Genome-Wide Characterization of the C-repeat Binding Factor (CBF) Gene Family Involved in the Response to Abiotic Stresses in Tea Plant (Camellia sinensis)

C-repeat (CRT)/dehydration responsive element (DRE)-binding factor CBFs, a small family of genes encoding transcriptional activators, play important roles in plant cold tolerance. In this study, a comprehensive genome-wide analysis was carried out to identify and characterize the functional dynamics of CsCBFs in tea plant (Camellia sinensis). A total of 6 CBF genes were obtained from the tea plant genome and named CBF1-6. All of the CsCBFs had an AP2/ERF DNA-binding domain and nuclear localization signal (NLS) sequence. CsCBF-eGFP fusion and DAPI staining analysis confirmed the nuclear localization of the CsCBFs. Transactivation assays showed that the CsCBFs, except CsCBF1, had transcriptional activity. CsCBF expression was differentially induced by cold, heat, PEG, salinity, ABA, GA, MeJA, and SA stresses. In particular, the CsCBF genes were significantly induced by cold treatments. To further characterize the functions of CsCBF genes, we overexpressed the CsCBF3 gene in Arabidopsis thaliana plants. The resulting transgenic plants showed increased cold tolerance compared with the wild-type Arabidopsis plant. The enhanced cold tolerance of the transgenic plants was potentially achieved through an ABA-independent pathway. This study will help to increase our understanding of CsCBF genes and their contributions to stress tolerance in tea plants.


INTRODUCTION
Abiotic stresses in the natural environment, including low temperature, drought, and high salinity, seriously affect the growth, development, distribution, and productivity of plants (Kulik et al., 2011). Plants have evolved a complex set of adaptive mechanisms to survive these adverse environmental conditions, and these mechanisms involve a number of biochemical and physiological changes (Ramanjulu and Bartels, 2002). Different studies have suggested that transcription factors can activate the expression of related genes to protect plants from adversity (Agarwal and Jha, 2010). CBFs/DREBs (C-repeat binding factors/dehydration responsive elementbinding factors) belong to the AP2/ERF (APETALA2/ethylene-responsive element-binding factor) transcription factor family (Riechmann and Meyerowitz, 1998) and have been reported to play pivotal roles in freezing tolerance and cold acclimation (Stockinger et al., 1997). The AP2/ERF family contains a highly conserved AP2/ERF domain that harbors a DNA-binding motif of ∼60 amino acids (Sakuma et al., 2002). When plants suffer from low temperatures, related regulatory proteins are modified to regulate the expression of the CBF gene. For example, the inducers of CBF expression 1 (ICE1), calmodulin-binding transcription activator 3 (CAMTA3) and brassinazole-resistant 1 (BZR1), positively regulate CBF expression, whereas MYB15, ethylene-insensitive 3 (EIN3), and 14-3-3 negatively regulate CBF expression (Liu et al., 2017). Subsequently, CBF/DREB1 proteins bind to the cold-and dehydration-responsive DNA regulatory element designated the CRT (C-repeat)/DRE (dehydration response element), which is present in the promoters of COR (cold-regulated) genes and contains the core motif of G/ACCGAC (Yamaguchi-Shinozaki and Shinozaki, 1994;Huang et al., 2012), and to the promoters of other cold responsive genes, such as COR15A, RD29A, and COR78 (Liu et al., 1998;Lucas et al., 2011;Akhtar et al., 2012), and stimulates their transcription (Baker et al., 1994;Jaglo et al., 2001;. Previous studies have identified a number of CBF/DREB1 genes and verified their functions in Arabidopsis thaliana (Novillo et al., 2007) and other plant species, such as cotton (Gossypium hirsutum) (Shan et al., 2007), wheat (Triticum aestivum) (Shen et al., 2003), rice (Oryza sativa) (Wang et al., 2008), maize (Zea mays) (Qin et al., 2004), soybean (Glycine max) (Kidokoro et al., 2015), and tomato (Lycopersicon esculentum) (Zhang et al., 2004). These findings suggested the conserved roles of CBF/DREB1 genes in the regulation of freezing tolerance across diverse plant species. Nevertheless, CBFs play different roles in responses to stress in different plant species (Ebrahimi et al., 2015). In addition to the cold response, CBFs could also respond to other abiotic stresses and hormones, such as heat, drought, salt, and abscisic acid (ABA) (Dubouzet et al., 2003;Xiao et al., 2006;Nada and Abogadallah, 2015).
Tea plants are the world's most important nonalcoholic beverage crop with a wealth of health benefits . The growth of tea plants is seriously affected by abiotic stresses , particularly extreme temperature and drought (Das et al., 2016;Liu et al., 2016;Hou et al., 2018;Zhou et al., 2018). Although several recent studies have demonstrated that CsCBF1 enables cold stress in tea plants , overexpression of CsDREB increases salt and drought tolerance in transgenic Arabidopsis thaliana . However, given that the CBF transcription factor is a polygenic family, it is still unclear whether there are other new CBF genes in tea plants that respond to low temperature. In the present study, we thoroughly investigated the CBF genes in tea plant using tea plant genome and transcriptome datasets, with the aim of providing novel insights into the functional dynamics of CBF genes in tea plant. The overall obtained results provide a foundation for additional studies of the biological functions of CsCBFs under abiotic stresses and a new perspective for resistance breeding in tea plants.

