Over-Expression of ERF38 Gene Enhances Salt and Osmotic Tolerance in Transgenic Poplar

Ethylene response factor (ERF) gene family plays an important role in abiotic stress responses. In this study, we isolated a salt-inducible ERF gene, ERF38 (Potri.006G138900.1), from the 84K poplar (Populus alba × Populus glandulosa) and investigated its functions in salt and osmotic tolerance. We identified that ERF38 protein was targeted to nucleus and had no self-activation. Results from yeast-one-hybrid indicated that the ERF38 protein can specifically bind to the dehydration responsive element (DRE). We then successfully transferred the ERF38 gene into the 84K poplar. Under respective salt and polyethylene glycol (PEG)-6000 stresses, four of the physiological traits, including peroxidase (POD) and superoxide dismutase (SOD) activities, soluble protein content, and proline content, increased significantly in the transgenic plants, compared to the wild type. Regarding the other two parameters, hydrogen peroxide (H2O2) and malondialdehyde (MDA) content, their increments in the transgenic lines under the stresses, which were compared to the water control, were significantly low than that of the wild type. In addition, reactive oxygen species (ROS) are scavenged in the transgenic lines under the stresses, but not in the wild type (WT). Interestingly, when challenged with the stresses, expression levels of a few genes associated with POD and SOD metabolism were significantly increased in the transgenic poplars. In all, evidence from morphological, physiological, and biochemical analyses indicated that over-expression of ERF38 gene can improve salt and osmotic tolerance in the transgenic poplar.


INTRODUCTION
Abiotic stresses, such as high salt, drought, and low temperature, seriously affect plant growth and development. Upon environmental stresses, plants sense the changes and transmit corresponding signal pathways, which result in activating or repressing expression of related genes. Expression of these genes can lead to a series of morphological, physiological, and biochemical reactions in response to changes of the stresses. For instance, plants can regulate the closure of the stomata to control the loss of water, thus alleviating salt or drought damage (He et al., 2018;Zhang et al., 2019). Furthermore, antioxidant enzymes in plants, including superoxide dismutase (SOD), peroxidase (POD), glutathione reductase (GR), will increase correspondingly, in order to reduce the contents of reactive oxygen species (ROS) that are induced by abiotic stresses (Scandalios, 2005;Yu et al., 2017;Zhu et al., 2019).
Transcription factors (TFs) impact plant stress responses significantly. TFs can regulate expression of their target genes by binding to specific cis-elements of the genes, in order to respond to the abiotic stresses (Singh et al., 2002;Chen and Zhu, 2004). Many studies indicated that through various mechanisms over-expression of several TF genes can enhance tolerance to abiotic stress. For example, over-expression of tomato SlNAC35 gene in transgenic tobacco increase root length, which leads to enhanced salt and drought tolerance . Over-expression of OsWR1 gene can increase wax content and enhance drought resistance of transgenic rice by reducing water loss (Wang et al., 2012). In addition, Zhang and colleagues found that over-expression of LeERF2/ TERF2 gene in tobacco and tomato can enhance the expression of ethylene synthesis gene, thereby activating the expression of some low-temperature related genes and enhancing the cold resistance of plants (Zhang and Huang, 2010).
As one of the largest plant-specific TFs, the ethylene response factor (ERF) family plays an important role in plant growth and development (Li et al., 2018a), lignification (Ma et al., 2017), seed germination , and plant defense (Onate-Sanchez et al., 2007). Moreover, many proteins from the ERF family are able to recognize specific deoxyribonucleic acid (DNA) motif sequences, such as the dehydration responsive element (DRE) motifs (GCCGAC) and the GCC-box (AGCCGCC). For instance, the protein encoded by gmERF3 gene can bind to the GCC-box and the DRE/C-repeat motifs, in order to regulate target gene expression (Zhang et al., 2009). In transgenic Arabidopsis, the protein encoded by MintMaRAP2-4 gene can regulate water logging by binding to various regulatory elements, such as the DRE motif, the jasmonic acid response element (GATGAATT) motif, and the GCC-box (Phukan et al., 2018). Furthermore, ERF genes play an indispensable role in plant responses to abiotic stresses. For instance, over-expression of tomato SlERF84 gene in Arabidopsis can increase salt tolerance (Li et al., 2018b). In wheat, the TaERF3 gene can increase salt and drought tolerance of transgenic wheat (Rong et al., 2014). Our previous studies indicated that transgenic poplars over-expressing ERF76 gene are superior to wild type (WT) in morphological and physiological traits under salt stress (Yao et al., 2016).
The poplar variety 84K is a hybrid strain from the cross of Populus alba and Populus glandulosa. In our previous studies, we found that poplar ERF38 gene had a high expression under salt stress (Yao et al., 2018). Thus we speculated that the poplar ERF38 gene play an important role in the response to abiotic stresses. We then successfully cloned the gene and genetically transformed it into the 84K poplar. Compared to the WT, the transgenic plants showed better physiological and morphological traits, and displayed superior tolerance to salt and osmotic stresses.

