Gibberellin mediates spermidine-induced salt tolerance and the expression of GT-3b in cucumber

Background Transcription factor GT-3b binds to the GT-1 element in the promoter of S -adenosylmethionine synthase ( SAMs ) gene, the key gene of spermidine (Spd) biosynthesis, to regulate its expression, thereby increasing the salt tolerance of cucumber plants. Furthermore, exogenous Spd increases the transcript level and protein abundance of SAMs, which promotes the accumulation of endogenous polyamine. However, whether Spd regulation of SAMs depending on GT-3b is largely unknown. Here, we investigated the potential mechanism of Spd in regulating of GT-3b in cucumber seedlings under salt stress. Results Exogenous Spd significantly increased the tolerance to salt stress in cucumber, while its effects were compromised when application of methylglyoxal bis-guanylhydrazone (MGBG), an inhibitor of Spd biosynthesis. GT-3b was significantly induced by Spd under salt stress. The promoter sequence of GT-3b was predicted that there contained cis-acting regulatory elements response to phytohormones, such as gibberellin (GA), salicylic acid (SA) and methyl jasmonate (MeJA). Foliar spray of GA 3 , SA and MeJA could induce the expression of GT-3b . Interestingly, exogenous Spd dramatically induced the expression of genes related to GA biosynthesis, increased the activity of gibberellin oxidase, and promoted the accumulation of GA 3 , but decreased in MGBG-treated plants. Furthermore, application of GA 3 increased the expression of GT-3b and salt tolerance, while blocked when treatment with paclobutrazol (PAZ), the GA biosynthesis inhibitor. In addition, Spd-induced salt tolerance was compromised in PAZ-treated plants. Conclusions Our results suggested that GA mediated Spd-induced salt tolerance and the expression of GT-3b in cucumber. These results provide new perspective for our

4 understanding the molecular mechanism of Spd in regulation of salt tolerance in plants.

Background
Transcription factors are key components that generate specific stress responses.
They manipulate the transcriptional expression of a variety of stress-tolerant genes and control a wide range of downstream metabolism pathways. Many transcription factors, such as NAC [1], MYB [2], ERF [3], WRKY [4] and zinc finger proteins [5], are involved in salt stress tolerance in plants [6]. The trihelix transcription factor is a unique transcription factor in plants and plays a key role in the process of plant stress signal transmission [7]. It can enhance the stress tolerance of plants by binding to cis-acting elements in the promoter region of stress-regulating genes and by regulating the expression of stress-related genes [7]. The trihelix family is divided into five subfamilies, GT-1, GT-2, GTγ, SH4 and SIP1, based on the number of DNA binding domains and the conserved amino acids in the GT protein domain [7]. Northern blotting indicates that the expression level of the GT-1 protein RML1 is significantly upregulated in rice seedling leaves infected with rice blast [8]. It wasfound that GT-4 interacts with TEM2 to regulate the salt response gene Cor15A to improve salt tolerance [9]. The GT-3b transcription factor is a trihelix transcription factor induced by NaCl and pathogens [10], which can interact with the GT-1 cis-element to promote the transcriptional expression of downstream genes and plays an important role in the salt tolerance of plants. Our previous work found that the GT-3b is induced by salt stress and binds to the GT-1 element in the promoter of S-adenosylmethionine synthase gene (CsSAMs) to trigger its expression in cucumber [11], thereby improving the salt tolerance of cucumber plants.
SAMs plays an important role in salt stress, and studies have confirmed that salt stress induces the expression of SAMs [11,12]. The expression of SAMs1 and SAMs3 genes is obviously induced by salt stress in tomato roots and can be maintained at a high level for a long period of time [13]. Plants overexpressing SAMs gene are more tolerant to salt and oxidative stress in comparison to wild-type plants [14].
Furthermore, proteomic analysis found that SAMs is upregulated stably during salt stress and is considered as marker protein for the salt stress response in chloroplasts [15]. Similarly, an increase in the expression of SAMs protein in wheat was observed using two-dimensional gel electrophoresis under salt stress [16].
Overexpression of SAMs from Pyropia tenera in E. coli significantly enhanced high salt tolerance [17].
SAMs catalyzes the reaction of ATP and l-methionine to generate Sadenosylmethionine (SAM), which is the main methyl donor in organisms and the precursor for the biosynthesis of ethylene and polyamine (PA). The interaction between the SAMs gene and the amino transferase gene in apples affects plant growth by promoting the accumulation of PA in plants [18]. Overexpression of SAMs gene activates the PA metabolic pathway, increases the level of PA, and confers salt tolerance to plants. It has been reported that salt tolerance of transgenic plants overexpressing SAMs2 is significantly improved, and the contents of spermidine (Spd) and spermine (Spm) in transgenic tobacco is significantly higher than those in wild-type tobacco [19]. In addition, overexpression of SAMs1 gene promotes the synthesis and accumulation of Spm and Spd, activates stress-related genes, enhances the expression of the antioxidant system, thereby increasing salt tolerance in tomato plants [20]. Similarly, overexpression of MfSAMs1 promotes the synthesis and oxidation of PAs, leading to improving antioxidant protection against H 2 O 2 [21,22]. As a stress-responsive protein, the expression of SAM is significantly inhibited by high concentrations of salts, while exogenous PA reverses the inhibitory effects of salt stress on SAMs at various levels, including enzymatic activity, gene and protein levels. Li [23] analyzed the effects of exogenous Spd on protein expression in cucumber leaves under salt stress and found that Spd increased the abundance of SAMs and other proteins related to antioxidant defense and detoxification. In addition, exogenous Spd can upregulate the expression of SAMs under salt stress and provide more methyl donors that can participate in the defense response by promoting the synthesis of SAM, and it may promote lignification in the vascular tissue of plants [12] and regulate water transport and ion-selective absorption under salt stress. However, whether Spd regulation the expression of SAMs is rely on GT-3b in still unclear.
In this study, we analyzed the mechanism of Spd regulation of GT-3b in response to salt stress in cucumber. The results showed that under salt stress, Spd upregulated the expression of GT-3b by increasing the level of endogenous GA in cucumber plants, which promoted the expression of GT-3b, resulting in enhanced the salt tolerance of cucumber plants. Our results provide new insight for understanding of the molecular mechanism of Spd enhancing tolerance to salt in plants.