Plant Materials and Stress Treatments
One-year-old tea cutting seedlings of the Shuchazao cultivar were planted in a pot and grown with a natural photoperiod in a greenhouse (12 h light and 12 h dark photoperiod, 25 • C temperature and 70% relative humidity)  at the State Key Laboratory of Tea Plant Biology and Utilization, Anhui Agricultural University (Hefei, China). For low and high temperature stresses, tea plants were grown at 4 and 38 • C in a plant growth chamber. For the GA, ABA, SA, and MeJA treatments, a working solution of 100 µM was foliar sprayed on plants . For drought and salinity, the plants were transferred to a 20% PEG6000 and 200 mM NaCl solution (Li et al., 2010;Yue et al., 2014). The second or third mature leaves were harvested for gene analysis 0, 4, 12, and 24 h (Agarwal M. et al., 2006) after treatments. After collection, the samples were immediately frozen in liquid nitrogen and stored at −80 • C for RNA extraction. Three biological replicates were conducted.

Identification of CsCBF Genes and Phylogenetic Construction
The protein sequences of AtCBFs were downloaded from the NCBI (https://www.ncbi.nlm.nih.gov/) database and used as queries to search homologous sequences against the published genomes of tea plant  with the BlastP program at an e-value of 10 −3 . The biophysical properties of the CsCBFs were computed using the online ProtParam tool (https://web.expasy.org/protparam/). The protein sequences of Camellia sinensis and Arabidopsis thaliana were aligned by DNAMAN6.0. Multiple sequence alignments of Camellia sinensis and Arabidopsis thaliana were analyzed using MEGA 6.0 (http://www.megasoftware.net/). The phylogenetic tree was constructed using the neighbor-joining algorithm with 1,000 replicates.

Characterization of CsCBF Genes and Proteins
The promoter sequences ∼2 kb upstream of the transcription start site of each CsCBF gene were identified, and the ciselements were analyzed by the online tool Plant CARE (http:// bioinformatics.psb.ugent.be/webtools/plantcare/html/). The fulllength amino acid sequences of CsCBF were entered into the MEME (http://meme-suite.org/tools/meme) analysis tool to find their conserved motifs. Parameters of MEME are following: number of different motifs: 10, Minimum/Maximum motif width: 6/100.