Poplar Materials and Growth Conditions
One-month-old poplar WT seedlings from 84K variety were grown with hydroponic culture for 30 days, without nutrients. Poplar seedlings with new leaves and roots were treated with 150 mM sodium chloride (NaCl) and 20% PEG-6000, respectively. Samples from young leaves, stems, and roots were collected at each time point of 0, 3, 6, 12, 24, and 48 h, with three biological replicates. Then the samples were frozen in liquid nitrogen and stored at -80°C in refrigerator.
Sterile poplar seedlings were grown on 1/2 Murashige and Skoog medium (MS), followed by rooting medium containing 0.01 mg/ml 1-naphthaleneacetic acid (NAA) and 0.1 mg/ml indole-3-butyric acid (IBA). Poplar leaves from 1-month-old seedlings were cut into 1 cm x 1 cm leaf disks and place on MS differentiation medium containing 0.1 mg/ml NAA and 0.04 mg/ml thidiazuron (TDZ) for differentiation. All materials were grown in a greenhouse at 25°C with 16/8-h light/dark cycles.

Phylogenetic analysis of Ethylene Response Factor Genes
Protein sequences of the ERF genes from nine different species were obtained from the National Center for Biotechnology Information database (http://www.ncbi.nlm.nih.gov/). Multiple sequence alignment was performed by use of clustalW (Thompson et al., 1994). Phylogenetic tree analysis was conducted by MEGA 6 (Tamura et al., 2013), using neighbor-joining method.

Cloning of ERF38 Gene
The coding sequence of ERF38 gene (Potri.006G138900.1) from Populus trichocarpa was employed to design primers. The complementary deoxyribonucleic acid (cDNA) of ERF38 gene was inserted into the pBI121 binary vector with XbaI and SacI restriction sites. The recombination vector was then transferred into Agrobacterium GV3101 by the freezing and thawing method (Holsters et al., 1978), followed by infection and transformation with Agrobacterium tumefaciens.

Transcriptional activation and Yeast Onehybrid assays
Transcriptional activation of the ERF38 protein was tested. Primers with EcoRI and BamHI restriction sites were designed to insert the full-length coding sequence of ERF38 gene into pGBKT7 to form pGBKT7-ERF38 fusion vector containing GAL4-DNA domain. The pGBKT7-ERF38 fusion vector, pGBKT7 vector (negative control), and pGBKT7-53/pGADT7-T (positive control) were transferred into yeast two-hybrid (Y 2 H) yeast cells, respectively. Then they were cultured on the SD/-Trp or SD/-Trp/-His/X-a-Gal medium at 30°C for 3-5 days.
The cDNA fragment of ERF38 was inserted into pGADT7 vector to form pGADT7-ERF38. The three tandem repeats of the DRE (GCCGAC) motif, and two mutant elements, M1 (GTCGGC) and M2 (GTTTAC), were combined to pAbAi to form bait reporter vectors, respectively. Respective reporters and p53-AbAi (negative control) vectors were transferred to yeast-one-hybrid yeast strain, followed by culture on SD/-Ura medium for 3-5 days. Positive clones were selected by smallscale transformation in the yeast transformation system. The transformations were diluted to 10, 100, and 1,000 times, respectively; and then spread on SD/-Leu and SD/-Leu/ AbA plates.

