Ethylene positively regulates cold tolerance in grapevine by modulating the expression of ETHYLENE RESPONSE FACTOR 057

Ethylene (ET) is a gaseous plant hormone that plays essential roles in biotic and abiotic stress responses in plants. However, the role of ET in cold tolerance varies in different species. This study revealed that low temperature promotes the release of ET in grapevine. The treatment of exogenous 1-aminocyclopropane-1-carboxylate increased the cold tolerance of grapevine. By contrast, the application of the ET biosynthesis inhibitor aminoethoxyvinylglycine reduced the cold tolerance of grapevine. This finding suggested that ET positively affected cold stress responses in grapevine. The expression of VaERF057, an ET signaling downstream gene, was strongly induced by low temperature. The overexpression of VaERF057 also enhanced the cold tolerance of Arabidopsis. Under cold treatment, malondialdehyde content was lower and superoxide dismutase, peroxidase, and catalase activities were higher in transgenic lines than in wild-type plants. RNA-Seq results showed that 32 stress-related genes, such as CBF1-3, were upregulated in VaERF057-overexpressing transgenic line. Yeast one-hybrid results further demonstrated that VaERF057 specifically binds to GCC-box and DRE motifs. Thus, VaERF057 may directly regulate the expression of its target stress-responsive genes by interacting with a GCC-box or a DRE element. Our work confirmed that ET positively regulates cold tolerance in grapevine by modulating the expression of VaERF057.


Results
Low temperature increased ET production in grapevine. The function of ET in cold tolerance varies in different plant species. ET biosynthesis increases or decreases under cold stress depending on the plant species 15,16,34,35 . To investigate ET production under cold stress in grapevine, we first determined ET biosynthesis in V. amurensis and 'Muscat Hamburg' plantlets that were subjected to low temperature (4 °C). A similar evolution of ET production was observed in both species throughout the treatment with low temperature (Fig. 1a). ET production rapidly increased after initiating the cold stress, reached the maximum at 8 h, and then decreased to the original level at 48 h (Fig. 1a). However, the maximum changes in the two grape species were slightly different (1.7-fold in V. amurensis and 2-fold in 'Muscat Hamburg'). These results indicated that ET biosynthesis in grapevine was remarkably enhanced at the early period of cold stress.
ACC is a direct product of ACS and its content can be used as an indicator of ACS activity. ACC content in the two grape species rapidly increased, reached the maximum at 8 h after initiating the cold stress, and then decreased (Fig. 1b). The greatest change in ACC content was up to 4.3-fold in V. amurensis and up to 2.6-fold in 'Muscat Hamburg' . Similar to the changes in ACC content, ACO activity under cold stress first increased and then decreased (Fig. 1c). Moreover, the fold changes in ACO activity were greater in V. amurensis than in 'Muscat Hamburg' . ACO is an enzyme that catalyzes the last step of the ET biosynthetic pathway by converting ACC into ET. The results of the evolution of both ACC content and ACO activity strongly support the increase in ET production in grapevines during the early stage of cold stress.

Application of ACC and AVG respectively enhanced and reduced the cold tolerance of grapevine.