Spd is crucial for salt tolerance in cucumber
The growth of cucumber seedlings subjected to salt stress was significantly inhibited compared with the control (CK) plants (Fig. 1a). The plant height, shoot fresh weight, root fresh weight, shoot dry weight and root dry weight decreased by 39.63%, 32.15%, 46.59%, 19.89% and 31.25%, respectively, in comparison to CK 7 plants and the above indexes in plants subjected to salt stress (S) with spraying of Spd (Spd+S) were 33.33%, 25.79%, 24.95%, 7.80%, and 18.18% higher than that in S-treated plants (Fig. 1b). The adverse effects-induced by salt stress were more significant in the plants treated with salt and methylglyoxal bis-guanylhydrazone (MGBG), an inhibitor of Spd biosynthesis, but was alleviated after spraying Spd (Fig.   1). Furthermore, the contents of malondialdehyde (MDA) and proline in cucumber leaves significantly increased under salt stress (Additional file 1: Figure S1). Foliar application of MGBG further promoted the accumulation of MDA and proline, while application of Spd significantly decreased the MDA and proline contents (Additional file 1: Figure S1). The results indicated that Spd could alleviate the inhibition of cucumber seedling growth by salt stress and reduce the degree of damage in plants.

Spd induced the expression of GT-3b under salt stress
It has been demonstrated that salt stress induced the expression of GT-3b, which mediates the biosynthesis of PAs to regulate salt tolerance [11]. To detect whether GT-3b is involved in Spd-induced the tolerance to salt, we used qPCR to analyze the expression patterns of GT-3b gene. As shown in Fig. 2, the expression level of GT-3b was upregulated under salt stress compared with the CK. The increased level of GT-3b in Spd+S plants was more great and reached to the peak at 12 h, while application of MGBG compromised the induction of GT-3b (Fig. 2). These results suggested that GT-3b was a salt stress response gene and was regulated by Spd.

