Overexpression of Transcription Factor GmTGA15 Enhances Drought Tolerance in Transgenic Soybean Hairy Roots and Arabidopsis Plants

: Soybean ( Glycine max ) is one of the important oil crops worldwide. In recent years, environmental stresses such as drought and soil salinization have severely deteriorated soybean yield and quality. We investigated the overexpression of the transcription factor GmTGA15 in response to drought stress in transgenic soybean hairy roots and Arabidopsis plants. The results of quantitative real time polymerase chain reaction (qRT-PCR) analyses showed that GmTGA15 was greatly induced by salt, PEG6000, salicylic acid (SA), gibberellic acid (GA), abscisic acid (ABA), and methyl jasmonate (MeJA) in soybean. In response to drought stress, the contents of both chlorophyll and proline were signiﬁcantly increased, while the content of malondialdehyde (MDA) was signiﬁcantly decreased in the soybean hairy roots with the overexpression of GmTGA15 in comparison to wild type (WT). Under the simulated drought conditions, the transgenic Arabidopsis plants showed signiﬁcantly longer roots and lower mortality than that of the wild type. These results suggest that GmTGA15 promotes tolerance to drought stress in both soybean and Arabidopsis plants. This study provides the scientiﬁc evidence for further functional analysis of soybean TGA transcription factors in drought stress and the breeding of drought-resistance crops. and MDA. The results showed that in all three lines, under the treatment of drought stress induced by 20% PEG6000, the contents of both proline and chlorophyll were signiﬁcantly increased, while the MDA was signiﬁcantly reduced in soybean hairy roots with the overexpression of GmTGA15 compared with the control (Figure 5).