To establish the relationship between ET production and cold tolerance, we measured the semi-lethal temperature (LT 50 ) of grapevine leaves treated with ET precursor (ACC) and ACS inhibitor (AVG). Application of exogenous ACC reduced LT 50 value from −9.2 °C to −11.3 °C in V. amurensis and from −7.3 °C to −10.2 °C in 'Muscat Hamburg' (Fig. 2). The application of AVG (100 μM) under cold stress largely inhibited ACC biosynthesis and ACO activity in grapevine (Fig. 1b,c), indicating that AVG can inhibit ET production in grapevine under low temperature. By contrast, application of exogenous AVG increased LT 50 values to −8.1 °C in V. amurensis and −5.7 °C in 'Muscat Hamburg' (Fig. 2). Previous results indicated that an increased ET production resulting from exogenous application of ET precursor enhances cold resistance, whereas a reduced cold tolerance was expected from suppression of ET production both in V. amurensis and 'Muscat Hamburg' . For AVG treatment, the plantlets were treated with 100 μM AVG under cold stress, and distilled water was used as control. Data are the mean values ± SE of three biological replicates. **and *indicate significant differences compared with the point of 0 h at P < 0.01 and P < 0.05 (Student's t-test), respectively. Plantlets of V. amurensis and 'Muscat Hamburg' were treated with 100 μM ACC or 100 μM AVG, and distilled water was used as control (CK). LT 50 values were calculated after the plants were treated with ACC and AVG for 8 h. Data are the mean values ± SE of three biological replicates. **and *indicate significant differences compared with the CK at P < 0.01 and P < 0.05 (Student's t-test), respectively. Isolation and characterization of VaERF057 and VvERF057. The ERF transcription factors found downstream of the ET signal pathway play an important role in response to abiotic stresses, including salt, drought, and cold stresses 18,36 . Our previous study revealed that ERF057 is significantly upregulated under cold stress 33 . We isolated the VaERF057 and VvERF057 genes from V. amurensis and 'Muscat Hamburg' , respectively. The putative protein sequences of VaERF057 and VvERF057 showed 97.5% identity with proteins containing the same AP2/ERF domain (Fig. S1). Analysis of the conserved AP2/ERF domain of 59 amino acids showed that the AP2/ERF domain shares over 90% identity with its homologs from different plant species (Fig. 3a). Based on the conventional classification 37 , the AP2/ERF domain is divided into two conserved segments (YRG and RAYD), and the protein formed contains three β -sheets and a single α -helix (Fig. 3a). Phylogenetic analysis further revealed the evolutionary relationship to the ERFs of Arabidopsis, suggesting that VaERF057 and VvERF057 belong to Group VII of the ERF family (Fig. 3b). Given that no significant differences exist between VaERF057 and VvERF057, the functional studies described below mainly focused on VaERF057. Three β -sheets and one α -helix of the AP2/ERF domain are marked above their corresponding sequences. The YRG and RAYD elements are indicated below the consensus sequence. The GenBank accession numbers are as follows: Malus domestica MdERF73 (XP_008369034), MdERF71 (XP_008386662), Arabidopsis thaliana AtERF73 (AT1G72360), ATERF71 (AT2G47520), Solanum tuberosum StERF73 (XP_006359865), Cucumis sativus CsERF73 (XP_004152238), and Prunus mume PmERF71 (XP_008235129). (b) Phylogenetic tree of VaERF057 and VvERF057 with ERFs from Arabidopsis. The Arabidopsis ERF sequences were downloaded from TAIR. The protein sequences were aligned using Clustal W2, and phylogenetic tree was generated by MEGA6.0 software using the neighbor-joining method. Increased expression levels of VaERF057 and VvERF057 were dependent on enhanced ET production under cold stress. The expression patterns of VaERF057 and VvERF057 in response to cold, ACC and AVG were analyzed by quantitative real time RT-PCR (qRT-PCR). Transcript levels of VaERF057 and VvERF057 rapidly increased after initiating cold stress and peaked at 12 h (6.2-fold) and 48 h (12.8-fold), respectively (Fig. 4a). Expression of VaERF057 and VvERF057 also rapidly increased in the plants after ACC application (Fig. 4b). The expression of VaERF057 increased by 8.2-fold at 2 h and VvERF057 increased by 15.3-fold at 4 h. The effects of cold stress on the expression of VaERF057 and VvERF057 were totally inhibited when the plants were subjected with AVG under cold condition (Fig. 4a,c). These results suggested that the response of VaERF057 and VvERF057 possibly relies on the increased ET release during cold stress. stress responses, we generated 10 T3 homozygous transgenic Arabidopsis lines, which were confirmed by genomic DNA PCR and qRT-PCR (data not shown). Three representative transgenic lines (OE1, OE2, and OE3) displaying high VaERF057 expression levels were used in subsequent experiments. VaERF057-overexpressing Arabidopsis plants displayed significantly improved cold tolerance (Fig. 5a). The average survival rate of the VaERF057-overexpressing plants was over 85% after cold stress at − 11 °C, whereas that of the wild-type (WT) and empty vector (EV) plants demonstrated only 21.4% and 23.8% survival rate, respectively (Fig. 5b).
