A Stress-Associated Protein, PtSAP13, From Populus trichocarpa Provides Tolerance to Salt Stress

The growth and production of poplars are usually affected by unfavorable environmental conditions such as soil salinization. Thus, enhancing salt tolerance of poplars will promote their better adaptation to environmental stresses and improve their biomass production. Stress-associated proteins (SAPs) are a novel class of A20/AN1 zinc finger proteins that have been shown to confer plants’ tolerance to multiple abiotic stresses. However, the precise functions of SAP genes in poplars are still largely unknown. Here, the expression profiles of Populus trichocarpa SAPs in response to salt stress revealed that PtSAP13 with two AN1 domains was up-regulated dramatically during salt treatment. The β-glucuronidase (GUS) staining showed that PtSAP13 was accumulated dominantly in leaf and root, and the GUS signal was increased under salt condition. The Arabidopsis transgenic plants overexpressing PtSAP13 exhibited higher seed germination and better growth than wild-type (WT) plants under salt stress, demonstrating that overexpression of PtSAP13 increased salt tolerance. Higher activities of antioxidant enzymes were found in PtSAP13-overexpressing plants than in WT plants under salt stress. Transcriptome analysis revealed that some stress-related genes, including Glutathione peroxidase 8, NADP-malic enzyme 2, Response to ABA and Salt 1, WRKYs, Glutathione S-Transferase, and MYBs, were induced by salt in transgenic plants. Moreover, the pathways of flavonoid biosynthesis and metabolic processes, regulation of response to stress, response to ethylene, dioxygenase activity, glucosyltransferase activity, monooxygenase activity, and oxidoreductase activity were specially enriched in transgenic plants under salt condition. Taken together, our results demonstrate that PtSAP13 enhances salt tolerance through up-regulating the expression of stress-related genes and mediating multiple biological pathways.


Introduction
Soil salinization is a major environmental constraint on plant growth, development, and production [1]. High salinization limits the ability of plants to extract water and disrupt ion distribution, and usually causes oxidative stresses [2,3]. Worldwide, there are at least 800 million hectares of land subjected to salinity, accounting for~6% of the land surface [3]. Due to intensive land use, lack of freshwater resources, and irrational irrigation, the salinized area is increasing. To overcome the detrimental effects of salt stress, plants have developed multiple complex strategies, including

Phylogenetic Analysis of SAPs in P. trichocarpa and Other Species
A total of 19 SAP genes were identified from P. trichocarpa genome, which is in keeping with previous research [19]. Based on their locations in the genome, we sequentially named them from PtSAP1 to PtSAP19 (Table S1). Conserved domain analysis revealed that two PtSAPs (-14 and -17) included one AN1 domain (AN1); PtSAP3 contained one A20 domain (A20); PtSAP13 contained two AN1 domains (AN1-AN1); PtSAP8 contained two AN1 domain, and two CH2-CH2 domains (AN1-AN1-CH2-CH2); and the remaining 14 SAP members included one A20 domain and one AN1 domain (A20-AN1) (Table S1, Figure 1 and Figure S1). To reveal the evolutionary pattern and phylogenetic relationships of SAP, an unrooted phylogenetic tree was constructed using the protein sequences of SAPs from P. trichocarpa and another eight species (Eucalyptus grandis, Manihot esculenta, Marchantia polymorpha, Oryza sativa, Selaginella moellendorffii, Zea mays, Arabidopsis thaliana, and Volvox carteri) ( Figure 1A). The result shows that the SAPs were classified into seven classes (Class A-Class G). The SAPs containing A20-AN1 were present in Class A-Class E, and almost all of SAPs in Class F and Class G contained AN1 and/or CH2 domains, including AN1, AN1-AN1, AN1-AN1-CH2, or AN1-AN1-CH2-CH2 ( Figure 1A). Except Class E, the 19 PtSAPs were unevenly distributed in the other six classes. The SAP members of each species were counted ( Figure 1B). The lower algae V. carteri had the minimum number of SAPs, with only two members, while P. trichocarpa had the maximum number of SAPs (19), followed by O. sativa (18), M. esculenta (16), A. thaliana (14), E. grandis (11), and Z. mays (11) (Figure 1B).  Table S1.