Analysis the expression patterns of GT-3b in response to phytohormones
To further investigate the role of GT-3b in Spd-induced salt stress tolerance, we analyzed its promoter and the expression patterns in response to different phytohormones. To identify the cis-acting element in GT-3b, the 1278 bp sequence 8 of the GT-3b promoter was isolated from genomic DNA, and the cis-acting regulatory elements were identified using the PlantCARE and PLACE online databases [24,25].
Bioinformatics analysis revealed that the putative cis-acting regulatory elements that regulate stress and defense gene expression in different plant species were existed in this region (Additional file 2: Figure S2 and Additional file 3: Table S1).
The phytohormone response elements were involved in these cis-acting elements, including the TGACG and CGTCA motifs, GARE motif and the TCA element, which are promoter fragments were treated with GA 3 , SA, and MeJA, which was followed by GUS histochemical staining and fluorometric assay analyses. After spraying GA 3 with the P1 vector containing the GARE motif, the staining was significantly deeper and the GUS activity was higher than other treatments (Fig. 4c,d), indicating that GA could activate GT-3b to participate in the stress defense response. Similarly, the GUS activity of the P2 fragment containing the CGTCA motif and the P3 fragment 9 containing the TCA element was consistent with the staining as showed significantly increased the GUS activity treated with MeJA or SA (Fig. 4c,d). These results indicated that the GARE motif, CGTCA motif and TCA element in the GT-3b promoter may be response to exogenous GA, MeJA, and SA, which was consistent with the finding by qPCR that GA, MeJA and SA could upregulate the expression of GT-3b.

Endogenous GA 3 content and gibberellin oxidase activity in cucumber leaves
The above results showed that Spd could regulate the expression level of GT-3b, and the transcription level of GT-3b was also regulated by GA, MeJA, and SA.
Therefore, we speculate that Spd regulates the expression of GT-3b by affecting GA, MeJA, and SA. To verify this hypothesis, we studied the effect of exogenous Spd treatment on the contents of endogenous GA 3 , SA and MeJA in cucumber leaves ( Fig. 5). The results showed that the content of GA 3 in leaves after Spd treatment was approximately 3.97-fold higher than that in CK plants (Fig. 5a). In contrast, the GA 3 content in MGBG treatment plants was significantly lower than that in CK plants ( Fig. 5a). Unlike GA 3 , the content of endogenous SA induced by Spd was not significantly different from that of the CK, while treated with MGBG significantly increased the content of SA (Fig. 5b). However, the content of endogenous MeJA in the cucumber leaves was too low to detect. On the basis of the effect of Spd on endogenous phytohormone contents, we speculate that the induction of the expression of GT-3b by Spd is most closely related to GA 3 .
To test whether Spd was involved in regulation of GA, we examined the effects of Spd on the expression of key genes in the GA anabolic pathways. The results showed that exogenous Spd upregulated the expression of CPS1, KS1, KO1, KAO1, KAO2, GA20ox1, GA20ox2, GA20ox3, GA20ox4 and GA20ox5, which are associated with GA biosynthesis (Fig. 6). The expression of GA3ox3, GA3ox4 and GA3ox5 was significantly upregulated in Spd-treated plants (Fig. 6), which promoted the conversion of GA 9 and GA 20 to GA 4 , GA 1 and GA 3 . Moreover, Spd upregulated the expression of GA2ox1, GA2ox2, GA2ox3, GA2ox4 and GA2ox7 (related to GA metabolism), which promoted the conversion of GA 1 and GA 4 to inactive gibberellin, while plants treated with the inhibitor of MGBG exhibited the opposite effects on the expression of GA anabolic pathways (Fig. 6). Furthermore, qPCR analysis showed that the GA receptor gene GID1 was significantly up-regulated under Spd treatment ( Fig. 6).
Interestingly, Spd enhanced the activity of GA3-oxidase, GA20-oxidase and GA2oxidase, while MGBG had the opposite effect (Fig. 7). These results suggested that exogenous Spd could promote the biosynthesis of GA through regulation the gene expression and enzyme activity of GA anabolic pathways.
To determine the role of Spd and GA on the regulation of GT-3b, we analyzed the effects of Spd and GA on the expression of GT-3b using qPCR. As shown in Fig. 8a, the transcript level of GT-3b increased by 6.81-fold in Spd-treated plants, but decreased by 25.73% in MGBG-treated plants, in comparison to CK plants. In addition, GA 3 upregulated the expression of GT-3b, while application of paclobutrazol (PAZ), an inhibitor of GA biosynthesis, significantly inhibited the expression of GT-3b (Fig. 8b). Therefore, Spd promoted the accumulation of GA in plants and GA could induce the expression of GT-3b, suggesting that the regulation of GT-3b by Spd might mediate by GA.

