Introduction

Plants are unable to escape unfavorable environmental conditions thus have to endure these environmental challenges such as soil salinity, drought and cold temperature during the course of their life. To maintain growth, development and productivity, plants have evolved elaborate mechanisms to tolerate and adapt to those environmental stresses (Zhu 2001; Ross et al. 2007; Hirayama and Shinozaki 2010; Xiang et al. 2013). When plants are exposed to these conditions, multiple stress-responsive genes, multiple complicated and precisely-regulated physiological and molecular networks and signaling pathways are involved in coping with these abiotic stresses to protect plants from further injury (Mittler 2006; Tang et al. 2012). Transcription factors including WRKY expanded and evolved in plants that play a central role in coordinating gene expression to maintain growth, development and productivity during the changing environments (Ross et al. 2007). The WRKY transcription factors contain one or two domains composed of the conserved amino acid sequence WRKYGQK at the N-terminus and a zinc-finger motif at the C-terminus (Eulgem et al. 2000; Rushton et al. 2012). The conserved WRKY domain plays important roles in recognizing and binding to the W-box in the promoter regions of target genes (Chen et al. 2012). The WRKY superfamily is classified into three distinct groups (I, II and III) based on the number of the WRKY domains and zinc-fingers feature. Group II can be further subdivided into five subgroups (IIa to IIe) depending on the presence of additional short and conserved structural motifs (Eulgem et al. 2000). WRKY proteins are able to activate or repress transcription of the target genes through directly binding to the W-box containing a core sequence (T)(T)TGAC(C/T) in the promoter (Eulgem et al. 2000). Previous studies revealed that WRKY transcription factors function in many processes including plant growth and development (Johnson et al. 2002; Zhang et al. 2011), senescence (Miao et al. 2004), response to abiotic stresses (Zhou et al. 2008; Niu et al. 2012; Luo et al. 2013), as well as plant defense (Xu et al. 2006; Zheng et al. 2006). As one of the largest transcription factor families in plants, there are 74 WRKY genes in Arabidopsis thaliana, more than 100 members in Oryza sativa (Ross et al. 2007; Rushton et al. 2010) and at least 43 putative WRKY genes that have been identified in Triticum aestivum (Niu et al. 2012).

Some WRKY genes responding to abiotic stress had been studied. In Arabidopsis, AtWRKY25 plays a role in thermo-tolerance by regulating heat shock proteins-related signaling pathway (Li et al. 2009). AtWRKY33 gene conferring salt stress tolerance and the AtWRKY63 (ABO3) gene involved in drought stress response are mediated by ABA signaling pathway (Jiang and Deyholos 2009; Ren et al. 2010). Arabidopsis WRKY18, WRKY40 and WRKY60 respond to abscisic acid and abiotic stress (Chen et al. 2010). AtWRKY70 and AtWRKY54 modulate osmotic stress tolerance by regulating stomatal aperture (Li et al. 2013). AtWRKY6 modulates PHOSPHATE1 expression in response to low Pi stress (Chen et al. 2009). AtWRKY57 confers drought tolerance by elevation of ABA level (Jiang et al. 2012). AtWRKY39 mediates response to heat stress by regulating the cooperation between the SA- and JA-activated signaling pathways in Arabidopsis (Li et al. 2010). The rice WRKY genes, OsWRKY24, -45, -72 and -77, are involved in ABA signaling (Xie et al. 2005). Rice OsWRKY11 enhances heat and drought tolerance in rice seedlings (Wu et al. 2009). Overexpression of OsWRKY30 results in increased drought tolerance by activation of MAP kinases and OsWRKY45 overexpression improves rice tolerance to drought and cold stresses (Tao et al. 2011; Shen et al. 2012). OsWRKY89 enhances ultraviolet B tolerance in rice (Wang et al. 2007). OsWRKY08 improves osmotic stress tolerance in Arabidopsis (Song et al. 2009). All these stress responsive WRKY genes are distributed mainly in group I, III, IIa, IIb, IIc and IId, whereas whether group IIe WRKY genes are involved in abiotic stress tolerance is still unknown.

