not be primarily responsible for the arrhythmic behavior of the mutant fly, which may instead result from the impaired activity of the dCLK/CYC heterodimer (6, 41). In contrast to CBP overexpression, pdf gene expression was negligibly affected by CBP knockdown. Considering that the pdf gene is strongly expressed in LNvs (41), it is possible that pdf gene expression may already be saturated in wild-type flies and/or that the sensitivity of our assay systems may be insufficient for detecting the subtle difference. Taken together, our data indicate that CBP overexpression abolishes the oscillating expression of clock genes by the dCLK/CYC heterodimer, thereby eliminating the locomotor rhythm of adult flies immediately after transfer to DD cycles. This notion is further supported by the previous observations that behavioral and molecular clocks in Clk mutant flies similarly display the arrhythmic phenotype (2, 3).

By in vitro experiments, we showed that CBP directly targets the PAS domain of dCLK, thereby inhibiting the dimerization, DNA binding, and transcriptional activity of the dCLK/CYC heterodimer. These unexpected data contrast with those for the mammalian circadian clock system, in which CBP/p300 associates with the CLOCK/BMAL1 heterodimer and augments its transcriptional activity (10, 15, 56). Although the protein interaction domains of CLOCK/BMAL1 and CBP/p300 were not extensively determined in the mammalian circadian clock system, this difference may originate from the evolutionary divergence of Clock orthologs in the circadian clock system. The PAS A domains of mammalian and Drosophila Clock genes (41% identity; 61% similarity) are less conserved than the PAS B domains (64% identity; 80% similarity), while the C-terminal glutamine-rich domains display little homology. Transient reporter assays in previous studies showed that the dCLK/CYC heterodimer activates transcription from E-box-containing reporters more robustly than the mammalian CLOCK/BMAL1 heterodimer (10, 12, 15, 56). When dCLK is fused to a GAL4 DNA-binding domain and subsequently tethered to a minimal promoter downstream of tandem GAL4-binding sites, it also strongly activates transcription from the artificial promoter (our unpublished observation); in contrast, mammalian CLOCK displays no transactivation activity under similar conditions (56). Furthermore, the ectopic expression of dCLK can induce the rhythmic expression of clock genes in misexpressed cells of adult fly brains or can cause developmental lethality (61), suggesting that the transcriptional activity of the dCLK/CYC heterodimer would be tightly controlled in a wild-type fly. A prior work showed that dCLK is bound to CYC throughout a daily cycle (4). However, dCLK may not be efficiently extracted from head homogenates under the experimental conditions of that study, since it was recently revealed that dCLK is constitutively expressed with the rhythmic phosphorylation state in fly head extracts (25, 60). We also observed that the extractability of dCLK protein from head homogenates is largely dependent on the buffer composition (our unpublished observation). Therefore, the expression level of dCLK would have been underestimated in previous reports. We propose that CBP constitutively restricts dCLK so that only a subpopulation of dCLK would form a functional complex with CYC and participate in transcriptional activity. This strategy would block the catastrophic induction of dCLK/CYC target genes before PER protein is sufficiently expressed to inhibit the transcriptional activity of the dCLK/CYC heterodimer, and it would enable dCLK/CYC target genes to maintain their highly oscillating expression throughout a daily cycle.

Taken together, our results provide genetic and molecular evidence that CBP may function as a negative regulator of the dCLK/CYC heterodimer. Since CBP participates in diverse biological processes, we cannot exclude the possibility that it also plays a functional role in other transcriptional feedback loops of core clock genes or in the input and output pathways of the Drosophila circadian clock system.


   ACKNOWLEDGMENTS
 
We thank J. Blau, S. Kay, C. Lee, and J. Ma for providing us with plasmid DNAs; R. Allada, K.-T. Min, J. Park, M. Rosbash, and F. Rouyer for transgenic and clock mutant flies; I. Edery and E. Y. Kim for the anti-dCLK antibody; M. Mannervik and A. Mazo for the anti-CBP antibody; and J. Park for the anti-PDF antibody.