Subcellular localization
The coding region of the ERF38 cDNA without stop codon was inserted into pBI121 vector containing green fluorescent protein (GFP), and driven by the 35S promoter, i order to form a 35S::ERF38::GFP fusion vector. The empty vector was used as a control (35S::GFP). Respective fusion vectors and 35S::GFP were transiently transformed into epithelial cells of the onions by particle bombardment. And the signal of GFP was observed under a laser confocal scanning microscope (LSM 700, Zeiss, Germany). Excitation wavelength used in 488 nm for GFP, and the wavelength range of captured light at 515-555 nm.

Generation of Transgenic Poplar
The transgenic poplar was obtained by leaf disk method (Nehra et al., 1990;Nietsch et al., 2017). Poplar leaves from 1-monthold seedlings were cut to 1 cm x 1 cm leaf disks and put on MS differentiation medium containing 50 mg/L kanamycin and 200 mg/L cephalosporin. The seedlings were then grown on rooting medium containing 50 mg/L kanamycin. The specific primers were designed to detect the positive transgenic poplars by both polymerase chain reaction (PCR) and reverse transcriptase PCR (RT-PCR).

Morphological and Physiological Measurements
The transgenic poplars were cultured for one month in rooting medium containing 0, 35, 50, 100 mM NaCl, respectively. Plant height, root length, and fresh weight of the plants at 1-month-old were measured with three biological replicates.
The seedlings at 1-month-old were transplanted into the soil in a greenhouse for another month, then they were treated with 150 mM NaCl (5 days) and 20% PEG-6000 (5 days), respectively. The morphological traits of both transgenic poplar and WT seedlings under normal and stress conditions were observed. Physiological parameters were measured, such as contents of proline, SOD, POD, H 2 O 2 , malondialdehyde (MDA), and soluble protein of transgenic poplar and WT (Yao et al., 2016;Wang et al., 2017). Each sample had three biological replicates.

histochemical Staining and Gene Expression analysis
Both 3,3′-diaminobenzidine (DAB) and nitrotetrazolium blue chloride (NBT) staining were used to detect the activities of hydrogen peroxide and superoxide in plants ( Kumar et al., 2013). Quantitative real-time PCR (RT-qPCR) was used to detect the relative expression levels of POD and SOD-related genes in transgenic plants under salt and PEG-6000 stresses (Zhang et al., 2019).

Reverse Transcriptase Quantitative Polymerase Chain Reaction
Total ribonucleic acid (RNA) from 84K seedlings was extracted by the RNA Isolation Kit (TAKALA, Dalian), and was reversely transcribed into cDNA using the reverse transcription kit (TAKALA, Dalian). The cDNA concentration of each sample was uniformly adjusted to the 100 ng/µl. RT-qPCR was performed using the Power SYBR Green Master mix method of ABI7500 Real Time System. The procedures include 95.0°C for 30 s, 95.0°C for 5 s, 60.0°C for 34 s, 95.0°C for 15 s, 60.0°C for 1 min, and 95.0°C for 15 s. The data was processed using 2 -∆∆Ct method (Dong et al., 2015). Each sample had three biological replicates. All primer sequences used in this study were shown in Supplementary Table S1 and S2.

RESUlTS alignment and Phylogenetic Tree analysis of Ethylene Response Factor Proteins
Comparisons of protein sequences from different species can help us to understand their phylogenetic relationships. The 465bp cDNA fragment of ERF38 gene was cloned from 84K poplar. Sequence similarity of ERF38 with the proteins from Populus trechocarpa, Populus euphratica, Vernicia montana, Manihot esculenta, Ricinus communis was 97.14, 94.16, 63.35, 65.10, and 61.33%, respectively. Evidence from multiple sequences alignment indicated that all proteins shared a conserved domain named as AP2 ( Figure 1A). In addition, phylogenetic tree analysis demonstrated that the ERF38 protein from 84K poplar was closely related to the counterparts from P. trechocarpa and P. euphratica, but had a distant genetic relationship with the proteins from Corchorus capsularis, Theobroma cacao, and M. esculenta ( Figure 1B).