As one of the major staple food crops, wheat (T. aestivum L.) genome consisting of the A, B and D genomes with about 16,000 Mb, is 40-fold larger than rice (Bhalla 2006; Tripathi et al. 2012) and is not sequenced completely, which hinders the identification and isolation of WRKY genes. In wheat, only a few WRKY genes involved in abiotic stress have been functionally characterized. Recently, four wheat WRKY genes (TaWRKY1, -2, -17 and -19) were reported to be induced by four or five stress treatments (salt, drought, cold, wounding and ABA) (Niu et al. 2012). Overexpression of TaWRKY2 exhibits tolerance to salt and drought stresses and TaWRKY19 confers tolerance to salt, drought and freezing stresses by regulating downstream genes in transgenic Arabidopsis (Niu et al. 2012). In addition, wheat TaWRKY10 confers tolerance to multiple abiotic stresses by regulating the osmotic balance, ROS scavenging and transcription of stress related genes in transgenic tobacco and TaWRKY79 enhances tolerance to both salinity and ionic stresses in Arabidopsis (Qin et al. 2013; Wang et al. 2013). Fortunately, the genome of Brachypodium distachyon is fully sequenced which is a phylogenetic sister group to wheat and has been used as a new model organism for wheat (Li et al. 2012a; Wen et al. 2014). Recently, 86 WRKY genes have been identified in B. distachyon (Tripathi et al. 2012; Wen et al. 2014). Studies of WRKY transcription factors from B. distachyon therefore will provide useful reference for their ortholog in wheat. However, the function of WRKY gene from B. distachyon is largely unknown. In this study, a group IIe WRKY gene designated as BdWRKY36 from B. distachyon was characterized and analyzed in regulating drought stress tolerance by activating the antioxidant system and the expression of stress-responsive genes.

Materials and methods

Plant materials, growth conditions and stress treatments

B. distachyon (Bd 21) seeds were surface-sterilized using 75 % (v/v) ethanol for 1 min and 1 % (v/v) mercury chloride for 5 min, and then were washed 3–5 times with sterile distilled water. Sterilized seeds were germinated on the filter paper in Petri dishes with sterile water under dark conditions. After germination, the Petri dishes were transferred to greenhouse under normal conditions with 16 h/light, 8 h/dark and 24 °C temperature for 14 days. For drought treatment, 14-day-old seedlings were dehydrated on filter paper at 24 °C with 30 % relative humidity. For NaCl treatment, seedlings were incubated in 200 mM NaCl solution. For treatments with signaling molecules, seedlings were sprayed with 100 µM ABA and 10 mM H2O2, respectively. For cold or heat treatment, seedlings were grown at 4 or 40 °C conditions, respectively. Leaves from control or treated plants were harvested after 0, 1, 3, 6, 12 and 24 h of treatments, and frozen in liquid nitrogen, and then stored at −70 °C for RNA preparation. For the organs differential expression analysis, roots, leaves, stems and spikes were collected from 3-month-old plants grown under greenhouse condition as described above.

Cloning and identification of BdWRKY36 gene

Total RNA from leaves of B. distachyon seedlings was extracted with Trizol reagent (Invitrogen, USA) according to the manufacturer’s protocol. The first strand cDNAs were synthesized using the first strand cDNA Synthesis kit (Fermentas, USA). Polymerase chain reaction (PCR) was performed with gene-specific primer pairs (P1; Table S1) designed by the software Primer Premier 5.0 based on the open reading frame (ORF) of BdWRKY36 from databases http://www.igece.org/WRKY/BrachyWRKY/BrachyWRKYIndex.html and http://plants.ensembl.org/Multi/blastview. Amplified products were cloned into pMD18-T vectors (TaKaRa, Japan) and were subsequently subjected to sequencing analysis. To verify the reliability of the results, the putative protein sequence was assessed with SMART (http://smart.embl-heidelberg.de/) and UniProt (http://www.uniprot.org/) databases. Multiple sequence alignment was performed using the putative protein sequence of BdWRKY36 and WRKY protein sequences from other species retrieved from GenBank by DNAMAN software. Phylogenetic tree was constructed using software MEGA 5.0 (Tamura et al. 2011).

Semi-quantitative RT-PCR

Reverse transcription-PCR (RT-PCR) was used to analyze the expression levels of BdWRKY36 in different organs of B. distachyon plants, at different time intervals (0, 1, 3, 6, 12 and 24 h) from B. distachyon seedlings and in the transgenic tobacco lines. PCR amplifications were performed as follows: 95 °C for 5 min, 28 cycles at 95 °C for 30 s, 60 °C for 30 s and 72 °C for 45 s, and the final extension at 72 °C for 5 min. Expression levels of target genes were normalized using BdUBC18 or NtUbiquitin as internal controls. The relative amount of the transcription was analyzed using Image Lab 4.1 analyzer software. Specific primers (P3–P5; Table S1) were designed using software Primer Premier 5.0.

Gene expression analysis by qRT-PCR

Expression of BdWRKY36 in B. distachyon seedlings with various treatments as well as relative transcription levels of ROS-scavenging and stress-responsive genes including NtPOX2, NtCAT, NtSOD, NtLEA5, NtNCED1 and NtDREB3 in transgenic tobacco lines were measured by quantitative RT-PCR (qRT-PCR) using SYBR Green qPCR Mix (ToYoBo, Japan). Specific primers (P2, P4–P5 and P13–P18; Table S1) used in qRT-PCR were designed to avoid the highly conserved sequences. The efficiency and specificity of the primer pairs were examined by melting curve, agarose gel electrophoresis, and sequencing analysis. The PCR parameters were as follows: 95 °C for 30 s, 40 cycles of 95 °C for 10 s, 59 °C for 10 s and 72 °C for 30 s according to the manufacturer’s protocol. Three biological replicates were performed. BdUBC18 and NtUbiquitin were used as the internal controls for B. distachyon and tobacco, respectively, which served as a benchmark to normalize tested genes. Expression levels for all the candidate genes were calculated and normalized using the 2−ΔΔCT method (Livak and Schmittgen 2001).

