Histone deacetylase OsHDA716 represses rice chilling tolerance by deacetylating OsbZIP46 to reduce its transactivation function and protein stability

Abstract

Low temperature is a major environmental factor limiting plant growth and crop production. Epigenetic regulation of gene expression is important for plant adaptation to environmental changes, whereas the epigenetic mechanism of cold signaling in rice (Oryza sativa) remains largely elusive. Here, we report that the histone deacetylase (HDAC) OsHDA716 represses rice cold tolerance by interacting with and deacetylating the transcription factor OsbZIP46. The loss-of-function mutants of OsHDA716 exhibit enhanced chilling tolerance, compared with the wild-type plants, while OsHDA716 overexpression plants show chilling hypersensitivity. On the contrary, OsbZIP46 confers chilling tolerance in rice through transcriptionally activating OsDREB1A and COLD1 to regulate cold-induced calcium influx and cytoplasmic calcium elevation. Mechanistic investigation showed that OsHDA716-mediated OsbZIP46 deacetylation in the DNA-binding domain reduces the DNA-binding ability and transcriptional activity as well as decreasing OsbZIP46 protein stability. Genetic evidence indicated that OsbZIP46 deacetylation mediated by OsHDA716 reduces rice chilling tolerance. Collectively, these findings reveal that the functional interplay between the chromatin regulator and transcription factor fine-tunes the cold response in plant and uncover a mechanism by which HDACs repress gene transcription through deacetylating nonhistone proteins and regulating their biochemical functions.

Figure 1.

OsHDA716 negatively regulates chilling tolerance in rice. A, D, G, and J) Chilling tolerance phenotype of corresponding wild-type plants and Oshda716-1 mutant (A), Oshda716-2 mutant (D), complementation lines Com-L1 (G), and OsHDA716 overexpression lines (J). Two-wk-old seedlings were incubated at 4 °C for the indicated days, followed by recovery for 7 or 14 d under normal conditions. Bars = 5 cm. Four independent experiments were performed, and representative images were given. HY, Hwayoung; DJ, Dongjin; Nip, Nipponbare. See also Supplementary Figs. S2, F to I and S3D. B, E, H, and K) Survival rates of various genetic background plants after cold treatment followed by recovery for 14 d as shown in A), D), G), and J), respectively. Values are means ± Sd (n = 4 independent experiments, with 12 seedlings per independent experiment). In B) and E), asterisks indicate significant differences compared with the corresponding wild type (**P < 0.01, Student's t test). In H) and K), different lowercase letters indicate significant differences (P < 0.05, one-way ANOVA with Tukey's HSD test). C, F, I, and L) Ion leakage of 2-wk-old plants after 4 °C treatment for 48 h. Values are means ± Sd (n = 4 independent experiments, with 6 seedlings per independent experiment). In C) and F), asterisks indicate significant differences compared with the corresponding wild type (**P < 0.01, Student's t test). In I) and L), different lowercase letters indicate significant differences (P < 0.05, one-way ANOVA with Tukey's HSD test).

Figure 2.