This work was supported by grants from the Brain Research Center (M103KV010003-06K2201-00310) of the 21st Century Frontier Research Program, funded by the Ministry of Science and Technology, Republic of Korea; from the Basic Research Promotion Fund (KRF-2005-201-C00035) of the Korea Research Foundation, funded by the Ministry of Education and Human Resources Development, Republic of Korea; and from the Korea Science and Engineering Foundation through the Protein Network Research Center at Yonsei University.


   FOOTNOTES
 
* Corresponding author. Mailing address: Department of Biological Sciences, Korea Advanced Institute of Science and Technology, Daejeon 305-701, South Korea. Phone: 82-42-869-2630. Fax: 82-42-869-5630. E-mail: jchoe{at}kaist.ac.kr Back

{triangledown} Published ahead of print on 23 April 2007. Back


   REFERENCES
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
1. Akimaru, H., Y. Chen, P. Dai, D. X. Hou, M. Nonaka, S. M. Smolik, S. Armstrong, R. H. Goodman, and S. Ishii. 1997. Drosophila CBP is a co-activator of cubitus interruptus in hedgehog signalling. Nature 386:735-738.[CrossRef][Medline]

2. Allada, R., N. E. White, W. V. So, J. C. Hall, and M. Rosbash. 1998. A mutant Drosophila homolog of mammalian Clock disrupts circadian rhythms and transcription of period and timeless. Cell 93:791-804.[CrossRef][Medline]

3. Allada, R., S. Kadener, N. Nandakumar, and M. Rosbash. 2003. A recessive mutant of Drosophila Clock reveals a role in circadian rhythm amplitude. EMBO J. 22:3367-3375.[CrossRef][Medline]

4. Bae, K., C. Lee, P. E. Hardin, and I. Edery. 2000. dCLOCK is present in limiting amounts and likely mediates daily interactions between the dCLOCK-CYC transcription factor and the PER-TIM complex. J. Neurosci. 20:1746-1753.[Abstract/Free Full Text]

5. Belvin, M. P., H. Zhou, and J. C. P. Yin. 1999. The Drosophila dCREB2 gene affects the circadian clock. Neuron 22:777-787.[CrossRef][Medline]

6. Blau, J., and M. W. Young. 1999. Cycling vrille expression is required for a functional Drosophila clock. Cell 99:661-671.[CrossRef][Medline]

7. Brand, A. H., and N. Perrimon. 1993. Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118:401-415.[Abstract]

8. Chang, D. C., and S. M. Reppert. 2003. A novel C-terminal domain of Drosophila PERIOD inhibits dCLOCK:CYCLE-mediated transcription. Curr. Biol. 13:758-762.[CrossRef][Medline]

9. Collins, B., E. O. Mazzoni, R. Stanewsky, and J. Blau. 2006. Drosophila CRYPTOCHROME is a circadian transcriptional repressor. Curr. Biol. 16:441-449.[CrossRef][Medline]

10. Curtis, A. M., S. B. Seo, E. J. Westgate, R. D. Rudic, E. M. Smyth, D. Chakravarti, G. A. FitzGerald, and P. McNamara. 2004. Histone acetyltransferase-dependent chromatin remodeling and the vascular clock. J. Biol. Chem. 279:7091-7097.[Abstract/Free Full Text]

11. Cyran, S. A., A. M. Buchsbaum, K. L. Reddy, M. C. Lin, N. R. Glossop, P. E. Hardin, M. W. Young, R. V. Storti, and J. Blau. 2003. vrille, Pdp1, and dClock form a second feedback loop in the Drosophila circadian clock. Cell 112:329-341.[CrossRef][Medline]

12. Darlington, T. K., K. Wager-Smith, M. F. Ceriani, D. Staknis, N. Gekakis, T. D. Steeves, C. J. Weitz, J. S. Takahashi, and S. A. Kay. 1998. Closing the circadian loop: CLOCK-induced transcription of its own inhibitors per and tim. Science 280:1599-1603.[Abstract/Free Full Text]

13. Doi, M., J. Hirayama, and P. Sassone-Corsi. 2006. Circadian regulator CLOCK is a histone acetyltransferase. Cell 125:497-508.[CrossRef][Medline]

