Overexpression of BplERD15 Enhances Drought Tolerance in Betula platyphylla Suk.

: In this study, we report the cloning and functional characterization of an early responsive gene, BplERD15 , from Betula platyphylla Suk to dehydration. BplERD15 is located in the same branch as Morus indica Linnaeus ERD15 and Arabidopsis Heynh ERD15 in the phylogenetic tree built with ERD family protein sequences. The tissue-speciﬁc expression patterns of BplERD15 were characterized using qRT-PCR and the results showed that the transcript levels of BplERD15 in six tissues were ranked from the highest to the lowest levels as the following: mature leaves (ML) > young leaves (YL) > roots (R) > buds (B) > young stems (YS) > mature stems (MS). Multiple drought experiments were simulated by adding various osmotica including polyethylene glycol, mannitol, and NaCl to the growth media to decrease their water potentials, and the results showed that the expression of BplERD15 could be induced to 12, 9, and 10 folds, respectively, within a 48 h period. However, the expression level of BplERD15 was inhibited by the plant hormone abscisic acid in the early response and then restored to the level of control. The BplERD15 overexpression (OE) transgenic birch lines were developed and they did not exhibit any phenotypic anomalies and growth deﬁciency under normal condition. Under drought condition, BplERD15-OE1 , 3 , and 4 all displayed some drought tolerant characteristics and survived from the drought while the wild type (WT) plants withered and then died. Analysis showed that all BplERD15-OE lines had signiﬁcant lower electrolyte leakage levels as compared to WT. Our study suggests that BplERD15 is a drought-responsive gene that can reduce mortality under stress condition.


Introduction
Drought stress is a severe environmental condition where plants are subjected to dehydration, resulting in loss in plant biomass productivity [1]. Due to wide-spreading and high-frequent occurrence, the loss caused by drought in crop yield is usually so high that it may exceed losses caused by all other environmental factors together [2,3]. When a drought occurs, various cellular signals are perceived and then conveyed through multiple pathways, for example, ionic and osmotic steady-state signaling pathways, damage control and repair response pathways, and growth regulation pathways [4]. Through these pathways, a series of physiological and biochemical reactions are activated or enhanced to produce gene products and various metabolites that can repair or prevent damages of cellular apparatuses, resulting in the survival in drought condition. In this process, the products of the drought-inducible genes can be largely classified into two categories: (1) stress tolerance proteins, which include chaperones, late embryogenesis abundant (LEA) proteins, osmotins, key enzymes for osmolyte biosynthesis, water channel proteins, and proline transporters, as well as detoxification several other ERD15 genes from other species, which include SpERD15 from S. pennellii [14], MiERD15 from Morus indica [29], GmERD15 from Glycine max [30] and VaERD15 from Vitis amurensis Rupr [15]. Following that, we built these genes into a phylogenetic tree using the neighbor-joining statistical method and the Poisson model in Mega X software with 1000 of bootstrap replications.

Cloning and Tissue-Specific Expression of BplERD15
The analysis of tissue-specific expression patterns of BplERD15 was performed using qRT-PCR. The samples were collected from multiple tissues including buds, young leaves, mature leaves, young stems, mature stems, and roots of B. platyphylla and frozen immediately into liquid nitrogen. The leaves from the first to third stem nodes were referred to as young leaves (YL), while the leaves of the fourth to sixth stem nodes were referred to as mature leaves (ML). Accordingly, the stems of the first to third stem nodes were referred to as young stems (YS), and the fourth to sixth stem nodes were referred to as mature stems (MS). Cetyltrimethylammonium bromide (CTAB)-based protocol [31] was used to extract RNA, which was reversely transcribed into cDNA. The cDNA acquired was then used for qRT-PCR with the Toyo Spinning Kit (TOYOBO SYBR qPCR Mix, QPS-201). The amplification conditions were as follows: 95 • C for 30 s, which was followed by 40 cycles of 95 • C for 15 s and 60 • C for 45 s, finally 95 • C for 15 s, 60 • C for 60 s, 95 • C for 30 s. Ubiquitin gene was chosen to be the internal reference. There were three biological replicates.

