Mulberry Transcription Factor MnDREB4A Confers Tolerance to Multiple Abiotic Stresses in Transgenic Tobacco

The dehydration responsive element binding (DREB) transcription factors have been reported to be involved in stress responses. Most studies have focused on DREB genes in subgroups A-1 and A-2 in herbaceous plants, but there have been few reports on the functions of DREBs from the A-3–A-6 subgroups and in woody plants. Moreover, mulberry trees are ecologically and economically important perennial woody plants, but there has been little research on its stress physiology, biochemistry and molecular biology. In this study, a DREB gene from the mulberry tree, designated as MnDREB4A, classified into the A-4 subgroup by our previous study, was selected for further characterization. Our results showed that the MnDREB4A protein was localized to the nucleus where it activated transcription. The promoter of MnDREB4A can direct prominent expression downstream of the β-glucuronidase (GUS) gene under heat, cold, drought and salt stress, and GUS staining was deepest after 12 h of stress treatment. The MnDREB4A-overexpression transgenic tobacco showed the improved growth phenotype under untreated conditions, such as greener leaves, longer roots, and lower water loss and senescence rates. Overexpression of MnDREB4A in tobacco can significantly enhance tolerance to heat, cold, drought, and salt stresses in transgenic plants. The leaf discs and seedlings of transgenic plants reduced leaf wilting and senescence rates compared to the wild type plants under the different stress conditions. Further investigation showed that transgenic plants also had higher water contents and proline contents, and lower malondialdehyde contents under untreated condition and stress conditions. Our results indicate that the MnDREB4A protein plays an important role in plant stress tolerance.


Introduction
Abiotic stresses, such as heat, cold, high-salt levels, and drought, present major challenges to sustainable food production because they reduce potential yields by up to 50% in crop plants including cold, waterlogged, drought and saline environments [37]. However, there has been little research on its stress physiology, biochemistry and molecular biology [38], and there is very little information available about the mulberry DBRE functions. In our previous study, we identified and characterized 30 DREB genes from the M. notabilis Genome Database [39], and their functions during abiotic stress were preliminarily predicted.
Based on our previous research, MnDREB4A was classified into A-4 subgroup and it can be induced by multiple abiotic stresses [39]. Furthermore, the promoter of this gene contained several cis-elements that were related to adverse stresses. Therefore, we selected MnDREB4A for further investigation. In this study, we characterized its subcellular localization, promoter activity and function. The tolerance of the transgenic tobacco harboring MnDREB4A to multiple abiotic stresses was also evaluated.

Subcellular localization of the MnDREB4A protein
MnDREB4A was combined with the N-terminal or C-terminal of green fluorescent protein (EGFP) to construct a fusion protein expression cassette. The MnDREB4A and EGFP were inserted into the Kpn I and EcoR I site of the plant expression vector pLGNL, which had been preserved in our laboratory. The gene specific primers that contained restriction enzyme sites are shown in S2 Table. The vectors that contained one of the MnDREB4A::EGFP fusion, EGFP::MnDREB4A fusion genes or the EGFP expression cassette were confirmed by sequencing and then transiently introduced into onion (Allium cepa L.) epidermal cells by Agrobacterium tumefaciens LBA4404, as described by Liu et al. [43]. The nuclei were stained by DAPI (4', 6-diamidino-2-phenylindole) and images were captured for each color channel with a fluorescence microscope (Nikon, Tokyo, Japan).

Construction of the MnDREB4A promoter-GUS fusion expression plasmid and Arabidopsis transformation
The 1500 bp 5 0 upstream region of MnDREB4A and the GUS sequence were amplified by PCR using the specific primers with restriction enzyme sites shown in S2 Table. The promoters were fused with the GUS gene and then the fusion construct was inserted into the Kpn I and EcoR I sites of the pCAMBIA2300 vector. The reconstructed plasmid (MnDREB4A pro::GUS) was confirmed by sequencing and transformed into Agrobacterium tumefaciens EHA 105 by the freeze-thaw method [44]. The positive Agrobacterium tumefaciens was transformed into Arabidopsis thaliana by the floral dip method [45]. Transgenic Arabidopsis were selected on 1/2 MS medium containing a final concentration of 50 mg/L kanamycin and confirmed by PCR analysis. The plasmid MnDREB4A pro::GUS was used as a positive control and the genomic DNA of wild type (WT) Arabidopsis was used as negative control in the PCR analysis.