14. Dunlap, J. C. 1999. Molecular bases for circadian clocks. Cell 96:271-290.[CrossRef][Medline]

15. Etchegaray, J. P., C. Lee, P. A. Wade, and S. M. Reppert. 2003. Rhythmic histone acetylation underlies transcription in the mammalian circadian clock. Nature 421:177-182.[CrossRef][Medline]

16. Fu, D., Y. Wen, and J. Ma. 2004. The co-activator CREB-binding protein participates in enhancer-dependent activities of bicoid. J. Biol. Chem. 279:48725-48733.[Abstract/Free Full Text]

17. Glossop, N. R., J. H. Houl, H. Zheng, F. S. Ng, S. M. Dudek, and P. E. Hardin. 2003. VRILLE feeds back to control circadian transcription of Clock in the Drosophila circadian oscillator. Neuron 37:249-261.[CrossRef][Medline]

18. Grima, B., E. Chelot, R. Xia, and F. Rouyer. 2004. Morning and evening peaks of activity rely on different clock neurons of the Drosophila brain. Nature 431:869-873.[CrossRef][Medline]

19. Hall, J. C. 2005. Systems approaches to biological rhythms in Drosophila. Methods Enzymol. 393:61-185.[CrossRef][Medline]

20. Hamblen, M. J., N. E. White, P. T. Emery, K. Kaiser, and J. C. Hall. 1998. Molecular and behavioral analysis of four period mutants in Drosophila melanogaster encompassing extreme short, novel long, and unorthodox arrhythmic types. Genetics 149:165-178.[Abstract/Free Full Text]

21. Helfrich-Forster, C. 2004. The circadian clock in the brain: a structural and functional comparison between mammals and insects. J. Comp. Physiol. A 190:601-613.[Medline]

22. Kalidas, S., and D. P. Smith. 2002. Novel genomic cDNA hybrids produce effective RNA interference in adult Drosophila. Neuron 33:177-184.[CrossRef][Medline]

23. Kaneko, M., and J. C. Hall. 2000. Neuroanatomy of cells expressing clock genes in Drosophila: transgenic manipulation of the period and timeless genes to mark the perikarya of circadian pacemaker neurons and their projections. J. Comp. Neurol. 422:66-94.[CrossRef][Medline]

24. Kim, E. Y., K. Bae, F. S. Ng, N. R. Glossop, P. E. Hardin, and I. Edery. 2002. Drosophila CLOCK protein is under posttranscriptional control and influences light-induced activity. Neuron 34:69-81.[CrossRef][Medline]

25. Kim, E. Y., and I. Edery. 2006. Balance between DBT/CKI{varepsilon} kinase and protein phosphatase activities regulate phosphorylation and stability of Drosophila CLOCK protein. Proc. Natl. Acad. Sci. USA 103:6178-6183.[Abstract/Free Full Text]

26. Koh, K., X. Zheng, and A. Sehgal. 2006. JETLAG resets the Drosophila circadian clock by promoting light-induced degradation of TIMELESS. Science 312:1809-1812.[Abstract/Free Full Text]

27. Kumar, J. P., T. Jamal, A. Doetsch, F. R. Turner, and J. B. Duffy. 2004. CREB binding protein functions during successive stages of eye development in Drosophila. Genetics 168:877-893.[Abstract/Free Full Text]

28. Kwok, R. P., J. R. Lundblad, J. C. Chrivia, J. P. Richards, H. P. Bachinger, R. G. Brennan, S. G. Roberts, M. R. Green, and R. H. Goodman. 1994. Nuclear protein CBP is a coactivator for the transcription factor CREB. Nature 370:223-226.[CrossRef][Medline]

29. Lee, C., K. Bae, and I. Edery. 1999. PER and TIM inhibit the DNA binding activity of a Drosophila CLOCK-CYC/dBMAL1 heterodimer without disrupting formation of the heterodimer: a basis for circadian transcription. Mol. Cell. Biol. 19:5316-5325.[Abstract/Free Full Text]

30. Lilja, T., D. Qi, M. Stabell, and M. Mannervik. 2003. The CBP coactivator functions both upstream and downstream of Dpp/Screw signaling in the early Drosophila embryo. Dev. Biol. 262:294-302.[CrossRef][Medline]