Analysis of the Expression of BplERD15 in Wild-Type B. Platyphylla
Drought experiments were simulated by adding various osmotica including polyethylene glycol (PEG) [32], mannitol, and NaCl to the growth media to decrease their water potentials. Two-month-old wild-type birch seedlings were irrigated with solutions containing 20% PEG6000, 200 mM Mannitol, or 200 mM NaCl. The aforementioned tissues (Roots, YS, MS, YL ML and Buds) were harvested at six time points: 0 h, 3 h, 6 h, 12 h, 24 h, and 48 h from the seedlings subject to different treatments. In order to test if BplERD15 was responsive to ABA, two-month-old birch seedlings were sprayed with a 100 µM solution of ABA and incubated for 0 h, 1 h, 3 h, 6 h, 12 h, 24 h, and 48 h. The materials we harvested were immediately frozen into liquid nitrogen. RNA was extracted from these samples and used for qRT-PCR to obtain the mRNA abundances of BplERD15 in different tissues in three biological replicates. Student's t-test was used to test the significance of the difference between a treatment and wild-type.

Plant Transformation
We designed primers with adaptors that contain specific restriction sites, and used birch cDNA as a template for PCR amplification of BplERD15; the PCR products were cloned into the binary vector called pROK2 upon a double-enzyme digestion of PCR products and vector sequence. The binary vector harboring BplERD15 was then transformed into Agrobacterium strain EHA105 by the freeze-thaw method [33]. The B. platyphylla transgenic lines were developed by the leaf disc method [22] with minor changes. First the transformed Agrobacterium stain EHA105 was cultured at 28 • C for overnight until the OD fell into the range 0.6-0.8. The vigorous birch leaves from cultured B. platyphylla plants were cut and soaked in the bacterium culture for 8 to 10 min. Then, the leaves were taken out and placed on a sterile paper to allow the excessive culture to be absorbed. The leaves were then transferred onto the culture plates containing the WPM medium with 0.8 mg/L 6-BA + 0.02 mg/L NAA + 2% (w/v) sucrose, pH 5.8-6.0. The leaves were cultured in the dark for three days before they were transferred onto the culture with WPM media containing kanamycin (50 mg/L) and timentin (400 mg/L). Calli were first seen in about two months. When seedlings grew to about 1 cm high, they were cut into segments, each inserted into tissue culture bottles containing 1/2 MS+ 0.02 mg/L NAA + 2% (w/v) sucrose + 400 mg/L timentin + 50 mg/L kanamycin; pH 5.8-6.0. When the seedlings grew large, DNA was extracted with Tiangen DNA extraction kit (TIANGEN, Beijing, China). The transformants were examined with PCR and transgene-specific primers. The expression levels of BplERD15 in different transgenic birch lines were analyzed by qRT-PCR. The primer sequences for PCR and qTR-PCR are shown in Table S1.

Drought Tolerance Assays of BplERD15 Overexpression Transgenic Lines
Three-month-old B. platyphylla transgenic lines were grown in a greenhouse under 16 h light/8 h dark and 25 • C. Before the drought experiment was performed, all plants were fully irrigated. After 15 days, the plants were subjected to dehydration. The photos were taken three days later after the rehydration we initiated.

Measurement of Electrolyte Leakage
Three-month-old transgenic lines with the highest expression of BplERD15 were selected and subjected to drought stress for 10 d together with WT plants. The leaves were harvested and used for measuring electrolyte leakage as described earlier [34]. Briefly, the equal sections from the leaf of each sample were harvested and placed into a clean beaker; 30 mL of deionized water was added and left under vacuum for 15 min. The electrolyte leakage was measured and defined as S1. The leaves were then heated to 90 • C and kept for 20 min before they were cooled down to room temperature. The electrolyte leakage was measured again and defined to be S2. The electrolyte leakage (EL) was calculated with the formula: EL = (S1/S2) × 100%.

Statistical Analysis
The Student's t-test was used to examine the differences between transgenic lines and WT plants, and the difference before and after stress treatment. The threshold for statistically significant differences was set to p < 0.05.

