Differential phosphorylation of Ca2+-permeable channel CYCLIC NUCLEOTIDE-GATED CHANNEL20 modulates calcium-mediated freezing tolerance in Arabidopsis

Abstract

Plants respond to cold stress at multiple levels, including increasing cytosolic calcium (Ca2+) influx and triggering the expression of cold-responsive genes. In this study, we show that the Ca2+-permeable channel CYCLIC NUCLEOTIDE-GATED CHANNEL20 (CNGC20) positively regulates freezing tolerance in Arabidopsis (Arabidopsis thaliana) by mediating cold-induced Ca2+ influx. Moreover, we demonstrate that the leucine-rich repeat receptor-like kinase PLANT PEPTIDE CONTAINING SULFATED TYROSINE1 RECEPTOR (PSY1R) is activated by cold, phosphorylating and enhancing the activity of CNGC20. The psy1r mutant exhibits decreased cold-evoked Ca2+ influx and freezing tolerance. Conversely, COLD-RESPONSIVE PROTEIN KINASE1 (CRPK1), a protein kinase that negatively regulates cold signaling, phosphorylates and facilitates the degradation of CNGC20 under prolonged periods of cold treatment, thereby attenuating freezing tolerance. This study thus identifies PSY1R and CRPK1 kinases that regulate CNGC20 activity and stability, respectively, thereby antagonistically modulating freezing tolerance in plants.

Figure 1.

CNGC20 promotes freezing tolerance in Arabidopsis. A and B) Freezing phenotypes (A) and survival rates (B) of Col-0, cngc20-1, and cngc20-1 proCNGC20:CNGC20-GFP complementation lines (cngc20-1 CNGC20 #4 and #20). Thirteen-day-old seedlings grown on a half-strength MS medium at 22 °C were subjected to gradient freezing treatment to −4 ℃ for 30 min (NA) or treated at 4 ℃ for 2 d, followed by gradient freezing treatment to −8 ℃ for 60 min (CA). C) Expression of CBF genes in Col-0, cngc20-1, and cngc20-1 proCNGC20:CNGC20-GFP complementation lines (cngc20-1 CNGC20 #4, #20) under cold stress. Relative transcript levels in untreated Col-0 seedlings were set to 1. D to F) A pseudocolor luminescence image of Ca²⁺-dependent photons emitted by 10-d-old Col-0, cngc20-1, and cngc20-1 CNGC20 complementation seedlings after treatment with 4 °C water (cold shock) or 22 °C water. Representative samples are shown in (D), and relative luminescence signal intensity is shown in (E). Apoaequorin-eYFP protein levels detected by an anti-GFP antibody are shown in (F). Actin2 was used as a control. G and H) CNGC20 positively regulates cold-induced Ca²⁺ influx. G) A time-course analysis of cytosolic-free calcium concentration [Ca²⁺]_cyt dynamics in 8-d-old Col-0, cngc20-1, and cngc20-1 CNGC20 complementation seedlings after treatment with 4 °C water (cold shock) or 22 °C water (the arrow shows the time point of treatment). Luminescence was recorded at 1^−s intervals. Quantification of the cold-induced [Ca²⁺]_cyt changes of (G) are shown in (H). Peak [Ca²⁺]_cyt indicates the highest [Ca²⁺]_cyt after treatment. Data are mean ± Sd (n = 10). Three independent experiments were done with similar results. In (B), (C), and (E), data are means ± Sd of 3 independent experiments, each with 3 technical replicates (B, C: n = 25 seedlings per replicate; E: n = 36 seedlings per replicate); in (G) and (H), data are means ± Sd of 10 seedlings; the different letters represent significant differences at P < 0.05 (1-way ANOVA and Tukey's multiple comparison test).

Figure 2.