Introduction
Plants often face abiotic and biological stresses such as drought, poisoning, salinealkali, and pathogenic bacteria, which severely affect the growth and yield of crops. In longterm evolution, plants are induced to maintain their own growth and development through stress-responsive genes. These genes are divided into two categories, one containing the cell metabolism genes and stress-resistance genes, and the other the genes encoding the transcription factors, protein kinases, and phosphatases [1,2]. Transcription factors regulate the expression of genes by specifically binding to cis-acting elements on the promoter, while a single transcription factor can simultaneously bind to the promoters of different target genes.
The TGACG motif-binding factor (TGA) is a transcription factor belonging to the subfamily D of the basic leucine zipper (bZIP) family, and specifically binds to the activation sequence 1 (as-1) with the core sequence of TGACGT [3]. Transcription factors play vital roles in plant stress resistance and growth by regulating the transcription level of target genes. For example, Katagiri et al. used the cauliflower mosaic virus (CaWV) 35S promoter to identify the first TGA gene (i.e., the transcription factor TGA1a gene) in tobacco and determined its recognition of TGACGT-specific sequences [4]. To date, TGA genes have been isolated and identified in many crop and model plants including Arabidopsis, rice, wheat, Begonia, persimmon, cotton, and papaya [5][6][7][8][9][10].
Each TGA transcription factor contains a zinc finger domain and the glutamine transcription activation domain. The zinc finger domain is an alpha helix-rotation-alpha helix structure with a dimer consisting of a leucine zipper region and a basic region located at the C-terminus, while the glutamine activation domain is an acidic transcription activation domain [11]. The TGA transcription factors recognize the cis-acting element TGACGT with its core sequence of ACGT bound to the basic domain, while the leucine zipper region participates in oligomerization. The N-terminus of the transcription factor contains the acidic activation region STDxDT forming a dimer with the alkaline region binding directly to DNA [11].
Studies have shown that TGA transcription factors are mainly involved in the processes of plant growth and development, resistance to diseases, and stress regulation. For example, members of TGA transcription factors often interact with the regulatory factors NPR1, GRX480, ERF72, ARR2, and SCL14, participating in various disease-resistance or stress-resistance signal transduction pathways, such as salicylic acid (SA), jasmonic acid (JA), and abscisic acid (ABA), ultimately enhancing the tolerance of plants to adverse stresses [12]. In Arabidopsis, both the TGA1 and TGA4 regulate not only the biosynthesis of SA and pipecolic acid (PA) by regulating the expression of genes SARD1 and CBP60g [13] but also the root response to nitrate [14]. Specifically, the nitrate signal transmitted by TGA1/TGA4 directly regulates the CAPRICE (CPC) root hair cell fate determination genes to increase root hair formation in Arabidopsis [15]. Studies have shown that AtTGA4 enhances drought resistance by increasing the nitrate transport and assimilation in Arabidopsis [16], while TGA7 may respond to plant drought stress by negatively regulating the expression of downstream gene AtBGI [17]. In soybean, three genes (i.e., GmbZIP44, GmbZIP62, and GmbZIP78) have demonstrated the ability to enhance salt tolerance and freezing tolerance of transgenic Arabidopsis as negative regulators of ABA signal transduction [18], while the bZIP transcription factor MdbZIP44 in apple regulates ABA to promote anthocyanin accumulation [19]. Furthermore, the overexpression of GmFDL19 enhances the drought and salt tolerance of soybean [20], while the nuclear protein GmbZIP110 shows transcriptional activation activities and plays an important role in response to salt stress in soybean [21]. Recently, studies have shown that GmTGAs play important roles in symbiotic nodulation and response to nitrogen utilization in soybean [22].
To date, a total of 27 GmTGA genes have been identified in the soybean genome with two of them (i.e., GmTGA15 and GmTGA17) up-regulated by 10 and 17 times, respectively, in the soybean seedlings under drought-stress treatment [23]. Furthermore, it has been demonstrated that the overexpression of GmTGA17 enhances both drought tolerance and salt tolerance in transgenic Arabidopsis and soybean hairy roots [23]. Although both GmTGA15 and GmTGA17 share a lot of molecular similarities, the variations between these two transcription factors warrant further studies to verify and characterize the functions of these and many other transcription factors identified previously [23]. Our previous studies have shown that GmTGA15 is a nuclear protein in soybean containing a highly conserved bZIP domain and a DOG1 structure and showing the characteristics of transcriptional activation [24]. Due to these shared molecular structures and characteristics with GmTGA17, it is speculated that GmTGA15 could enhance tolerance to both drought and salt stresses in crop plants. To date, the functional studies of GmTGA15 in crop plants under environmental stresses are sparse.
In our study, we investigated the molecular functions of the overexpression of GmTGA15 in response to drought stress in transgenic soybean hairy roots and Arabidopsis plants. Our results showed that GmTGA15 significantly enhances drought tolerance in both soybean hairy roots and transgenic Arabidopsis plants. This study provides valuable information for the further functional analysis of soybean TGA transcription factors in drought stress and the breeding of drought-resistance crops.

Plant Materials, Growth Conditions, and Treatments
Plants of both soybean (Glycine max cultivar Williams 82) and Arabidopsis thaliana were grown in a mixture of nutrient soil and vermiculite (3:1) inside the growth chamber with a humidity of 50% under a daily light illumination (300 µmol·m −2 ·s −1 for soybean and 120 µmol·m −2 ·s −1 for Arabidopsis) of 14 h (7 a.m. to 9 p.m.) at 25 • C, then dark for 10 h (9 p.m. to 7 a.m.) at 22 • C. The young roots of soybean plants with the first 3-leaflet compound leaf unfolded were selected as the experimental materials.