Expression Profiles of PtSAP Genes
Revealing the expression profiles in various tissues under stress conditions will provide clues for gene potential functions; thus, we firstly analyzed the expression patterns of PtSAPs. Based on the public database of Populus, the expression profiles of PtSAPs in different tissues and developmental processes, including leaf expansion, germination of bud, male/female development, and different nitrogen nutrition treatment in root and stem, were investigated ( Figure S2). Three PtSAPs (-2, -10, and -11) were more highly expressed in root than other tissues, and three PtSAPs (-4, -11, and -13) were induced under ammonia, nitrate, or urea condition in root ( Figure S2). Only five PtSAPs (-4, -10, -11, -12, and -13) were highly expressed in the middle of male catkin development. In the process of pre-dormant bud I to fully open bud, the expression of three PtSAPs (-2, -5, and -13)  Table S1.

Expression Profiles of PtSAP Genes
Revealing the expression profiles in various tissues under stress conditions will provide clues for gene potential functions; thus, we firstly analyzed the expression patterns of PtSAPs. Based on the public database of Populus, the expression profiles of PtSAPs in different tissues and developmental processes, including leaf expansion, germination of bud, male/female development, and different nitrogen nutrition treatment in root and stem, were investigated ( Figure S2). Three PtSAPs (-2, -10, and -11) were more highly expressed in root than other tissues, and three PtSAPs (-4, -11, and -13) were induced under ammonia, nitrate, or urea condition in root ( Figure S2). Only five PtSAPs (-4, -10, -11, -12, and -13) were highly expressed in the middle of male catkin development. In the process of pre-dormant bud I to fully open bud, the expression of three PtSAPs (-2, -5, and -13) increased early and then declined, which indicates that these genes might be involved in dormant bud formation ( Figure S2).

Expression Patterns of PtSAP13
According to the expression level of PtSAP13 under salt treatment, we chose PtSAP13 for further function analysis. The full length CDS of PtSAP13 was 450 bp and it encoded a 149 amino acid protein with two AN1 domains (Table S1). The molecular weight of PtSAP13 was 16.97 kDa and its predicted isoelectric point was 9.02, respectively (Table S1). In order to further determine the expression patterns of PtSAP13, the promoter fragment was amplified and constructed into the pMDC164 vector to fuse with the reporter gene GUS (namely, pPtSAP13::GUS), and transgenic Arabidopsis was generated. GUS staining in Arabidopsis revealed that GUS signal was detected in roots in 1-day-old seedlings ( Figure 3A). Subsequently, 3-day-old seedlings showed high GUS signal in cotyledons and roots ( Figure 3B), as well as detected in 19-day-old seedlings ( Figure 3C) and 1-month-old seedlings ( Figure 3D). In addition, GUS signaling was also detected in stigma, stamen, and siliques ( Figure  3E,F). When a 10-day-old pPtSAP13::GUS transgenic plant was treated with salt for 12 h, strong GUS activity was detected ( Figure 3H), which confirms the results of the expression profile of PtSAP13.

Expression Patterns of PtSAP13
According to the expression level of PtSAP13 under salt treatment, we chose PtSAP13 for further function analysis. The full length CDS of PtSAP13 was 450 bp and it encoded a 149 amino acid protein with two AN1 domains (Table S1). The molecular weight of PtSAP13 was 16.97 kDa and its predicted isoelectric point was 9.02, respectively (Table S1). In order to further determine the expression patterns of PtSAP13, the promoter fragment was amplified and constructed into the pMDC164 vector to fuse with the reporter gene GUS (namely, pPtSAP13::GUS), and transgenic Arabidopsis was generated. GUS staining in Arabidopsis revealed that GUS signal was detected in roots in 1-day-old seedlings ( Figure 3A). Subsequently, 3-day-old seedlings showed high GUS signal in cotyledons and roots ( Figure 3B), as well as detected in 19-day-old seedlings ( Figure 3C) and 1-month-old seedlings ( Figure 3D). In addition, GUS signaling was also detected in stigma, stamen, and siliques ( Figure 3E,F). When a 10-day-old pPtSAP13::GUS transgenic plant was treated with salt for 12 h, strong GUS activity was detected ( Figure 3H), which confirms the results of the expression profile of PtSAP13.