ERD Phylogenetic Analysis
The ORF (open reading frame) of BplERD15 is 480 bp long and thus encodes a protein with 159 amino acids (Figure 1a). With this protein sequence, we used 16 ERD protein sequences from A. thaliana, and several other ERD15 protein sequences from other plant species. We then built a phylogenetic tree ( Figure 1b). We found that BplERD15, BpeERD15, and MiERD15 had the closest distance and were clustered together. The multiple alignment analysis ( Figure 1c) showed that BplERD15 shared 100% and 50.56% similarity to BpeERD15 from B. pendula and MiERD155 from M. indica, respectively. In addition, BplERD15 shared 45.98% and 44.31% similarity with SpERD15 from S. pennellii and AtERD15 from A. thaliana, respectively.

Tissue-Specificity and Drought Stress Response of BplERD15 in WT Plants
The analysis of tissue-specific expression patterns of BplERD15 was performed using qRT-PCR, and the results are shown in Figure 2a. The transcript levels of BplERD15 in six tissues are ranked from the highest to the lowest in the following sequence: mature leaves (ML) > young leaves (YL) > roots (R) > buds (B) > young stems (YS) > mature stems (MS). In addition, the WT plants were subjected to drought treatment, and the results are shown in Figure 2b-d. It is obvious that under different drought stress conditions, BplERD15 positively and differentially responded to PEG, Mannitol, and NaCl stresses. BplERD15 transcript level was progressively enhanced by PEG treatment, with a slight decrease at the 24 h, and eventually reached its maximal level (12 folds) at the 48 h (Figure 2b).
Under mannitol stress, BplERD15 transcript level was steadily up-regulated from 0 to 12 h period, and increased up to 9-fold as compared to 0 h (Figure 2c), and after that, it declined all the way to 48 h. Under salt stress, BplERD15 transcript level peaked at 3 h where it had a more than 9-fold increase (Figure 2d). Though the transcript level of BplERD15 started to decrease after 3 h, it remains higher than that of 0 h. We also applied ABA treatment. At the first time, the expression level of BplERD15 was significantly down-regulated at 1 h and started to increased but was still significantly down-regulated at 3 h ( Figure 2e). In subsequent time points, the expression level of BplERD15 was not significant compared to that of wild-type.

Tissue-Specificity and Drought Stress Response of BplERD15 in WT Plants
The analysis of tissue-specific expression patterns of BplERD15 was performed using qRT-PCR, and the results are shown in Figure 2a. The transcript levels of BplERD15 in six tissues are ranked from the highest to the lowest in the following sequence: mature leaves (ML) > young leaves (YL) > roots (R) > buds (B) > young stems (YS) > mature stems (MS). In addition, the WT plants were subjected to drought treatment, and the results are shown in Figure 2b-d. It is obvious that under different drought stress conditions, BplERD15 positively and differentially responded to PEG, Mannitol, and NaCl stresses. BplERD15 transcript level was progressively enhanced by PEG treatment, with a slight decrease at the 24 h, and eventually reached its maximal level (12 folds) at the 48 h (Figure 2b).
Under mannitol stress, BplERD15 transcript level was steadily up-regulated from 0 to 12 h period, and increased up to 9-fold as compared to 0 h (Figure 2c), and after that, it declined all the way to 48 h. Under salt stress, BplERD15 transcript level peaked at 3 h where it had a more than 9-fold increase (Figure 2d). Though the transcript level of BplERD15 started to decrease after 3 h, it remains higher than that of 0 h. We also applied ABA treatment. At the first time, the expression level of BplERD15 was significantly down-regulated at 1 h and started to increased but was still significantly down-regulated at 3 h (Figure 2e). In subsequent time points, the expression level of BplERD15 was not significant compared to that of wild-type.

Overexpression of BplERD15 in Transgenic Birch Lines
In this study, five independent overexpression lines of BplERD15 were generated and validated by PCR (Figure 3a) and the expression levels of BplERD15 in these lines were quantified with qRT-PCR (Figure 3b). It is obvious that the expression levels of BplERD15 in OE1, OE3, OE4, and OE5 were significantly higher than that of wild-type. The expression level of BplERD15 in OE2 was not significantly different from that of wild-type. For all analyses conducted hereafter, we used three transgenic lines, OE1, OE3, and OE4, which had the highest expression.