Plasmid construction and subcellular localization analysis

The BdWRKY36 coding sequence without the stop codon was amplified using primers (P6; Table S1) containing BglII/SpeI restriction sites. The PCR products were purified and digested with BglII/SpeI restriction enzymes and then were inserted into the plant binary vector pCAMBIA1304 under the control of Cauliflower mosaic virus 35S (CaMV 35S) promoter to generate a fusion construct pCAMBIA1304-BdWRKY36-GFP. For protein localization assays, the recombinant plasmid pCAMBIA1304-BdWRKY36-GFP and the vector pCAMBIA1304-GFP as a control were transformed into onion epidermal cells, respectively, by particle bombardment methods using a gene gun (BIO-RAD, USA). After transformation, the onion epidermal cells were incubated on MS medium at 25 °C under dark condition for 16 h. Subsequently, nuclei were stained with 100 µg/ml 4′,6-diamidino-2-phenylindole (DAPI) (Solarbio, China) in phosphate-buffered saline for 10 min, and then the localization of green fluorescent proteins in transformed cells was observed by fluorescence microscope (Olympus, Japan).

Analysis of transcriptional activation in yeast cells

To examine the transcriptional activation activity, the complete coding sequence of BdWRKY36, N-terminal (from the 1st to the 132th aa), N-terminal-WRKY domain (from the 1st to the 199th aa) and C-terminal (from 200th aa to the end of C-terminus) fragments were amplified using specific primers (P9–P12; Table S2) containing EcoRI/BamHI restriction sites, respectively. The PCR products digested by restriction enzymes EcoRI/BamHI were cloned into the pGBKT7 vector digested with the same restriction enzymes to create pGBKT7-BdWRKY36, pGBKT7-BdWRKY36-N and pGBKT7-BdWRKY36-N-WRKY, and pGBKT7-BdWRKY36-C, respectively. These constructs and the pGBKT7 vector as a control were transformed into yeast strain AH109 according to the manufacturer’s protocol (Clontech, USA). The transformants were screened on the SD/Trp− and SD/Trp−/His− mediums, respectively. Transcriptional activation was evaluated based on the growth status of yeast cells after incubating SD/Trp−/His− medium with X-α-gal at 30 °C for 3 days.

Tobacco transformation and generation of transgenic lines

The recombinant plasmid pCAMBIA1304-BdWRKY36-GFP and the vector pCAMBIA1304-GFP as a control were introduced into Agrobacterium tumefaciens strain EHA105 and then transformed into tobacco (Nicotiana tabacum L. cv Samsun) by Agrobacterium-mediated leaf disk transformation method according to Horsch et al. (1985). The seeds from transgenic plants were screened on MS medium with 30 mg/L of hygromycin. The T1 BdWRKY36-overexpressing tobacco plants were confirmed by PCR assay with specific primers (P6–P8; Table S1) for BdWRKY36, GFP and HptII genes, respectively. The control plants transformed with GFP only containing vector were confirmed by PCR analysis using specific primers for GFP and HptII genes. Three independent transgenic T2 lines (OE-6, OE-14, OE-18), which almost all survived on MS medium with 30 mg/L of hygromycin were used for further experiments. BdWRKY36 expression in the three independent T2 lines was analyzed with semi-quantitative RT-PCR using primers (P3; Table S1) with NtUbiquitin as an internal control.

Drought stress tolerance assay for transgenic tobacco

For drought stress tolerance assay, the seeds of wild type (WT), vector control (VC) and transgenic lines(OE-6,OE-14, OE-18)were surface-sterilized using 75 % (v/v) ethanol for 1 min and 10 % (v/v) H2O2 for 15 min and washed 3–5 times with sterile water. The sterilized seeds grew on MS medium for 10 days and then were transplanted into plastic pots filled with soil and sand (3:1) under normal conditions for 20 days. Some plants were still cultured under normal conditions and the others were subjected to water depriving for 20 days and then were rewatered for 10 days to evaluate the survival rate.

Measurements of physiological–biochemical parameters

For drought treatment, groups of 30-day-old transgenic and control plants were withheld water for 20 days. The control groups were watered regularly. Then the leaves from control and transgenic plants under normal or drought treatments were sampled to detect the physiological–biochemical parameters.

For measurement of relative water content (RWC), the leaves were detached immediately and weighed as fresh weight (FW) followed by incubating in distilled water for 4 h at room temperature under normal light, and then the turgid weight (TW) was recorded. The samples were then transferred into an oven at 80 °C for 48 h and the dry weights (DW) were recorded. RWC was calculated as follows: RWC (%) = [(FW − DW)/(TW − DW)] × 100 according to Hu et al. (2013).