Plasmid Construction and Agrobacterium Rhizogenes-Mediated Transformation of Soybean Hairy Roots
In our previous study [24], the GmTGA15 gene, with a length of 1089 base pairs containing eight exons and seven introns, was cloned from soybean plants of the cultivar Williams 82. Our results showed that the protein GmTGA15 was composed of 362 amino acids with a molecular weight of 41.01 kDa belonging to the family of bZIP transcription factors and was located in the nucleus, showing evident transcriptional activation activities. The full-length coding sequence of GmTGA15 amplified was inserted into the BamH I site of a pCAMBIA3301 vector using the forward primer (5 -TCCAGCTCCAGGATCCATGAATTCAGCATCCCCGCA-3 ) and the reverse primer (5 -GAGAAAGCTTGGATCCCTAAGCAGGTTCCCGGGGTCTAT-3 ). The pCAMBIA3301-GmTGA15 plasmid was then transformed into soybean hairy roots by the high-efficiency Agrobacterium rhizogenes-mediated method, as described previously [25].

Expression Analysis by qRT-PCR
For expression analysis, seedlings of soybean were exposed to a net floating on the surface of a solution containing 20% PEG6000, 200 mM NaCl, 100 µM abscisic acid (ABA), 200 µM gibberellic acid (GA), 2 mM salicylic acid (SA), and 50 µM Methyl jasmonate (MeJA), respectively. The selection of these four types of phytohormones as indicators of drought stress was based on previous studies [26,27]. The plant materials were sampled at 0, 1, 3, 6, 12, and 24 h, respectively, immediately frozen in liquid nitrogen, and kept in storage at −80 • C. Total RNA was extracted from the seedlings of soybean using the plant total RNA kit (TaKaRa, Shiga, Japan). To study the relative expression patterns of GmTGA in soybean, total RNA samples were also extracted from the different tissues or organs (i.e., roots, stems, leaves, flowers, pods, and seeds) at four different developmental stages in soybean, including the seedling stage (without branches on the stems), branching stage (with branches on the stems), flowering stage (with the presence of flowers), and podding stage (with the presence of pods and seeds). Samples were generally collected in the morning and were immediately frozen in liquid nitrogen, then kept in storage at −80 • C. The reverse transcription for the first strand cDNA synthesis for each plant sample was performed with 2 µg of total RNA using the first-strand DNA synthesis kit (TaKaRa, Shiga, Japan). The qRT-PCR analyses were performed on a MX3000P Real-Time system containing 12.5-µL 2× SYBR Premix Ex Taq (TaKaRa, Shiga, Japan) and 9.5-µL double distilled water using the GmTGA15 specific forward primer 5 -TGGACACTGGGTGAATGAGC-3 and reverse primer 5 -GTCAGGGGTTCAATCAGGGG-3 . The procedure to conduct the melting curve analysis was as follows: 1 cycle for 30 s at 95 • C, followed by 30 cycles for 5 s at 95 • C and 20 s at 60 • C, and 1 cycle for 60 s at 95 • C, 30 s at 55 • C, and 30 s at 95 • C. The inner reference gene EF1-α (forward primer 5 -TGCAAAGGAGGCTGCTAACT-3 and reverse primer 5 -CAGCATCACCGTTCTTCAAA-3 ) was used to normalize the data. The quantitative variations of gene expression between the examined replicates were evaluated by the 2 −∆∆Ct method as described previously [28]. Three independent biological repeats were performed to ensure accurate statistical analysis.

Drought Treatment in Transgenic Soybean Hairy Roots
The transgenic soybean hairy roots with the overexpression of GmTGA15 and the control group were selected to conduct the abiotic stress tolerance assays. Recombinant vectors pCAMBIA3301-GmTGA15 were transformed into the soybean hairy roots by Agrobacterium rhizogenes K599-mediated transformation [25]. The transgenic soybean hairy roots were then treated with 10% (w/v) PEG6000 solution for 1 week, while the control group continued to grow under normal conditions without the treatment of PEG6000. As the three indicators of drought stress, the contents of malondialdehyde (MDA), chlorophyll, and proline in leaves were measured to evaluate the drought tolerance of the soybean plants [29].