Overexpression of BplERD15 in Transgenic Birch Lines
In this study, five independent overexpression lines of BplERD15 were generated and validated by PCR (Figure 3a) and the expression levels of BplERD15 in these lines were quantified with qRT-PCR ( Figure 3b). It is obvious that the expression levels of BplERD15 in OE1, OE3, OE4, and OE5 were significantly higher than that of wild-type. The expression level of BplERD15 in OE2 was not significantly different from that of wild-type. For all analyses conducted hereafter, we used three transgenic lines, OE1, OE3, and OE4, which had the highest expression.

Overexpression of BplERD15 Confers Enhanced Drought Tolerance
The transgenic lines of BplERD15 and WT plants were grown in a greenhouse until they were three-months old, which is when they were used for the drought stress experiment. Both transgenic lines and WT were well irrigated before the drought experiment was initiated. The transgenic lines were then subjected to dehydration with a duration of 15 days. The plants were fully rehydrated for three days, and they were photographed, as shown in Figure 4a. It is obvious that the wild-type plants showed a severe wilting symptom while all three BplERD15-OE lines survived from the extended drought treatment.
We measured the electrolyte leakage in the leaves of all three BplERD15 transgenic lines with WT plants as comparison. It was found that the transgenic lines had significantly lower electrolyte leakage than WT plants (Figure 4b).

Overexpression of BplERD15 Confers Enhanced Drought Tolerance
The transgenic lines of BplERD15 and WT plants were grown in a greenhouse until they were three-months old, which is when they were used for the drought stress experiment. Both transgenic lines and WT were well irrigated before the drought experiment was initiated. The transgenic lines were then subjected to dehydration with a duration of 15 days. The plants were fully rehydrated for three days, and they were photographed, as shown in Figure 4a. It is obvious that the wild-type plants showed a severe wilting symptom while all three BplERD15-OE lines survived from the extended drought treatment.
We measured the electrolyte leakage in the leaves of all three BplERD15 transgenic lines with WT plants as comparison. It was found that the transgenic lines had significantly lower electrolyte leakage than WT plants (Figure 4b).
three days, and they were photographed, as shown in Figure 4a. It is obvious that the wild-type plants showed a severe wilting symptom while all three BplERD15-OE lines survived from the extended drought treatment.
We measured the electrolyte leakage in the leaves of all three BplERD15 transgenic lines with WT plants as comparison. It was found that the transgenic lines had significantly lower electrolyte leakage than WT plants (Figure 4b).

Discussion
Several studies have shown that ERD genes play important roles in various abiotic stresses that include but are not limited to salt [35], drought [36], and freezing [13], as well as protein metabolic processes [37]. For example, ZmERD3 gene expression is induced by abiotic stress treatments (such as PEG, NaCl, ABA, and low temperature) [35]. Owing to the inhibition of ABA signaling, the