For measurement of ion leakage (IL), about 0.5 g tobacco leaves were cut into strips and incubated in 15 mL tubes with distilled water for 12 h at 26 °C. After incubation, initial electrolyte conductivity (C1) was measured with a conductivity meter (DDBJ-350, China) followed with the samples boiled for 30 min. We did not measure the electrolyte conductivity (C2) until the samples were cooled to room temperature. IL was calculated according to the formula: IL (%) = C1/C2 × 100 based on the method described by Jiang and Zhang (2001).

For measurement of chlorophyll content, the chlorophyll of tobacco leaves was determined with UV spectrophotometry method according to Yang et al. (2009). ROS accumulation of O2 − and H2O2 in situ was detected based on histochemical staining by nitroblue tetrazolium (NBT) and 3,3′-diaminobenzidine (DAB), respectively, as described by Pan et al. (2012). H2O2 content was assayed using Detection Kit (A064, Jiancheng, China) according to the instruction of manufacturer. The activities of peroxidase (POD), catalase (CAT) and superoxide dismutase (SOD) were measured by a spectrophotometer. About 0.5 g tobacco leaves was homogenized in 5 mL extraction buffer with 0.05 M phosphate buffer (pH 7.8), then the samples were centrifuged at 10,000 g at 4 °C for 15 min to collect the supernatant that was used to determine enzyme activities of POD, CAT and SOD using Detection Kit (A084, A007 and A001, Jiancheng, China) according to the instruction of manufacturers.

Statistical analysis

Statistical analyses were performed using the software in Excel and Photoshop CS5 (Microsoft and Adobe, USA). Mean values ± SD are calculated from at least two replicates, and significant differences between transgenic and control plants in measured parameters are given at P < 0.01 and P < 0.05 based on Student’s t test.

Results

Cloning and sequence analysis of BdWRKY36

The coding sequence of the BdWRKY36 gene was cloned from the leaves of B. distachyon seedlings by RT-PCR. The ORF of BdWRKY36 is 1,200 bp, encoding a protein with 399 amino acid residues. Its predicted relative molecular mass is 42.3 kDa with isoelectric point of 9.01. Blast analysis showed that BdWRKY36 shared a high degree of sequence identity with WRKYs from other plant species: 72 % sequence identity with OsWRKY37 and 58 % with OsWRKY66 from O. sativa, 50 % with AtWRKY35 and 47 % with AtWRKY14 from A. thaliana. Amino acid sequence alignment by homology search from the GenBank database showed that BdWRKY36 was homologous to WRKY transcription factors, especially with regards to the WRKY domains. BdWRKY36 contains a single WRKYGQK motif, a C2H2-type zinc finger and a nuclear localization signal “KRRK” sequence (Fig. S1). Phylogenetic analysis revealed that BdWRKY36 was more closely related to OsWRKY37 and both of them belong to the group IIe WRKY (Fig. S2).

Expression analysis of BdWRKY36 in response to abiotic stresses and signaling molecules

To determine the expression profile of BdWRKY36 in different organs, we performed semi-quantitative RT-PCR with the total RNA from different organs of B. distachyon. BdWRKY36 was ubiquitously expressed in all organs examined, including roots, stems, leaves and spikes, and exhibited higher expression levels in stems and leaves (Fig. 1a). In addition, we also determined the expression pattern of BdWRKY36 at different time intervals (0, 1, 3, 6, 12 and 24 h) by RT-PCR using the total RNA extracted from leaves of 2-week-old B. distachyon seedlings. The results showed that there were no expression changes of BdWRKY36 at different time intervals under normal conditions (Fig. S3). Therefore, the sample at 0 h was used as a control in the following expression assay. To explore the response of BdWRKY36 to various abiotic stress conditions and signaling molecules, we carried out qRT-PCR using the total RNA extracted from 2-week-old B. distachyon seedlings subjected to various treatments at different time intervals (0, 1, 3, 6, 12 and 24). Under drought stress, the expression of BdWRKY36 was increased 5.4-fold at 1 h and 2-fold at 6 h followed by a decrease (Fig. 1b-i). Cold (4 °C) treatment enhanced BdWRKY36 transcripts about 2.5-fold at 3 h and 3-fold at 6 h followed by a recovery (Fig. 1b-ii). BdWRKY36 transcripts were also slightly upregulated by heat treatment (Fig. 1b-iii). When treated with 200 mM NaCl, the expression of BdWRKY36 was not decreased until 6 h treatment (Fig. 1b-iv). In addition, under the treatment of ABA, the expression levels of BdWRKY36 was decreased slightly after 1 h and then remained at low level until 12 h (Fig. 1b-v). In case of treatment with H2O2, there was not significant change for BdWRKY36 expression from 1 to 24 h (Fig. 1b-vi). These results suggested that BdWRKY36 expression could be upregulated by drought, cold and heat treatments and downregulated by NaCl and ABA treatments.