Tempo-Spatial Expression Pattern of ERF38 Gene
Evidence from gene expression analysis indicated that ERF38 gene displayed a tempo-spatial pattern that is similar to NaCl and PEG stresses (Figure 2). It is clear that ERF38 gene is mainly expressed in leaves and stems, but relative lowly expressed in roots. And the relative expression level of ERF38 gene reached a peak at 12 h, followed by a decrease at 24 h (Figure 2).

ERF38 Protein Was localized to the Nucleus
As shown in Figure 3, the fluorescence signal of 35S::ERF38::GFP only appears in the nucleus, while the 35S::GFP appeared in the cytoplasm, nucleus and cell membrane. This indicates that the ERF38 protein is localized to the nucleus of plant cell.

ERF38 Protein had No Self-activation and Specifically Bind to the Dehydration Responsive Element Motif
The coding sequence of ERF38 was inserted into pGBKT7 in vivo, and transformed into the Y 2 H with a negative control and a positive control. As shown in Figure 4A, pGBKT7-ERF38 is able to grow normally on the SD/-Trp medium, as with the positive and negative controls. The positive control grew normally and turned blue on the SD/-Trp/-His/X-a-Gal medium, however, both pGBKT7-ERF38 and the negative control did not grow, indicating that the ERF38 protein has no self-activation.
Yeast one-hybrid was conducted to verify whether the ERF38 protein binds to the DRE element. Results indicated that the yeast cells were able to grow normally on the SD/-Leu medium, but not on the SD/-Leu medium containing 100 mg/ml AbA. Only the yeast strain, harboring ERF38 protein binding with the DRE motif, can grow on the SD/-Leu/ AbA. Negative control and other mutant motifs cannot grow normally with AbA ( Figure 4B). Therefore, it appears that ERF38 can specifically bind to the DRE motif.
FIGURE 2 | Tempo-spatial expression patterns of the ERF38 gene under stresses. After 150 mM NaCl and 20% polyethylene glycol-6000 treatments at each time point of 0, 3, 6, 12, 24, and 48 h, the roots, stems, and leaves were collected and used for reverse transcriptase quantitative polymerase chain reaction. The data was processed using the 2 -ΔΔCt method. The error bars represent the standard deviation. Asterisks indicate significant differences between transgenic lines and wild type lines (t test, *P < 0.05).

Molecular Identification of Transgenic Plants
As shown in Figure 5A, the exogenous ERF38 fragment was only identified in the transgenic poplar lines, but not in the WT seedlings. The transgenic seedlings can grow root normally in the rooting medium containing 50 mg/ml kanamycin and 200 mg/ml cephalosporin, however, the WT seedlings cannot grow root in the same medium ( Figure 5B). The relative expression level of ERF38 in transgenic poplar lines was significantly higher than that in the WT under both normal and stress conditions ( Figure 5C).

Morphological Changes in Response to abiotic Stresses
Seedlings from two transgenic lines and the WT were treated with 0, 35, 50, and 100 mM NaCl, respectively. Our data showed that under normal conditions, the plant height, root length, and fresh weight of transgenic poplar were 1.04, 1.12, and 1.05 times than that of WT, respectively. Under 35 mM NaCl conditions, the plant height, root length, and fresh weight of transgenic poplars were 1.07, 1.14, 1.11 times than those of WT, respectively. Under 50 mM NaCl conditions, the values were 1.15, 1.13, 1.15 times, respectively. And under 100 mM NaCl conditions, the ratio FIGURE 4 | Transcription activation activity assay and protein-deoxyribonucleic acid interaction analysis. (a) All three strains are able to grow normally on SD/-Trp medium, but only positive control can grow and turn blue on the SD/-Trp/-His/X-a-Gal medium. (B) Yeast one-hybrid assay was used to verify ERF38 protein specifically binding to the dehydration responsive element (DRE) motif, but not to the two DRE mutants.
To compare the transgenic lines and the WT under both abiotic stresses, we treated the 1-month-old plants with respective 150 mM salt and 20% PEG-6000 for 1 week. As shown in Figure  6B, the transgenic poplar and WT grew normally and there was no morphological difference under the control condition. Under the salt stress, however, the WT seedlings became wilting with dropping leaves, while the transgenic lines showed no obvious symptom ( Figure 6B). In contrast, under the PEG-6000 stress, the WT plant died, while the transgenic lines still survive with observable symptoms (Figure 6B).