Southern blotting and GUS staining of transgenic Arabidopsis
Genomic DNAs from the WT and the different transgenic Arabidopsis lines were prepared from leaves using the cetyltrimethyl ammonium bromide method (CTAB) [46]. Exactly 20 μg of DNA per sample was digested with EcoR I overnight at 37°C. The WT Arabidopsis genomic DNA was used as the negative control. The digested DNAs were separated on 0.8% (w/v) agarose gel using electrophoresis and then transferred to a positively charged Hybond-N membrane (Roche Diagnostics GmbH, Mannheim, Germany). The GUS probe, labeled with digoxigenin (DIG), was synthesized by PCR followed the manufacturer's instructions (Roche Diagnostics GmbH, Mannheim, Germany). The Southern blot was finally performed following the pre-hybridization, washing and chemiluminescent detection procedure [47].
GUS activity was assessed histochemically using kanamycin-resistant T1 and T2 generation transgenic Arabidopsis that contained MnDEEB4A pro::GUS. The staining solution for the GUS assay was prepared as described by Rae et al. [48]. The plants were incubated with the staining solution at 37°C for about 12 h. The 7-day-old seedlings in each sample were exposed to 40°C (heat), 4°C (cold), 150 mM NaCl or 20% PEG6000 (drought) treatment. The samples were subjected to GUS staining after 0, 1, 3, 6, 9, 12, and 24 h of exposure. The GUS staining was quantified by ImageJ (National Institutes of Health, Maryland, USA). All results are representative of three independent experiments.

Construction and transformation of the vector for tobacco transformation
The MnDREB4A sequence, flanked by a Kpn I and a Spe I restriction site, were amplified by PCR using the specific primers shown in S2 Table. The obtained DNA fragments were digested by corresponding restriction enzymes and then inserted between the Kpn I and Spe I sites of the plant expression vector pLGNL under the control of the cauliflower mosaic virus (CaMV) 35S promoter. The recombinant plasmid (CaMV35S::MnDREB4A) was confirmed by sequencing and transformed into Agrobacterium tumefaciens EHA 105 using the freeze-thaw method [44]. Then the positive Agrobacterium tumefaciens was transformed into tobacco (Nicotiana tabacum L.) plants using the leaf disc method [49]. Transgenic tobacco was selected for on 1/2 MS medium containing a final concentration of 50 mg/L kanamycin and confirmed by GUS staining, and genomic PCR. To determine the expression levels of the MnDREB4A in transgenic tobacco plants, total RNA was extracted with RNAiso plus (Takara, Dalian, China), the first strand cDNA was synthesized using the Perfect Real Time version of the PrimerScript™ RT reagent Kit with gDNA Eraser (Perfect Real Time) (TaKaRa, Dalian, China). The qRT-PCR procedure was the same as the one described in our previous study [38]. The NtActin gene (GenBank Accession number: U60489) was used as an internal reference, the forward and reverse primers for amplification were NtActin -F (5'-TCACAGAAGCTCCTCCTAATCCA -3') and NtActin -R (5'-GAGGGAAAGAACAGCCTGAATG-3'). Besides, the seeds of the confirmed transgenic lines were cultured in MS medium containing 1 mg/L 6-BA for further research.

Difference analysis between transgenic lines and WT tobacco under untreated conditions
The leaves (taken from the same position on each sample), roots and water loss rates for the transgenic line and WT tobacco were compared under the same growth conditions. In addition, the leaves were cut into 1 cm diameter discs, which were then cultured with 1/2 MS nutrient solution in a climate chamber (Ningbo Southeast Instrument Corporation, Ningbo, China) (25°C, 16 h day/8 h night). The roots were pulled out and cleaned carefully so that they did not break. To determine the water loss rate, the leaves were weighed (FW) immediately and incubated at 30°C with 40% relative humidity and then weighed at designated times (Wt). The water loss rate was calculated as follows: Water loss rate ¼ ðFW À WtÞ = FW Â 100%: Each measurement was repeated three times.