OsHDA716 impairs cold-induced Ca²⁺ influx. A and B) Extracellular Ca²⁺ influx upon cold shock in live roots of corresponding wild-type plants and Oshda716-1 (A) and OsHDA716 overexpression lines (B). Three independent experiments were performed with similar results, and values are means ± Sd (n = 8) in A) and (n = 6) in B) from 1 experiment. The background indicates the duration of cold treatment. HY, Hwayoung; Nip, Nipponbare. C and D) Significance testing of the mean maximal Ca²⁺ influx as indicated in A) and B), respectively. Three independent experiments were performed with similar results, and values are means ± Sd (n = 8) in C) and (n = 6) in D) from 1 experiment. In C), asterisks indicate significant differences compared with the corresponding wild type (**P < 0.01, Student's t test). In D), different lowercase letters indicate significant differences (P < 0.05, one-way ANOVA with Tukey's HSD test). E) Distribution of [Ca²⁺]_cyt in the root tips of HY and Oshda716-1 mutant after cold shock. Two-d-old roots with or without cold shock were stained with 5 μm Fluo-4 AM, and fluorescence signals were captured using the laser scanning confocal microscope. Bars = 100 μm. Three independent experiments were performed with similar results, and representative images were given. F) Relative fluorescence intensity in response to cold shock as indicated in E). Fluorescence over a 350 × 600 μm window for HY and 350 × 450 μm window for Oshda716-1 mutant was collected at the similar region (indicated in the dot box) in the root tips described in E). Arrows indicated the fluorescence peak in HY and Oshda716-1 mutant, respectively. G) Expression levels of cold stress-related genes COLD1, OsDREB1A, OsCPK6, and OsCPK13 in various genetic background plants in response to cold stress. Three independent experiments were performed with similar results, and values are means ± Sd (n = 3) from 1 experiment. Different lowercase letters indicate significant differences (P < 0.05, one-way ANOVA with Tukey's HSD test). OsACTIN1 was used as the internal control.

Figure 3.

OsHDA716 physically interacts with OsbZIP46. A) Yeast 2-hybrid assay to verify the interaction between OsHDA716 and OsbZIP46. DDO, SD/-Leu-Trp; QDO, SD/-Ade-His-Leu-Trp. B) In vitro GST pull-down assay for OsHDA716–OsbZIP46 interaction. GST-tagged OsHDA716 (GST-OsHDA716) was used as a bait, and pull-down of OsbZIP46-His was detected by the anti-His antibody. C) OsHDA716–OsbZIP46 interaction indicated by BiFC assay in leaf epidermal cells of N. benthamiana. Bars = 50 μm. D) OsHDA716–OsbZIP46 interaction indicated by SLCA in N. benthamiana leaves. E) Semiendogenous Co-IP analysis showing the interaction of OsbZIP46-His with endogenous OsHDA716-GFP. The purified OsbZIP46-His recombinant protein was incubated with protein extracts from OsHDA716-GFP plants for 2 h, and then, the immunoprecipitation was performed with the anti-GFP antibody. The immunoprecipitated proteins were detected with anti-His and anti-GFP antibodies, respectively. The immunoblot analysis of Histone H3.1 (H3.1) was used as a control. F) OsHDA716–OsbZIP46 interaction by in vivo Co-IP assays. OsHDA716-GFP and OsbZIP46-3 × FLAG were transiently coexpressed in N. benthamiana, and then, the immunoprecipitation was performed with the anti-GFP antibody. The immunoprecipitated proteins were detected with anti-FLAG and anti-GFP antibodies, respectively.

Figure 4.

OsHDA716 can deacetylate OsbZIP46 in vitro and in vivo. A) OsHDA716-modulated K deacetylation in OsbZIP46 in E. coli. OsbZIP46-His was coexpressed with GST-OsHDA716 or GST alone in E. coli BL21 (DE3). The acetylation levels of purified OsbZIP46-His recombinant protein were determined with the anti-acLys antibody, with 2 technical replicates. B) In vitro deacetylation assays of OsbZIP46 mediated by OsHDA716. Assays of in vitro deacetylation reaction were performed using the different combinations of purified proteins as indicated, with or without TSA. The acetylation levels of GST-OsbZIP46 were determined with anti-acLys antibody. See also Supplementary Fig. S11A. C and D) OsHDA716-modulated K deacetylation of OsbZIP46 in plants. Protein extracts from OsbZIP46-GFP plants were incubated with extracts from different genotype plants of OsHDA716, with or without TSA treatment, and the immunoprecipitated OsbZIP46-GFP was analyzed by immunoblotting using anti-acLys and anti-GFP antibodies, respectively. Histone 3.1 (H3.1) was used as a loading control. Two-wk-old seedlings were used. E) Schematic representation of OsbZIP46 showing the acetylation sites determined by mass spectrometry. The purified OsbZIP46-His recombinant was used for mass spectrometry analysis, and the positions of detected K acetylation sites were indicated. See also Supplementary Fig. S12. F) Amino acid substitutions (K263R, K281R, K294R, K310R) reduce OsbZIP46 acetylation levels. OsbZIP46-His or its mutated version was coexpressed with GST-OsHDA716 in E. coli BL21 (DE3), and the purified OsbZIP46-His and its mutated versions were detected by immunoblotting with the anti-acLys antibody. OsbZIP46^KBDR, amino acid substitutions (K263R, K281R, K294R) in the DNA-binding domain of OsbZIP46. For immunoblotting analysis, 2 independent experiments were performed, and similar results were obtained. The relative intensity of protein bands was calculated from 2 independent experiments by using ImageJ.