Generation of Transgenic Arabidopsis Plants
The function of enhancing drought tolerance by GmTGA15 was further evaluated in transgenic Arabidopsis plants with the heterologous overexpression of GmTGA15. The CDS sequence of GmTGA15 was amplified and cloned into the pCAMBIA3301 vector. The recombinant expression vector containing pCAMBIA3301-GmTGA15 was transformed into the Agrobacterium tumefaciens strain Agl0 and subsequently transformed into Arabidopsis ecotype Col-4 by the floral dip method [30]. The harvested seeds were screened by Basta tolerance, while the T3-transformed plants were confirmed by qRT-PCR analysis and used for further study based on the expression level of GmTGA15. Three independent transgenic lines of Arabidopsis were used to evaluate their drought tolerance. For the root growth analysis, 5-day-old seedlings initially grown in 1/2 MS medium were transferred to 1/2 MS medium containing 6% (m/v) PEG6000 for vertical growth under a photoperiod of 16-h with light of 120 µmol·m −2 ·s −1 and 8 h in dark at 22 • C. The lengths of roots were measured 1 week before and after the treatment of 6% PEG6000. The relative root elongation (RRE) was calculated as (root elongation in drought treatment/root elongation in control) × 100. For the growth under the drought stress in soil, 2-week-old seedlings were not watered until phenotypic differences on leaves were observed. Then, plants were watered and recovered for 1 week to calculate the survival rates under the drought-stress treatment.

Measurement of Chlorophyll, Proline, and Malondialdehyde (MDA)
The content of free proline in plants was determined following the methods as described previously with slight modifications [31]. Briefly, 0.2 g of plant sample were homogenized in 7 mL of 3% sulfosalicylic acid using a mortar. The 2-mL supernatant was transferred to a new tube with the extraction solution containing 4 mL of acid ninhydrin, 2 mL sulfosalicylic acid, and 2 mL of glacial acetic acid. The reaction mixture was boiled in a water bath at 100 • C for 1 h then stored at 4 • C for 30 min. Then, a total of 4 mL of toluene was added to the solution and shaken to extract the upper red product. The proline content for each sample was measured based on the absorbance at 520 nm. The content of MDA in plants was determined following the methods as described previously with minor modifications [32]. Specifically, 0.1 g leaf sample with 1 mL of 0.05 M phosphate buffer (pH 7.8) and a small amount of quartz sand added in a mortar was ground into a homogenate, which was poured into a 15-mL centrifuge tube, then the mortar was washed with 4 mL of phosphate buffer with the solution transferred to the centrifuge tube for centrifugation at 10,000× g and 4 • C for 20 min. The content of chlorophyll in plants was determined following the methods as described previously with slight modifications [33]. Briefly, 0.1 g soybean leaves were cut into 1-mm long sections with a razor blade then completely submerged under 2 mL of dimethyl sulfoxide for extraction of chlorophyll in a 65 • C water bath in the dark until leaves became whitish or transparent. Then, 8 mL 80% acetone was added to the cooled extraction solution. The chlorophyll content for each sample was measured based on absorbance at 520 nm. All experiments were repeated with three replicates.

Data Analysis
All data were presented as the mean of three biological replicates ± standard deviation (SD). One-way ANOVA tests using DPS 11.0 edition for windows were conducted to identify the significant difference between the treatments, with the p values set to 0.01 or 0.05 [34].