Effect of Salt Stress on Seed Germination in Transgenic PtSAP13 Arabidopsis Lines
To gain insight into the function of PtSAP13 in salt stress tolerance, transgenic Arabidopsis lines with overexpressing PtSAP13 were generated. Three transgenic lines (Line 6, Line 15, and Line 30) with high expression of PtSAP13 were used for its stress tolerance investigation ( Figure 4A). On the normal 1/2 MS medium, there was no difference in germination rates between the WT and PtSAP13overexpressing plants, and the germination rate was nearly 100% ( Figure 4E). After being cultured on 200 mM NaCl medium for 7 days, the germination rates of the three transgenic lines (Line 6, Line 15, and Line 30) were 74.5%, 80.7%, and 84.0%, respectively; whereas the WT exhibited a lower germination rate, with nearly 45.4%, than transgenic lines ( Figure 4D). This result indicates that overexpression of PtSAP13 increased the seed germination under salt stress.

Effect of Salt Stress on Seed Germination in Transgenic PtSAP13 Arabidopsis Lines
To gain insight into the function of PtSAP13 in salt stress tolerance, transgenic Arabidopsis lines with overexpressing PtSAP13 were generated. Three transgenic lines (Line 6, Line 15, and Line 30) with high expression of PtSAP13 were used for its stress tolerance investigation ( Figure 4A). On the normal 1/2 MS medium, there was no difference in germination rates between the WT and PtSAP13-overexpressing plants, and the germination rate was nearly 100% ( Figure 4E). After being cultured on 200 mM NaCl medium for 7 days, the germination rates of the three transgenic lines (Line 6, Line 15, and Line 30) were 74.5%, 80.7%, and 84.0%, respectively; whereas the WT exhibited a lower germination rate, with nearly 45.4%, than transgenic lines ( Figure 4D). This result indicates that overexpression of PtSAP13 increased the seed germination under salt stress. Error bars indicate the standard deviations, and * indicate significant differences compared with WT at the p < 0.05 level.

PtSAP13 Transgenic Arabidopsis Seedling Resistance to Salt Stress
One-week-old WT and PtSAP13-overexpressing plants were transferred to 1/2 MS solid medium containing 0 mM (normal) or 150 mM NaCl (salt stress). No significant phenotypic difference was observed between the WT and transgenic plants under normal medium ( Figure 4C). Under salt condition, the growth of the WT and transgenic plants was significantly inhibited, but the growth situation of the transgenic plants was better than that of the WT plants ( Figure 4C). Both average fresh weight and root length of the transgenic plants were more than the WT plants ( Figure 4E,F). Furthermore, similar results were also obtained in the soil-grown WT and transgenic plants ( Figure  5A,B). The relative electric conductivity in the transgenic plants was lower than in the WT plants ( Figure 5C), whereas the proline content was higher in the transgenic plants than in the WT plants under salt condition ( Figure 5D). The peroxidase (POD), Superoxide dismutase (SOD), and catalase (CAT) activities of the transgenic plants were higher than that of the WT plants after salt treatment ( Figure 5E-G), indicating that scavenging of reactive oxygen species (ROS) was enhanced by PtSAP13

PtSAP13 Transgenic Arabidopsis Seedling Resistance to Salt Stress
One-week-old WT and PtSAP13-overexpressing plants were transferred to 1/2 MS solid medium containing 0 mM (normal) or 150 mM NaCl (salt stress). No significant phenotypic difference was observed between the WT and transgenic plants under normal medium ( Figure 4C). Under salt condition, the growth of the WT and transgenic plants was significantly inhibited, but the growth situation of the transgenic plants was better than that of the WT plants ( Figure 4C). Both average fresh weight and root length of the transgenic plants were more than the WT plants ( Figure 4E,F). Furthermore, similar results were also obtained in the soil-grown WT and transgenic plants ( Figure 5A,B). The relative electric conductivity in the transgenic plants was lower than in the WT plants ( Figure 5C