Physiological Changes and Gene Expression in Response to Both Salt and Osmotic Stresses
We measured six physiological parameters of both the transgenic lines and the WT, under the two abiotic stresses and a water control (Figures 7A-F). In general, the two stresses displayed a similar physiological profiling pattern, with significant increases in the measurements, compared to the water control. Under the control condition, physiological traits, including POD and SOD activities, soluble protein content, and proline content, were significantly higher in the transgenic poplar than that in the WT (Figures 7A-D). But transgenic lines and the WT displayed no significant difference in MDA and H 2 O 2 contents (Figures 7E, F). Under each of the stresses, four of the physiological traits, including POD and SOD activity, soluble protein content, and proline content, increased significantly in the transgenic plants, compared to the WT (Figures 7A-D). Regarding the other two physiological traits (H 2 O 2 and MDA content), the increments in the transgenic lines under the stresses, which is compared to the water control, were significantly low than that of the WT, suggesting an improved ROS environment in the transgenic lines, compared to the WT (Figures 7E, F).
Furthermore, relative expression levels of the genes relevant to POD and SOD metabolism in the transgenic lines and the WT were measured by RT-qPCR. Under the control condition, the expression levels of such five genes were significantly higher in the transgenic lines than that in the WT (Figure 7G). Similarly, under salt and PEG-6000 stress conditions, their expression levels were still higher in the transgenic lines than that in the WT ( Figure 7G). These lines of evidence indicated that ERF38 gene may regulate the expression of the POD and SOD-related genes.
Similarly, we measured relative expression levels of four abscisic acid (ABA)-responsive genes (RD29B, ZmRD22B, PtPYRL1, and ZmPTF1) in the transgenic lines and the WT under normal growth and PEG-6000 stress conditions by RT-qPCR. These genes all can increase the plant tolerance by participating in the ABA signaling pathway (Yamaguchi-Shinozaki and Shinozaki, 1994;Yu et al., 2016;Phillips and Ludidi, 2017;Li et al., 2019b). Our results displayed that under the control condition, the expression levels of genes had no significant difference (Supplementary Figure S1). However, under the PEG-6000 condition, the expression of genes was significantly lower in the WT than in the transgenic poplars (Supplementary Figure S1).

