Figure 1.
Differential effect of cycloheximide on light-induced gene expression.
(A–F) Quantitative RT-PCR analysis (qRT-PCR) of light inducible genes in PAC-2 cells in the presence (red traces) or absence (black traces) of cycloheximide (CHX) during 8 hours of light exposure (left panels) or constant darkness conditions (right panels). Cells were maintained for 3 days in DD prior to the experiment. 1 h before sampling, cells were treated with CHX (10 µg/ml). Each gene is indicated above its respective panels. Yellow and black bars above each panel indicate the light and dark periods, respectively. Relative mRNA levels are plotted on the y-axes and were set arbitrarily as 1 at time-point 0 hrs for each gene. Endogenous β-actin mRNA levels were not influenced by light or cycloheximide treatment and so these were used to normalize the expression of each gene (see Figure S1 B). Time (hrs) is plotted on the x-axes. In each panel, points are plotted as the means of three independent experiments +/− SD. All statistical analyses (t-test and two-way ANOVA) are presented in Table S2. The blocking of protein synthesis by cycloheximide treatment of PAC-2 cells was confirmed in Figure S1 A.
Figure 2.
Role of AP-1 enhancer elements in light-induced expression of cry1a.
(A) Schematic representation of the 1.3 kb cry1a promoter. The 53 bp exon 1 is indicated by a green rectangle. The transcription start site (TSS) at position −688 bp and the ATG at position +1 bp are indicated. Violet rectangles denote the three AP-1 sites (AP-1 #1 at position −1168 bp, AP-1 #2 at position −702 bp and AP-1 #3 at position −416 bp). (B – E) Representative real time bioluminescence assays of PAC-2 cells transfected with the following constructs (B) cry1a-Luc. (C) cry1a-Luc (black trace) and cry1a AP1 mut -Luc (green trace). (D) AP1-Luc. (E) AP1-Luc in the presence (red trace) or absence (black trace) of 50 ng/ml of the phorbol ester TPA. The black arrow indicates the time of TPA or DMSO-control addition. In each panel relative bioluminescence is plotted on the y-axis and time (hrs) on the x-axis. Each time-point represents the mean of at least four independently transfected wells +/− SD from a single experiment. Each experiment was performed a minimum of three times. Yellow and black bars above each panel represent the light and dark periods, respectively.
Figure 3.
Identification of the light-responsive region within the cry1a promoter.
(A) Schematic representation of cry1a-Luc and deletion constructs 1 to 17 (grey bars) (see also Table S5). The 186 bp cry1a light responsive region (cry1a LRR) is indicated by dotted lines and black arrows (region between −401 bp and −215 bp). (B–C) Representative real time bioluminescence assays from PAC-2 cells transfected with cry1a-Luc (black trace) and cry1a-Luc Deletion 12 or cry1a-Luc Deletion 13 (green traces) (Figure 3B and C). In both cry1a-Luc Deletions 12 and 13, the phase of rhythmic expression is significantly shifted (for both p<0.0001, t-test). In deletions 12 and 13, the increase in luciferase activity that anticipates the onset of the light phase is indicated by a horizontal black arrow. (D) Representative real time bioluminescence assay from PAC-2 cells transfected with cry1a-Luc (black trace) and cry1a LRR-Luc (red trace). In each panel relative bioluminescence is plotted on the y-axis and time (hrs) on the x-axis. Each time-point represents the mean of at least four independently transfected wells +/− SD from a single experiment. Each experiment was performed a minimum of three times. Yellow and black bars above each panel represent the light and dark periods, respectively. Statistically significant differences are indicated by an asterisk (*) and are reported in Table S3.
Figure 4.
A single functional D-box is necessary and sufficient for the light response of the cry1a gene.