Fig. 1
figure 1

RT-PCR and qRT-PCR analysis of expression profiles of BdWRKY36. a The expression analysis of BdWRKY36 in different organs by semi-quantitative RT-PCR. Total RNA was isolated from roots (R), stems (S), leaves (L) and spikes (SP), respectively. BdUBC18 gene was used as an internal control. b The expression analysis of BdWRKY36 under different treatments by real-time RT-PCR analysis. Total RNA was isolated from leaves of B. distachyon seedlings treated with (i) drought, (ii) cold (4 °C), (iii) heat (40 °C), (iv) 200 mM NaCl, (v) 100 µM ABA, (vi) 10 mM H2O2 at indicated times. Data are mean ± SD calculated from three replicates. Asterisks indicate significant difference between the control (0 h) and the treatments (*P < 0.05; **P < 0.01). Three biological experiments were performed, which produced similar results

Full size image

BdWRKY36 is localized in nucleus

To clarify the subcellular localization of the BdWRKY36, the ORF of BdWRKY36 was fused to the N-terminus of the green fluorescent protein (GFP) reporter gene under the control of the Cauliflower mosaic virus (CaMV) 35S promoter. We performed transient expression assays using onion epidermal cells with the construct BdWRKY36-GFP fusion and GFP alone by particle bombardment methods. As shown in Fig. 2, GFP alone distributed throughout the whole cell, whereas BdWRKY36-GFP fusion protein accumulated in the nucleus only. Additionally, the green fluorescence and blue (DAPI staining) were both confined to the nucleus in BdWRKY36-GFP-expressing cells, suggesting the nucleus location of BdWRKY36. These results clearly demonstrated that BdWRKY36 is localized in the nucleus, which is consistent with its predicted function as a transcription factor.

Fig. 2
figure 2

Subcellular localization of BdWRKY36-GFP The fusion protein 35S-BdWRKY36-GFP (pCAMBIA1304-BdWRKY36-GFP) and 35S-GFP (pCAMBIA1304-GFP as control) were transiently expressed in onion epidermal cells and visualized under bright and fluorescence field before or after DAPI staining with fluorescence microscopy after bombardment of 16 h. a Fluorescence image of GFP control (b) the corresponding DAPI staining for the nucleus (c) merged fluorescence and DAPI image (d) the corresponding bright field (e) fluorescence image of BdWRKY36-GFP, (f) the corresponding DAPI staining for the nucleus, (g) merged fluorescence and DAPI image, (h) the corresponding bright field. Scale bars = 100 µm. Three biological experiments were performed, which produced similar results

Full size image

C-terminal domain of BdWRKY36 shows transcriptional activity

To investigate whether BdWRKY36 possesses transcriptional activation activity, the ORF of BdWRKY36 was fused to the GAL4 DNA binding domain (GAL4BD) in the vector pGBKT7 and the construct was transformed into the AH109 yeast strain. As shown in Fig. 3, the reporter genes His and LacZ were not expressed in the yeast cell transformed with the GAL4BD-BdWRKY36 construct as well as GAL4BD as a control, indicating that the full length ORF of BdWRKY36 had no transcriptional activity or was repressed by some repression region in itself in the yeast cell.

Fig. 3
figure 3

Transactivation activity of the BdWRKY36 protein in yeast. a The schematic diagram demonstrating the BdWRKY36 cDNA fragments encoding different portions of BdWRKY36 that were fused to the yeast vector pGBKT7. b Transactivation activity analysis of BdWRKY36 was performed using yeast strain AH109. The transformants were screened on the SD/Trp− and SD/Trp−/His−mediums. The transformants were examined for growth in the presence of X-α-gal. Three biological experiments were performed, which produced similar results

Full size image

To further check the transactivation region of BdWRKY36, GAL4BD-N-terminus, GAL4BD-N-WRKY domain and GAL4BD-C-terminus were constructed and transformed into the yeast strain AH109. The yeast cell growth on the His− medium and LacZ staining indicated that the transactivation region of BdWRKY36 is located at the C-terminal region of BdWRKY36 (beginning from 200th aa to the end of C-terminus (Fig. S1) only, whereas its full length, N-terminal domain and N-term-WRKY domain had no transcriptional activity (Fig. 3).

Generation of transgenic tobacco overexpressing BdWRKY36

To further investigate the biological function of BdWRKY36, transgenic tobacco plants overexpressing BdWRKY36 under the control of CaMV 35S promoter were generated. A total of 26 independent transgenic lines (T1) were obtained by 30 mg/L hygromycin resistance selection and confirmation by genomic PCR using primers specific to BdWRKY36, HptII and GFP. Among the T1 lines, three (OE-6, OE-14 and OE-18) lines segregated at a ratio of ~ 3:1 based on 30 mg/L hygromycin resistance and showed higher expression levels of BdWRKY36 by RT-PCR analysis (Fig. 4a). In this experiment, tobacco plants transformed with the empty vector alone were also subjected to similar analysis.