Physiological and abiotic stress tolerance analysis of transgenic tobaccos
The transgenic lines and WT tobacco samples were grown on MS medium until root formation and then transferred into soil in a climate chamber (25°C, 16 h day/8 h night). After about 8 weeks, the tobacco plants were approximately 15 cm high. Then they were subjected to heat (40°C), cold (4°C), drought (20% PEG6000), and salt (400 mM NaCl) stresses until significant differences appeared between the transgenic line and WT tobacco. When this occurred, the leaves from the same positions on the sample were collected for relative water content (RWC), free proline and malonaldehyde (MDA) content measurements. Each treatment was replicated three times. To explore the effects of the various treatments on leaves, 1 cm diameter leaf discs were cut from healthy and fully expanded leaves of transgenic and WT tobacco plants. Then the leaf discs were treated with heat, cold, drought, and salt stresses, respectively.
After the abiotic stress treatments, the fresh weights (FW) of the transgenic line and WT tobacco leaves were immediately recorded. The leaves were soaked for 12 h in distilled water at room temperature under a constant light and the turgid weight (TW) was recorded. Total dry weight (DW) was recorded after drying for 24 h at 80°C. RWC was calculated as follows: RWC ð%Þ ¼ ½ðFW À DWÞ = ðTW À DWÞ Â 100% [50].
For the proline content analysis, exactly~0.5 g of fresh leaves was homogenized in 10 ml of 3% sulfosalicylic acid and centrifuged at 2,000×g for 15 min. The resulting 2 ml extract was incubated with 2 ml of ninhydrin reagent, which contained 2.5% (w/v) ninhydrin, 60% (v/v) glacial acetic acid and 40% 6 M phosphoric acid, and 2 ml of glacial acetic acid, at 100°C for 40 min. The reaction was terminated in an ice bath and 5 ml of toluene was added. The samples were then vortexed. The reaction mixture was incubated at 23°C for 24 h and the absorbance of the supernatant at 520 nm was determined by a spectrometer (TECHCOMP, Shanghai, China). Then the proline content was calculated as described previously [51].
The MDA content was determined as described by Hodges et al. [52]. Around 0.5 g of fresh leaves was homogenized in 10 ml 10% trichloroacetic acid (TCA) and centrifuged at 12,000×g for 10 min. Then 2 ml 0.6% thiobarbituric acid (TBA) dissolved in 10% TCA was added to a 2 ml aliquot of the supernatant. The mixture was heated in boiling water for 30 min and then quickly cooled in an ice bath. After centrifugation at 2,500×g for 10 min, the absorbance of the supernatant at 450, 532, and 600 nm was determined with a spectrometer (TECHCOMP, Shanghai, China). Then the MDA content was calculated as described previously [52].

Statistical analysis
The results were organized and analyzed by Excel 2013 (Microsoft, Redmond, USA), Origin 7.0 (OriginLab, MA, USA) and SPSS Statistics 17.0 (SPSS Inc., Chicago, IL, USA) software. The means were compared by Student's t-test were calculated to find the significant difference between different lines. Mean values that were significantly different within treatment from each other were indicated by asterisks.

Bioinformation analysis of the MnDREB4A gene
Multiple sequence analysis showed that all deduced proteins exhibited the typical features of DREB proteins, including an AP2 domain, and YRG, WLG, and RAYD motifs (Fig 1a). Moreover, these proteins also had the conserved valine (V14) at the 14th position and glutamic acid (E19) at the 19th position in the AP2 domain. The AP2 domain displayed a higher degree of amino acid identity than the N-termini and C-termini of the proteins.
To gain insight into the nature of the AP2 domain of mulberry DREB protein and its homologues from other species, including Populus trichocarpa, Vitis vinifera, Malus domestic, Gossypium hirsutum, Arabidposis thaliana, Zea mays, and Oryza sativa, sequence logos were produced to examine the level of conservation at each residue position. The results revealed that 46 (70.8%) of the conserved amino acids were identical in the AP2 domain, and the residues located in other positions displayed varying levels of conservation (Fig 1b). In addition to V14 and E19, there were four arginine (R), two tryptophan (W) and a valine (V) participating in the interaction between the DREB proteins and specific DNA sequences (Fig 1b). Apart from the AP2 domain, two other conserved motifs were also found (Fig 1b), but the functions of these motifs require further investigation.
Based on the SWISS-MODEL online prediction, the MnDREB4A encoding a protein containing three β-folded sheets and one α-helix (S1a Fig). The α-helix packed approximately parallel to the β-folded sheets. The V14 and E19 residues, which allow the protein to bind DNA, were located in the second β-folded sheet (β-2) (S1b Fig). When the model was rotated at certain angles, the second β-folded sheet (β-2) was in the most favorable position to bind the DNA (S1c Fig), which was consistent with that of other reported DREB genes [25].  (Fig 2), which implied that the MnDREB4A proteins were nucleus-localized proteins. This result was consistent with the protein structure prediction.