Discussions
Several studies have shown that ERD genes play important roles in various abiotic stresses that include but are not limited to salt [35], drought [36], and freezing [13], as well as protein metabolic processes [37]. For example, ZmERD3 gene expression is induced by abiotic stress treatments (such as PEG, NaCl, ABA, and low temperature) [35]. Owing to the inhibition of ABA signaling, the overexpression of ERD15 in Arabidopsis leads to reduced tolerance to drought stress [36]. Compared with wild-type plants, the ERD10 mutant has reduced tolerance to cold stress [13]. ERD1, also referred to as ClpD, is an ATP-dependent chaperone. ERD1 functions as a component in the plant plastid Clp machinery, which comprises a hetero-oligomeric ClpPRT proteolytic core, ATP-dependent chaperones ClpC and ClpD, and an adaptor protein, and plays crucial roles in maintaining protein homeostasis [37]. In this study, we constructed a phylogenetic tree with the BplERD15 and BpeERD15 protein sequences of B. pendula, together with 16 ERD protein sequences of A. thaliana and four other ERD15 protein sequences from other plant species. The distances among BplERD15, BpeERD15, and MiERD15 were the shortest. A previous study has shown that MiERD15 can be induced by drought stress, ABA treatment, salinity, and temperature extremes [29], which indicates that BplERD15 may be an effector of multiple stresses and ABA too. In our study, BplERD15 was found to be a positive regulator of drought, but its expression was induced by several osmotica that include PEG, mannitol, and NaCl. Surprisingly, the expression level of BplERD15 was inhibited under ABA treatment. AtERD15 is a negative regulator of abscisic acid responses in A. thaliana [38]. An overexpression of AtERD15 reduces ABA sensitivity and drought tolerance in A. thaliana. The wild S. pennellii (SpERD15) was most closely related to AtERD15 (Figure 1b). Transgenic lines overexpressing SpERD15 manifested stress tolerance to dehydration, salinity, and cold. They exhibited an accumulation of soluble sugars and proline, and a limited lipid peroxidation [14]. Overexpression transgenic lines of VaERD15 from Chinese wild V. amurensis showed robust cold tolerance [15]. The ERD15 from sweet potato (Ipomoea batatas (L.) Lam.), IbERD15, has been reported to play an important role in the response to drought stress [39].
Drought stress affects phenotypical traits such as plant height, root length, leaf area, plant biomass, and root stomata area [40]. In addition, drought stress can result in considerable structural alterations in mitochondria, chloroplast, and vacuole [41]. Plants usually survive drought stress through a series of physiological [42], cellular [41], and molecular adaptation mechanisms [5,43]. The physiological adaptation is usually accompanied with significant changes in oxidative and antioxidant metabolism, and an escalation of proline content and scavenging capacity of reactive oxygen species (ROS) through transgenic approach always leads to augmented stress tolerance [44,45]. Sometimes stress can increase the levels of some metabolites such as glucose, proline, and corilagin [46]. As reported, chloroplastically localized Os3BGlu6 significantly affects cellular ABA pools, which changes drought tolerance in rice [47]. Since the expression level of BplERD15 in the leaves was the highest, we speculate that it may contribute to the accumulation of soluble osmotic compounds and limit membrane peroxidation to improve the drought stress tolerance. In addition, plants under drought and salt stress share some common signaling transduction pathways [48], indicating the existence of common effector genes in response to both stresses [49,50]. BplERD15 may be such a gene because it could be induced by both osmotica and salt (Figure 2c,d), implying that it may be located downstream of a common signaling transduction pathway [51]. The AtERD15 in A. thaliana has been recently reported to be a negative regulator of ABA but it was induced by ABA and salicylic acid (SA), as well as by wounding and pathogenic infection [38]. ABA plays an important role in the drought stress and the plants that are subjected to drought release a large amount of ABA [52]. ABA reduces the stomatal conductance and alter many physiological processes, resulting in a progressive decrease of the net photosynthetic rate (Pn) and stomatal conductance (G s ) under drought stress. Application of ABA enhances the expression of some members of the same ERD group (ERD10 and 14) [12] but have no effect on others (ERD2, 8, and 16) [10].
In response to dehydration, significant physiological changes such as electrolyte leakage can occur [53]. Overexpression of some drought stress tolerance genes can counteract such a change. For example, overexpression of BpERF2 or BpMYB102 in birch significantly reduced the electrolyte leakage, and thereby increased the tolerance to drought stress [20]. We found that the electrolyte leakages were all significantly lower in the three transgenic lines overexpressing BplERD15 than in wild-type (Figure 4b), suggesting that the BplERD15 gene plays a determining role in the greater survival rates of transgenic lines under drought stress treatment.

Conclusions
BplERD15 is a positive regulator of drought stress response and tolerance. Tissue-specific expression analysis indicates that it has the highest expression level in mature leaves and the second highest expression in young leaves. Transgenic birch lines overexpressing BplERD15 showed significantly improved drought tolerance. BplERD15 could be induced by other osmotica, suggesting that it could be used as a wide-spectrum regulator for enhancing stress tolerance in transgenic plants. This study provides some functional basis of BplERD15 and we believe it is instrumental for genetic engineering of plants for enhanced stress tolerance to both drought and other abiotic stresses. Our findings indicate that BplERD15 is a common effector to multiple osmotica in addition to drought and thus future research should focus on characterizing its upstream signal pathways so that we could use it precisely in fighting for various stresses.