CPK28-mediated Ca2+ signaling regulates STOP1 localization and accumulation to facilitate plant aluminum resistance

Abstract

The transcription factor SENSITIVE TO PROTON RHIZOTOXICITY 1 (STOP1) functions as a crucial integrator of plant responses to various stresses, including aluminum (Al) stress. Its stability and accumulation are modulated by stress-specific post-translational mechanisms such as phosphorylation and ubiquitination. However, the upstream signaling mechanisms governing these modifications remain poorly understood. Here, we reveal that Ca²⁺ signaling and Ca²⁺-dependent phosphorylation are essential for Al stress-responsive regulation of STOP1. Al exposure specifically induces rapid, spatio-temporally defined biphasic Ca²⁺ signals in Arabidopsis roots and concomitantly activates the Ca²⁺-dependent kinase CPK28. Al-activated CPK28 phosphorylates STOP1 at Ser163, a modification that promotes the nuclear localization of STOP1 and prevents its degradation by inhibiting its interaction with the F-box protein RAE1. This phosphorylation enhances STOP1 accumulation and Al resistance. Our findings identify Ser163 phosphorylation as a key molecular switch and establish a Ca²⁺-CPK28-STOP1 signaling axis critical for plant adaptation to Al stress.

Fig. 1. Al stress induces specific Ca²⁺ dynamics in root cells.

a, b False-color representations depicting cytoplasmic Ca²⁺ concentrations in entire roots (a) and root tips (b) at selected informative time points following exposure to 20 μM AlCl₃. Plants expressing the ultrasensitive ratiometric Ca²⁺ reporter protein GCaMP6f-mCherry under the UBQ10 promoter were cultivated for 8 days on ½ MS medium at pH 4.8 and were incubated in containing imaging buffer (5 mM KCl, 10 mM MES and 500 µM CaCl₂, pH 4.8). Buffer exchange for Al exposure (20 μM AlCl₃, 5 mM KCl, 10 mM MES and 500 µM CaCl₂, pH 4.8) occurred at time point 0 min. The region of interest (ROI, indicated by a red circle) was used for data analysis in c. Scale bars, 100 μm. c Quantitative determination of cytoplasmic Ca²⁺ dynamics within the ROI indicated in b. The original data for the –Al graph are provided in Supplementary Fig. 1. The graphs illustrate respective changes in ΔR/R₀ (indicative for cytoplasmic Ca²⁺ accumulation) over a period of 30 min. Data are presented as means ± SE (n = 4).

Fig. 2. Mutation of CPK28 diminishes expression of STOP1-regulated genes and STOP1 accumulation, and reduces Al resistance.

a Representative photographs (left panel) and quantitative data (right panel) of Al resistance in WT and cpk28 mutants. Seedlings were grown on gel medium (pH 5.0) soaked with a nutrient solution (pH 3.6) containing 0, 0.75, 1, or 1.25 mM Al for 7 d; relative root length was measured to evaluate Al resistance. Scale bars, 1 cm. b Decreased malate exudation in cpk28 mutants compared to WT. Seedlings were treated with or without 20 μM Al at pH 4.8 for 12 h and the root exudates were analyzed for the malate concentrations (pmol plant^-1 hour^-1). c cpk28 mutants accumulate more Al than WT in response to Al treatment. Seedlings were exposed to a 0.5 mM CaCl₂ solution containing 0 or 30 µM Al at pH 4.8 for 12 h and then the Al content in the roots was measured. d Decreased expression of ALMT1, MATE and ALS3 in cpk28 mutants. Seedlings were treated with 0 or 30 μM Al at pH 4.8 for 12 h and root tips (0-1 cm) were collected for RT-qPCR analysis. e Representative immunoblot (left panel) and quantitative data (right panel) of STOP1-HA levels. Seedlings WT and cpk28-2 carrying the pSTOP1:STOP1-HA transgene were pretreated with a 0.5 mM CaCl₂ solution at pH 4.8 for 6 h and then exposed to the same CaCl₂ solution containing 0 or 30 μM Al at pH 4.8 for 12 h, and root tips (0-1 cm) were collected for the immunoblot analysis. Actin protein was served as an internal control. For the quantification of STOP1-HA relative levels, STOP1-HA protein levels were initially standardized against their corresponding Actin controls, followed by normalization to the protein level in WT under the −Al condition. f Representative photographs of GUS staining in roots of a pCPK28:GUS transgenic line. Seedlings were treated with 0 or 30 μM Al at pH 4.8 for 12 h. Scale bars, 200 µm. The experiment was repeated three times with similar results. Data shown in (a) are means ± SD of the relative root length (WT/cpk28-2/cpk28-6: n = 28/28/35); data shown in (b–e) are means ± SD of three biological replicates. Different letters in (a–d) indicate significantly different means (P < 0.05, ANOVA followed by Tukey test). Comparisons in (e) were performed using an unpaired two-tailed t-test (^***P < 0.001).