Fig. 4
figure 4

Analysis of the enhanced drought tolerance in transgenic lines. a The expression of BdWRKY36 in transgenic lines (OE-6, OE-14, OE-18). The tobacco NtUbiquitin gene was used as internal reference. WT wild type, VC vector control; OE-6, OE-14 and OE-18, three independent transgenic tobacco lines overexpressing BdWRKY36. b The photographs of WT, VC and transgenic lines before drought, after water withholding for 20 days, and after rewatering for 10 days. c The survival rates of WT, VC and transgenic lines after rewatering for 10 days. Data are mean ± SD calculated from three replicates. Asterisks indicate significant difference between WT and three transgenic lines (**P < 0.01). Two biological experiments were performed, which produced similar results

Full size image

Overexpression of the BdWRKY36 gene enhances drought tolerance in transgenic tobacco

The expression of BdWRKY36 was upregulated under drought stress in B. distachyon (Fig. 1), indicating that BdWRKY36 may play a role in drought stress response. Thus, 30-day-old tobacco plants were used to test drought tolerance. After water withholding for 20 days, all the leaves of WT and VC plants severely wilted and curled, whereas some leaves of transgenic plants remained expanded (Fig. 4b). When rewatered for 10 days, only 10 % of the WT and 12.5 % of the VC plants survived, while the survival rate of the transgenic lines OE-6, OE-14 and OE-18 was 45, 68.75 and 36.25 %, respectively (Fig. 4c). These results suggested that overexpression of BdWRKY36 enhances drought stress tolerance in transgenic tobacco plants.

Overexpression of BdWRKY36 improves RWC and chlorophyll level, decreases IL and H2O2 in transgenic tobacco under drought stress

Enhanced drought tolerance in transgenic plants promoted us to analyze the physiological-biochemical differences between transgenic plants and controls. Relative water content (RWC) that gives a credible evaluation of the plant water status is a relevant tool for the measurement of drought tolerance. Transgenic lines showed higher RWC than WT plants after 20 days drought treatment (Fig. 5a). In addition, chlorophyll content which is an indicator to plant living status was measured in transgenic and WT plants. As shown in Fig. 5b, transgenic lines presented significantly higher chlorophyll content than WT plants under drought treatment. IL, also an important indicator of membrane injury, was significantly lower in transgenic lines compared with WT plants under drought stress (Fig. 5c), indicating that transgenic plants suffered less membrane damage than WT plants. Furthermore, H2O2 content was lower in transgenic lines than that in WT plants under drought treatment (Fig. 5d). These results suggested that BdWRKY36-overexpressing plants can keep more RWC and were less injured under drought stress condition.

Fig. 5
figure 5

Analysis of RWC (a), chlorophyll content (b), IL (c) and H2O2 content (d) in WT and transgenic lines (OE-6, OE-14) under normal and drought conditions. Thirty-day-old tobacco plants were deprived of water for 20 days, and then leaves were collected to measure RWC, IL, chlorophyll and H2O2. Data are mean ± SD calculated from three replicates. Asterisks indicate significant difference between the WT and the two transgenic lines (*P < 0.05, **P < 0.01). Two biological experiments were performed, which produced similar results

Full size image

Overexpression of BdWRKY36 reduces ROS accumulation and increases SOD, POD and CAT activities under drought stress

To confirm the function of BdWRKY36 in reducing ROS levels in transgenic tobacco plants under drought stress, we examined the accumulation of endogenous H2O2 and superoxide radical anions (O2 −) in transgenic plants and controls using DAB and NBT staining, respectively. As shown in Fig. 6, overexpression of BdWRKY36 decreased the accumulation of brown (DAB staining) and blue (NBT staining) pigments compared with WT and VC after 20 days drought stress. Since the antioxidant enzymes play vital roles in ROS scavenging and influence cellular ROS balance in the antioxidant defense system. Therefore, the activities of three important antioxidant enzymes SOD, POD and CAT were also measured in the leaves of transgenic and WT plants. As shown in Fig. 7, higher SOD, POD and CAT activities were observed in transgenic plants than that in WT plants under drought conditions. These results suggested that overexpression of BdWRKY36 increases SOD, POD and CAT activities under drought stress.