Figure 5.

OsHDA716 reduces the protein stability of OsbZIP46 through mediating K deacetylation. A) In vitro cell-free degradation assays showing degradation of GST-OsbZIP46 in protein extracts from different genotypes of OsHDA716 in the presence of ATP. B) Effect of OsHDA716 on OsbZIP46 protein stability in vivo. Total protein extracted from OsbZIP46-GFP and OsbZIP46-GFP Oshda716-1 transgenic plants was incubated with 10 mm ATP for indicated time at room temperature. The protein levels of OsbZIP46-GFP were detected with the anti-GFP antibody. C) Effect of OsHDA716 on OsbZIP46 protein stability in N. benthamiana. OsbZIP46-3 × FLAG was transiently coexpressed with OsHDA716-GFP or GFP alone in N. benthamiana leaves. Total protein was extracted and then incubated with 10 mm ATP for indicated time. The protein levels of OsbZIP46 and OsHDA716 were detected with anti-FLAG and anti-GFP antibodies, respectively. H3.1 was used as loading control. The relative quantification of OsbZIP46-3 × FLAG protein intensities was calculated from 3 independent experiments. D) Effect of OsbZIP46 deacetylation on its ubiquitination. OsbZIP46-GFP was immunoprecipitated from OsbZIP46-GFP or OsbZIP46-GFP Oshda716-1 transgenic plants with the anti-GFP antibody, and then, the immunoprecipitated OsbZIP46-GFP was detected by immunoblotting using anti-acLys and anti-Ubi antibodies, respectively. E) In vitro cell-free degradation assays showing degradation of GST-OsbZIP46 or GST-OsbZIP46^KBDR in the protein extracts from wild-type plants, in the presence of 10 mm ATP. F) OsHDA716 protein levels under cold treatment. OsHDA716-3 × FLAG transgenic seedlings were treated under 4 °C for the indicated time, and OsHDA716-3 × FLAG protein was detected by using the anti-FLAG antibody, with H3.1 as a loading control. See also Supplementary Fig. S13. G) OsbZIP46 acetylation levels as well as protein levels in OsbZIP46-GFP and OsbZIP46-GFP Oshda716-1 seedlings under cold stress for the indicated time. OsbZIP46-GFP was immunoprecipitated from total extracts of OsbZIP46-GFP and OsbZIP46-GFP Oshda716-1 plants with the anti-GFP antibody and detected with anti-acLys and anti-GFP antibodies. For immunoblotting analysis, 3 independent experiments in A) and C) and 2 independent experiments in E) and F) were performed, and similar results were obtained. The relative intensity of protein bands was calculated from 2 or 3 independent experiments by using ImageJ. H3.1 was used as a loading control.

Figure 6.

