Introduction

Plants are continuously exposed to various stresses in their natural environment. To acclimate to several kinds of stress conditions at the molecular, cellular, and at the whole plant levels, plants have evolved intricate mechanisms.

Plant growth and development are highly affected by these environmental stresses, such as cold, drought, and salt stresses. Although many tropical plants are cold-sensitive and significant losses of yield are observed under freezing temperatures, most temperate plants have ability to tolerate to freezing temperatures following a period of exposure to non-freezing temperatures. This process is known as cold acclimation. The expression of many genes is induced by the cold signaling pathway during acclimation to cold stresses. These genes include those encoding the following: anti-freezing proteins; molecular chaperones; proteins for respiration; proteins involved in metabolism of carbohydrates, lipids, and antioxidants; and factors that putatively function in freezing-induced dehydration tolerance1. C-repeat (CRT)/dehydration responsive element (DRE) cis-elements are found in the promoters of many cold- and dehydration-responsive genes that encode proteins helping plants to respond to cold2,3. The CBF (CRT-binding factor)/DREB1 (DRE-binding protein) proteins are a family of AP2-type transcription factors, which binds to the core sequence CCGAC of CRT/DRE cis-elements to induce expression of cold- and dehydration-responsive genes4,5. Because the expression of CBF/DREB1 factors is also induced by cold temperatures4,5, other transcription factors are required for the induction of CBF/DREB1 expression in response to cold stress.

Several factors involved in the regulation of CBF/DREB expression have been genetically identified in Arabidopsis1. The ice1 mutation was identified using a genetic screen for mutants, which down-regulated expression of CBF3:LUC reporter gene6. The expression of CBF3/DREB1A and its downstream target genes was down-regulated in the ice1 mutant; accordingly, the ice1 plants significantly exhibited chilling and freezing sensitive6. ICE1, MYC-like bHLH (basic-helix-loop-helix) transcription factor, recognizes the canonical MYC cis-elements (CANNTG) in the CBF3/DREB1A promoter6 to regulate expression of CBF3/DREB1A. A genome-wide transcript profiling demonstrated that the ice1 mutation affected the expression levels of 204 genes among 939 cold-regulated genes7. On the other hand, the expression of CBF1/DREB1B, not CBF3/DREB1A, was primarily influenced by ICE2, a homolog of ICE18. Therefore, ICE1 and ICE2 play important roles in the transcriptional regulation of CBF/DREB1 genes.

Understanding the function of ICE1 and its regulation of cold-responsive genes would contribute to elucidation of ICE1-dependent cold response pathway. In the present study, we performed a yeast 2-hybrid screen and focused on two ICE1-interacting factors, MYC67 and MYC70. Because the myc67 or myc70 mutant exhibited cold tolerance and MYC67- or MYC70-overexpression lines exhibited cold sensitive, these factors play negative role in regulation of cold tolerance. The MYC67 and MYC70 were able to bind to the MYC cis-elements in the promoter of CBF3/DREB1A. The ChIP assay revealed that binding activity of ICE1 to the MYC cis-elements was increased in the myc67 or myc70 mutants. Increased binding activity of ICE1 may enhance cold responsive gene transcription and tolerance. These results suggest that MYC67 or MYC70 inhibits interaction between ICE1 and MYC cis-elements, leading to negative regulation of cold tolerance.

Results

Identification of MYC67 and MYC70 as ICE1-interacting proteins

To identify molecular factors that regulate cold signaling, a yeast 2-hybrid screen was performed using ICE1 as the bait. Because full-length ICE1 and N-terminal region of ICE1 activated the transcription of a GAL4-responsive reporter gene in yeast, we used the C-terminal region of ICE1 (ICE1-C, amino acid residues 267–494) as the bait. Three types of prey libraries from ABRC were screened9,10,11, and 38 clones were confirmed to positively express HIS3, ADE2, and MEL1. A sequence analysis of the prey plasmids revealed that four of the encoded proteins were bHLH-type transcription factors. One of these clones encodes FAMA (At3g24140), which was previously reported to interact with ICE112. The other clones encode bHLH transcription factors, MYC67 and MYC70, which belong to the same subfamily as FAMA13. m2 and m3 encode MYC67 (amino acids 3–358 of At3g61950), and m4 encodes MYC70 (amino acids 246–359 of At2g46810) (Fig. 1A–C).