31. Lim, C., H. Sohn, Y. Gwack, and J. Choe. 2000. Latency-associated nuclear antigen of Kaposi's sarcoma-associated herpesvirus (human herpesvirus-8) binds ATF4/CREB2 and inhibits its transcriptional activation activity. J. Gen. Virol. 81:2645-2652.[Abstract/Free Full Text]

32. Lim, C., Y. Gwack, S. Hwang, S. Kim, and J. Choe. 2001. The transcriptional activity of cAMP response element-binding protein-binding protein is modulated by the latency associated nuclear antigen of Kaposi's sarcoma-associated herpesvirus. J. Biol. Chem. 276:31016-31022.[Abstract/Free Full Text]

33. Lin, Y., G. D. Stormo, and P. H. Taghert. 2004. The neuropeptide pigment-dispersing factor coordinates pacemaker interactions in the Drosophila circadian system. J. Neurosci. 24:7951-7957.[Abstract/Free Full Text]

34. Ludlam, W. H., M. H. Taylor, K. G. Tanner, J. M. Denu, R. H. Goodman, and S. M. Smolik. 2002. The acetyltransferase activity of CBP is required for wingless activation and H4 acetylation in Drosophila melanogaster. Mol. Cell. Biol. 22:3832-3841.[Abstract/Free Full Text]

35. Majercak, J., D. Kalderon, and I. Edery. 1997. Drosophila melanogaster deficient in protein kinase A manifests behavior-specific arrhythmia but normal clock function. Mol. Cell. Biol. 17:5915-5922.[Abstract]

36. Marek, K. W., N. Ng, R. Fetter, S. Smolik, C. S. Goodman, and G. W. Davis. 2000. A genetic analysis of synaptic development: pre- and postsynaptic dCBP control transmitter release at the Drosophila NMJ. Neuron 25:537-547.[CrossRef][Medline]

37. McDonald, M. J., and M. Rosbash. 2001. Microarray analysis and organization of circadian gene expression in Drosophila. Cell 107:567-578.[CrossRef][Medline]

38. Naruse, Y., K. Oh-hashi, N. Iijima, M. Naruse, H. Yoshioka, and M. Tanaka. 2004. Circadian and light-induced transcription of clock gene Per1 depends on histone acetylation and deacetylation. Mol. Cell. Biol. 24:6278-6287.[Abstract/Free Full Text]

39. Nawathean, P., and M. Rosbash. 2004. The doubletime and CKII kinases collaborate to potentiate Drosophila PER transcriptional repressor activity. Mol. Cell 13:213-223.[CrossRef][Medline]

40. Panda, S., J. B. Hogenesch, and S. A. Kay. 2002. Circadian rhythms from flies to human. Nature 417:329-335.[CrossRef][Medline]

41. Park, J. H., C. Helfrich-Forster, G. Lee, L. Liu, M. Rosbash, and J. C. Hall. 2000. Differential regulation of circadian pacemaker output by separate clock genes in Drosophila. Proc. Natl. Acad. Sci. USA 97:3608-3613.[Abstract/Free Full Text]

42. Peng, Y., D. Stoleru, J. D. Levine, J. C. Hall, and M. Rosbash. 2003. Drosophila free-running rhythms require intercellular communication. PLoS Biol. 1:E13.[Medline]

43. Petruk, S., Y. Sedkov, S. Smith, S. Tillib, V. Kraevski, T. Nakamura, E. Canaani, C. M. Croce, and A. Mazo. 2001. Trithorax and dCBP acting in a complex to maintain expression of a homeotic gene. Science 294:1331-1334.[Abstract/Free Full Text]

44. Renn, S. C., J. H. Park, M. Rosbash, J. C. Hall, and P. H. Taghert. 1999. A pdf neuropeptide gene mutation and ablation of PDF neurons each cause severe abnormalities of behavioral circadian rhythms in Drosophila. Cell 99:791-802.[CrossRef][Medline]

45. Reppert, S. M., and D. R. Weaver. 2002. Coordination of circadian timing in mammals. Nature 418:935-941.[CrossRef][Medline]