Overexpression of PtSAP13 Induced the Expression of Stress-Related Genes
To further investigate the regulatory mechanism of PtSAP13, RNA-seq analysis was performed for the WT and    (Figure 7). A considerable proportion of the GO terms, including multiple responses to abiotic or biotic stresses, responses to various hormones, glutathione metabolic process, oxidationreduction process, phenylpropanoid biosynthesis and metabolic process, and toxin catabolic and metabolic process, were mainly enriched in WT-Salt/WT-Normal and SAP13-Salt/SAP13-Normal, implying that these pathways were universal and common response mechanisms under salt stress in both the WT and PtSAP13-overexpressing plants (Figure 7). The GO terms of flavonoid biosynthesis and metabolic process, regulation of response to stress, response to ethylene, and cellular response to hypoxia were uniquely enriched only in SAP13-Salt/SAP13-Normal. In addition, DEGs were enriched in 29 MF GO terms (Figure 7). The dioxygenase activity, glucosyltransferase activity, monooxygenase activity, and oxidoreductase activities acting on CH-OH group of donors and NAD or NADP as acceptor were uniquely enriched only in SAP13-Salt/SAP13-Normal (Figure 8). These To gain insight into the major functional categories represented by the DEGs, GO enrichment analysis was performed in four comparison sets (WT-Salt/WT-Normal, SAP13-Normal/WT-Normal, SAP13-Salt/SAP13-Normal, and SAP13-Salt/WT-Salt). The results show that DEGs were enriched in 83 BP GO terms (Figure 7). A considerable proportion of the GO terms, including multiple responses to abiotic or biotic stresses, responses to various hormones, glutathione metabolic process, oxidation-reduction process, phenylpropanoid biosynthesis and metabolic process, and toxin catabolic and metabolic process, were mainly enriched in WT-Salt/WT-Normal and SAP13-Salt/SAP13-Normal, implying that these pathways were universal and common response mechanisms under salt stress in both the WT and PtSAP13-overexpressing plants (Figure 7). The GO terms of flavonoid biosynthesis and metabolic process, regulation of response to stress, response to ethylene, and cellular response to hypoxia were uniquely enriched only in SAP13-Salt/SAP13-Normal. In addition, DEGs were enriched in 29 MF GO terms (Figure 7). The dioxygenase activity, glucosyltransferase activity, monooxygenase activity, and oxidoreductase activities acting on CH-OH group of donors and NAD or NADP as acceptor were uniquely enriched only in SAP13-Salt/SAP13-Normal (Figure 8). These results suggested that PtSAP13-overexpressing plants might possess specific regulatory pathways to enhance their salt tolerance (Figure 8).

Discussion
The number of SAP genes increased from lower algae (two SAPs in V. carteri) through early land plants (from three SAPs in M. polymorpha to six SAPs in S. moellendorffii), to higher plants (from 11 SAPs in O. sativa to 19 SAPs in P. trichocarpa). P. trichocarpa had the maximum number of SAPs, suggesting that PtSAP genes underwent expansion through whole genome duplication. Previous studies have found that the HTK1 (high affinity K + transporter 1) family in P. euphratica [37] and Hsfs (Heat shock transcription factors) family in P. trichocarpa [38] are also expanded to defense against abiotic stresses. P. trichocarpa is widely distributed across western North America, with wide geographical and environmental gradients [39]. In addition, compared with herbaceous plants, P. trichocarpa has a long life span, thus it might have evolved sophisticated adaptive systems that need the participation of more stress-related genes to ensure it survives in prolonged and repeated environmental stresses.