Figure 1
figure 1

MYC67 and MYC70 interact with ICE1. Yeast 2-hybrid analysis. (A) A schematic representation of the bait (top) and prey (bottom) combinations. The yeast strains contained pAS2-ICE1-C (amino acids 267–494) as bait together with pACT-MYC67 m2 (amino acids 3–358), pACT2-MYC67 full (amino acids 1–358), pACT2-MYC70 m4 (amino acids 246–359), or pACT2-MYC70 full (amino acids1–359) as prey. pAS2-1 and pACT2 were used as negative controls. (B) The yeast strains were grown on synthetic dropout (SD) plates without leucine and tryptophan (−LW) or on selection plates (SP), which consisted of an SD plate containing X-α-gal and lacking leucine, tryptophan, histidine and adenine. (C) The domain structures of MYC67 and MYC70. Acidic, basic, helix-loop-helix (HLH), ZIP domains are presented. Black bars indicate the region of MYC67 m2 (amino acids 3–358) or MYC70 m4 (amino acids 246–359), respectively. (D) BiFC analysis indicated interaction between ICE1 and the MYC proteins. ICE1 was fused to N-terminal EYFP (enhanced yellow fluorescent protein), and the MYC proteins were fused to C-terminal EYFP. Each combination of constructs was transfected into Arabidopsis protoplasts. Differential interference contrast (DIC) images and the YFP fluorescence from the BiFC analysis are shown along with the merged image. The scale bar indicates 10 μm.

Full size image

To confirm the interactions of full-length MYC67 and MYC70 with ICE1-C, the cDNAs were amplified by RT-PCR. The full-length coding region of MYC67 was amplified from a cDNA library that was derived from 10-day-old seedlings, but MYC70 was not amplified from the same cDNA library. To reevaluate the gene structure, the genomic sequence of MYC70 was analyzed using Eukaryotic GeneMark (http://opal.biology.gatech.edu/GeneMark/eukhmm.cgi). The open reading frame predicted by Eukaryotic GeneMark did not match the sequence of At2g46810.1 (Fig. S1A). The predicted coding region of MYC70 (At2g46810.3) was amplified with a pair of primers specific to the predicted ORF, indicating that the predicted ORF was transcribed in vivo. Hereafter, we used At2g46810.3 (AB678434.1 in GenBank database) as the MYC70 full-length ORF (Fig. S1B). The full-length MYC67 and MYC70 proteins were interacted with ICE1-C (Fig. 1B). By using At2g46810.3 for phylogenic analysis (Fig. S2), MYC67 and MYC70 have similar amino acid sequences.

To assess in vivo interactions between ICE1 and the MYC proteins, a bimolecular fluorescence complementation (BiFC) assay was performed14. ICE1 was fused to the N-terminal region of enhanced yellow fluorescence protein (nEYFP), and the MYC proteins were fused to the C-terminal region of EYFP (cEYFP). These constructs were transformed into Arabidopsis mesophyll protoplasts and transiently expressed. Fluorescence at nuclei was detected when ICE1 was co-expressed with MYC67 or MYC70 but not done when ICE1 was co-expressed with a vector (Fig. 1D). These results indicate that ICE1 can interact with these MYC proteins in vivo.

To determine which region of ICE1 interacts with the MYC proteins, several regions of ICE1 were used in a yeast 2-hybrid assay (Fig. 2A). MYC67 m2 and MYC70 m4 were able to interact with region C (amino acids 381–494) of ICE1 (Fig. 2B), suggesting that bHLH region of ICE1 is not required for the interaction of these MYC proteins. Because MYC67 m2 also interacted with region G, the region for interaction with MYC67 was determined to be amino acids 381–409 of ICE1. On the other hand, interaction with MYC70 m4 required the C-terminal region of ICE1 (Fig. 2B).

Figure 2
figure 2

Interactions between ICE1, MYC67, and MYC70. (A) A schematic representation of the deletion constructs for ICE1. The indicated regions of ICE1 were cloned into the bait vector pGBKT7. (B) The interaction of ICE1 with MYC67 or MYC70 was assessed using the yeast 2-hybrid assay. The ICE1 bait plasmid was transformed with either pACT2 (vector control), pACT2-MYC67 m2 (amino acids 3–358), or pACT2-MYC70 m4 (amino acids 246–359). Yeast grown on SD media without leucine and tryptophan (upper, −LW) and on selection plates (lower, SP) are shown. (C) MYC67 and MYC70 form homodimers and interact with each other. Yeast strains containing full-length pGBKT7-MYC67 or pGBKT7-MYC70 as bait were transformed with pACT2 (vector control), full-length pACT2-MYC67 or full-length pACT2-MYC70 as prey. Yeast was incubated on −LW or on SP. (D) Only ICE1 was able to interact with the C-terminal region of MYC70 in the yeast 2-hybrid assay. MYC70 full was interacted with MYC67, MYC70, and ICE1, but MYC70 m4 was able to interact with ICE1. (E) The effect of ICE1 point mutations, S403A or K393R, on the interactions with MYC67 and MYC70.