PSY1R associates with CNGC20 to enhance freezing tolerance. A) The kinase domain of PSY1R (PSY1R-KD) interacts with CNGC20C in an in vitro pull-down assay. Proteins were detected by using immunoblotting with anti-His and anti-GST antibodies. B) PSY1R interacts with CNGC20 in Arabidopsis protoplasts in a co-IP assay. Proteins were incubated with GFP beads and detected by using immunoblotting with anti-GFP and anti-MYC antibodies. C) PSY1R interacts with CNGC20 in Arabidopsis protoplasts in a BiFC assay. The interaction signal was detected by using confocal microcopy. Scale bars, 10 μm. D) Cold induces the phosphorylation of PSY1R. Phosphorylation was detected with an anti-pThr antibody; PSY1R-GFP was detected with an anti-GFP antibody (left). PSY1R-GFP seedlings were treated at 4 °C for 15 min and treated or not treated with CIAP at 37 °C for 30 min (right). The immunoblot results were quantified using ImageJ software. The phosphorylation level of PSY1R-GFP without cold treatment was set to 1.0. E and F) Freezing phenotypes (E), survival rates, and ion leakage (F) of Col-0, psy1r-1, and psy1r-1 proPSY1R:PSY1R-GFP complementation lines (psy1r-1 PSY1R #1 and #2). G) Survival rates and ion leakage of Col-0, psy1r-1, and psy1r-1 proPSY1R:PSY1R^K831A-GFP (kinase-dead form) seedlings (K831A #39 and #67). In (A) to (D), 3 independent experiments were performed with similar results. In (F) and (G), data are means ± Sd of 3 independent experiments, each with 3 technical replicates (n = 25 seedlings per replicate); the different letters represent significant differences at P < 0.05 (1-way ANOVA and Tukey's multiple comparison test).

Figure 3.

PSY1R phosphorylates and promotes the Ca²⁺ channel activity of CNGC20. A and B) Identification of residues in N terminus (A) and C terminus (B) of CNGC20 that are phosphorylated by PSY1R in an in vitro kinase assay. Recombinant GST-PSY1R-KD was incubated with the N terminus of GST-CNGC20 (20N), its variants, or GST (A), and the C terminus of His-CNGC20 (20C), its variants, or His (B). A phosphorylation signal was detected by using autoradiography. Protein loading controls were stained by using CBB. C and D) PSY1R phosphorylates CNGC20 upon cold exposure in phos-tag assays. C) Total proteins were extracted from CNGC20-GFP seedlings, which were treated at 4 °C for indicated times and subsequently treated or not treated with λPP at 30 °C for 30 min. D) Total proteins were extracted from CNGC20-GFP, psy1r-1 PSY1R, and CNGC20^2A-GFP seedlings, which were treated at 4℃ for indicated times. Proteins were separated by 6% phos-tag-PAGE and 10% SDS–PAGE and detected with an anti-GFP antibody. Actin2 was used as equal loading controls. E to G) PSY1R promotes the channel activity of CNGC20. E) Confocal fluorescence images of Xenopus oocytes expressing different combinations of CNGC20-YFP or its variants with PSY1R-mCherry. YFP and mCherry signals were observed by using confocal microscopy. Scale bars, 200 μm. F) PSY1R promotes the channel activity of CNGC20 but not PSY1R^K831Aor CNGC20^2A. The current–voltage relationship was recorded in Xenopus oocytes in the presence of 30 mm CaCl₂. Xenopus oocytes injected with water served as a control. G) Absolute values of current amplitudes at −160 mV under the indicated concentrations of 30 mm CaCl₂. H and I) PSY1R positively regulates cold-induced Ca²⁺ influx. H) A time-course analysis of [Ca²⁺]_cyt dynamics in 8-d-old Col-0, psy1r-1, and psy1r-1 PSY1R complementation seedlings after treatment with 4 °C water (cold shock) or 22 °C water (the arrow shows the time point of treatment). Quantification of the cold-induced [Ca²⁺]_cyt changes shown in (I). Data are mean ± Sd (n = 10). In (A) to (D), 3 independent experiments were performed with similar results. In (F) and (G), data are means ± Sd of 10 cells; in (H) and (I), data are means ± Sd of 10 seedlings; the different letters represent significant differences at P < 0.05 (1-way ANOVA and Tukey's multiple comparison test).

Figure 4.

PSY1R is required for CNGC20 in regulating plant freezing tolerance. A and B) Freezing phenotypes (A) and survival rates (B) of Col-0, cngc20-1, and cngc20-1 proCNGC20:CNGC20^T61A,T526A-GFP seedlings (20-2A #32 and #35). C and D) Freezing phenotypes (C), survival rates (D) of Col-0, cngc20-1, psy1r-1, and cngc20-1 psy1r-1 double mutant seedlings. E and F) Freezing phenotypes (E), survival rates (F) of Col-0, psy1r-1, proSuper:CNGC20-GFP (20-OE), and psy1r-1 CNGC20-GFP (psy1r 20-OE) seedlings. In (B), (D), and (F), data are means ± Sd of 3 independent experiments, each with 3 technical replicates (n = 25 seedlings per replicate); the different letters represent significant differences at P < 0.05 (1-way ANOVA and Tukey's multiple comparison test).

Figure 5.