(A) Schematic representation of cry1a LRR- Luc and sub-deletion constructs 1 to 13 (dark grey bars). The red rectangle denotes the putative E-box while the three yellow ellipses represent the putative D-boxes. The region delimited by cry1a LRR-Luc Sub-Deletions 5 and 6 is indicated by red arrowheads and red dotted lines. This region includes the light responsive D-box. (B–E) Representative real time bioluminescence assays from transfected PAC-2 cells. The identity of the transfected constructs and their colour codes are indicated above each panel. In each panel relative bioluminescence is plotted on the y-axis and time (hrs) on the x-axis. Each time-point represents the mean of at least four independently transfected wells +/− SD from a single experiment. Each experiment was performed a minimum of three times. Yellow and black bars above each panel represent the light and dark periods, respectively. Statistically significant differences are reported in Table S3.
Figure 5.
Light induced D-box enhancer activity requires de novo protein synthesis.
(A–C and E) qRT-PCR analysis of luciferase mRNA expression in PAC-2 cells transfected with different heterologous luciferase reporter constructs, in the presence (red traces) or absence (green traces) of CHX during 8 hours of light exposure or DD conditions (+CHX, blue traces, −CHX, black traces). (D) qRT-PCR analysis of endogenous per1b expression in PAC-2 cells in the presence or absence of CHX during 8 hours of light exposure or DD conditions (colour coded the same as in panels A–C and E). Each construct is indicated above its respective panel. Relative mRNA levels are plotted on the y-axis and were set arbitrarily as 1 at time-point 0 hrs. Time (hrs) is plotted on the x-axis. In each panel, points are plotted as means of three independent experiments +/− SD. All statistical analyses (two-way ANOVA) are presented in Table S2 B or in the results section.
Figure 6.
Regulation of the cry1a D-box by PAR bZip transcription factors.
(A) Schematic representation of the experimental design. Black and yellow bars represents 12 hours dark and light periods respectively, while the dark grey bar denotes the subjective day period under constant darkness. Arrows indicate sampling time points where ZT and CT represent zeitgeber times and circadian times respectively (ZT0 represents “lights on”). (B–G) qRT-PCR analysis of PAR bZip gene expression in PAC-2 cells under LD (pink traces) and DD (black traces) conditions. Each gene is indicated above its respective panel. Relative mRNA levels are plotted on the y-axis and ZT or CT times on the x-axes. In each panel, points are plotted as the means of three independent experiments +/− SD. Yellow and black bars above each panel represent the light and dark periods, respectively. The statistical significance of rhythmic expression was assessed by t-test analysis in Table S2 C. (H) In vitro luciferase assay of PAC-2 cells co-transfected with expression constructs encoding the six PAR bZip factors and the cry1a LRR-Luc or cry1a LRR D-box mut-Luc reporters (dark grey and green bars, respectively). Each expression construct is indicated below its respective bars. Relative bioluminescence levels (%) are plotted on the y-axis where the highest value measured during the experiment is set arbitrarily as 100%. The results are plotted as the means of three independent experiments performed in triplicate, +/− SD. Each independent experiment was standardized for transfection efficiency using a β-galactosidase assay. The statistical significance of levels of transactivation was assessed by t-test analysis with * p<0.05, ** p<0.001, *** p<0.0001.
Figure 7.
Contribution of de novo protein synthesis to light–induced clock gene expression.
(A) Under normal conditions light exposure triggers expression of the gene encoding the PAR bZip factor, TEF-1. This in turn binds to D-boxes in the cry1a and per2 promoters and trans-activates gene expression. In parallel, light also entrains the circadian clock. Via binding of the CLOCK–BMAL complex, the clock regulates the E-box in the per2 promoter and thereby contributes to light induced gene expression [21]. The clock also regulates expression of the additional PAR bZip factors (PAR) that contribute to D-box driven transcription. (B) Upon light exposure and coincident inhibition of de novo protein synthesis by treatment with cycloheximide (+CHX), translation of TEF-1 and the other PAR bZip factors is prevented. Therefore, light-driven transactivation via the D-box enhancer of the cry1a promoter is abolished. However, light-induced expression of the per2 promoter persists due to regulation by the E-box. Specifically, upon cycloheximide treatment the core clock machinery directs increased activation via the E-box in a light dependent manner. We speculate that this up-regulation of E-box driven expression may also influence other clock-regulated genes including those encoding the PAR bZip factors.