Full size image

The MYC proteins were also found to form homodimers and interact with each other (Fig. 2C). Although the C-terminal portion (amino acids 246–359, m4) of MYC70, which lacks the bHLH domain, was unable to interact with full-length MYC67 or full-length MYC70 (Fig. 2D). On the other hand, MYC70 m4 interacts with ICE1 (Fig. 2D). This result suggests that homodimer of MYC70 and heterodimer formation between MYC67 and MYC70 requires bHLH domain of MYC70.

Several key amino acid residues are present in the C-terminal region of ICE1, including K393 and S403, which are required for the sumoylation and stabilization of the protein, respectively15,16. Because the C-terminal region of ICE1 was found to be necessary for interaction with MYC67 and MYC70 (Fig. 2B), the interaction of ICE1(K393R) and ICE1(S403A) with MYC67 and MYC70 was examined. These mutated versions of ICE1 were able to interact with MYC67 and MYC70 (Fig. 2E), suggesting that the mutations do not affect interaction.

MYC67 and MYC70 negatively regulate cold tolerance

The homozygous T-DNA insertion lines SALK_009770 and SALK_069605 for MYC67 and MYC70, respectively, were isolated (Fig. S3A). Semi-quantitative RT-PCR revealed that transcripts for the respective MYC genes were not detected in these myc mutants (Fig. S3B). The growth of myc67 or myc70 plants was similar to that of wild-type plants under normal conditions. MYC67- or MYC70-overexpressing lines were also generated for further functional analysis (Fig. S3C).

Three-week-old wild-type and myc plants were subjected to −7 °C for 4 hr after cold acclimation. After a one-week recovery period at room temperature, the mutant plants exhibited a higher survival rate than the wild-type plants (Fig. 3A,B). Because MYC67 and MYC70 have similar amino acid sequences (Fig. S2), the double mutant was generated. The myc67 myc70 double mutant exhibited similar phenotype as did each mutant (Fig. 3B). On the other hand, MYC-overexpression transgenic plants that were subjected to −6 °C for 4 hr after cold acclimation were more sensitive to freezing temperature than wild-type plants (Fig. 3C,D). These results indicate that MYC67 and MYC70 negatively regulate cold tolerance. To confirm complementation of the myc67 or myc70 mutation with MYC67 or MYC70, respectively, MYC67pro::FLAG:MYC67 in myc67 or MYC70pro::FLAG:MYC70 in myc70 was generated (Fig. S3). The plants with similar expression level of MYC67 or MYC70, compared to that in wild type, was used for freezing assay (Fig. 3E,F), demonstrating that each MYC gene was able to complement each mutation.

Figure 3
figure 3

MYC67 and MYC70 negatively regulate freezing tolerance. (A) The freezing tolerance of myc67 and myc70 after cold acclimation. Representative wild-type (i), myc67 (ii) and myc70 (iii) plants that were treated with −7 °C are shown. (B) Quantification of the survival rates for the cold-acclimated wild-type, myc67, and myc70 plants after a −7 °C freezing treatment. The data indicate the means and standard errors (n = 6). The tolerance of myc67, myc70, and myc67 myc70 mutants was significantly different from that of wild-type plants (t-test, P < 0.05). (C) MYC67-, and MYC70-overexpressing plants exhibited freezing sensitivity. The plants shown are representative wild-type plants (i), MYC67-overexpressing plants #4 (ii) and #5 (iii), and MYC70-overexpressing plants #1 (iv) and #6 (v). (D) Quantification of the survival rates for the cold-acclimated MYC67-, and MYC70-overexpressing plants after a −6 °C freezing treatment. The data indicate the means and standard errors (n = 6). The survival ratios of MYC67-, and MYC70-overexpressing plants were significantly different than that of wild-type plants (t-test, P < 0.05). (E) MYC67pro:: FLAG:MYC67 or MYC70pro:: FLAG:MYC70 was expressed in myc67 or myc70 mutant, respectively. The mutant phenotype was complemented. The plants shown are representative wild-type (i), myc67 plants with MYC67pro::FLAG:MYC67 #2-1 (ii) and #2-4 (iii), and myc70 plants with MYC70pro:: FLAG:MYC70 #3-1 (iv) and #5-2 (v). (F) Quantification of the survival rates for the cold-acclimated myc67 plants with MYC67pro::FLAG:MYC67 and myc70 plants with MYC70pro:: FLAG:MYC70 after freezing treatment at −6 °C. The data indicate the means and standard errors (n = 6).