To further investigate the physiological changes under cold treatment in the VaERF057-overexpressing Arabidopsis, we determined the four important physiological indexes, namely, MDA, SOD, POD, and CAT, which reflect plant cold tolerance. No significant differences were observed among transgenic lines and the controls (WT and EV) under normal conditions (Fig. 6). However, under cold stress, MDA content was significantly lower in transgenic plants than in WT and EV plants (Fig. 6a), whereas SOD, POD, and CAT activities were significantly increased in the transgenic plants compared with that in the WT and EV plants ( Fig. 6b-d). These results suggested that the tissues of the VaERF057-overexpressing plants were less injured as reflected in their low MDA index, and this phenomenon should increase SOD, POD, and CAT activities to improve tolerance of Arabidopsis to cold stress.

Transcriptome analysis of the VaERF057-overexpressing Arabidopsis revealed altered expression of genes involved in biotic and abiotic stress responses.
To determine the possible direct or indirect targets of VaERF057, we performed whole transcriptome sequencing (RNA-Seq) of two biological replicates of WT and three transgenic Arabidopsis lines mixture (OE1, OE2, and OE3) under normal conditions. A total of 140 significantly differentially expressed genes demonstrating changes of more than 2-fold were found in the transgenic lines compared with the WT plants. Among these altered genes, 89 were upregulated and 51 were downregulated (Table S2). To confirm the RNA-Seq results, we verified the 23 significantly differentially expressed genes, including 16 upregulated and 7 downregulated genes by using qRT-PCR (Fig. S3). All of these verified genes showed similar changes in expression as revealed by RNA-Seq and qRT-PCR, indicating the reliability of the transcription profile analysis.
To gain insight into the functions of these genes, we performed Gene Ontology (GO) term enrichment analysis by using agriGO v1.2 in default parameters 38 . The most significantly enriched biological processes among these significantly differentially expressed genes in transgenic Arabidopsis are involved in responding to stimuli (including biotic and abiotic stimuli, cold, heat, water deprivation, hypoxia, wounding, pathogen, and osmotic and oxidative stresses), metabolic processes, biological regulation, and signal transduction (Fig. S2).
Among the significantly differentially expressed genes, 39 genes are involved in cold, drought, salt, hypoxia, and pathogen stresses; 32 of these genes were upregulated and 7 were downregulated (Table 1). Many key genes involved in these stress tolerances were identified. For instance, CBF1, CBF2, and CBF3, which encode a small family of transcriptional activators that play an important role in freezing tolerance and cold acclimation in Arabidopsis 39,40 , were upregulated by 3.9-, 2.4-, and 4.5-fold, respectively. NCED3 (upregulated by 2.7-fold) is a key gene in the ABA biosynthesis pathway and enhances plant tolerance to abiotic stresses by regulating endogenous ABA biosynthesis 41,42 . These results showed that VaERF057 overexpression can regulate expression of many genes involved in biotic and abiotic stress response. The enhanced cold tolerance of the VaERF057-overexpressing Arabidopsis suggested that VaERF057 directly or indirectly regulates expression of these stress-responsive genes.

VaERF057 demonstrated binding activity to GCC-box and DRE motifs. Previous studies have
shown that some ERF proteins can bind to GCC-box or DRE motif 24,25,43 . Especially, AtERF71/HRE2, the closest homologue of VaERF057 in Arabidopsis, can bind to both GCC-box and DRE motifs 43 . To determine whether VaERF057 can also bind to GCC-box and DRE motifs, we cloned four tandem copies of GCC-box, DRE, or their mutants into pAbAi vector, which was interacted with the pGADT7 activation domain vector harboring VaERF057 by using Y1H system. Figure 7 shows that VaERF057 can specifically bind to both GCC-box and DRE motifs but failed to bind to all of the mutated GCC-box and DRE motifs. Among the 140 significantly differentially expressed genes, 24 genes contain GCC-box and 54 genes contain at least one DRE core motif upstream of their 1 kb promoter region (Table S2). These results suggested that VaERF057 regulates the expression of target genes by binding to GCC-box and DRE motifs in their promoter regions.