OsbZIP46 positively regulates the chilling tolerance in rice by promoting cold-induced Ca²⁺ influx. A and D) Chilling tolerance phenotype of corresponding wild-type plants and Osbzip46 mutant (A) and OsbZIP46 overexpression lines (D). Two-wk-old seedlings were incubated at 4 °C for 6 or 7 d, followed by recovery for 7 or 14 d under normal conditions. Bars = 5 cm. Four independent experiments were performed, and representative images were given. DJ, Dongjin; Nip, Nipponbare. See also Supplementary Fig. S14, E and F. B and E) Survival rates of different genotype plants of OsbZIP46 after cold treatment followed by recovery for 14 d as shown in A) and D), respectively. Values are means ± Sd (n = 4 independent experiments, with 12 seedlings per independent experiment). In B), asterisks indicate significant differences compared with the corresponding wild type (**P < 0.01, Student's t test). In E), different lowercase letters indicate significant differences (P < 0.05, one-way ANOVA with Tukey's HSD test). C and F) Ion leakage of 2-wk-old plants after 4 °C treatment for 48 h. Values are means ± Sd (n = 4 independent experiments, with 6 seedlings per independent experiment). In C), asterisks indicate significant differences compared with the corresponding wild type (**P < 0.01, Student's t test). In F), different lowercase letters indicate significant differences (P < 0.05, one-way ANOVA with Tukey's HSD test). G and H) Extracellular Ca²⁺ influx upon cold shock in live roots of corresponding wild-type plants and Osbzip46 mutant (G) and OsbZIP46 overexpression lines (H). Three independent experiments were performed with similar results, and values are means ± Sd (n = 5) from 1 experiment. The background indicates the duration of cold treatment. DJ, Dongjin; Nip, Nipponbare. I) Significance testing of the mean maximal Ca²⁺ influx as indicated in G) and H), respectively. Three independent experiments were performed with similar results, and values are means ± Sd (n = 5) from 1 experiment. Asterisks indicate significant differences compared with the corresponding wild type (**P < 0.01, Student's t test). J) Expression levels of cold stress-related genes COLD1, OsDREB1A, OsCPK6, and OsCPK13 in various genetic background plants in response to cold stress. Three independent experiments were performed with similar results, and values are means ± Sd (n = 3) from 1 experiment. Different lowercase letters indicate significant differences (P < 0.05, one-way ANOVA with Tukey's HSD test). OsACTIN1 was used as the internal control.

Figure 7.

OsbZIP46 deacetylation mediated by OsHDA716 represses rice chilling tolerance. A) Chilling tolerance phenotype of wild-type plants, Oshda716-1, OsbZIP46-OE Oshda716-1, and OsbZIP46 overexpression plants. Two-wk-old seedlings were incubated at 4 °C for 7 d, followed by recovery for 7 or 14 d under normal conditions. Bars = 5 cm. Four independent experiments were performed, and representative images were given. HY, Hwayoung. See also Supplementary Fig. S17B. B) Survival rates of various genetic background plants after cold treatment followed by recovery for 14 d as shown in A). Values are means ± Sd (n = 4 independent experiments, with 12 seedlings per independent experiment). C) Ion leakage of 2-wk-old plants after 4 °C treatment for 48 h. Values are means ± Sd (n = 4 independent experiments, with 6 seedlings per independent experiment). D) Expression of cold stress-related genes COLD1, OsDREB1A, OsCPK6, and OsCPK13 in various genetic background plants in response to cold stress. OsACTIN1 was used as the internal control. Three independent experiments were performed with similar results, and values are means ± Sd (n = 3) from 1 experiment. E) Chilling tolerance phenotype of wild-type plants and transgenic plants overexpressing OsbZIP46 or OsbZIP46^KBDR. Two-wk-old seedlings were incubated at 4 °C for 7 d, followed by recovery for 7 or 14 d under normal conditions. Bars = 5 cm. Four independent experiments were performed, and representative images were given. See also Supplementary Fig. S17D. F) Survival rates of different genotype plants after cold treatment followed by recovery for 14 d as shown in E). Values are means ± Sd (n = 4 independent experiments, with 8 seedlings per independent experiment). G) Ion leakage of 2-wk-old plants after 4 °C treatment for 48 h. Values are means ± Sd (n = 4 independent experiments, with 6 seedlings per independent experiment). H) Expression of cold stress-related genes in various genetic background plants in response to cold stress. OsACTIN1 was used as the internal control. Three independent experiments were performed with similar results, and values are means ± Sd (n = 3) from 1 experiment. In B), C), D), F), G), and H), different lowercase letters indicate significant differences (P < 0.05, one-way ANOVA with Tukey's HSD test).