Full size image

Several cold-responsive genes are ICE1-target genes7. Thus, quantitative RT-PCR analysis was performed. The cold-induced transcript accumulation of CBF/DREB1 genes and their dependent cold-responsive genes, including COR47, COR15A, KIN1, and P5CS2, was increased in the myc67 and myc70 mutants (Fig. 4A). Additionally, the expression of these genes was reduced in MYC67- and MYC70-overexpression lines (Fig. 4B), suggesting that these MYC genes act as negative regulators of cold signaling. The expression of MYC67 was decreased after cold treatment, whereas expression level of MYC70 was not significantly altered in the presence or absence of cold treatment (Fig. 4C). Each mutation did not affect expression of another MYC gene. In the ice1-D mutant, MYC70 was highly expressed in normal condition and was decreased after cold treatment (Fig. 4D). Expression level of MYC70 was increased in the ice1-D mutant.

Figure 4
figure 4

MYC67 and MYC70 negatively regulate the expression of cold-responsive genes. (A) The relative mRNA transcript levels of CBF/DREB1 genes, COR47, COR15A, KIN1, and P5CS2 in wild-type, myc67 and myc70 seedlings were determined by quantitative RT-PCR analysis. Ten-day-old seedlings grown at 23 °C were incubated at 4 °C for the indicated amounts of time. The data indicate the means and SD (n = 3). (B) The relative mRNA transcript levels in wild-type, MYC67-, and MYC70-overexpressing seedlings were determined by quantitative RT-PCR. The data indicate the means and SD (n = 3). (C) The expression levels of MYC67 and MYC70 after cold treatment were measured by quantitative RT-PCR analysis. (D) Expression level of MYC67 and MYC70 after cold treatment in wild type and the ice1-D mutant. Asterisks indicates a significant difference from wild-type or plants with a vector at each point (p < 0.05) as determined by Student’s t-tests.

Full size image

Because ICE1 is involved in stomatal development12, the stomata on the abaxial surfaces of rosette leaves of the myc67 or myc70 were observed. The mature guard cells of the myc mutants were indistinguishable from wild-type guard cells (Fig. S4).

Expression and localization of MYC67 and MYC70

The expression pattern of each MYC gene was analyzed using promoter:GUS fusion transgenes. Six-week-old plants harboring MYC67pro:GUS or MYC70pro:GUS was treated with or without low temperature. Then, MYC67pro:GUS or MYC70pro:GUS expression was observed in base of rosetta leaves and floral organs (Fig. 5A,B). In mature leaves, only MYC70pro::GUS expression was observed (Fig. 5B). Then, expression pattern was compared after cold treatment. Before or after cold treatment, expression pattern of MYC67pro::GUS and MYC70pro::GUS was not altered (Fig. 5A,B). Expression level of MYC70pro::GUS was slightly increased, probably because mRNA level of MYC70 was not altered under cold treatment (Fig. 4C) and GUS protein may be accumulated after cold treatment. Even though no signal of MYC67pro::GUS was observed in 6-week-old rosetta leaves, the expression was observed around wound sites with normal condition (Fig. S5). It is suggested that expression of MYC67 is response to wound.

Figure 5
figure 5

Expression and localization of MYC67 and MYC70. (A) Six-week-old plants expressing MYC67pro:GUS was treated without (i–iii) or with (iv–vi) low temperature for 6 h. MYC67pro:GUS expression in mature rosetta leaves (i,iv), base of rosetta leaves (ii, v), and floral organs (iii,vi). (B) Six-week-old plants expressing MYC70pro:GUS was treated without (i–iii) or with (iv–vi) low temperature for 6 h. MYC70pro:GUS expression in mature rosetta leaves (i, iv), base of rosetta leaves (ii,v), and floral organs (iii, vi). (C) Subcellular localization of the GFP-MYC proteins. The nuclei were stained with 20 mg/ml PI. The scale bar indicates 20 μm. (D) GFP fluorescence at the root tips. Seven-day old seedlings were treated at 4 °C for 3 h, and GFP fluorescence was observed by confocal microscopy. The scale bar indicates 150 μm.

Full size image

To evaluate the localization of the MYC proteins, transgenic plants expressing GFP-MYC fusion proteins under the control of the constitutive CaMV 35 S promoter were prepared. GFP fluorescence was observed in the nuclei, indicating that MYC67 and MYC70 proteins are localized to the nucleus (Fig. 5B). All GFP-MYC transgenic plants were subjected to low-temperature conditions (4 °C) for 3 h to assess the effects on GFP emission. The intensities of GFP-MYC67 and GFP-MYC70 fluorescence were similar to those in non-treated plants (Fig. 5C).