CRPK1 phosphorylates and facilitates the degradation of CNGC20. A) CRPK1 interacts with CNGC20 in the split-ubiquitin membrane yeast 2-hybrid assays. Yeast growth on an SC medium lacking Trp and Leu or an SC medium lacking Trp, Leu, His, and Ade is shown. Yeast transformed with pAI-Alg5 and pCCW-Alg5 was used as a positive control, and yeast transformed with pDL2-Alg5 and pCCW-Alg5 was used as a negative control. B) CRPK1 interacts with CNGC20 in a BiFC assay. CRLK1 was used as a negative control. Scale bars, 10 μm. C) CRPK1 interacts with CNGC20C in an in vitro pull-down assay. Proteins were detected by using immunoblotting with anti-His and anti-GST antibodies. D) CRPK1 interacts with CNGC20 in a co-IP assay. Total proteins were extracted from the samples and incubated with GFP beads. The proteins were detected by using immunoblotting with anti-GFP and anti-MYC antibodies. E) Identification of phosphorylated residues of CNGC20C in an in vitro kinase assay. Recombinant CRPK1-MBP-His was incubated with GST-CNGC20C, CNGC20C^3A (T560A, T613A, S646A), CNGC20C^5A (T560A, T613A, S617A, S618A, S646A), CNGC20C^7A (T560A, T613A, S617A, S618A, S646A, T675A, S763A), or GST. Phosphorylation of CNGC20C was detected by using autoradiography (AR). Protein loading controls were stained by using CBB. F and G) A cell-free degradation assay showing that CNGC20 degradation is facilitated by CRPK1. CNGC20-GFP and CNGC20^7A-GFP proteins were detected by using immunoblotting with an anti-GFP antibody. Actin2 was used as a control. MG132 (50 μm) was used to examine protein degradation via the 26S proteasome pathway. Representative images are shown in (F), and relative protein levels of CNGC20-GFP are shown in (G). The immunoblot results were quantified using ImageJ software. The level of CNGC20-GFP in Col and crpk1-1 without ATP treatment was set to 1.0. H and I) An immunoblot analysis of CNGC20 protein levels in Col-0 and crpk1-1 seedlings under cold stress or room temperature. CNGC20-GFP was detected by using immunoblotting with an anti-GFP antibody. Actin2 was used as a control. Representative images are shown in (H), and relative protein levels of CNGC20-GFP are shown in (I). In (A) to (E), 3 independent experiments were performed with similar results. In (G) and (I), data represent means ± Sd of 3 independent experiments (n = 20 seedlings per replicate). The asterisks represent significant differences compared with Col-0 or CNGC20-GFP under the same treatment (*P < 0.05, **P < 0.01, and ***P < 0.001 by Student's t-test).

Figure 6.

Phosphorylation of CNGC20 by CRPK1 attenuates CNGC20-mediated freezing tolerance. A and B) Freezing phenotypes (A) and survival rates (B) of Col-0, cngc20-1, and cngc20-1 proCNGC20:CNGC20^7A-GFP seedlings (20-7A #1 and #2). C and D) Freezing phenotypes (C) and survival rates (D) of Col-0, cngc20-1, and cngc20-1 proCNGC20:CNGC20^7D-GFP seedlings (20-7D #1 and #2). E and F) Freezing phenotypes (E) and survival rates (F) of Col-0, cngc20-1, crpk1-1, and cngc20-1 crpk1-1 seedlings. In (B), (D), and (F), data are means ± Sd of 3 independent experiments (n = 25 seedlings per replicate), each with 3 technical replicates; the different letters represent significant differences at P < 0.05 (1-way ANOVA and Tukey's multiple comparison test).

Figure 7.

A working model depicting the function of PSY1R and CRPK1 in regulating CNGC20 in Arabidopsis response to cold stress. Under normal temperature conditions, the Ca²⁺ concentration in the cytosol ([Ca²⁺]_cyt) remains at a low steady-state level, resulting in low basal expression levels of CBFs and their target genes. At the early stage of cold stress, cold-activated RLK PSY1R phosphorylates and enhances the channel activity of CNGC20, thereby elevating [Ca²⁺]_cyt. As a result, the activation of calcium-dependent protein kinases (CPKs) and an abundant expression of CBF and COR genes lead to enhanced freezing tolerance. During the prolonged cold treatment, CNGC20 undergoes CRPK1-mediated phosphorylation for its degradation via the 26S proteasome pathway, resulting in a compromised cold response in Arabidopsis. The solid arrows indicate direct regulatory relationships, and the dotted arrows indicate indirect regulation. P stands for phosphorylation modification, U stands for ubiquitination modification, and X stands for transcriptional repression.