Subcellular Localization of CsCBFs
The full-length ORFs of CsCBF sequences were cloned into the pK7WGF2.0 vector containing the enhanced green fluorescent protein (eGFP) reporter gene by the gateway method. The isolated DNA was transformed into Agrobacterium strain GV3101. Six 35S::eGFP-CsCBF constructs and the control plasmids without CsCBF coding sequences were separately infiltrated into tobacco (Nicotiana tabacum) leaves by the Agrobacterium-mediated genetic transformation method. DAPI (4' ,6-diamidino-2-phenylindole dihydrochloride) was used to identify nuclei. The cells of transformed tobacco leaves were observed by an Olympus IX81 fluorescence microscope (Olympus, Japan).

Transactivation Assay Analysis
Each CsCBF gene was amplified and cloned into the pGBKT7 vector. Six pGBKT7-CsCBF vectors, the pGBKT7-AtCBF2 vector (positive control) and the pGBKT7 vector (negative control) were separately transformed into the Y2HGold yeast strain. The transformed yeast cells were incubated on SD/-Trp, SD/-Trp/-His-Ade, and SD/-Trp-His-Ade-x-gal plates at 30 • C for 3 d.

RNA Isolation and Quantitative RT-PCR
Total RNA was extracted using the RNAprep Pure Plant Kit (Tiangen, Beijing, China). A Nanodrop 2000 spectrophotometer (Thermo Scientific, Wilmington, USA) was used to measure the concentration of isolated RNA, and the quality was assessed using 1.2% formaldehyde-agarose gel electrophoresis. cDNA was synthesized for qRT-PCR by the PrimeScript TM RT Reagent Kit with gDNA Eraser (Takara, Tokyo, Japan) and diluted 10-fold for PCR amplification.
The specific primers for qRT-PCR were designed by Primer Premier 5 software and synthesized by Sangon Biotech Co. (Shanghai, China). The qRT-PCR reaction program was performed under the following conditions: 95 • C for 30 s, followed by 40 cycles at 95 • C for 5 s, and 60 • C for 30 s. The reaction volume was 25 µL, which contained 4 µL of diluted cDNA, 6.5 µL of deionized water, 12.5 µL of SYBR R Premix Ex Taq TM II (Tli RNaseH Plus; TaKaRa), 1 µL of forward primer, and 1 µL of reverse primer. CsGAPDH was used as the reference gene . The relative gene expression levels were calculated using the comparative 2 − Ct method (Livak and Schmittgen, 2001). Regarding the heatmaps, the 2 − Ct values of the transcripts of the CsCBF genes were normalized as the log2fold change. In the stress-treated plant samples, the values were normalized to plant samples of the 0 h treatment and expressed as a log2-fold change. qRT-PCR experiments were conducted with three independent total RNA samples.

Functional Analysis of CsCBF3-Overexpressing Transgenic Arabidopsis thaliana
Expression of CsCBF3 shows significant changes during cold treatment and we selected it for overexpression. The ORF of CsCBF3 was cloned into the pBI121 vector. The construct was transformed into Agrobacterium strain GV3101 by electroporation, and the Arabidopsis plants were transformed using the floral dip method. Arabidopsis ecotype Columbia-0 (col-0) was used as the wild type in this study. Transformed plants were selected on the basis of their resistance to kanamycin, and 4-weeks-old homozygous T3 plants were used for further experiments. Two transgenic lines were treated at −4 and −8 • C for 12 h and then grown under 25 • C conditions for analysis of survival rate.

Identification and Characterization of CBF Genes in Tea Plant
We initially identified 10 CBF genes from genome and transcriptome databases of tea plant. Six of them were successfully cloned and deposited into the NCBI GenBank database under the accession numbers CsCBF1 (EU563238.1), CsCBF5 (MH165878.1), and CsCBF6 (MN544638.1). Alignment of the sequences obtained from cDNA and genomic DNA indicated that the CsCBF genes were intronless. The CDS length of the CsCBFs ranged from 720 to 879 bp, and the genes encoded proteins with lengths varying from 239 to 292 amino acids. The molecular weights were between 26.43 and 32.86 kDa, and the pI-values ranged from 4.84 to 8.09. Most of the CsCBF proteins presented grand average hydropathicity (GRAVY) values of <0, implying their hydrophilic nature (Table 1).