MYC67 and MYC70 bind to MYC cis-elements in the CBF3/DREB1A promoter

There are five possible MYC-recognition elements in the CBF3/DREB1A promoter, and ICE1 interacts specifically with these MYC cis-elements6. Because MYC67 and MYC70 have bHLH domains, it is likely that these MYC proteins recognize MYC cis-elements.

To investigate the binding activity of MYC proteins at the promoter of CBF3/DREB1A in vivo, a chromatin immunoprecipitation (ChIP) assay was performed using an anti-DYKDDDDK tag monoclonal antibody and the immunoprecipitated DNA was amplified (Fig. 6A,B). After cold treatment, the binding activity of MYC67 and MYC70 to the CBF3/DREB1A promoter was increased (Fig. 6B). It is probably because negative regulators are also important in cells for fine tuning. When ChIP assay was performed using an anti-ICE1 antibody in the wild-type or myc mutant, the binding activity of ICE1 to the CBF3/DREB1A promoter was increased in the myc67 or myc70 mutant (Fig. 6C). Because no canonical MYC cis-element in the region a, no significant binding activity was detected (Fig. 6B,C). These results suggest that the binding activity of ICE1 is suppressed by MYC67 or MYC70, probably leading to suppression of some downstream cold responsive genes and tolerance.

Figure 6
figure 6

MYC67 and MYC70 were able to bind to the promoter region of CBF3/DREB1A. (A) A schematic representation of the promoter region of CBF3/DREB1A and the region that was amplified after the ChIP analysis. (B) The wild-type and transgenic plants used for the ChIP assay were treated at 4 °C for 3 h. The immunoprecipitated DNA with anti-DYKDDDDK tag monoclonal antibody was quantified by quantitative PCR using primers in the promoter region. The data are the average of technical triplicates for the quantitative PCR (mean ± SD). (C) The wild-type and myc mutants were treated at 4 °C for 3 h, and ChIP analysis was performed with anti-ICE1 antibody.

Full size image

Discussion

In the present study, we isolated bHLH-type transcription factors, MYC67 and MYC70, that function as ICE1 interactors. T-DNA insertions in these genes improved cold tolerance and enhanced the expression of cold-regulated genes. Furthermore, the overexpression of MYC67 and MYC70 decreased cold tolerance and down-regulated cold-regulated genes. These results indicate that MYC67 and MYC70 act as negative regulators of cold signaling and tolerance in Arabidopsis. Additionally, these MYC proteins recognize the MYC cis-elements in the promoters of CBF/DREB1 genes. The binding activity of ICE1 to the promoter of CBF3/DREB1A may be suppressed by MYC67 and MYC70, leading to reduction of cold tolerance.

The MYC proteins were able to bind to the MYC cis-elements in the promoters of CBF/DREB1 genes (Fig. 6), which were previously shown to be bound by ICE16. These MYC proteins were also able to form homodimers and heterodimers (Fig. 2C). The heterodimerization of bHLH proteins can either stimulate or inhibit transcription. Previous characterization of the Myc/Max/Mad network has shown that Max can form heterodimers with either Myc or Mad in vertebrates. The Myc-Max heterodimer acts as a transcriptional activator, whereas the Mad-Max heterodimer acts as a transcriptional repressor17. Another reason may be the heterodimer of MYC67 and MYC70 cover MYC cis-elements to inhibit binding of ICE1 to MYC cis-elements. Because the myc67 myc70 double mutant exhibited similar cold tolerance as did each mutant (Fig. 3B), it is suggested that hetero-dimerization of MYC67 and MYC70 may be required for suppression of cold tolerance. According to ChIP analysis, binding activity of ICE1 to MYC cis-elements was increased in cold-treated plants and in the myc67 or myc70 mutant (Fig. 6C). Together, these cold-induced changes of binding activity may contribute to the induction of cold signaling and cold acclimation and MYC67 and MYC70 may inhibit activity of ICE1 to suppress cold signaling and tolerance.

We previously identified K393 and S403 as important residues in ICE115,16. K393 is a target for sumoylation that is mediated by the SUMO E3 ligase SIZ1 and the sumoylation of ICE1 is important for the inhibition of ubiquitylation and cold tolerance15. The substitution of S403 for alanine in ICE1 enhances the following: the stabilization of ICE1; the expression of cold-regulated genes, including CBF3/DREB1A, COR47, and KIN1; and cold tolerance16. The interaction between ICE1 and MYC67 or MYC70 was observed (Fig. 2E), suggesting that the interaction may be unaffected by posttranslational modification. Although ICE1 primarily affects the expression of CBF3/DREB1A6, ICE2 primarily influences the expression of CBF1/DREB1B and not CBF3/DREB1A8. The increased expression of CBF1/DREB1B in the myc67 or myc70 mutant (Fig. 4A) suggest that MYC67 and MYC70 may interact with ICE2 or affect its activity. Among 1256 cold-induced genes, 11% of these genes were upregulated by overexpression of CBF/DREB118. The low-temperature regulatory network is highly interconnected by CBF/DREB1 and other transcription factors18. Regulation of other transcription factors by ICE1, MYC67, or MYC70 is to be elucidated.