11
To investigate whether GA mediated Spd-induced salt tolerance, we first analyzed the effect of Spd on the content of GA 3 under salt stress. The content of GA 3 was significantly increased both in S and Spd+S plants, but the increased level was more profound in Spd+S plants (Fig. 9). In contrast, the content of GA 3  To further determine the role of GA 3 in Spd-mediated salt tolerance, we analyzed the effect of GA 3 , PAZ and Spd on the growth of cucumber seedlings under NaCl stress. The results showed that the growth of cucumber seedlings subjected to salt stress was significantly inhibited and the inhibition was alleviated after spraying GA 3 (Fig. 10). However, the adverse effects of salt stress were more serious when plants treatment with PAZ (Fig. 10). The growth parameters in plants treatment with PAZ and Spd were slight higher than the single PAZ-treated plants under salt stress ( Fig. 10). Therefore, these results indicated that GA mediated Spd-induced salt tolerance in cucumber seedlings.

Discussion
A large number of studies have reported that exogenous Spd can improve the tolerance of plants to salt stress [26][27][28]. Exogenous Spd could enhance the accumulation of proline and antioxidant enzymes to increase tolerance to salt stress [26][27][28]. Spd reduces the content of MDA, increases the content of chlorophyll, and regulates PSII activity in wheat seedlings under salt stress [29]. Spd inhibits stearic acid, linoleic acid and linolenic acid in salt-stressed plants, and reduces the total free fatty acids and the contents of JA and SA to improve salt stress tolerance [30].
In addition, exogenous Spd significantly increases the contents of free Spd and Spm [29]. It is found that MGBG enhances the adverse effects of salt stress [31]. In this study, the effects of exogenous Spd and MGBG on the growth and physiological indicators of resistance in cucumber seedlings under salt stress were analyzed, and the results were in accord with the results of previous studies. Wang et al. [32] described the molecular mechanism of exogenous Spd on improving hightemperature tolerance based on the expression of miRNA in cucumber. Sang et al. [33] used proteomics studies to analyze the effects of Spd on alleviating hightemperature stress in tomato. The molecular mechanism of Spd on enhancing hightemperature tolerance has been studied via transcriptomics and proteomics.
However, the molecular mechanism of Spd on improving salt tolerance in plants is poorly understood. Here, we found that Spd increased salt tolerance through promoted the accumulation of GA, which further induced the expression of GT-3b in cucumber.

Spd induces the expression of salt-responsive gene GT-3b
GT-3b is a salt stress protein induced by salt stress and pathogens [10]. We found that Spd significantly increased the expression level of GT-3b gene (Fig. 2). A study found that the AtGT-3b protein specifically binds to the promoter of SCaM-4 gene to confer NaCl responsiveness [10]. Our previous study found that the GT-3b transcription factor specifically binds to the salt response element GT-1 in the promoter of CsSAMs to regulate its expression [11]. Studies have shown that in Lycoris [34], wheat [15] and tomato [12], transcription of the SAMs gene is elevated during salt stress. Exogenous PAs can enhance the metabolism and accumulation of endogenous PAs [35,36]. Wu et al. [37] indicated that exogenous Spd can induce 13 endogenous PAs production under salt stress. In the present study, we found that Spd induced the expression of GT-3b under salt stress (Fig. 2). Thus, exogenous Spd may upregulate the expression of SAMs under salt stress via GT-3b. The accumulation of SAMs promotes the synthesis of PAs [38], which is beneficial for the enhancement of the tolerance of plants to salt stress [39,40].