Expression Patterns of GmTGA15 under Abiotic Stresses and Hormone Treatments
To investigate the expression patterns of GmTGA15 in response to abiotic stresses and hormone treatments, quantitative real-time PCR (qRT-PCR) was performed to analyze the transcript abundance of GmTGA15 in soybean plants of cultivar Williams 82 ( Figure 1). To further analyze the expression patterns of GmTGA genes, the transcript levels of GmTGA15 were analyzed in six different tissues and organs of soybean plants at four different developmental stages ( Figure 2).
Various expression patterns of GmGAT15 in soybean hairy roots were observed under different abiotic stresses and hormone treatments. GmTGA15 was quickly induced by the treatment of 200 mM NaCl with the total RNA of GmTG15, increasing up to the highest level of~3.3-fold over 3 h in roots (in comparison to that of the control) before starting to gradually decline for the rest of the experiments over 24 h ( Figure 1A). Under the treatment by PEG6000, the expression level of GmTGA15 gene showed an overall upward trend, increasing up to the highest level of~7.2-fold over 24 h in roots compared to the control ( Figure 1B). Similar to the treatment of NaCl, under the treatment of 2 mM SA, the expression of GmTGA15 increased up to the highest level of~5.1-fold over 3 h in comparison to that of the control, then declined until the end of the experiment ( Figure  1C). The treatments of both 100-µM ABA ( Figure 1D) and 50-µM MeJA ( Figure 1E) stresses increased the expression levels of GmTGA15 in roots~2-fold in comparison to that of the control groups over 24 h, with varied levels of expression prior to the end of the experiments. Under the treatment of 200 µM GA, the expression of GmTGA15 increased quickly up to the highest level of~3.8-fold over 1 h compared to that of the control, then started to gradually decline to the level comparable with that of the control as it reached 24 h ( Figure 1F). a 65 °C water bath in the dark until leaves became whitish or transparent. Then, 8 mL 80% acetone was added to the cooled extraction solution. The chlorophyll content for each sample was measured based on absorbance at 520 nm. All experiments were repeated with three replicates.

Data Analysis
All data were presented as the mean of three biological replicates ± standard deviation (SD). One-way ANOVA tests using DPS 11.0 edition for windows were conducted to identify the significant difference between the treatments, with the p values set to 0.01 or 0.05 [34].

Expression Patterns of GmTGA15 under Abiotic Stresses and Hormone Treatments
To investigate the expression patterns of GmTGA15 in response to abiotic stresses and hormone treatments, quantitative real-time PCR (qRT-PCR) was performed to analyze the transcript abundance of GmTGA15 in soybean plants of cultivar Williams 82 (Figure 1). To further analyze the expression patterns of GmTGA genes, the transcript levels of GmTGA15 were analyzed in six different tissues and organs of soybean plants at four different developmental stages (Figure 2).   Figure 1A). Under the tr ment by PEG6000, the expression level of GmTGA15 gene showed an overall upw trend, increasing up to the highest level of ~7.2-fold over 24 h in roots compared to control ( Figure 1B). Similar to the treatment of NaCl, under the treatment of 2 mM SA expression of GmTGA15 increased up to the highest level of ~5.1-fold over 3 h in com ison to that of the control, then declined until the end of the experiment ( Figure 1C). treatments of both 100-µM ABA ( Figure 1D) and 50-µM MeJA ( Figure 1E) stresses creased the expression levels of GmTGA15 in roots ~2-fold in comparison to that of control groups over 24 h, with varied levels of expression prior to the end of the exp ments. Under the treatment of 200 µM GA, the expression of GmTGA15 increased qui up to the highest level of ~3.8-fold over 1 h compared to that of the control, then sta to gradually decline to the level comparable with that of the control as it reached ( Figure 1F).
The expression patterns of GmTGA15 were investigated in different tissues or org of the soybean plants, including the root, stem, leaf, flower, pod, and seed samples, at different developmental stages. The results of the qRT-PCR analysis showed that the pression levels in different tissues or organs during different developmental stages va greatly, indicating that GmTGA15 was expressed constitutively with different funct in different tissues or organs during different developmental stages in soybean (Figur Among the four developmental stages, GmTGA showed the highest expression level ing the seedling stage, followed by the podding stage ( Figure 2). In the seedling st roots showed the highest expression level among all examined organs, namely, ~68higher than those in either stems or leaves (Figure 2A). In the branching stage, the exp sion levels in both stems and leaves were similar and ~3-fold lower than that of roots ( ure 2B), while the expression level in flowers was ~6-fold higher than those in either st The expression patterns of GmTGA15 were investigated in different tissues or organs of the soybean plants, including the root, stem, leaf, flower, pod, and seed samples, at four different developmental stages. The results of the qRT-PCR analysis showed that the expression levels in different tissues or organs during different developmental stages varied greatly, indicating that GmTGA15 was expressed constitutively with different functions in different tissues or organs during different developmental stages in soybean ( Figure 2). Among the four developmental stages, GmTGA showed the highest expression level during the seedling stage, followed by the podding stage ( Figure 2). In the seedling stage, roots showed the highest expression level among all examined organs, namely,~68-fold higher than those in either stems or leaves (Figure 2A). In the branching stage, the expression levels in both stems and leaves were similar and~3-fold lower than that of roots ( Figure 2B), while the expression level in flowers was~6-fold higher than those in either stems or leaves in the flowering stage ( Figure 2C). In the podding stage, the expression levels in pods and seeds were~12-fold and~14-fold higher than that of stems, respectively, while the roots showed the highest expression level at~32-fold that of the stems ( Figure 2D).