Phosphorylation of ICE1 at S278 by OST1/SnRK2.6 kinase enhances stability of ICE1 through interfering with the interaction between HOS1 and ICE119. High-density protein microarray analysis identified 570 phosphorylation substrates of MPK (MAP kinase)20. According to these data, ICE1 is phosphorylated by MPK5 and MPK6. In the previous study, MKK2 (MAP kinase kinase 2) phosphorylates MPK4 and MPK6 in response to cold and the overexpression of a constitutively active MKK2 in plants confers cold and salt stress tolerance21. These results suggest that phosphorylation of ICE1 activates its transactivation activity. On the other hand, neither MYC67 nor MYC70 was found to be a substrate of MPK in the protein array20, suggesting that regulatory mechanism may be different.

In summary, MYC67 and MYC70 interact with ICE1 and act as negative regulators of cold signaling. Because the ICE1-dependent cold signaling cascade is conserved in several plants, such as tomato22,23, rice24, and wheat25, similar MYC interacting proteins may regulate cold tolerance through the expression of CBF/DREB and cold-regulated genes in various species.

Methods

Yeast 2-hybrid assay

The C-terminal region of ICE1 (amino acids 267–494) was amplified from pMAL-ICE16 with 5a-1 and MAL-3 primers. The PCR product was digested with SmaI and SalI, cloned into SmaI and SalI sites of the pAS2 vector (Table S1). pAS2-ICE1-C was transformed into the AH109 Saccharomyces cerevisiae yeast strain (Matchmaker GAL4 Two-Hybrid System 3, Clontech) and incubated on selective medium lacking tryptophan (SD-W). cDNA libraries (CD4-10, CD4-22, CD4-30) were obtained from the Arabidopsis Biological Resource Center and then transformed into the AH109 yeast strain containing pAS2-ICE1-C, which was then incubated on SD lacking tryptophan, leucine, histidine, and adenine (SD-WLHA). Colonies were transferred to SD-WLHA with 20 μg/ml of X-α-Gal (5-bromo-4-chloro-3-indolyl α-D-galactopyranoside) to confirm the expression of the MEL1 reporter gene, which is positive for α-galactosidase activity. The prey vector was obtained from yeast and transformed into yeast containing pAS2-ICE1-C or pAS2 (the vector control); the transformed yeast was then incubated on SD-WLHA containing X-α-Gal. Prey vectors were sequenced when the yeast harboring the prey vector and pAS2-ICE1-C survived and the yeast with the prey vector and pAS2 did not.

The full-length coding regions of the MYC67 and MYC70 prey vectors were constructed as follows. At2g46810.3 was amplified by PCR using primers MYC70-4 and MYC70-5 (Table S1). The coding regions of MYC67 and MYC70 were amplified by PCR using gene-specific primers (MYC67-1, MYC67-2, MYC70-7, and MYC70-2; Table S1), digested with XmaI and SalI, and introduced into XmaI- and XhoI-digested pACT2 or into XmaI- and SalI-digested pGBKT7. The plasmids of the bait and prey constructs were transformed into the yeast AH109 strain. The positive colony was obtained from SD-WLHA with X-α-Gal.

To construct the ICE1-deletion clones (A-H) and ICE1-mutant clones (K393R and S403A), portions of wild-type ICE1, ICE1(K393R) and ICE1(S403A) were amplified by PCR using specific primers (Table S1), digested with EcoRI and BamHI, and introduced into pGBKT7 at the corresponding restriction sites.

BiFC assay

The coding region of ICE1 was amplified using primers AtICE1-17 and AtICE1-3 (Table S1). The PCR products were digested with EcoRI and BamHI and then inserted into the EcoRI and BamHI sites of the pSAT6-nEYFP-C1 vector14. To make the cEYFP-MYC plasmids, the coding regions of MYC67 and MYC70 were amplified with primers MYC67-7, MYC70-11, and NOS-transR (Table S1). The resultant PCR products were digested with EcoRI and SalI and inserted into the EcoRI and SalI sites of the pSAT6-cEYFP-C1 vector14. The constructs were transiently expressed in Arabidopsis leaf protoplasts, which were prepared from two-week-old wild-type seedlings using Cellulase Ononzuka R-10 and Macerozyme R-10 (Yakult Pharmaceutical) as previously described26. The plasmid DNA of nEYFP-ICE1 and cEYFP-MYC was introduced into Arabidopsis protoplasts by polyethylene glycol-mediated transfection27. After an overnight incubation at 22 °C, the fluorescent and differential interference contrast (DIC) images were observed with a microscope (IRBE-4, Leica).