Promoter activity analysis
To examine whether the abiotic stress responsiveness of the MnDREB4A gene was transcriptionally controlled by 5 0 upstream regions, transgenic Arabidopsis were generated by The key residues that are essential for the highly specific interactions with and for binding to the DRE/CRT element are highlighted by colorful circles and squares. One-week-old T2 seedlings were treated with heat (40°C), cold (4°C), drought (20% PEG6000), and salt (150 mM NaCl). After the different treatments, the strongest GUS expressions were observed in the leaves and roots of TS2 (Fig 3), TS1, TS9, and TS10 (S4a-S4c Fig), and lower in TS6 (data not shown). The GUS activity was the highest after 12 h treatment and then declined. Although the GUS accumulation levels in the different transgenic lines were different, their response trends under abiotic stresses were identical. We also found that the GUS reporter was mainly expressed in the leaves and roots and it preferentially localized to the vasculature of various tissues.

The difference between transgenic line and WT tobacco under normal conditions
To investigate MnDREB4A function, the recombinant plasmid (CaMV35S::MnDREB4A) (Fig  4a) was transformed into tobacco and produced transgenic plants. These transgenic lines were selected on selection medium containing kanamycin. The transgenic plants were analyzed further by PCR using genomic DNA as the template, and 35S forward and gene-specific reverse primers (Fig 4b). The expression levels of the MnDREB4A gene in the transgenic lines and the control plants (WT) were examined by qRT-PCR analyses (Fig 4c). Expression of MnDREB4A in transgenic lines 1, 5 and 7 (designated as OE1, OE5, and OE7) was over 300-fold greater than in the control plants (Fig 4c). The OE1 transgenic line was chosen for further experiments.
The transgenic OE1 tobacco and the WT were cultivated under the same conditions. However, when compared to the WT under normal conditions, the OE1 tobacco plants showed obvious phenotypic leaf and root differences. For example, the OE1 leaves were smaller and greener than the WT leaves (Fig 5a) and the OE1 roots were longer than the WT roots (Fig 5b). Furthermore, the OE1 plants exhibited lower rates of water loss at each time point than the WT after leaf detachment (Fig 5c). When the leaf discs had been cultured in 1/2 MS nutrient solution for about 1 month, the discs from the WT plants became yellow and senescence was more obvious than for the transgenic tobacco OE1 disks (Fig 5d). All these results indicated that the OE1 transgenic tobacco should have a stronger tolerance to adverse conditions than the WT.

Overexpression of MnDREB4A enhanced tolerance to multiple abiotic stresses in transgenic tobacco
The OE1 and WT leaf discs and seedlings were tested with various abiotic stresses to characterize the function of MnDREB4A transgenic line under heat, cold, drought, and salt stress. Fig 6a  shows that the largest areas of damage were observed in WT tobacco leaf discs after 4 days of heat (40°C) treatment, while OE1 discs suffered only slight damage. After 1 month of cold (4°C) or drought (20% PEG6000) treatment, the WT discs exhibited more obvious chlorosis and senescence than the OE1 discs (Fig 6a). Similarly, the WT discs showed more severe water loss and senescence compared to the OE1 discs after salt (400 mM NaCl) treatment (Fig 6a). When OE1 and WT seedlings were subjected to the treatments, no evident morphological differences were observed between the transgenic OE1 line and the WT seedlings during the first few days of treatment. After heat treatment for 4 days, the WT plants exhibited severe symptoms of water loss and significant wilting, but only slight wilting was observed in some of the OE1 leaves (Fig 6b). When the plants were exposed to 4°C for 6 days, the leaves of the WT seedlings began wilting, whereas the leaves of the OE1 seedlings continued to grow well (Fig  6b), and when the seedlings were treated with 20% PEG6000 (drought) for 8 days, the WT seedlings also showed more obvious leaf wilting and senescence (Fig 6b). After 3 weeks of exposure to 400 mM NaCl, the OE1 seedlings had lower rates of leaf yellowing than the WT plants, and the OE1 plants grew taller and reached the flowering stage sooner than the WT plants (Fig  6b). The phenotype characterization suggested that overexpression of MnDREB4A enhanced The OE1 roots were longer than the WT roots. (c) The OE1 plants showed lower rates of water loss at each time point than the WT after leaf detachment and incubation at 30°C and 40% relative humidity. The significantly different values between the WT and OE1 plants are indicated by asterisks (n = 3). A single asterisk indicates a significant difference at P < 0.05 and two asterisks indicates a significant difference at P < 0.01. (d) The leaf discs were cultured in 1/2MS nutrient solution for about a month. The discs from WT plants became yellow and senescence was more obvious than on the transgenic tobacco OE1 leaf discs. The top line is the leaf discs before culturing, and the bottom line is the leaf discs after 1 month's culturing. heat, cold, drought and salt stress tolerance. Some important physiological indices were also measured in order to investigate the physiological differences between OE1 and WT plants. The OE1 transgenic line showed remarkably higher levels of proline (Fig 6c) and increased RWC (Fig 6c) compared to the WT plants, but lower levels of MDA (Fig 6c) under the different abiotic stresses. These results were also confirmed in the OE7 transgenic tobacco (S5 Fig). In conclusion, the physiological characterization results suggested that the abiotic stress tolerance of transgenic tobacco was considerably enhanced because of expeditious accumulation of stress relieving substances compared to the WT plants.