GmTGA15 Improves Tolerance to Drought Stress in Transgenic Soybean Hairy Roots
The successful construction of the plant expression vector pCAMBIA3301-GmTGA15 and its transformation into soybean hairy roots was confirmed by PCR ( Figure S1). The functions of GmTGA15 under drought stress were further investigated by performing abiotic stress assays induced by PEG6000 in Agrobacterium rhizogenes-mediated soybean hairy roots. The successful amplifications of 35S promoter and the bar gene located on the vector pCAMBIA3301 were observed in the GmTGA15 transgenic soybean hairy roots but not in the Agrobacterium rhizogenes-transformed soybean hairy roots (Figure 3).
The overexpression of GmTGA15 was evaluated by the qRT-PCR analysis, revealing that the expression level of GmTGA15 was significantly higher (i.e., overexpressed) in three lines of soybean hairy roots with the overexpression of GmTGA15 by more than 4-fold compared to the transgenic Agrobacterium hairy roots (Figure 4). Due to their significantly increased expression levels of GmTGA15, these three lines of soybean plants were selected for further treatments of drought stress in order to study the change of proline, chlorophyll, Agronomy 2021, 11, 170 7 of 13 and MDA. The results showed that in all three lines, under the treatment of drought stress induced by 20% PEG6000, the contents of both proline and chlorophyll were significantly increased, while the MDA was significantly reduced in soybean hairy roots with the overexpression of GmTGA15 compared with the control (Figure 5).

GmTGA15 Improves Tolerance to Drought Stress in Transgenic Soybean Hair
The successful construction of the plant expression vector pCAMBIA330 and its transformation into soybean hairy roots was confirmed by PCR (Fi functions of GmTGA15 under drought stress were further investigated by pe otic stress assays induced by PEG6000 in Agrobacterium rhizogenes-mediated s roots. The successful amplifications of 35S promoter and the bar gene located pCAMBIA3301 were observed in the GmTGA15 transgenic soybean hairy ro the Agrobacterium rhizogenes-transformed soybean hairy roots (Figure 3). The overexpression of GmTGA15 was evaluated by the qRT-PCR analy that the expression level of GmTGA15 was significantly higher (i.e., over three lines of soybean hairy roots with the overexpression of GmTGA15 by fold compared to the transgenic Agrobacterium hairy roots (Figure 4). Due to cantly increased expression levels of GmTGA15, these three lines of soybea selected for further treatments of drought stress in order to study the chan chlorophyll, and MDA. The results showed that in all three lines, under the drought stress induced by 20% PEG6000, the contents of both proline an were significantly increased, while the MDA was significantly reduced in s roots with the overexpression of GmTGA15 compared with the control (Figu