Discussion
This study explored the function of ET and the roles of its downstream gene VaERF057 in grapevine during cold stress. Our results demonstrated that cold stress enhanced the production of ET and the expression of ET signal pathway downstream genes, such as VaERF057 and VvERF057, in grapevine. Enhanced ET synthesis is essential in cold tolerance both in cold-hardy (V. amurensis) or cold-sensitive (V. vinifera 'Muscat Hamburg') grapevine.
Our findings are consistent with previous results 15,16,18,35 wherein an increase of ET level positively regulates the cold tolerance of grapevine. In addition, ET biosynthesis is required for the accumulation of antifreeze proteins in winter rye (Secale cereale) during cold acclimation 44 . ET has also been proposed to protect mitochondrial activity in Arabidopsis under cold stress 45 . However, an opposite conclusion was reached in other studies on  34 , and dwarf bean (Phaseolus vulgaris) 46 . Especially in Arabidopsis, Catala et al. 15 reported that ET production increases under cold treatment and that ET positively regulates cold tolerance, whereas Shi et al. 14 obtained the opposite results. These different results can be largely caused by the different growth conditions employed in these studies. Catala et al. 15 used soil-grown plants, whereas Shi et al. 14 used seedlings grown on Murashige and Skoog (MS) medium in Petri dishes, implying a growth condition of high relative humidity. Indeed, ET production is inhibited in response to cold stress under high relative humidity 47,48 . In our research, the grapevine plantlets were grown on half-strength MS medium in a conical flask covered with a sealing film that allows good aeration. The relative humidity in our study was higher than the relative humidity the soil-grown plants were exposed to but lower than that to which the plants grown on MS medium in Petri dishes were exposed. Therefore, a critical value of relative humidity possibly inhibits ET production under cold stress. The function of ET in cold stress in different species exposed to different conditions still requires further investigation.
ERF transcription factors, which are located downstream of the ET signal pathway, were reported to respond to various biotic and abiotic stresses in several species. In Arabidopsis, AtERF1−5 genes were differentially regulated by ET and by abiotic stresses, such as cold, drought, wounding, high salinity, or pathogen. AtERF1, AtERF2, and AtERF5 function as activators, whereas AtERF3 and AtERF4 act as repressors by downregulating not only the basal transcription levels of a reporter gene but also the transactivation activity of other transcription factors 49,50 . Many other ERFs were identified as stress-responsive through detection of their expression under various stress conditions, although functional studies on ERF genes remain limited. Transcript analysis in 'Cabernet Sauvignon' grapevines revealed 13 ET-responsive transcripts upregulated and 7 downregulated after cold treatment, whereas ERF057 wasn't included 51 . This study isolated the Group VII ERF genes (VaERF057 and VvERF057) from grapevine, and these genes can be induced by cold stress and exogenous ET. AVG can also totally inhibit the inductive effect of cold stress on the expression of these genes. Low temperature promoted the release of ET, and the ET increased the expression of VaERF057 and VvERF057. Therefore, the enhanced expression of VaERF057 under cold stress was possibly directly induced by ET, the release of which was promoted by low temperature (Fig. 8).
As transcription factors, the AP2/ERFs play important roles in stress response by regulating the expression of downstream stress-related genes. For instance, overexpression of CBF genes in plants enhances cold tolerance by regulating the expression of cold-responsive genes [52][53][54] . TERF2/ERF2 overexpression can also increase cold tolerance in rice, tomato, and tobacco by activating expression of cold-related genes 18,55 . In the present study, VaERF057 overexpression enhanced cold tolerance in transgenic Arabidopsis. These physiological adjustments confer cold tolerance to the transgenic Arabidopsis. Moreover, RNA-Seq analysis showed that VaERF057 overexpression in Arabidopsis regulated the expression of the cold-induced regulatory genes, including AtCBF1−3, WRKY33, WRKY70, NCED3, and ERF5 under normal condition. An increasing number of evidence indicates that ERF transcription factors play critical regulatory roles in response to stress by interacting with GCC-box or DRE motif to activate expression of targeted stress-responsive genes 24,56,57 . AtERF71/HRE2 is the closest homologue of VaERF057 in Arabidopsis. Binding of AtERF71/HRE2 to GCC-box and DRE motifs was detected by electrophoretic mobility shift assay (EMSA), fluorescence measurement and surface plasmon resonance spectroscopy (BIAcore) experiments 43 . In this study, we provided evidence that VaERF057 can bind to both GCC-box and DRE motifs by Y1H system. Our results indicate that VaERF057 in Arabidopsis possibly directly upregulates expression of its targeted genes by interacting with these cis-elements. The accumulated expression of VaERF057 and VvERF057 was also found in other stress conditions, such as drought and high salinity (data not shown). Further studies should be conducted to overexpress or to knock down VaERF057 and VvERF057 in grapevine and thus elucidate their functions in stress response.