Discussion
The number of SAP genes increased from lower algae (two SAPs in V. carteri) through early land plants (from three SAPs in M. polymorpha to six SAPs in S. moellendorffii), to higher plants (from 11 SAPs in O. sativa to 19 SAPs in P. trichocarpa). P. trichocarpa had the maximum number of SAPs, suggesting that PtSAP genes underwent expansion through whole genome duplication. Previous studies have found that the HTK1 (high affinity K + transporter 1) family in P. euphratica [37] and Hsfs (Heat shock transcription factors) family in P. trichocarpa [38] are also expanded to defense against abiotic stresses. P. trichocarpa is widely distributed across western North America, with wide geographical and environmental gradients [39]. In addition, compared with herbaceous plants, P. trichocarpa has a long life span, thus it might have evolved sophisticated adaptive systems that need the participation of more stress-related genes to ensure it survives in prolonged and repeated environmental stresses.
The expression patterns of PtSAPs under salt stress were analyzed. We found that most of PtSAPs were induced under salt stress, and similar results have been found in SAPs from other species. For example, Gossypium hirsutum GhSAPs (-7A, -7D, -8A, -11D, and -16D) and P. euphratica PeuSAPs (-1, -2, -8/14, and -16/17) were significantly induced under salt treatment [15]. Among these PtSAPs, PtSAP13 exhibited more significant up-regulation than other PtSAPs under salt stress, suggesting that PtSAP13 might be involved in salt tolerance in P. trichocarpa. Structural analysis found that PtSAP13 protein contained two AN1 domains without AN20. The specific role of AN1 domain is still poorly known. Two SAP proteins, including Arabidopsis AtSAP13 and T. aestivum TaSAP17-D, that contain two AN1 domains have been reported to enhance the salt tolerance in transgenic plants [33]. In our study, overexpression of PtSAP13 also improved salt tolerance of transgenic plants, as evidenced by the better growth in fresh weight and root length, as well as increased activities of antioxidative enzymes than WT plants under salt treatment. These results further demonstrate that SAPs containing AN1-AN1 had conserved function and played essential roles in salt tolerance.
Salt stress can trigger two types of stresses, including osmotic stress and ion toxicity. Early effects of salt stress are to prevent plants from absorbing water, and quickly cause decrease of growth rate [40]. During this process, many physiological and transcriptional responses are identical to those caused by water stress [13,40,41]. After excessive amounts of Na + and Cl − enter the plants, ion toxicity will eventually give rise to premature senescence of leaves and limit plant growth and development [13]. To adapt to salt stress, plant have evolved a complex regulatory network that mediates multiple genes [13,41]. In our study, numerous DEGs (2879 DEGs in PtSAP13-overexpressing plants and 3641 DEGs in WT plants) were detected after salt treatment, and these DEGs were enriched in multiple biological pathways. These results confirm that plant response and tolerance to salt stress is a complex system. Transcriptome data revealed that some stress-related genes, such as BZIP60, DI19-4, GPX8, NADP-ME2, RAS1, WRKYs, GSTUs, and MYBs, were up-regulated in the PtSAP13-overexpressing plants, but did not change in the WT plants under salt stress. BZIP60, a member of basic leucine zipper transcription factor, is involved in plant immunity and abiotic stress responses [42]. Overexpression of TabZIP60 from T. aestivum can enhance the resistance of transgenic Arabidopsis to abiotic stresses and enhance ABA sensitivity [43]. DI19-4 is a member of the Drought-Induced gene family, its expression in Arabidopsis is induced by salt and dehydration [44,45]. Overexpression of OsDI19-4 from O. sativa increases stress tolerance by enhancing ROS-scavenging activity [44,45]. GPX8participates to withstand oxidative damage and/or to modulate oxidative signaling caused by reactive oxygen species. In the mutant of gpx8, the plant exhibits increased sensitivity to oxidative damage by enhancing the oxidized proteins, while overexpression of Arabidopsis AtGPX8 increases the oxidative tolerance [46]. The expression of O. sativa NADP-ME2 is induced by NaHCO 3 , NaCl, and PEG stresses. Overexpression from O. sativa and SbNADP-ME from Sorghum bicolor enhances tolerance of transgenic plants' salt stress [47,48]. RAS1 is associated with salt tolerance and ABA sensitivity using QTL mapping in Arabidopsis [49]. Although there are no reports of WRKYs (-14, -24, -35, and -65) functions in salt tolerance, multiple members of the WRKY family, including Arabidopsis AtWRKYs (-11, -17, -20, and -25) and G. hirsutum GhWRKYs (-17, -34, -39-1, and -41), have been confirmed to enhance plant salt tolerance [50][51][52][53][54][55]. The expression levels of these genes increased in the PtSAP13-overexpressing plants under salt stress, suggesting that PtSAP13 might enhance plant salt tolerance by collaborating with these stress-related genes.
Salt stress causes excessive generation of ROS, which gives rise to oxidative damage to DNA, proteins, and lipids [56]. To cope with the oxidative damage, plants activate defense systems mediated by enzymatic and non-enzymatic antioxidants to maintain ROS homeostasis [56]. Enzymatic antioxidants, such as POD, SOD, and CAT, play important roles in scavenging ROS and maintaining redox equilibrium [57]. In our study, the activities of these enzymatic antioxidants were higher in PtSAP13-overexpressing plants than in the WT plants under salt stress, indicating that the ability of scavenging ROS was enhanced by PtSAP13 in transgenic plants. Flavonoids have been identified as non-enzymatic antioxidants to scavenge ROS and protect plants from oxidative damage [58]. Overexpression of Vitis amurensis NAC26 and Antirrhinum AmDEL significantly increases the accumulation of flavonoids and enhances salt tolerance in transgenic plants [59,60]. In PtSAP13-overexpressing plants, flavonoid biosynthesis and metabolic processes were uniquely enriched under salt stress, which might help mediate the salt tolerance of transgenic plants.
In addition, water and nutrients were transported in conductive tissues to maintain plant growth. When plants are subjected to salt stress, their nutrient homeostasis is disrupted [61]. Improvement of the ability to absorb nutrients and increase amounts of essential nutrients will help plants resist the salt stress [61,62]. In our study, PtSAP13 was strongly expressed in the conductive tissues (Figure 3), and its expression was induced under nitrate and urea condition ( Figure S2). Thus, we speculated that PtSAP13 might also be involved in uptake and/or transport of nutrients. We also found that the nutrient-related biological processes, including the pathways of celluar response to nutrient levels and response to starvation, were enriched in SAP13-Normal/WT-Normal and SAP13-Salt/WT-Salt. Thus, these results further indicate that PtSAP13 might enhance salt tolerance by regulating nutrient-related processes.