Figure 8.

OsbZIP46 deacetylation mediated by OsHDA716 reduces the transactivation of OsDREB1A and COLD1 and represses the binding to target promoters. A) Schematic diagrams of OsDREB1A and COLD1 promoters. The vertical lines represent the ACGTG motifs recognized by OsbZIP46. B) ChIP-qPCR assays showing that OsHDA716 prevents OsbZIP46 binding to OsDREB1A and COLD1 promoters. Chromatin was immunoprecipitated with the anti-GFP antibody, and the precipitated DNA was quantified by qPCR experiment. DNA enrichment was calculated as a percentage of input DNA. Each sample contained 2 g leaves. Values are means ± Sd (n = 3 independent experiments). Asterisks indicate significant differences (**P < 0.01, ***P < 0.001, ns, not significant, Student's t test). C) EMSA showing OsbZIP46 deacetylation mediated by OsHDA716 reduces the DNA-binding ability. OsDREB1A_pro-probe 1, an oligonucleotide probe including the ACGTG motifs in OsDREB1A promoter (−462 to −501 bp). COLD1_pro-probe 2, an oligonucleotide probe including the ACGTG motifs in COLD1 promoter (−228 to −267 bp). OsbZIP46^KBDR, amino acid substitutions (K263R, K281R, K294R) in the DNA-binding domain in OsbZIP46. D and E) OsHDA716-mediated OsbZIP46 deacetylation represses the transcriptional activity reflected by the LUC/REN ratio (D) and LUC luminescence imaging (E) in N. benthamiana leaves. Values are means ± Sd (n = 3 independent experiments), and different lowercase letters indicate significant differences (P < 0.05, one-way ANOVA with Tukey's HSD test). In E), 3 independent experiments were performed, and representative images were given. REN, renilla luciferase; LUC, firefly luciferase.

Figure 9.

OsHDA716 functions in histone deacetylation of the target genes of OsbZIP46. A) OsHDA716 reduces histone acetylation in plants. Histone acetylation levels (H3K9) were detected by immunoblotting with the anti-H3K9ac antibody in different genotype plants of OsHDA716. Histone 3.1 (H3.1) was used as a loading control. The relative quantification of acetylated H3K9 protein intensities was calculated from 3 independent experiments. Band intensities were measured using ImageJ. B) ChIP-qPCR assays showing OsHDA716 reduces the enrichment of H3K9ac in OsDREB1A and COLD1 promoters. Chromatin was immunoprecipitated with the anti-H3K9ac antibody. The precipitated DNA was quantified by qPCR, and DNA enrichment was calculated as a percentage of input DNA. Each sample contained 2 g leaves. Values are means ± Sd (n = 3 independent experiments), and asterisks indicate significant differences compared with the wild type (*P < 0.05, **P < 0.01, ***P < 0.001, ns, not significant, Student's t test).

Figure 10.

Proposed model deciphering the functional interaction between OsHDA716 and OsbZIP46 to regulate rice cold stress response. Under normal conditions, OsbZIP46 deacetylation mediated by OsHDA716 leads to the protein degradation and inhibits the transcriptional regulation of OsDREB1A and COLD1; OsHDA716 also represses the expression of OsbZIP46 target genes via histone deacetylation. Chilling stress induces OsHDA716 degradation, leading to acetylated OsbZIP46 accumulation with increased transactivation function, which then activates the downstream target genes to confer chilling tolerance in rice. Ac, lysine acetylation.