In summary, low temperature can promote the release of ET and endogenous ET can enhance the cold tolerance of grapevine. We isolated a group VII stress-responsive ERF gene (VaERF057), which is located downstream of the ET signal pathway and is implicated as a positive regulator of cold tolerance. VaERF057 overexpression enhanced the cold tolerance of Arabidopsis. This finding indicated that VaERF057, when induced by ET, may function as a key regulator that improves the cold tolerance of grapevine. VaERF057 may directly upregulate the expression of its target stress-responsive genes by interacting with GCC-box or DRE/CRT element in grapevine. On the basis of our results, we proposed a comprehensive model of ET-dependent signal transduction in grapevine under cold stress (Fig. 8). Further studies should be performed to determine other components related to VaERF057 and to gain a clear-cut silhouette of the major hub in the network.

Methods
Plant materials and treatments. Micropropagated V. amurensis and V. vinifera 'Muscat Hamburg' plantlets were grown on half-strength Murashige and Skoog (1/2 MS, pH 5.8) solid medium containing 30 g L −1 of sucrose and 0.7% agar in conical flasks (120 mL) in a growth chamber at 26 °C (Fig. S3). The average photosynthetic photon flux was 100 μmol m −2 s −1 with a 16 h light and 8 h dark cycle. Six-week-old plantlets were subjected to cold stress or supplied with plant growth regulators. For cold treatment, whole plants in conical flasks were placed in an illuminated incubation chamber at 4 °C. For treatments with plant growth regulators, plantlet leaves were sprayed with 100 μ μM 1-aminocyclopropane-1-carboxylate (ACC, an ethylene (ET) precursor) or 100 μM aminoethoxyvinylglycine (AVG, an ACC synthase inhibitor), whereas distilled water was used as control. The shoot apex with one well developed leaf was harvested at appropriate times. To reduce the differences between individuals, we collected the samples from three independent replicate plantlets at a single time point, with three biological replicates per treatment. After the collection, all samples were immediately frozen in liquid nitrogen and then stored at −70 °C.
Seeds of Arabidopsis thaliana (ecotype Columbia-0, Col-0) were surface-sterilized with 6% sodium hypochlorite for 5 min followed by three washes with sterile water and then plated on 1/2 MS solid medium containing 0.7% agar. One-week-old seedlings were transferred from plates into pots filled with peat moss in a greenhouse under controlled environmental conditions, at 22 °C, under 50-60% relative humidity, and under 130 μmol photons m −2 s −1 with a 16 h/8 h light/dark cycle.
Determination of ET production. Six-week-old plants grown in conical flasks (120 ml) under standard conditions (26 °C) or exposed to low temperature (4 °C) were maintained open for 1 h. The conical flasks were subsequently sealed with rubber stoppers to collect ET produced within 1 hour at 0, 2, 4, 8, 24, and 48 h. Gas (1 ml) was withdrawn from each conical flask using a gas-tight syringe and left for 5 min to equilibrate. The gas samples were subsequently injected into a gas chromatograph (Agilent 7890A) equipped with an HP-5 column and a flame ionization detector. The detector and injector were operated at 180 °C and 120 °C, respectively, whereas the oven temperature was set at 110 °C. The carrier gas was nitrogen at a flow rate of 40 ml min −1 . ET production (μl g −1 FW h −1 ) was determined by comparison with an ET standard. All measurements were obtained in triplicate from three independent samples.