Plant Materials and Salt Treatments
Seeds of A. thaliana (ecotype Columbia) which conserved in Chinese Academy of Forestry were surface-sterilized with 75% (v/v) ethanol twice, followed by 100% (v/v) ethanol once in clean bench. After being dried on filter paper, the seeds were sowed on 1/2 Murashige-Skoog (MS) solid medium containing 0.8% agar. After being vernalized at 4 • C for 3 days in the dark, the seeds were grown at 20-22 • C under long-day conditions (16/8 h light/dark) in medium or soil.
Two-month-old clones of P. trichocarpa were water-cultured using Hoagland solution with 150 mM NaCl for salt treatment. Fully matured leaves from six clones were immediately harvested after 0, 1, 6, 12, and 48 h of salt treatment and stored at −80 • C for further expression analysis. The dosages of the salt treatment were determined based on previous study [63]. This experiment included three biological replicates.

RNA Isolation and Quantitative Real-Time PCR (qRT-PCR) Analysis
An RNeasy Plant Mini Kit (Qiagen, Hilden, Germany) was used for extracting RNA from plant samples, according to the manufacturer's instructions. During the extraction process, DNase I (Qiagen, Hilden, Germany) was used to degrade genomic DNA. The SuperScript III first-strand synthesis system (Life technologies, Carlsbad, CA, USA) was used to synthesize the first-strand cDNA with 3 µg mRNA. The qRT-PCR was performed on the Roche LightCycler 480 (Roche Applied Science, Penzberg, Germany) using SYBR Premix Ex Taq (Takara, Kusatsu, Japan). Each sample was performed in three biological replicates and four technical replicates. AtActin2 (AT3g18780) and PtActin (Potri.001G309500) were used as reference genes in Arabidopsis and P. trichocarpa, respectively. The 2 −∆∆CT method was used to calculate the relative expression levels of each gene [64]. The primers used for qRT-PCR in this study are listed in Table S2.

Generation of Transgenic A. thaliana Plants
The coding sequence of PtrSAP13 was amplified from the cDNA of P. trichocarpa using gene-specific primers (Table S2). The amplified fragment was cloned into the pDONR222 vector (Life technologies, USA) to produce pENTR for sequencing. The correct sequence was sub-cloned into pMDC32 to construct the plasmid p35S::PtSAP13, which was used to generate overexpressing transgenic plants.
The promoter sequence of PtSAP13 was amplified from the genomic DNA of P. trichocarpa and cloned into the pMDC164 vector to construct the plasmid pPtSAP13::GUS, which was used to study the expression pattern of PtSAP13. The correct constructs described above were transformed into Agrobacterium GV3101 by electroporation and transform A. thaliana using the floral dip method [65]. The positive transgenic plants were selected on 1/2 MS plates containing 25 mg/L hygromycin and PCR identification. More than 30 independent transgenic lines of pPtSAP13::GUS and 31 independent transgenic lines of p35S::PtSAP13 were obtained, respectively. Finally, three transgenic lines of pPtSAP13::GUS were used for GUS staining, according to previous research [66]. Three transgenic lines with high abundance of PtSAP13 were used for further salt tolerance analysis and RNA-Seq analysis.