Fig. 6
figure 6

Analysis of ROS accumulation in the WT, VC and transgenic lines (OE-6, OE-14 and OE-18) under drought conditions. Thirty-day-old tobacco plants were deprived of water for 20 days and then leaves were collected to detect H2O2 and O2 − by DAB (upper) and NBT (lower) staining, respectively. Two biological experiments were performed, which produced similar results

Full size image
Fig. 7
figure 7

Analysis of SOD (a), POD (b) and CAT (c) activities in the WT and transgenic lines (OE-6, OE-14) under normal and drought conditions. Thirty-day-old tobacco plants were deprived of water for 20 days and then leaves were collected to detect SOD, POD and CAT activities. Data are mean ± SD calculated from three replicates. Asterisks indicate significant difference between the WT and the two transgenic lines (* P < 0.05, **P < 0.01). Two biological experiments were performed, which produced similar results

Full size image

Overexpression of BdWRKY36 affects the expression of ROS-scavenging and stress-responsive genes under drought stress

To further understand the molecular mechanism underlying the drought resistance conferred by BdWRKY36, the expression levels of ROS-scavenging and stress-responsive genes were detected in the transgenic (OE-14) and WT plants by qRT-PCR under normal and drought stress. Genes including NtSOD, NtPOX2 and NtCAT involved in ROS scavenging, NtLEA5 related to stress defense, NtNCED1 involved in ABA biosynthesis and the modulatory gene NtDREB3 were selected for this assay. Under normal conditions, there was no obvious difference for the expression of tested genes between transgenic lines and WT. Under drought conditions, transcription levels of ROS-scavenging genes NtSOD, NtPOX and NtCAT as well as stress-responsive genes NtLEA5, NtNCED1 and NtDREB3 were higher in the transgenic lines compared with WT. These results indicated that overexpression of BdWRKY36 enhances the expression of ROS-scavenging and stress-responsive genes under drought conditions.

Discussion

Lots of WRKY transcription factor genes have been identified and the function of some has been characterized from numerous higher plant species since the first WRKY gene was identified from sweet potato by Ishiguro and Nakamura (1994). However, there is no report about functional characterization of WRKY transcription factor from B. distachyon, which is now used as a model for the grasses especially for the wheat research recently. In the present study, we cloned a WRKY gene designated as BdWRKY36 from B. distachyon. Sequence analysis indicated that BdWRKY36 contains WRKYGQK and C2H2-type zinc finger motifs and belongs to group IIe WRKY. The result of subcellular localization of BdWRKY36-GFP indicated that BdWRKY36 is located in the nucleus (Fig. 2), which is consistent with previous studies on WRKY transcription factors from other species (Luo et al. 2013; Wang et al. 2013). Transcriptional activation analysis showed that the transcriptional activity is in its C-terminal region of BdWRKY36 only, whereas its full length, N-terminal domain and N-term-WRKY domain have no transcriptional activity (Fig. 3), which is similar with previous report on OsbZIP46, the full length of which also has no transactivation activity while its D domain deletion mutant has the activity (Tang et al. 2012). It seems that N-terminal domain or N-term-WRKY domain may have an important role in the regulation of the transactivation activity of BdWRKY36 and that the activation of BdWRKY36 may need posttranslational level modifications or need to be activated by some unknown upstream proteins. For instance, Arabidopsis ABO3 (AtWRKY63) needs some ABA-triggered post-translational modification, or need to cooperate with other ABA-related proteins for its capacity to mediate plant responses to abscisic acid and drought tolerance (Ren et al. 2010). Phosphorylation of OsWRKY33 by OsBWMK1 increases its W-box binding activity to mediate SA-dependent defense responses (Koo et al. 2009). The senescence-related AtWRKY53 protein needs MEKK1 for its W-box binding activity (Miao et al. 2007). AtWRKY33 functions in the MPK4-mediated JA/ET-activating signaling pathways (Zheng et al. 2006). These data suggest that the regulation of BdWRKY36 transactivation activity is important for its functionality.

qRT-PCR results showed that BdWRKY36 expression was downregulated by NaCl and ABA, but not changed by H2O2 treatment (Fig. 1b). However, transcription levels of BdWRKY36 were upregulated by drought, cold and heat, in particular the drought stress, suggesting that BdWRKY36 may play a role in abiotic stress responses. The stronger upregulation of the BdWRKY36 by drought stress treatment promoted us to analyze its function in drought tolerance by overexpressing BdWRKY36 in transgenic tobacco. Phenotype analysis of BdWRKY36-overexpressing transgenic lines showed that expression of BdWRKY36 in tobacco greatly enhances their ability to survive a very strong drought stress. (Fig. 4b). This result is agreement with previous studies on the functions of some WRKY genes. For instance, overexpressions of wheat TaWKRY2 and TaWRKY19 in transgenic Arabidopsis enhanced drought tolerance by regulating downstream genes (Niu et al. 2012). In addition, expression of soybean WRKY20 in Arabidopsis enhanced drought tolerance by regulating ABA signaling (Luo et al. 2013). Overexpression of TaWRKY10 in transgenic tobacco resulted in enhanced drought stress tolerance by improving proline and soluble sugar contents and reducing ROS accumulation (Wang et al. 2013). Although functional assays using water withheld not necessarily indicate that a plant has tolerance to moderate drought (Skirycz et al. 2011), it is still a beneficial trait of choice in agricultural practice especially in arid regions.