Determination of ACC content. Determination of ACC content was performed according to the method described by Zhao et al. 17 with some minor modifications. In brief, the tissues were ground with a mortar and pestle in 2 ml of 80% ethanol until homogenized. After centrifugation at 12,857 × g for 10 min, the supernatant liquid was dried using a centrifugal dryer. The residue was dissolved in 1 ml of deionized water, and 0.3 ml of sample to be assayed was added in a 7 ml vial in an ice bath. Subsequently, 100 μl of HgCl 2 (10 mM) was added and then the vial was sealed with a rubber serum cap. Afterwards, 100 μl of cold mixtures of 5% NaOCl and saturated NaOH (2:1, v/v) was injected into the vial and the mixture was kept on ice for 30 min. After shaking vigorously, ET production was determined as described above. The efficiency of conversion of ACC into ET was calculated by adding authentic ACC (1 mM) as an internal standard. ACC content was calculated from the quantity of ET liberated and from the conversion efficiency.
Determination of ACO activity. ACO activity was detected according to the method described by Zhao et al. 17 with some minor modifications. Briefly, the tissues were ground with a mortar and pestle in extraction buffer containing 100 mM Tris (pH 7.2), 30 mM sodium ascorbate, 10% (w/v) glycerol, and 5% (w/v) polyvinylpolypyrrolidone (PVPP). After centrifugation at 18,514 g for 30 min at 4 °C, the supernatant liquid was used for enzyme activity analysis. Crude extract (200 μl) was then mixed with 1.7 ml of extraction buffer (without PVPP), 100 μl of FeSO 4 (1 M), and 100 μl of ACC (20 mM) in a 7-ml capped vial. After incubation at 30 °C for 1 h, ET production was determined as described above.
Electrolyte leakage assay. Electrolyte leakage was measured to verify the effect of exogenous ACC and AVG treatment on cold tolerance of grapevine. After treatment with ACC (100 μM) and AVG (100 μM) for 8 h, leaf discs were obtained from V. amurensis and 'Muscat Hamburg' plantlets. The 10 ml tubes containing five leaf discs were placed in a low-temperature illuminated incubation chamber at 0 °C for an hour. The temperature was subsequently reduced at 2 °C h −1 . The tubes were then removed at −6 °C, −8 °C, −10 °C, and −12 °C for V. amurensis and at −4 °C, −6 °C, −8 °C, and −10 °C for 'Muscat Hamburg' . The leaf discs were thawed overnight at 4 °C in the dark, and 3 ml of deionized water was added into each tube. After shaking at 200 rpm for 2 h at 25 °C, the electrical conductivity of the samples was first measured. The samples were then autoclaved at 121 °C for 20 min, and the final conductivity was measured. Relative electric conductivity is the ratio of initial conductivity before autoclaving to the final conductivity after autoclaving. The semi-lethal temperature (LT 50 ) was calculated by logistic equation. Each point was repeated for five times.
Cloning and sequence analyses of VaERF057 and VvERF057. To clone the ERF057 gene (GSVIVT01028050001) of V. amurensis and 'Muscat Hamburg' , we designed a pair of primers (ERF057-ORF-F/ ERF057-ORF-R) based on the sequence of V. vinifera 'Pinot Noir' (PN40024) genomes (Table S1). The open reading frame (ORF) sequences of VaERF057 and VvERF057 were amplified by polymerase chain reaction (PCR). An analysis of the VaERF057 and VvERF057 protein structures was performed using Smart (http://smart. embl-heidelberg.de/). Sequence alignments of the AP2/ERF domain with the other ERFs from different plant species was done by DNAMAN 7.0. Phylogenetic analysis was performed using MEGA6.0 software by using the neighbor-joining method, and the internal branch support was estimated using 1000 bootstrap replicates.

Gene expression analysis by quantitative real time RT-PCR (qRT-PCR).
Total RNA was extracted from the collected samples by using Column Plant RNAOUT 2.0 Kit (Tiandz, Beijing, China) according to the manufacturer's instructions. RNase-free DNase I (Promega, USA) was used to degrade any DNA present in the extracted RNA. cDNA was synthesized using SuperScript III Reverse Transcriptase (Invitrogen, USA) with Oligo-dT according to the manufacturer's instructions. qRT-PCR was performed in an ABI SteopOneplus TM Real-Time PCR System (Applied Biosystems, USA) using FastStart Universal SYBR Green Master (Roche, Shanghai, China). Each reaction was replicated three times for each biological sample, using a total of three biological replicates. The relative expression and standard errors were calculated using Biogazelle qbasePLUS as described in our previous study 58 .