Fig. 3. Overexpression of CPK28 increases expression of STOP1-regulated genes and STOP1 accumulation, and enhances Al resistance.

a Overexpression of CPK28 increases the expression of STOP1-regulated genes. Seedlings of WT and two CPK28 overexpression lines (CPK28-OX) were treated with 0 or 30 μM Al at pH 4.8 for 12 h and then the roots were collected for RT-qPCR analysis. b Representative photographs (left panel) and quantitative data (right panel) of Al resistance in WT and CPK28 overexpression lines. Seedlings were grown on gel medium (pH 5.0) soaked with a nutrient solution (pH 3.6) containing 0, 0.75, 1, or 1.25 mM Al for 7 d; relative root length was used to evaluate Al resistance. Scale bars, 1 cm. c Representative immunoblot (left panel) and quantitative data (right panel) of STOP1-HA levels. Seedlings of WT and CPK28 overexpression lines carrying the pSTOP1:STOP1-HA transgene were treated with 0 or 30 μM Al at pH 4.8 for 12 h and the root tips (0–1 cm) were then collected for immunoblot analysis. Actin protein served as an internal control. d Increased malate release in CPK28 overexpression lines compared to WT. Seedlings were treated with 0 or 20 μM Al at pH 4.8 for 12 h and the root exudates were analyzed for the malate concentrations (pmol plant^-1 hour^-1). Data shown in (a, c, d) are means ± SD of three biological replicates; data shown in (b) are means ± SD of the relative root length (WT/CPK28-OX1/CPK28-OX2: n = 24/24/26). Different letters in (a–d) indicate significantly different means (P < 0.05, ANOVA followed by Tukey test).

Fig. 4. CPK28 interacts with and phosphorylates STOP1.