Analysis of Salt Tolerance
For the seed germination assays, T3 homozygous seeds (100-150 seeds) of transgenic lines and wild-type (WT) were sown on 1/2MS medium containing 0 mM (normal) or 200 mM NaCl (salt stress). After being vernalized at 4 • C for 3 days in the dark, seeds were transferred under normal conditions, and seed germination was counted after 1 week. Germination was scored as rupture of the seed coat.
For the salt tolerance experiments of seedlings, 1-week-old PtSAP13-overexpressing and WT plants were cultured on 1/2 MS medium containing 0 mM or 150 mM NaCl for 2 weeks, and the fresh weight and root length were measured. In addition, after growth on 1/2 MS normal medium for 2 weeks, the transgenic and WT plants were transferred on soil for 4 weeks, and then irrigated with 30 mL of 200 mM NaCl for 2 weeks, and followed with pure water for 7 days. The relative electric conductivity, proline content, and enzyme activities were measured after salt tolerance experiments. The relative electric conductivity was determined by conductivity meter (DDS-307). The proline content was analyzed according to Dubois et al. [67]. The activities of SOD, POD, and CAT were measured according to Li et al. [63]. All experiments were performed in three biological replicates. The data were analyzed with a t-test to detect significant differences.

RNA-Seq Analysis of Transgenic Arabidopsis
The seeds from WT and transgenic plants were grown under normal condition for 1 week. Subsequently, the seedlings were transferred into 1/2 MS medium containing 0 mM or 150 mM NaCl for 2 weeks. Finally, the whole plants of WT under normal and salt conditions were collected and marked as WT-Normal and WT-Salt, respectively; three independent transgenic lines under normal and salt conditions were equally pooled and marked as SAP13-Normal and SAP13-Salt, respectively.
Total RNA was also extracted using RNeasy Plant Mini Kit (Qiagen, Hilden, Germany) and treated with DNase I (Qiagen, Hilden, Germany). The RNA purity and integrity were detected by the NanoPhotometer spectrophotometer and Agilent2100 Bioanalyzer, and the RNA quantity was measured by Qubit2.0 Fluorometer (Life Technologies, New York, USA). The cDNA libraries were constructed and sequenced on the Illumina HiSeq2500 sequencing platform by Gene Denovo Biotechnology Co. (Guangzhou, China), with paired-end sequencing and read lengths of 150 bp. Each sample was performed for three biological replicates. Quality-control procedure was performed to filter adapter sequences and low-quality reads using fastp (version 0.18.0) (https://github.com/ OpenGene/fastp) [68]. The clean data were obtained for subsequent analyses. Clean reads were mapped to the reference genome using HISAT2 (version 2.4) (http://ccb.jhu.edu/software/hisat/index.shtml) [69]. Gene expression levels were calculated as reads per kilobase of transcript sequence per million base pairs sequenced (FPKM) using StringTie (version 1.3.1) (https://ccb.jhu.edu/software/stringtie/)) [70]. The differentially expressed genes (DEGs) were identified by DESeq2 (https://bioconductor.org/ packages/release/bioc/html/DESeq2.html) [71]. Genes with |log 2 FoldChange| >1.0 and false discover rate (FDR) <0.05 were considered differentially expressed. To study the biological function of DEGs, gene set enrichment with gene ontology (GO) terms was performed using Blast2GO (version 5.2.5) (https://www.blast2go.com/) [72]. Fischer's exact test was used to assess the significance of GO categories. p-values were corrected by Benjamini-Hochberg FDR and the corrected p-value cut-off was 0.05. RNA-Seq data have been deposited in the NCBI Sequence Read Archive under accession number PRJNA579744.

Conclusions
In our study, we isolated PtSAP13 from P. trichocarpa and characterized its function in salt tolerance. PtSAP13 was highly induced by salt stress and constitutive expression of PtSAP13 increased the tolerance to salt stress. The PtSAP13-overexpressing plants enhanced salt tolerance by improving ROS-scavenging ability. Moreover, RNA-seq data showed that some stress-related genes were significantly induced in PtSAP13-overexpressing plants, suggesting that PtSAP13 may paly a role in plant salt tolerance by regulating the expression of these genes, although detailed work will be needed to identify the exact regulatory mechanism of PtSAP13 in stress tolerance in poplar.

Conflicts of Interest:
The authors declare no conflicts of interest.