Overexpression of GmTGA15 Improved the Drought Tolerance in Transgenic Arabidopsis Plants
Having demonstrated the function of enhancing drought tolerance in soybean, GmTGA15 was then overexpressed in Arabidopsis plants grown in both medium (containing 6% PEG6000 simulating drought-stress conditions) and soil to further confirm its roles in drought tolerance. Both the T3 transgenic Arabidopsis lines and wild type (WT) were sown in 1/2 MS medium with and without 6% PEG6000, respectively. In the absence of 6% PEG6000, the total root length and survival rates showed no significant difference between transgenic and WT plants of Arabidopsis ( Figure 6A,B). In contrast, when exposed to 1/2 MS medium with 6% PEG6000 for 8 days, the growth of transgenic and WT plants was strongly inhibited, while the transgenic Arabidopsis plants showed significantly longer root lengths than those of the WT plants. Under drought-stress conditions, the WT plants either shriveled severely or died in two weeks, while the transgenic Arabidopsis plants just started to shrivel slightly and showed significantly lower mortality than that of the WT plants ( Figure 6B,C). to 1/2 MS medium with 6% PEG6000 for 8 days, the growth of transgenic and WT plants was strongly inhibited, while the transgenic Arabidopsis plants showed significantly longer root lengths than those of the WT plants. Under drought-stress conditions, the WT plants either shriveled severely or died in two weeks, while the transgenic Arabidopsis plants just started to shrivel slightly and showed significantly lower mortality than that of the WT plants ( Figure 6B,C).

Discussion
Studies have shown that the cis-elements in the promoter region of a gene are closely related to the biological function of the gene [35]. Our previous analysis of the cis-acting elements showed that the promoter region of GmTGA15 contained the pathogen-related cis-elements such as ERE, W-box, P-box, or WUN-motif, indicating that GmTGA15 is involved in the defense of pathogens [24]. This observation is consistent with the involvement of TGA genes in response to pathogen attacks [36]. Furthermore, we identified the cis-elements related to stresses in the promoter region of GmTGA15 in response to