Scientific RepoRts | 6:24066 | DOI: 10.1038/srep24066 Generation of overexpression transgenic Arabidopsis. The full-length ORF of VaERF057 containing the restriction sites of Kpn I and Xba I was ligated into pCAMBIA 1301s driven by the CaMV 35S promoter and transformed into Arabidopsis using the floral dip method. The positively transformed plants were selected by screening successive generations on hygromycin B (50 mg L −1 ) and confirmed by genomic DNA PCR and qRT-PCR. Three T3 homozygous transgenic lines were selected for further analyses.

Stress treatments of transgenic Arabidopsis.
Three-week-old WT and transgenic Arabidopsis were subjected under cold tolerance assays. To evaluate cold tolerance, we placed the plants in a growth chamber set at −1 °C for 8 h in the dark. The chamber was then cooled at a rate of 1 °C h −1 until −11 °C was reached. After exposure to −11 °C for 3 h, the plants were thawed at 4 °C for 12 h in the dark and then transferred into standard condition. The survival rate was scored 3 days later. Each sample contained 14 seedlings, and each experiment was performed in three biological replicates.
Physiological analysis of the cold stress response of the Arabidopsis plants. For physiological analysis, 3-week-old seedlings of WT and transgenic Arabidopsis plants were treated at 4 °C (cold stress) for 4 days, and the plants grown under normal condition served as control. Superoxide dismutase (SOD), peroxidase (POD), and catalase (CAT) activities, along with malondialdehyde (MDA) contents, were measured according to the protocol described by Shin et al. 59 . Each sample was pooled from three seedlings, and each experiment was performed in three biological replicates.
RNA-Seq analysis. RNA-Seq analysis was performed using 3-week-old WT and transgenic Arabidopsis mixture under normal condition. Total RNA was extracted using Column Plant RNAOUT 2.0 Kit (Tiandz, Beijing, China) as described above, and mRNA sequencing libraries were constructed using the TruSeq RNA Sample Preparation Kit (Illumina, USA). Sequencing was performed on an Illumina HiSeq3000 platform. Materials of the libraries were pooled from six WT plants and three transgenic Arabidopsis lines (OE1, OE2, and OE3). The RNA-Seq experiments were performed in two biological replicates. The total reads were mapped to the TAIR10 Arabidopsis genome reference by using the TopHat software 60 . The number of sequencing reads generated from each sample was converted into fragments per kilobase of transcript per million fragments mapped. Significantly differentially expressed transcripts between WT and transgenic lines were detected from two biological replicates by using Cuffdiff with default Benjamini-Hochberg correction for multiple-testing, based on a False Discovery Rate (FDR) < 0.05. Functional annotations of genes and AGI symbols were retrieved from TAIR10 datasets. RNA-Seq data were submitted to NCBI and can be accessed under the GEO accession number GSE75456.
Yeast one-hybrid assay. Yeast one-hybrid assay was performed using the Matchmaker Gold Yeast One-Hybrid Library Screening System (Clontech) to examine the ability of VaERF057 to bind with GCC-box and DRE motifs. The ORF of VaERF057 was fused in-frame with the GAL4 activation domain (AD) in a pGADT7-AD vector to generate the prey plasmid (pGAD-VaERF057). A synthesized DNA fragment harboring four tandem copies of the GCC-box, DRE or their mutants, and mGCC or mDRE was cloned into the pAbAi vector to construct the baits. The BstBI-cut bait vectors were transformed into the Y1HGold yeast strain. After selecting the transformants on SD/− Ura plates and determining the minimal inhibitory concentration of Aureobasidin A (AbA) for the bait strains, we introduced pGAD-VaERF057 into the Y1HGold strain. Positively co-transformed yeast cells were determined by spotting serial dilutions (1:1, 1:10, and 1:100) of yeast on SD/− Leu medium supplemented with 400 ng ml −1 AbA and cultured at 30 °C for 3 days. Positive (pGAD-p53 + p53-AbAi) and negative (pGADT7-AD + p53-AbAi) controls were processed in the same manner.