Phosphorylation of a central clock transcription factor is required for thermal but not photic entrainment

Abstract

Transcriptional/translational feedback loops drive daily cycles of expression in clock genes and clock-controlled genes, which ultimately underlie many of the overt circadian rhythms manifested by organisms. Moreover, phosphorylation of clock proteins plays crucial roles in the temporal regulation of clock protein activity, stability and subcellular localization. dCLOCK (dCLK), the master transcription factor driving cyclical gene expression and the rate-limiting component in the Drosophila circadian clock, undergoes daily changes in phosphorylation. However, the physiological role of dCLK phosphorylation is not clear. Using a Drosophila tissue culture system, we identified multiple phosphorylation sites on dCLK. Expression of a mutated version of dCLK where all the mapped phospho-sites were switched to alanine (dCLK-15A) rescues the arrythmicity of Clk(out) flies, yet with an approximately 1.5 hr shorter period. The dCLK-15A protein attains substantially higher levels in flies compared to the control situation, and also appears to have enhanced transcriptional activity, consistent with the observed higher peak values and amplitudes in the mRNA rhythms of several core clock genes. Surprisingly, the clock-controlled daily activity rhythm in dCLK-15A expressing flies does not synchronize properly to daily temperature cycles, although there is no defect in aligning to light/dark cycles. Our findings suggest a novel role for clock protein phosphorylation in governing the relative strengths of entraining modalities by adjusting the dynamics of circadian gene expression.

Figure 1. Blocking phosphorylation at multiple phospho-sites on dCLK prevents global phosphorylation but does not impair several key clock-relevant activities.

(A) Schematic diagram of dCLK protein. Phosphorylation sites on dCLK identified in this study are indicated as red vertical lines. Horizontal line at bottom indicates relative positions of amino acid residues. (B, C) S2 cells were transiently transfected with 500 ng of pMT-HA-dClk (WT) or pMT-HA-dClk-16A (16A), either singly or in combination with 500 ng of pAct-cyc-V5 or pAct-per-V5 as indicated. Expression of dCLK was induced 24 hr after transfection by adding 500 µM CuSO₄ to the medium. Cells were harvested 24 hr after induction, and protein extracts were first subjected to immunoprecipitation using anti-HA (12CA5) antibody (B), the anti-epitope tag antibodies (V5 or HA) as indicated on top of the blots (C). Immune complexes were directly analyzed by immunoblotting (C) or further incubated in the absence (−) or presence (+) of λ phosphatase followed by immunoblotting (B). (D, E) S2 cells were transiently co-transfected either singly with pMT-dClk-V5 (WT) and pMT-dClk-16A-V5 (16A) (D), or in combination with increasing amount of pAct-per (E). Shown are the average values from three independent experiments for relative E box dependent luciferase activity in the absence (−) or presence (+) of pMT-dClk-V5. *p<0.05; error bars denote SEM.

Figure 2. (A–E) The p{dClk-15A};Clk ^out flies manifest short period behavioral rhythms.

Each panel represents the average activity of male flies for a given genotype during the third and fourth day of 12 hr light∶12 hr dark entrainment (LD) followed by 4 days of constant darkness (DD). White vertical bars represent locomotor activity during light phase and black vertical bars represent locomotor activity during dark phase in LD. Gray vertical bars represent locomotor activity during the subjective light phase in DD. White horizontal bars and black horizontal bars below each panel indicate 12 hr periods of lights-on and lights-off, respectively. Arrowheads indicate the times in a daily cycle when trough levels of activity were attained following the evening bout of activity. Standard error of the mean is indicated as dots above each bars.

Figure 3. The levels of dCLK are substantially higher and hypo-phosphorylated at all times of the day in p{dClk-15A};Clk ^out flies.

(A–C) Adult flies of the indicated genotype were collected at different times of day (ZT), head extracts prepared and directly analyzed for immunoblotting (A) or processed for immunoprecipitation with anti-V5 Ab (B). β-Actin (ACTIN) served as a loading control. Immune complexes were further incubated in the absence (−) or presence (+) of λ phosphatase and immunoblotted with anti-V5 antibody, as indicated (B). Filled arrowheads denote hyper-phosphorylated isoforms of dCLK and open arrowhead denotes hypo-phosphorylated isoforms of dCLK (A). (C) Relative levels of dCLK were determined by measuring staining intensities using image J software. Shown are the average values from three independent experiments. (D) Total RNA was extracted from fly heads, and quantitative real-time RT-PCR was performed to measure the relative levels of dClk transcripts. Shown are the average values from three independent experiments using p{dClk-15A}, 6M;Clk ^out flies. Error bars denote SEM.

Figure 4. Higher amplitude rhythms of per mRNA and protein in p{dClk-15A};Clk ^out flies.

Adult flies of the indicated genotype were collected at the indicated times (ZT) during a day and total RNA (A, B) or protein extracts (C to F) prepared. (A, B) Quantitative real-time RT-PCR was performed to measure the relative levels of per (A) or tim (B) transcripts. Shown are the average values from three independent experiments. (C to F) Immunoblotting was performed using anti-PER (Rb1) or anti-TIM (TR3) Ab. O-GlcNAc transferase (OGT) served as a loading control. Relative levels of PER and TIM proteins were determined by measuring band intensities of immunoblot using image J software (D, F). Shown are the average values from three independent experiments using p{dClk-15A}, 6M;Clk ^out flies. *p<0.05; error bars denote SEM.

Figure 5. Impaired behavioral entrainment of p{dClk-15A};Clk ^out flies in daily temperature cycles.

Adult male files of the indicated genotype were entrained in 12 h∶12 h temperature cycles of 24°C∶29°C in the absence (A–F) or presence (G–J) of constant light. (A–C, G, H) Each panel represents the daily average activity beginning on the third day of TC followed by 7 consecutive days. Orange vertical bars represent locomotor activities during the thermo phase and black vertical bars represent locomotor activities during the cryo phase. Red and blue horizontal bars indicate thermo- and cryo-phases, respectively. (D–F, I, J) Red and blue shades indicate thermo-and cryo-phases, respectively. The vertical black bars on each row of the actogram depict fly activity (measured in 30 min intervals). HD, hash density of the actogram (for example, HD = 10 signifies that 10 activity events are required to produce a hash mark). To better visualize rhythmic behavior, each row of an actogram was double plotted. To better visualize the progressive advancement of the main activity bout in p{dClk-15A};Clk ^out flies, a vertical line was drawn across the activity offsets.

Figure 6. Molecular rhythms in p{dClk-15A};Clk ^out flies show increased alterations after prolonged entrainment to temperature cycles.

Adult flies of the indicated genotype were entrained in 12 hr∶12 hr of 24°C∶29°C temperature cycle in the absence (A–C) or presence of light (D, E). During the third (A, B, D) and sixth day (C, E) of TC, flies were collected and protein (A) or RNA (B–E) was extracted from fly heads. Protein extracts were analyzed by immunoblotting using anti-V5 Ab to probe dCLK. Quantitative real-time RT-PCR was performed to measure the relative levels of tim mRNA. Shown are the average values from three independent experiments using p{dClk-15A}, 6M;Clk ^out flies. *p<0.05; error bars denote SEM. Red horizontal bars represent thermo phase, blue horizontal bars represent cryo phase, black horizontal bars represents constant dark conditions, and white horizontal bars represents constant light conditions.