histochemical Staining
Both NBT and DAB staining were used to detect the contents of hydrogen peroxide and superoxide, respectively. As shown in Figure 8, there was no significant difference in NBT staining between the transgenic seedlings and the WT under the control condition. After salt and PEG-6000 treatments, however, staining area of the transgenic seedlings was significantly smaller than that of the WT seedlings. Similar trend was observed for the DAB staining (Figure 8). These results indicate that accumulation of ROS in the transgenic lines was lower than that in WT.  (Bolt et al., 2017), tomato (Pan et al., 2012), rice (Jisha et al., 2015), wheat (Mondini et al., 2015). In this study, we identified a saltinduced and nucleus-targeted gene from the ERF gene family, ERF38. The gene encodes 155 amino acids, with the 14th amino acid in the conserved domain of valine acid, which indicates that ERF38 gene belongs to DREB subfamily. Evidences from Y 2 H indicated that ERF38 protein is not self-activating, and it also can specifically bind to the DRE element. Many ERF transcription factors can alter the morphological characteristics of plants. For example, over-expression of ERF76 gene in poplar changed leaf type, and increased root length and plant height of transgenic poplar (Yao et al., 2016). Overexpressing of chrysanthemum CmERF053 gene increased lateral buds and adventitious roots of transgenic chrysanthemums (Nie et al., 2018). In our study, we verified that the expression of ERF38 gene was significantly inducible after high salt and osmotic stress treatments. And the transgenic poplars over-expressing ERF38 gene displayed morphological advantages, compared to the WT poplars.
High salt and drought are main abiotic stresses threatened plant growth and development. There will be large amounts of reactive oxygen free radicals (including O 2 -, H 2 O 2 , hydroxide anions, etc.) accumulated in plants under the two stress conditions. Excessive ROS have a strong toxic effect on macromolecular substances, such as plant cell membrane systems, proteins and nucleic acids (Sarvajeet Singh and Narendra, 2010;Zhang et al., 2011). MDA content is the embodiment of the degree of membrane peroxidation in plant cells, and the higher its content, the more serious damage to the plant cell membrane (Tamirisa et al., 2014;Mellacheruvu et al., 2015).
In this work, we found that under the stresses, the contents of MDA and H 2 O 2 in the transgenic poplars were lower than that in the WT. In addition, evidence from histochemical staining demonstrated that ROS accumulation in the transgenic poplars was lower than that in the WT. These lines of evidence suggested that ERF38 gene might decrease membrane lipid peroxidation, in order to maintain the stability of membrane structure and reduce ROS accumulation, thereby improving poplar tolerance to salt and osmotic stresses.
In contrast, POD and SOD are the most important antioxidant enzymes to help eliminate extra ROS in plants, so that ROS in plants are maintained at a low level, to increase the tolerance of plants under adverse conditions Mittler et al., 2004;Feng et al., 2016). In the present study, compared to the WT poplars, the transgenic poplars possessed higher POD and SOD activities and the expression of POD and SOD-related genes was significantly up-regulated in the transgenic poplars under the two stresses, suggesting that the ERF38 gene can increase the relevant antioxidant enzymes in the transgenic poplars in response to salt and osmotic tolerance. Moreover, plants can produce osmotic regulators, such as proline and soluble proteins, in response to abiotic stresses (Wani and Gosal, 2011;de Freitas et al., 2019). High levels of proline and soluble protein are beneficial to maintenance of cell structure and function in plants, thus improving plant tolerance Xiong et al., 2018). In addition, proline in plants can effectively synergize with SOD enzymes to scavenge the ROS (Szabados and Savourcb, 2010;Rejeb et al., 2014). Our data indicated that transgenic poplars produce more proline and soluble protein than WT poplars, which provides another evidences supporting the function of ERF38 gene in salt tolerance and osmotic tolerance.
ABA is an important signal molecule for abiotic stress, and it has important biological functions in abiotic stresses, such as salt and drought Leon et al., 2014;Saxena et al., 2016). Abiotic stress factors can induce the synthesis of ABA (Xiong et al., 2002). It can change the stomata closure to regulate the transpiration rate and regulate the synthesis of osmotic substances to improve the tolerance of plants (Liu et al., 2005;Tuteja, 2007). The expression of some genes can also increase the plant tolerance by participating in the ABA signaling pathway (Li et al., 2019a;Li et al., 2019b). All evidence suggests that ABA is a key hormone in response to abiotic stress. In the present study, we found that the expression levels of ABA-responsive genes in transgenic poplars had a significantly elevated compare to WT poplars under PEG-6000 stress condition. However, we did not investigate whether or not the ERF38 gene can regulate the ABA signaling pathway, therefore it will be incorporated into future studies, in order to scrutinize molecular functions of the ERF38 gene.

CONClUSIONS
In the study, we isolated a salt-inducible ERF gene, ERF38, from P. alba × P. glandulosa, followed by transferring it into poplar. Evidence from yeast-one-hybrid indicated that ERF38 protein can specifically bind to the DRE element. And the transgenic poplar lines over-expressing ERF38 gene have advantages in the morphological, physiological, and biochemical traits, compared to WT poplars. Furthermore, expression levels of several POD and SOD-related genes were significantly higher in the transgenic lines than that in the WT. All the results indicated that over-expression of ERF38 gene can improve salt and osmotic tolerance of transgenic poplar by regulating the expression of stress-related genes.

DaTa aVaIlaBIlITY STaTEMENT
All datasets for this study are included in the article/ Supplementary Material.

aUThOR CONTRIBUTIONS
TJ and BZ designed research. ZC conducted experiments and data analysis and wrote the manuscript. XZ and KZ performed in data analysis. RL and WY revised the manuscript. All authors read and approved the manuscript.