Sequence Alignment and Phylogenetic Analysis
We investigated the amino acid characteristics of CBF proteins of C. sinensis and A. thaliana. Alignment of the amino acid sequences of the CsCBFs and AtCBFs revealed that the CsCBFs contained one highly conserved DNA-binding domain (AP2 domain) consisting of 59 amino acids. A putative nuclear localization signal (NLS) sequence (PKKRAGRKKFK) was detected in the N-terminal region. In addition, the C-terminal regions of the CsCBF proteins were highly diverged, particularly CsCBF6 (Figure 1).
A phylogenetic tree was constructed using the neighborjoining method with the MEGA program to explore the evolutionary relationships among CBF homologs in C. sinensis and A. thaliana (Figure 2). The 60 collected DREBs from C. sinensis and A. thaliana were clustered into six groups. All 10 CsCBF proteins were clustered together with AtDDF1-2 and AtCBF1-4 in Group A1. The result showed that CsCBFs belongs to DREB-A1 subfamily.

Motif Analysis
To explore the diversification of CBF genes in tea plants, we examined the conserved motifs within the CsCBFs proteins using the MEME program (Figure 3). Ten motifs were identified. Multilevel consensus sequences and the E-value of each motif are shown. The results revealed conserved motif distribution among CsCBF proteins. Motif 1 and motif 2 were found in all CBFs. Motifs 3-6 could be found in CsCBF1, 2, 3, and 5. Motif 7 was mainly identified in the C-terminal regions of CBF1, 2, and 3. Motif 8 was unique to CsCBF6 and motif 9 was unique to CsCBF4. Motif 10 was distributed in CBF1, 2, and 3. Motif analysis supports the results of the phylogenetic analysis.

Subcellular Localization of CsCBF Proteins
To investigate the subcellular localization of CsCBF proteins, six 35S::eGFP-CsCBF constructs, one for each CsCBF, and a 35S::eGFP (positive control) construct were generated and  transiently expressed in tobacco leaves. As shown in Figure 4, eGFP alone resulted in a diffuse distribution of green fluorescence throughout the entire cell. In contrast, CsCBF-eGFP proteins localized predominantly to the nucleus, which was further confirmed by 4 ′ ,6-diamidino-2-phenylindole (DAPI) staining. The results indicated that CsCBFs are nuclear localized proteins.  Table S1.

Transactivation Assay of CsCBFs
To examine the transcriptional activities of CsCBFs, full length sequences of six CsCBFs and AtCBF2 (positive control) were fused to the vector pGBKT7 containing the GAL4 DNA-binding domain and subsequently transformed into yeast. The yeast cells harboring pGBKT7-CsCBF2, pGBKT7-CsCBF3, pGBKT7-CsCBF4, pGBKT7-CsCBF5, pGBKT7-CsCBF6, and pGBKT7-AtCBF2 grew well on the selection media without Trp, His or adenine (SD/-Trp-His-Ade) and were positive for αgalactosidase activity. The yeast cells with the empty vector pGBKT7 (negative control) and pGBKT7-CsCBF1, which is missing the transcriptional activation domain GAL4 AD, were unable to grow on the same medium. The results indicated that CsCBF2, CsCBF3, CsCBF4, CsCBF5, and CsCBF6 have transcriptional activity, while CsCBF1 has no transcriptional activity (Figure 5).