Plant materials and growth conditions

The Arabidopsis T-DNA insertion mutants myc67-1 (SALK_009770) and myc70-1 (SALK_069605) were obtained from the Arabidopsis Biological Resource Center at Ohio State University. T-DNA insertion was confirmed by genotyping with primers SALK009770L, SALK009770R, SALK069605L, SALK069605R, and Salk_LBa1 (Table S3).

The T7 tag was introduced into the CaMV35S-Nos vector by inverse PCR using primers T7-1 and T7-2 (Table S1). The resultant CaMV35S-T7tag-Nos vector was digested with HindIII and EcoRI, and the fragment was introduced into the plant binary vector pCAMBIA1300 to generate pCAMH-35S-T7tag-Nos. The coding regions of MYC67 and MYC70 were excised from pGBKT7-MYC67 and pGBKT7-MYC70, respectively, by digestion with XmaI and SalI. The resultant fragments were introduced into the XmaI and XhoI sites of pCAMH-35S-T7tag-Nos. pCAMH-35S-T7-MYC67 and pCAMH-35S-T7-MYC70 were transformed into Arabidopsis (Col-0 or the respective mutant for complementation) by Agrobacterium (strain GV3101)-mediated transformation16. The transgenic lines were selected on 1/2 MS agar plates with 30 μg/mL hygromycin. The transgenic plants were selected, and quantitative RT-PCR was performed using gene-specific forward primers (MYC67-5 and MYC70-10) and NOS-transR (Table S2) to determine the expression levels of each transgene (Fig. 3E).

The whole plant freezing assay was performed as previously described16. Cold-acclimated plants were treated at −6 °C or −7 °C for 4 hr in a programmed incubator (IN602, Yamato, Japan). After the freezing treatment, the plants were stored overnight at 4 °C and then transferred to 23 °C. The survival ratio was determined one week after the freezing treatment.

RNA preparation and quantitative RT-PCR

Wild-type, myc mutant, and MYC-overexpressing plants were grown on 1/2 MS agar plates at 23 °C for 10 d under a long-day photoperiod (16 h light/8 h dark). The plants were then subjected to cold treatments at 4 °C. Total RNA was isolated using TRIZOL reagent (Invitrogen) according to the manufacturer’s instructions16. cDNA was synthesized from 2 μg of total RNA using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, USA). The primers used for real-time PCR are listed in Table S2. Real-time PCR amplification and detection were carried out using SYBR Premix ExTaq (Takara Bio, Japan) on a Thermal Cycler Dice TP800 (Takara Bio, Japan). The relative transcript abundance was calculated using the comparative CT method as previously described15, with Actin2/7 as the control standard. Values representative the means ± SD (n = 3) from representative experiments from three or more independent experiments.

Microscopic Analysis

The coding region of GFP was amplified with the primers GFPN-1 and GFPN-2 (See Supplemental Table S1). The resultant PCR product was inserted into the SmaI site of the CaMV35S-Nos vector to make CaMV35S-GFPN-Nos. The plasmid DNA with GFP in the sense direction was digested with HindIII and EcoRI, and the fragment was introduced into the plant binary vector pCAMBIA1300 to make pCAMH-35S-GFPN-Nos. The coding regions of MYC67 and MYC70 were excised from pGBKT7-MYC67 and pGBKT7-MYC70, respectively, by digestion with XmaI and SalI. The fragments were then introduced into the XmaI and XhoI sites of pCAMH-35S-GFPN-Nos.

The GFP-MYC67 and GFP-MYC70 plasmids were introduced into Agrobacterium and transformed into wild-type Arabidopsis plants (Col-0). T2 transgenic lines that were resistant to hygromycin were chosen for the analysis of GFP expression. The fluorescence of the GFP-MYC fusion proteins was observed with a confocal laser-scanning microscope (TCS SP2, Leica).