a Interaction of CPK28 with STOP1 in a yeast two-hybrid assay. Yeast cells (OD₆₀₀ = 1.0) co-expressing CPK28-BD and STOP1-AD were 10-fold serially diluted and grown on SD without Leu /Trp (SD − LT) and SD without Leu/Trp/His/Ade (SD − LTHA) mediums for 3 d at 30°C. b Interaction of CPK28 with STOP1 in a split-LUC assay. The CaMV 35S promoter-driven construct pairs indicated in the figure were co-expressed in N. benthamiana leaves, and the interaction-dependent luciferase activity was determined. c Interaction between CPK28 and STOP1 in a BiFC assay. STOP1-nYFP or NLP8-nYFP was co-transformed with CPK28-cYFP or CPK7-cYFP in N. benthamiana leaves. The interaction-dependent YFP signal and the plasma membrane-localized marker BAK1-mCherry were observed by a confocal microscopy. The right panel of each figure displays pixel intensity profiles of GFP (green) and mCherry (red) along the dashed white arrow in the left panel. Scale bars, 40 μm. The experiment was repeated three times with similar results. d Co-immunoprecipitation of STOP1 with CPK28. CPK28-HA was co-expressed with STOP1-FLAG or GFP-FLAG in rae1rah1 protoplasts. Crude protein extracts were immunoprecipitated with anti-HA magnetic beads and then detected with anti-FLAG antibody. The experiment was repeated three times with similar results. e Phosphorylation assays of WT and mutated STOP1 by CPK28. WT or mutated versions of recombinant His-tagged STOP1 were incubated with MBP-tagged CPK28, and phosphorylated STOP1 was visualized by autoradiography after gel electrophoresis (upper panel). Recombinant CPK28 and STOP1 were detected by Coomassie brilliant blue (CBB) staining (lower panel). f, g Effects of Al alone (f) and/or La (g) on CPK28 kinase activity. Roots of pCPK28:CPK28-FLAG seedlings were treated with 0 or 30 μM Al at pH 4.8 for 20, 60, or 180 min in (f), or treated with 0 or 30 μM Al and/or 0 or 50 μM La at pH 4.8 for 60 min in (g), and then subjected to CPK28-FLAG immunoprecipitation. Recombinant STOP1-GFP-His (middle panel) was incubated with the CPK28-FLAG IP products for the phosphorylation assay (upper panel). CPK28-FLAG was detected using anti-FLAG antibody (lower panel). The experiments in (f and g) were repeated three times with similar results. Data shown in (e) are means ± SD of three biological replicates. Different letters in (e) indicate significantly different means (P < 0.05, ANOVA followed by Tukey test).

Fig. 5. Phosphorylation of STOP1 at Ser163 increases STOP1 level and Al resistance.

a, b Effect of S163A, S296A or S399A mutations in STOP1 (a) and S163A or S163D mutations in STOP1 (b) on ALMT1 promoter-driven LUC activity. 35S:STOP1^WT, 35S:STOP1^S163A, 35S:STOP1^S163D, 35S:STOP1^S296A, 35S:STOP1^S399A, or 35S:GFP were co-expressed with pALMT1:LUC reporter gene and pUBQ10:GUS internal control into stop1 protoplasts. LUC activity relative to GUS activity in the protoplasts was measured and compared. c Effect of S163A or S163D mutations on STOP1 accumulation. 35S:STOP1^WT-LUC, 35S:STOP1^S163A-LUC, or 35S:STOP1^S163D-LUC were co-expressed with pUBQ10:GUS internal control into stop1 protoplasts. d–g Effect of S163A mutation (d, e) or S163D mutation (f, g) on STOP1 accumulation (d, f) and Al resistance (e, g). h Effect of S163A mutation on STOP1 phosphorylation. pIMAGO was used to detect phosphorylated STOP1. Transgenic roots of pSTOP1:STOP1^WT-HA (STOP1^WT), pSTOP1:STOP1^S163A-HA (STOP1^S163A), or pSTOP1:STOP1^S163D-HA (STOP1^S163D) in the stop1 background were treated with 0 or 30 μM Al at pH 4.8 for 12 h and with 0 or 30 μM Al and 50 μM MG132 at pH 4.8 for 3 h for immunoblotting and immunoprecipitation of STOP1-HA, respectively, or grown on gel medium (pH 5.0) soaked with a nutrient solution (pH 3.6) containing 0, 0.75, 1.0, or 1.25 mM Al for 7 d to evaluate their Al resistance. Actin protein was served as an internal control. Scale bars, 1 cm. Data shown in (a–d, f, h) are means ± SD of three biological replicates; data shown in (e, g) are means ± SD of the relative root length (e, STOP1^WT/STOP1^S163A-1/STOP1^S163A-2/cpk28-2: n = 24/27/26/27; g, STOP1^WT/STOP1^S163D-1/STOP1^S163D-2: n = 31/30/33). Different letters in (a–h) indicate significantly different means (P < 0.05, ANOVA followed by Tukey test).