Discussion
To survive different abiotic stresses, plants have evolved intricate mechanisms that respond and adapt to these stresses at the molecular, cellular and whole-plant level [1]. Transcription factors (TFs) that regulate signal transduction pathways in response to heat, cold, drought and salt have attracted considerable attention. In our previous study, 30 DREB genes were identified in M. notabilis [39]. The MnDREB4A genes were classified into the A-4 subgroup and shown to respond to heat, cold, drought, and salt stresses. In this study, the MnDREB4A gene was further characterized to explore the function of this gene in response to various stresses.
Our previous work suggested that the response to different abiotic stresses by MnDREB4A may be related to the cis-elements in the promoter [39,53]. The promoter analysis showed that the 1500 bp 5'-upstream region of the MnDREB4A contained different kinds of cis-elements that were related to abiotic stresses (S6 Fig) [43], so we constructed the MnDREB4A pro::GUS vector and introduced it into Arabidopsis. After the stress treatments, the GUS activity of the transgenic Arabidopsis seedlings increased and reached its highest level after 12 h treatment (Fig 3 and S4 Fig). This result provided additional evidence that MnDREB4A played a role in plant resistance to abiotic stresses.
Studies have demonstrated that overexpression of the stress-inducible TF genes in transgenic plants is an effective strategy for improving abiotic stress tolerance [54]. Transgenic plants were generated in this study that overexpressed the MnDREB4A gene in tobacco (Fig 4). However, some reports have found that gene overexpression may cause dwarf plant phenotypes [55], but our MnDREB4A transgenic line showed improved growth under normal conditions, such as greener leaves, longer roots, and reduced damage and senescence, etc. (Fig 5). When exposed to abiotic stresses, the transgenic line had lower leaf wilting and senescence rates than the WT plants (Fig 6). These results suggested that overexpression of MnDREB4A gene can improve tolerance to abiotic stresses.
To explore the possible mechanisms underlying improved stress tolerance, several experiments were conducted to monitor the changes in physiological processes associated with plant responses to stresses. RWC is often used to measure plant water status under abiotic stresses [56]. In this study, transgenic tobacco had a higher RWC than the WT plants. Plants also accumulate several metabolites that can help to prevent detrimental changes [57]. Proline is a common osmolyte in plants, and proline accumulation is associated with plant responses to stress [58]. Abiotic stress causes lipid peroxidation, which leads to MDA accumulation [59]. Therefore, MDA content could be used as a measure of the damage caused by abiotic stresses [60]. Although physiological status changes are not the only way that plants cope with damage, plants with higher proline and lower MDA contents do show improved stress tolerance [56] and these physiological changes improved transgenic tobacco responses to adverse conditions.

Conclusion
In conclusion, our results clearly demonstrate that MnDREB4A is a stress inducible transcription factor in mulberry, and MnDREB4A can respond to heat, cold, drought and salt stress. Transgenic tobacco overexpressing MnDREB4A displayed a higher tolerance to abiotic stresses than the WT plants. Our results suggest that these functions are achieved by altering the plant's physiological status, such as accumulating water and proline, and reducing plant MDA contents, etc. This is probably because transgenic tobacco can efficiently respond to abiotic stresses. The present study provides some insights into the function of MnDREB4A from the A-4 subgroups of DREB proteins in mulberry plants and lays a foundation for the further understanding of their mechanisms underlying plant responses to abiotic stresses.

Author Contributions
Conceived and designed the experiments: X-QL C-YL A-CZ. Performed the experiments: X-QL C-YL QG MZ B-NC. Analyzed the data: X-QL C-YL MZ. Contributed reagents/materials/ analysis tools: A-CZ Z-HX. Wrote the paper: X-QL C-YL A-CZ.