Expression Analysis of CsCBFs Under Various Abiotic Stresses
We investigated the expression profile of CsCBF genes in response to eight stress treatments (cold, heat, PEG, salinity, ABA, GA, MeJA, and SA) by qRT-PCR (Figure 6). Under cold treatment, all CsCBFs showed high expression levels during 24 h. Notably, the expression of CsCBF1,2,3,4, and 5 was much higher than that of CsCBF6. The expression of CsCBF1, 2, and 3 increased a 100-fold or even 1000-fold throughout the entire incubation time, while the expression of CsCBF4, 5, and 6 increased with processing time but reached a maximum at 12 h. Under heat stress, CsCBF1, 2, and 5 presented an expression trend of rising first and then falling, while the expression of CsCBF4 and 6 declined during processing, and the expression of CsCBF3 fell to 12 h and then returned to normal. Under PEG stress, CsCBF2 had a strong increasing response.   GAPDH was used as a housekeeping gene. The mean expression values were again normalized using logarithm with the base of two. The color bar in all heat maps represents the expression values: red represents upregulation, black represents no significant difference in expression, and green denotes downregulation.
and 12 h, respectively, and finally returned to the control level. CsCBF2 and 4 both increased and reached a maximum at 4 and 12 h, respectively. CsCBF3 showed no response to GA. CsCBF5 showed a downward adjustment in the later stages of processing. Under MeJA treatment, the expression of CsCBF1, 2, 3, 4, and 5 was downregulated. CsCBF6 rose in 4 h and then decreased with time. Under SA treatment, CsCBF1, 3, and 5 were downregulated and reached a minimum at 12 h. CsCBF2 was positive in the first 4 h and then decreased to negative over time. CsCBF4 was downregulated and reached its minimum at 12 h, then it returned to the control level. CsCBF6 was negatively regulated and then positively regulated. In general, the results illustrated that cold, ABA and GA could induce high expression of most of the CsCBF genes.

Overexpression of CsCBF3 in Arabidopsis thaliana Improves Tolerance to Cold Stress
To confirm the in vivo functions of the CsCBF3 gene during low-temperature stress in plants, we transferred CsCBF3 into A. thaliana. The expression of CsCBF3 was detected using realtime PCR assay in overexpressed (OE) plants but not in wildtype (WT) plants ( Figure 7A). Two overexpressed lines (OE-9 and OE-14) were treated in low temperature conditions. Thirty seeds of wild-type and overexpressed plants were selected for treatment. Under normal growth conditions (25 • C), there were no obvious differences in survival rate between the WT and transgenic plants. A 3-days recovery after low-temperature treatment at −4 and −8 • C for 12 h, the survival rates of WT plants significantly decreased compared to those of transgenic plants ( Figure 7B). Recovery for 3 d after −4 • C treatment, ∼31% of wild-type plants survived, while 73% of OE-9 survived. Recovery for 3 d after −8 • C treatment, the survival rates of wild-type plants and OE-14 were 11 and 58%, respectively ( Figure 7C).
In addition, we detected the expression of downstream target ABA-dependent stress-induced genes (AtRD29B, AtRAB18, AtABI1, and AtABI2) and ABA-independent stress-induced genes (AtCOR15a and AtRD29A) to explore the potential CsCBF3-associated regulatory pathway. Under unstressed conditions, the expression of AtCOR15a and AtRD29A in transgenic plants was significantly higher than that in wildtype plants. The expression of AtRD29B, AtRAB18, AtABI1, and AtABI2 showed slight differences in WT and OE plants (Figure 8). These results indicated that CsCBF3 may affect the expression of ABA-independent stress-induced genes to increase plant tolerance to cold stress.