Fig. 6. CPK28 regulates STOP1 accumulation and Al resistance through phosphorylating STOP1 at Ser163.

a Effect of cpk28-2 mutation on STOP1 phosphorylation. Roots of WT and cpk28-2 harboring the pSTOP1:STOP1^WT-HA were treated with 0 or 30 μM Al and 50 μM MG132 at pH 4.8 for 3 h, and then root tips (0-1 cm) were collected for immunoprecipitation of STOP1-HA and subsequent phosphorylation detection using pIMAGO. b, c Effect of S163A mutation on CPK28-triggered STOP1-regulated pALMT1:LUC (b) or 35S:STOP1^WT-LUC (c) activities. 35S:CPK28 or 35S:GFP were co-expressed with 35S:STOP1^WT or 35S:STOP1^S163A and with pALMT1:LUC reporter gene, or co-expressed with 35S:STOP1^WT-LUC or 35S:STOP1^S163A-LUC into stop1 protoplasts at pH 5.7. pUBQ10:GUS was also introduced into the protoplasts as an internal control. LUC activity relative to GUS activity in the protoplasts was measured and compared. d, e Introduction of pSTOP1:STOP1^S163D-HA (STOP1^S163D) transgene rescues decreased STOP1 accumulation (d) and Al-sensitive phenotype (e) in cpk28-2 mutant. Roots of pSTOP1:STOP1^WT-HA/WT (STOP1^WT/WT), pSTOP1:STOP1^WT-HA/cpk28-2 (STOP1^WT/c28), pSTOP1:STOP1^S163D-HA/WT (STOP1^D/WT), and STOP1^S163D/cpk28-2 (STOP1^D/c28) were pretreated with a 0.5 mM CaCl₂ solution (pH 4.8) for 6 h, followed by treatment with the same solution containing 0 or 30 μM Al (pH 4.8) for 12 h for immunoblot analysis of STOP1-HA, or grown on gel medium (pH 5.0) soaked with nutrient solution (pH 3.6) containing 0, 0.75, 1.0, or 1.25 mM Al for 7 d to evaluate their Al resistance. Actin protein was served as an internal control. Scale bars, 1 cm. Data shown in (a–d) are means ± SD of three biological replicates; data shown in (e) are means ± SD of the relative root length (STOP1^WT/WT, STOP1^WT/c28, STOP1^D/WT, STOP1^D/c28: n = 30/30/29/34). Comparisons in (a) were performed using an unpaired two-tailed t-test (***P < 0.001). Different letters in (b–e) indicate significantly different means (P < 0.05, ANOVA followed by Tukey test).

Fig. 7. Phosphorylation of STOP1 at Ser163 inhibits RAE1-mediated degradation of STOP1.

a Inhibition of cpk28 mutation-induced STOP1 degradation by MG132, an inhibitor of the 26S proteasome. WT and cpk28-2 seedlings carrying the pSTOP1:STOP1-HA transgene were treated with 0 or 30 μM Al and/or 50 μM MG132 at pH 4.8 for 3 h. The experiment was repeated three times with similar results. b, c Mutations of both RAE1 and RAH1 rescue the Al-sensitive phenotype (b) and the reduced expression of STOP1-regulated genes (c) in cpk28-2. For Al resistance evaluation, WT, cpk28-2 (c28), rae1 rah1 (e1 h1), and cpk28-2 rae1 rah1 (c28 e1 h1) seedlings were grown on gel medium (pH 5.0) soaked with a nutrient solution (pH 3.6) containing 0, 0.75, 1, or 1.25 mM Al for 7 d. For the expression analysis, the seedlings were exposed to 0 (–Al) or 30 μM Al (+Al) at pH 4.8 for 12 h and then root tips (0-1 cm) were collected for RNA extraction and expression analysis of ALMT1, MATE and ALS3. Scale bars, 1 cm. d Inhibition of the S163A mutation-induced STOP1 degradation by MG132. The experiment was repeated three times with similar results. e, f Effect of S163A or S163D mutations on the interaction between STOP1 and RAE1 lacking the F-box domain (RAE1ΔF) in pull-down (e) and Co-IP (f) assays. For the pull-down assay, recombinant GST-RAE1ΔF was incubated with STOP1-His, STOP1^S163A-His or STOP1^S163D-His at 4 °C for 4 h. For the Co-IP assay, RAE1ΔF-HA was co-transfected with STOP1^WT-FLAG, STOP1^S163A-FLAG or STOP1^S163D-FLAG in rae1rah1 protoplasts. Crude protein extracts were incubated with anti-FLAG magnetic beads at 4 °C for 4 h. Data shown in (c, e, f) are means ± SD of three biological replicates; data shown in (b) are means ± SD of the relative root length (WT/c28/e1 h1/c28 e1 h1: n = 28/28/30/31). Different letters in (b, c, e and f) indicate significantly different means (P < 0.05, ANOVA followed by Tukey test).