The promoter region of MYC67 (1086 bp upstream from the initiation codon) was amplified using gene-specific primers (MYC67pro-pCAMBIA1391Z-F and MYC67pro-pCAMBIA1391Z-R; Table S1). The PCR product was digested with EcoRI and NcoI and were inserted into pCAMBIA1391Z with In-Fusion HD Cloning Kit (Takara Bio, Japan) to generate MYC67pro-GUS. The promoter region of MYC70 (1474 bp upstream from the initiation codon) was amplified using gene-specific primers (MYC70.2pro-pCAMBIA1391Z-F and MYC70.2pro-pCAMBIA1391Z-R; Table S1). The PCR products was inserted into EcoRI and NcoI and were inserted into pCAMBIA1391Z to generate MYC70pro-GUS. GUS staining was performed as previously described28. Representative seedlings were photographed using a microscope (IRBE-4, Leica).

The DIC images were used for the quantitative and qualitative examination of the epidermal phenotypes. Seven-day-old seedlings were fixed in 1:9 (v:v) acetic acid:ethanol overnight, rinsed with water and then placed in chloral hydrate (1:8:1 glycerol:chloral hydrate:water v:w:v). The samples were visualized using a microscope (IRBE-4, Leica). To score the stomatal number, the number of cells in 0.25 mm2 regions of wild-type and mutant leaves were counted, and the means were calculated.

Chromatin Immunoprecipitation (ChIP) Assay

To investigate whether MYC proteins were interacted with the promoter region of CBFs, ChIP assay was performed. Transgenic plants expressing MYC67pro::FLAG:MYC67 or MYC70pro::FLAG:MYC70 were generated. FLAG fused with MYC67 or MYC70 was amplified with the primers, pRI201-3xFLAG-F, 3xFLAG-R, 3xFLAG-MYC-F, and MYC-pRI201-R (Table S1), then PCR products were introduced into pRI201-AN (Takara Bio, Japan) digested with NdeI and SalI with In-Fusion HD Cloning Kit (Takara Bio, Japan). After construction, 35 S promoter was exchanged with MYC67pro or MYC70pro, which was amplified with the primers, pRI201-MYC67pro-F and pRI201-MYC67pro-R, or pRI201-MYC70.2pro-F and pRI201-MYC70.2pro-R, respectively. The DNA region of MYC67pro::FLAG:MYC67:HSPterminator or MYC70pro::FLAG:MYC70:HSPterminator was amplified with the primers pCsV-MYC67pro-F and pCsV-HSPter-R, or pCsV-MYC70.2pro-F and pCsV-HSPter-R, respectively, and the PCR products were introduced into PvuII digested pCsV1300 with In-Fusion HD Cloning Kit. The constructs were designated as MYC67pro::FLAG:MYC67-CsV and MYC70.2pro::FLAG:MYC70.2-CsV, which were transformed into myc67 or myc70 mutants, respectively. Expression level of FLAG:MYC67 or FLAG:MYC70 was examined by RT-PCR with the primers 3xFLAG-F and 3xFLAG-R (Fig. S3).

Two-week-old seedlings were treated with or without cold (4 °C) for 3 h, and the seedlings were subsequently harvested and cross-linked with buffer A (0.4 M sucrose; 10 mM Tris-HCl, pH 8; 1 mM EDTA; 1 mM PMSF; 1% formaldehyde). The cross-linking reaction was stopped with 125 mM glycine. The tissues were ground with liquid nitrogen, resuspended with lysis buffer (50 mM HEPES, pH 7.5; 150 mM NaCl; 1 mM EDTA; 1% Triton X-100; 0.1% deoxycholate; 0.1% SDS; 1 mM PMSF; and 1X protease inhibitor cocktail II), and sonicated (Microson XL-2000 sonicator, Qsonica, LLC., USA) to achieve an average fragment size of 0.1 to 1.0 kb. After sonication, centrifugation was performed, and the supernatants were incubated overnight at 4 °C with anti-DYKDDDDK tag monoclonal antibody (Wako Pure Chemical Industies, Japan) or anti-ICE1 rabbit polyclonal antibody generated from peptide CQEDGSSQFKPMLEG (Scrum Inc., Japan). The immunoprecipitation was performed using Magna ChIP A Chromatin Immunoprecipitation Kits (Millipore).

Quantitative PCR was performed with THUNDERBIRD SYBR Premix (Toyobo) using a real-time PCR Thermal Cycler Dice (Takara) as previously described16. The relative quantities of immunoprecipitated DNA fragments were calculated as the percentage of input chromatin that was immunoprecipitated using the comparative CT method. The primer sequences are listed in Supplemental Table S4.

Accession numbers

Sequence data from this article can be found in the GenBank/EMBL databases under the following accession number: MYC70, At2g46810.3 (AB678434.1). Other sequence data can be found in the Arabidopsis Genome Initiative database under the following accession numbers: ACT2/7, At5g09810 (NP_196543); ICE1, At3g26744 (NP_001030774); MYC67, At3g61950 (NP_001325697).