DISCUSSION
Low temperature is a major abiotic factor that limits crop productivity. When plants suffer from cold stress, CBF/DREB transcription factors are triggered and regulate ∼12% of the cold responsive transcriptome, showing important roles in cold tolerance (Sun et al., 2014). To date, two CsCBF genes' biological function have been reported in tea plant. Wang et al. (2012) and Ban et al. (2017) found that CsCBF1 was not expressed at normal temperature (20 • C) but was significantly induced at low temperature (4 • C). In addition, Wang et al. (2012) used a DNAbinding assay to demonstrate that CsCBF1 can specifically bind to the CRT/DRE cis-element, suggesting that CsCBF1 can regulate downstream genes containing CRT elements such as COR15 and RD29A. Wang et al. (2017) showed that overexpression of a CsDREB gene in Arabidopsis thaliana plants could increase the salt and drought tolerance of transgenic plants. Wang et al. (2019) identified five CsCBFs from tea plant genome sequences , namely CsCBF1-5. However, we failed to clone CsCBF2 (TEA010423), which might be caused by draft nature of the current genome assembly with low assembly quality. The CsCBF1 identified by Wang et al. (2012) was consistent with that identified in this study but totally different from that identified by Wang et al. (2019) (Data Sheet S3). Similarly, we were not only unable to find homologous gene of the CsCBF3 identified by Yin et al. (2016) in the current tea plant genome assembly, but also failed to clone it in this study. We here successfully cloned a total of 6 CsCBFs (CsCBF1-6) based on the latest version of tea plant genome annotation . Of them, CsCBF1-4 were in agreement with those identified by Wang et al. (2019), and CsCBF5 and 6 were newly discovered (Data Sheet S3). Sequence analysis showed that CsCBFs contain an AP2 DNA-binding domain and two signature motifs of the CBF family (Figure 1). These results showed that CsCBF1∼6 typically belong to the CBF family. The C-terminus of CsCBF6 is longer than those of the other CsCBFs, and it may cause functional divergence in plants under cold temperatures. Expression of CsCBF6 is much less than other CsCBFs verified this (Figure 6).
It is widely accepted that transcription factors must be present in the nucleus to perform their functions (Wang et al., 2008;Yang et al., 2011). Bioinformatics analysis showed the presence of nuclear localization signal (NLS) sequences in CsCBFs. In our vivo targeting experiment using a CsCBFsfused GFP as a florescent marker demonstrated that the fusion protein was localized to the nucleus of tobacco leaf, suggesting that CsCBFs are nuclear proteins and functions as transcription factors. The results are consistent with the findings in cotton (Shan et al., 2007) and eggplant (Zhou et al., 2018). We designed a yeast single hybrid experiment to verify the transcriptional activities of CsCBFs. The results showed that CsCBF2-6 had transcriptional activity in yeast Y2HGold cells, but not CsCBF1. This finding is similar to the findings of Sakuma and Zhao (Sakuma et al., 2006;Zhao X. et al., 2012). A plausible explanation is probably because the secondary structure of the CsCBF1 protein itself is abnormal or CsCBF1 is required to be activated by a posttranslational modification.
CBF encodes a member of the DREB subfamily A-1 of ERF/AP2 transcription factor family. CBF/DREB genes from different plant species may have inconsistent expression profiles in response to various stresses (Zhou et al., 2016). There are six members in this subfamily, including CBF1, CBF2, CBF3, and CBF4 in Arabidopsis thaliana. AtCBF1-3 gene is involved in response to low temperature and abscisic acid. AtCBF4 gene is involved in response to drought stress and abscisic acid treatment, but not to low temperature (Novillo et al., 2004). EgCBF3 and FeDREB1 can be upregulated not only by cold but also by osmotic and high-salt stresses (Ebrahimi et al., 2015;Fang et al., 2015). CaDREBLP1 is not upregulated by low temperature but by dehydration and salt (Hong and Kim, 2005). We examined the expression patterns of CsCBF genes in relation to various environmental stresses. At normal growth temperatures, CsCBF genes is not transcribed, or is transcribed at lower levels, while CsCBF genes except CsCBF4,6 are rapidly, transiently and strongly induced by cold stress. CsCBF genes were induced to varying degrees by other abiotic stress treatment including exposure to high temperature, drought, exogenous hormones, or salinity. CsCBF4 had a strong response to salinity stress, which was similar to the study by Wang et al. (2017) in CsDREB. AtCBF2 gene is involved in a negative regulatory or feedback circuit of the CBF pathway (CBF2/DREB1C is a negative regulator of CBF1/DREB1B and CBF3/DREB1A expression and plays a central role in stress tolerance in Arabidopsis). Whether there is redundancy or feedback circuit function among CsCBFs genes needed to be further research.