Fig. 8. CPK28-mediated phosphorylation at Ser163 enhances nuclear localization of STOP1 protein.

a–c Impact of cpk28-2 mutation (a), S163A mutation (b), and S163D mutation (c) on STOP1 subcellular localization. Seedlings of WT and cpk28-2 lines carrying pSTOP1:STOP1^WT-HA (STOP1^WT) transgene, or stop1 mutant line carrying pSTOP1:STOP1^WT-HA (STOP1^WT), pSTOP1:STOP1^S163A-HA (STOP1^S163A) or pSTOP1:STOP1^S163D-HA (STOP1^S163D) transgenes were treated with 0 or 30 μM Al and 50 μM MG132 at pH 4.8 for 3 h. Root tips (0-1 cm) in (a) or entire roots in (b, c) were collected for isolation of protein fractions: total (T), nuclear (N) and cytosolic (C), followed by immunoblot analysis of STOP1-HA. Histone and Tubulin served as nuclear and cytoplasmic markers, respectively. d, e Quadruple phospho-dead mutation deceases STOP1 accumulation (d) and Al resistance (e) compared to single or triple phospho-dead mutations. Transgenic roots of STOP1^WT, STOP1^S163A (STOP1^1A), pSTOP1:STOP1^{T386A,S448A,S486A}-HA (STOP1^3A), and pSTOP1:STOP1^{S163A, T386A, S448A, S486A} -HA (STOP1^4A) in the stop1 background were treated with a 0.5 mM CaCl₂ solution containing 0 or 30 μM Al at pH 4.8 for 12 h for immunoblotting of STOP1-HA (d), or grown on gel medium (pH 5.0) soaked with a nutrient solution (pH 3.6) containing 0, 0.75, 1.0, or 1.25 mM Al for 7 d to evaluate their Al resistance e. Actin protein was served as an internal control. Scale bars, 1 cm. Data shown in (a–d) represent means ± SD of three biological replicates; data shown in (e) are means ± SD of the relative root length (STOP1^WT/STOP1^1A/STOP1^3A/STOP1^4A−1/STOP1^4A-2: n = 30/30/30/34). Different letters in (a–e) indicate significantly different means (P < 0.05, ANOVA followed by Tukey test).

Fig. 9. Model depicting CPK28-mediated Ca²⁺ signaling in regulating STOP1 subcellular localization and protein accumulation.

Al stress induces a cytosolic Ca²⁺ signature, possibly by activating unknown plasma membrane Ca²⁺ channels. Elevated Ca²⁺ levels activate the plasma membrane-localized kinase CPK28, which phosphorylates STOP1 at the Ser163. This phosphorylation event facilitates STOP1 translocation into the nucleus, inhibiting its interaction with the F-box protein RAE1, and enhances STOP1 accumulation in the nucleus. Consequently, expression of STOP1-regulated genes (e.g., ALMT1, MATE, ALS3) is upregulated, leading to increased Al resistance. The model was originally created by the authors using Adobe Illustrator 2020.