Discussion
Studies have shown that the cis-elements in the promoter region of a gene are closely related to the biological function of the gene [35]. Our previous analysis of the cis-acting elements showed that the promoter region of GmTGA15 contained the pathogen-related ciselements such as ERE, W-box, P-box, or WUN-motif, indicating that GmTGA15 is involved in the defense of pathogens [24]. This observation is consistent with the involvement of TGA genes in response to pathogen attacks [36]. Furthermore, we identified the cis-elements related to stresses in the promoter region of GmTGA15 in response to drought, salt, and hormones ABA, SA, GA, and MeJA. Therefore, it is speculated that GmTGA15 is involved in these responses.
It is well-known that environmental stresses such as salt, drought, high temperature, and oxidation affect the growth and development of plants, while TGA transcription factors participate in the response of plants to these stresses. Our results showed that GmTGA15 was induced to show varied expression patterns under different environmental stresses (i.e., drought and salt) and exogenous hormones (i.e., ABA, GA, SA, and MeJA), with the greatest response to drought, suggesting that GmTGA15 plays different roles under different abiotic stresses and hormones ( Figure 1). Furthermore, our results of the qRT-PCR analysis showed that the expression levels in different tissues or organs during different developmental stages varied greatly. For example, roots showed the highest expression level among all examined organs during the four developmental stages, and were~68-fold higher than those in either stems or leaves during the seedling stage (Figure 2A). These results were consistent with those reported recently [23], further suggesting the existence of the diverse tissue-or organ-specific regulatory mechanisms (Figure 2). Studies have extensively shown that plant bZIP transcription factors are induced and expressed by plant hormones such as SA, MeJA, ABA, and ethylene (ETH) [37][38][39][40]. Furthermore, it has been reported that different TGA transcription factors play different roles, showing varied specificities and responding differently to different adverse conditions. For example, Zhang et al. [41] showed that the wheat TabZIP1 gene is rapidly induced by exogenous MeJA and ETH, but not influenced by SA. Our results showed that exogenous SA and GA rapidly induce the expression of GmTGA15 at high levels, in contrast with the slow expression under the treatment of exogenous ABA and MeJA and indicating that GmTGA15 may participate in SA and GA signal pathways to regulate gene expression in response to the adverse stresses ( Figure 1).
Studies have shown that drought generates osmotic stress, leading to an increase in the amount of reactive oxygen species (ROSs), lipid peroxidation, protein oxidation, nucleic acid damage, and the activation of programmed cell death, while transgenic plants demonstrate enhanced drought tolerance by eliminating reactive oxygen species [42]. In order to study the molecular mechanisms of enhanced drought tolerance by GmTGA15 in genetically modified soybean hairy roots, the plant expression vector pCAMBIA3301-GmTGA15 was constructed and then transformed into soybean hairy roots by Agrobacterium rhizogenes-mediated methods. The GmTGA15 transgenic soybean hairy roots were treated under drought stress induced by 20% PEG6000, with the contents of malondialdehyde (MDA), proline, and chlorophyll evaluated ( Figure 5). As an indicator of the destruction of ROSs under stress conditions, the contents of MDA in transgenic plants were significantly reduced under salt and osmotic stresses [43]. As an osmotic substance, the accumulation of a large amount of proline plays an important role in stabilizing the macromolecules and cell membranes in transgenic plants under salt and osmotic stresses [44,45]. Furthermore, as an important functional substance absorbing and transmitting light energy in plant photosynthesis, the level of chlorophyll content, which is changed by drought stress, directly affects the performance and yields of photosynthesis in plants [46]. For example, studies have shown that chlorophyll content was decreased significantly by an elevated degree of drought stress [47]. Our results showed that the contents of both proline and chlorophyll were increased, while the contents of malondialdehyde was decreased, in the soybean hairy roots with an overexpression of GmTGA15. These results indicate that GmTGA15 increases the drought tolerance of soybean hairy roots by elevating the contents of proline and chlorophyll and lowering the contents of malondialdehyde. These results further indicate that GmTGA15 is probably involved in the regulation of the synthesis of proline, chlorophyll, and MDA. In our study, enhanced drought tolerance by GmTGA15 was further confirmed by its heterologous overexpression in transgenic Arabidopsis plants. Specifically, transgenic Arabidopsis plants showed significantly longer root lengths and lower mortality under drought-stress conditions than those of WT plants. These results are consistent with those reported previously showing enhanced tolerance to drought stress in both transgenic Arabidopsis plants and soybean hairy roots by the transcription factor GmTGA17 [23]. We note that due to their shared molecular similarities and functions in enhancing drought tolerance in soybean and Arabidopsis plants, it is speculated that there might be a synergistic effect between GmTGA15 and GmTGA17. Further studies are necessary to verify this speculation.

Conclusions
Our investigation of the overexpression of the transcription factor GmTGA15 in response to drought stress in transgenic soybean hairy roots and Arabidopsis plants has demonstrated that both abiotic stresses (drought and saline) and hormone treatments (SA, ABA, MeJA, and GA) greatly induced the expression of GmTGA15 in soybean plants. Results showed that the contents of both chlorophyll and proline were significantly increased, while the content of malondialdehyde (MDA) was significantly decreased in the soybean hairy roots with the overexpression of GmTGA15 in comparison to wild type in response to drought stress. Furthermore, the transgenic Arabidopsis plants with an overexpression of GmTGA15 showed significantly advantageous phenotypes (i.e., longer roots and lower mortality) compared to that of the wild type under the simulated drought conditions, suggesting that GmTGA15 promotes tolerance to drought stress in both soybean and Arabidopsis plants. Due to the severe deterioration caused by the various environmental factors (e.g., drought and soil salinization), it is important to breed drought-resistance crops. This study provides the scientific evidence for further functional analysis of soybean TGA transcription factors in the drought stress and breeding of drought-resistance crops.