Numerous studies have demonstrated that the expression of CBFs is regulated by GA, JA, ABA, ETH, and brassinosteroids (BRs) (Shan et al., 2007;Hu et al., 2013;Eremina et al., 2015;Barrero-Gil and Salinas, 2017). Different numbers plant hormone-responsive cis-elements were detected in the CsCBF promoters including abscisic acid, MeJA, ethylene, gibberellin, salicylic acid, auxin-responsive element. Results consistent with Wang et al. (2019). The molecular regulatory mechanisms of CsCBFs at the crossroads of plant hormone signaling in cold stress response need to be further elucidation. Light responsiveness motifs were also found in CsCBF promoters. Previous studies have showed that light is required for many cold-responsive genes, and there is a complex cross-talk between light and cold (Catalá et al., 2011). Large numbers of stress responsiveness cis-elements were found in CsCBF promoters, giving the reason tea plant can defend the cold stress. Sequence analysis of the CsCBF promoters revealed the existence of different numbers of MYB and MYC binding site, suggesting that induction of the CsCBF genes in response to low temperature is involved in the regulation of transcription factors, such as ICE1 and MYB15 in Arabidopsis (Chinnusamy et al., 2003;Agarwal M. et al., 2006).
To confirm in vivo functions of the CsCBF3 gene during lowtemperature stress in plants, we ectopically expressed CsCBF3 into Arabidopsis. The results showed that in the case of CsCBF3 overexpression, transgenic plants showed enhanced resistance to cold damage. We also observed overexpression of CsCBF3 resulting in delayed flowering and dwarfism. Under a cold environment, 4-weeks-old overexpression plants had a much higher survival rate than wild-type plants. Similar function was observed in AtCBF1-3, GhDREB1, SmCBF, LpCBF3 (Xiong and Fei, 2006;Novillo et al., 2007;Shan et al., 2007;Zhou et al., 2018). To clarify how CsCBF3overexpressing transgenic plants cope with low temperature stress, we examined the relative expression level of the downstream target gene of CBF identified in Arabidopsis plants. The results showed that the genes belonging to the ABAindependent pathway had a higher expression than the ABAdependent genes when transgenic Arabidopsis plants were exposed to cold stress. Zhang et al. (2004) demonstrated the LeCBF1 in a heterologous system could activate the Arabidopsis cold related (COR) genes involved in increasing freezing tolerance, but that LeCBF1 in tomato plants did not upregulate equivalent genes (Zhang et al., 2004). The function of CsCBF3 overexpression in tea plant can't be studied because of genetic transformation system has not been established. We speculated that CsCBF3 overexpression in transgenic plants improved cold tolerance mainly through the ABA-independent pathway, which was consistent with CBFs belonging to the ABAindependent pathway. CONCLUSION CBF/DREB transcription factors were identified in tea plants. CsCBF proteins were localized to the nucleus. CsCBFs had transcriptional activity except CsCBF1. CsCBF gene expression could be affected by abiotic stress and plant hormones. Ectopic expression of CsCBF3 in Arabidopsis induced cold tolerance, and the mechanism of CsCBF3 regulation of downstream target genes was mainly the ABA-independent pathway.

AUTHOR CONTRIBUTIONS
YL and EX designed the study. ZH, QB, JH, XZ, YC, JM, and ML conducted the experiments and analyzed the data. ZH and QB prepared the manuscript. All authors consent to the manuscript.

ACKNOWLEDGMENTS
The authors are thankful to Prof. Zhaoliang Zhang and for providing plant expression vector PK7WGF2.0 and PGBKT7.

SUPPLEMENTARY MATERIAL
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpls.2020. 00921/full#supplementary-material Table S1 | Primers used in this study. Table S2 | Name of genes in this study.
Data Sheet S1 | The amino acid sequences of CBF genes used in phylogenetic tree construction. Data Sheet S2 | Expression level of CsCBFs under abiotic stresses. Data Sheet S3 | Expression level of downstream target genes of CsCBF3 in Arabidopsis.