Cold-induced degradation of core clock proteins implements temperature compensation in the Arabidopsis circadian clock

Abstract

The period of circadian clocks is maintained at close to 24 hours over a broad range of physiological temperatures due to temperature compensation of period length. Here, we show that the quantitative control of the core clock proteins TIMING OF CAB EXPRESSION 1 [TOC1; also known as PSEUDO-RESPONSE REGULATOR 1 (PRR1)] and PRR5 is crucial for temperature compensation in Arabidopsis thaliana. The prr5 toc1 double mutant has a shortened period at higher temperatures, resulting in weak temperature compensation. Low ambient temperature reduces amounts of PRR5 and TOC1. In low-temperature conditions, PRR5 and TOC1 interact with LOV KELCH PROTEIN 2 (LKP2), a component of the E3 ubiquitin ligase Skp, Cullin, F-box (SCF) complex. The lkp2 mutations attenuate low temperature-induced decrease of PRR5 and TOC1, and the mutants display longer period only at lower temperatures. Our findings reveal that the circadian clock maintains its period length despite ambient temperature fluctuations through temperature- and LKP2-dependent control of PRR5 and TOC1 abundance.

Fig. 1.. Circadian rhythms in period mutants grown at various temperatures.

(A) Mean traces of 16 seedlings expressing the CCA1:LUC reporter in the wild type and in prr5 toc1. (B) Period length and Q₁₀ values of wild type and prr5 toc1. Note that a sample with a fitting error value exceeding 0.05 was discarded due to incorrect period determination (n = 10 to 16 for wild type and n = 3 to 20 for prr5 toc1). (C) Q₁₀ values of period mutants. Traces and period lengths of the mutants are shown in fig. S1.

Fig. 2.. Quantitation of CCA1 expression and PRR5 and TOC1 proteins in plants incubated at 12°, 22°, and 28°C.

(A) Relative expression of CCA1 at 12°, 22°, and 28°C (four biological replicates). The mean value of CCA1 relative to IPP2 in the wild type at 12°C was normalized to 1.0. (B) TOC1 mRNA levels in 35Sp:TOC1-F and PRR5 in 35Sp:PRR5-F seedlings 24 hours after temperature shifts to 12°, 22°, and 28°C (top). TOC1-F and PRR5-F protein accumulation in seedlings at 12°, 22°, and 28°C. Coomassie brilliant blue (CBB)–stained gels and Western blot probed with anti-FLAG antibody (WB: FLAG) are shown. TOC1-F or PRR5-F band intensity of three biological replicates (bottom graphs). (C) TOC1-F and PRR5-F protein accumulation in seedlings treated with 26S proteasome inhibitor MG132. (D) Ubiquitination of TOC1-F or PRR5-F proteins in seedlings at 12°, 22°, and 28°C. Protein samples immunoprecipitated with anti-FLAG antibody (IP: FLAG) were detected with anti-FLAG (WB: FLAG) or anti-ubiquitin (WB: UBQ) antibodies. The band intensity of the strongest signal in the gel was normalized to 1. Letters in the graphs show statistical differences as determined by Tukey-Kramer test. Double and single asterisks indicate statistical differences P < 0.01 and P < 0.05, respectively, as determined by Student’s t test. Arrow heads indicate non-specific bands.

Fig. 3.. Interaction of PRR5 and TOC1 proteins with LKP2.

(A) Volcano plot of proteins in the TOC1-F–immunoprecipitated (IP) fraction from plants grown at 12°C versus 28°C, analyzed by LC-MS/MS (left). Three independent biological replicates were used. The x axis shows the log 2 fold change (FC) in plants at 28°C versus 12°C. The y axis shows the log₁₀ of the false discovery rate (FDR) value. Number of detected peptides corresponding to TOC1, PRR5, LKP2, AT4G02480, AT1G43800, and ZTL in TOC1-F–immunoprecipitated fractions from plants grown at 12° or 28°C (right). (B) Volcano plot of PRR5-F–immunoprecipitated fraction from plants grown at 28°C versus 12°C (left), and the six proteins in PRR5-F–immunoprecipitated fractions (right). (C) Venn diagrams showing number of proteins enriched in TOC1-F– and PRR5-F–immunoprecipitated fractions in plants grown at 12° or 28°C. (D) LKP2-GFP protein was immunoprecipitated from 35Sp:PRR5-F/lkp2-cr3/LKP2-GFP plants grown at 12° or 28°C, and LKP2-GFP and PRR5-FLAG proteins in immunoprecipitated fractions were detected (left). Three independent biological replicates were performed for the quantification (right). The band intensity of the 12°C sample was normalized to 1.

Fig. 4.. Effects of low temperature on PRR5 and TOC1 in the lkp2 mutants.

(A) lkp2 mutations were created in 35Sp:TOC1-F (lkp2-cr1 and lkp2-cr2) or 35Sp:PRR5-F plants (lkp2-cr3). A ztl mutation was created in 35Sp:PRR5-F (ztl-cr1). Empty boxes indicate deletions. Prediction of LKP2 and ZTL proteins in the mutants. (B) TOC1-F protein accumulation in wild type and lkp2 mutants at 12° and 22°C (left). Three independent biological replicates were used for the quantification (right). (C) PRR5-F protein in wild type, lkp2-cr3, and lkp2-cr3/LKP2-GFP complementation lines (right), and quantification data from three independent replicates (right). (D) PRR5-F protein in wild type, lkp2-cr3, ztl-cr1, and lkp2-cr3/ztl-cr1 at 12°C. (E) PRR5-F protein in wild type, lkp2-cr3, ztl-cr1, and lkp2-cr3/ztl-cr1 at 22°C. Arrowheads show nonspecific bands. The band intensity exhibiting the strongest signal in the gel was normalized to 1. Letters in the quantification graphs indicate statistical differences as determined by Tukey-Kramer test. bp, base pair.

Fig. 5.. Temperature dependency of circadian period in lkp2 mutants.

(A) lkp2-cr4 mutant were created in CCA1:LUC (left). Prediction of LKP2 protein in lkp2-cr4 (right). (B) Mean traces of CCA1:LUC reporter (n = 16). (C) Period length and Q₁₀ value (n = 13 to 16). (D) Comparisons of period lengths among wild type, lkp2-cr4, and lkp2-cr4/LKP2-GFP (n = 15 for wild type, n = 16 or 14 for lkp2-cr4, and n = 11 for lkp2-cr4/LKP2-GFP). (E) Comparisons of period lengths among wild type, lkp2-cr4, ztl-3, and lkp2-cr4/ ztl-3 (n = 15 or 16 for wild type, n = 16 for lkp2-cr4, n = 14 for ztl-3, and n = 8 or 14 for lkp2-cr4/ztl-3). Significant differences were determined using the Tukey-Kramer test. (F) Schematic model of the molecular basis of temperature compensation by PRR5 and TOC1. PRR5 and TOC1 are targeted for degradation by LKP2 and ZTL at lower temperatures and by ZTL at high temperatures.