The deteriorative effects of drought on plants may include leaf wilting, reduction in leaf area, leaf abscission, changes in chlorophyll content, RWC and IL, generation of ROS, and accumulation of free radicals which disrupt cellular homeostasis by reacting with lipids, proteins, pigments and nucleic acids resulting in lipid peroxidation, membrane damage, and inactivation of enzymes, thus affecting cell viability (Roychoudhury et al. 2013). Thus, we carried out a set of experiments to analyze physiological and biochemical parameters of transgenic plants, and the results showed that the enhanced tolerance of transgenic lines to drought stress is related to higher contents of chlorophyll and RWC, but lesser IL and H2O2 (Fig. 5). Furthermore, transgenic plants accumulated lower levels of H2O2 and O2 − under drought stress compared with controls (Fig. 6). These results indicate that BdWRKY36 enhancing plant tolerance to drought is likely through reduction of membrane injury by inhibiting ROS accumulation or scavenging ROS production. To minimize cellular damage caused by oxidative stress, plants have to maintain their ROS pools at low levels (Foyer and Shigeoka 2011). To regulate ROS homeostasis, plants have evolved a complex antioxidant system that includes several enzymes, which scavenge ROS and protect cells against oxidative stress (Huang et al. 2010; Miller et al. 2010). Thus, the levels of antioxidant enzyme activity including SOD, POD and CAT were analyzed, and the results showed that the transgenic plants have higher SOD, POD and CAT activity than that in WT under drought stress (Fig. 7). This implies that ROS scavenging systems in transgenic plants might work more effectively compared to WT. Taken together, these data demonstrate that overexpression of BdWRKY36 gene may function in activation of antioxidant defense system, which results in transgenic plants suffering from less ROS-mediated injury under drought stress. These results are consistent with previous studies on some WRKY genes in other plants, which function in activation of antioxidant system under abiotic stress. For example, in a recent study, overexpression of the DgWRKY3 gene in tobacco confers tolerance to salt stress by increasing SOD, POD and CAT activities of antioxidant system (Liu et al. 2013). In addition, overexpression of the SlWRKY gene in tobacco has been shown tolerance to salt and drought stresses due to the elevation of SOD and POD activity (Li et al. 2012b) (Fig. 8).

Fig. 8
figure 8

Analysis of expression levels of ROS-related and stress-responsive genes in the WT and the transgenic line (OE-14) by qRT-PCR under normal and drought conditions. Thirty-day-old tobacco plants were deprived of water for 20 days and then leaves were used to detect gene expression. Data are mean ± SD of three replicates. Asterisks indicate significant difference between normal and drought treatment (*P < 0.05, **P < 0.01). Two biological experiments were performed, which produced similar results

Full size image

To further understand the regulatory roles of BdWRKY36 at transcriptional levels under drought condition, the expression patterns of some ROS-scavenging and stress-responsive genes were investigated. It was found that the transcript levels of the genes encoding ROS-scavenging enzymes including NtSOD, NtPOX and NtCAT were upregulated in the transgenic lines under drought condition, which may contribute to the improved antioxidant system. Moreover, ABA plays crucial roles in adaptive responses to environmental stress including drought. Biochemical and genetic evidence revealed that ABA biosynthesis under stress is regulated by the rate-limiting enzyme, 9-cis-epoxycarotenoid dioxygenase (NCED) (Huang et al. 2010). The expression of NtNCED1 was enhanced in transgenic plants compared with the WT under drought condition, indicating that ABA biosynthesis may be enhanced by BdWRKY36 overexpression. NtLEA5 encodes group five late embryogenesis abundant (LEA) proteins, which functions in dehydration tolerance by binding water, stabilizing labile enzymes, and protecting cellular and macromolecular structures (Hundertmark and Hincha 2008). Higher level transcription of NtLEA5 in the transgenic plants under drought stress implied that more LEA proteins might be synthesized in transgenic lines, which contributes to transgenic plants maintaining higher RWC and less membrane damage.

Taken together, under drought condition, the expression levels of ROS-scavenging and stress-responsive genes were significantly upregulated in transgenic plants than that in WT plants. However, no significant difference in the expression of these genes was found between the BdWRKY36-overexpressing and WT tobacco plants under normal condition, despite the fact that the constitutive promoter (35S) was used to drive BdWRKY36. Similar results were reported such as DgWRKY3 (Liu et al. 2013). The possible explanation is that BdWRKY36 may mediate the activation of such genes accompanied by other stress-responsive regulators (Sun et al. 2010; Yang et al. 2012; Liu et al. 2013).

In conclusion, the findings of this study demonstrate a role for BdWRKY36, group IIe WRKY in enhancing drought stress tolerance. BdWRKY36 confers tolerance to drought stress through improving antioxidant system and upregulating the transcriptional levels of ROS-scavenging and stress-responsive genes. These findings will help our understanding of the mechanisms of environmental stress on plants and highlight the role of WRKY transcription factor in the activation of antioxidant system and transcriptional regulation. In the future, exploring the direct targets of BdWRKY36 involved in drought stress tolerance will shed light on the molecular mechanism of BdWRKY36-mediated stress tolerance.