Endogenous GA regulates the transcription level of GT-3b
The predicted stress-related cis-elements may be necessary for the regulation of stress-induced gene expression [1]. In this study, we found that the 1278 bp promoter fragment of GT-3b contained the TGACG and CGTCA motifs associated with MeJA signaling, the GARE motif involved in responses to GA and the TCA element involved in SA responses [41]. Interestingly, GA, SA, and MeJA can increase the degree of GUS staining and enhance GUS activity (Fig. 4), indicating that the GA, SA, and MeJA binding motifs in the GT-3b promoter might play an important role in its response to exogenous GA, SA and MeJA.
It has been reported that transcription factors are regulated by multiple phytohormones. Some NAC transcription factors are affected by auxin [42], ethylene [43], abscisic acid [44], GA [45], MeJA [46], SA [47] and cytokinin [48]. These phytohormones participate in signal transduction. The phytohormone-related motifs may play a vital role in regulating the expression of GT-3b when bound by the corresponding phytohormones [49]. In this study, we treated cucumber seedlings with GA 3 , SA and MeJA and found that the relative expression level of GT-3b gene was upregulated (Fig. 3). Furthermore, GA 3 could rapidly induce the expression of the GT-3b gene, indicating that GA, SA and MeJA can regulate the transcription level of GT-3b. By measuring the contents of endogenous phytohormones, we found that only the content of GA 3 increased after Spd treatment (Fig. 5). qPCR analysis found that GA 3 can upregulate the expression level of GT-3b. These results indicate that Spd regulates GT-3b expression and is most closely related to GA.

Spd promotes the accumulation of endogenous GA under salt stress
It has been reported that Spd increases the concentration of GA in white clover under drought conditions [50] and regulates the changes in endogenous PA and phytohormones to reduce the structural and functional destruction of photosynthetic organs [51]. Similarly, we found that exogenous Spd significantly increased the content of endogenous GA 3 under salt stress, upregulated the expression of genes related to GA biosynthesis, enhanced the activity of GA3oxidase and GA20-oxidation, and decreased the activity of GA2-oxidase associated with GA degradation (Fig. 6,7,9). Taken together, our results indicated that Spd may regulate the GT-3b transcription factor by enhancing the accumulation of GA. conditions. When the third leaves of the cucumber seedlings were fully expanded, the following treatment was carried out.

Salt stress and Spd treatment
To investigate the effect of Spd on the alleviation of salt stress injury in cucumber and the expression of GT-3b, cucumber seedlings were divided into two groups; one group was subjected to salt stress via irrigation with 50 ml 100 mM NaCl solution, and the other group was irrigated with the same amount of water. At the same time, replicates per treatment and 10 seedlings per replicate. The samples were collected at the indicated time, frozen in liquid nitrogen and stored at -80 °C for further analysis. After treatment for 7 d, the growth parameters were measured by the method of Wang et al. [32] and Tang et al. [52].

Phytohormone treatment
To clarify the characteristics of the response of GT-3b to phytohormones, 100 μM GA 3 , SA and MeJA solution were sprayed on both sides of cucumber leaves. The same amount of deionized water was sprayed as a control. To verify whether the GT-3b promoter responds to phytohormones, the tobacco plants at the 5-leaf stage were injected with an Agrobacterium tumefaciens promoter solution and sprayed with 100 μM GA 3 , SA and MeJA, respectively. After treatment for 1 d, tobacco leaves were collected for GUS chemical tissue staining and GUS fluorescence quantitative analysis.

Promoter cloning and sequence analysis
The GT-3b promoter sequence was obtained from the NCBI database (GenBank No: NC_026657.1). The 1278 bp sequence upstream of the start codon (ATG) of GT-3b was cloned from cucumber genomic DNA for further analysis. The online search tools PlantCARE (http:// bioinformatics.psb.ugent.be/webtools/plantcare/html) and PLACE (http://www.dna.affrc.go.jp/PLACE/signalscan.Html) were used to predict the cis-acting regulatory elements.

Plasmid construction and transient tobacco transformation
On the basis of the putative cis-acting elements responsible for phytohormones, four different sequences of varying lengths from the GT-3b promoter fragments were amplified using PCR with the specific primers (Additional file 4: Table S2) and inserted into the pBI121 vector, and the plasmids were named P1, P2, P3 and P4, respectively. All constructs were introduced into the Agrobacterium tumefaciens strain EHA105 using electroporation after sequencing. The activated Agrobacterium tumefaciens was injected in the tobacco leaves. The activated Agrobacterium tumefaciens with the pBI121 vector was used as a positive control, and the infiltration medium was used as a negative control.

Histochemical staining and fluorometric analysis of the GUS enzyme
The injected tobacco leaves were stained in a staining solution (25 mg mL -1 X-gluc, 50 mM sodium phosphate, pH 7.0, 10 mM Na 2 EDTA, pH 8.0, 0.1% Triton X-100, 0.5 mM potassium ferricyanide, and 0.5 mM potassium ferricyanide) overnight at 37 °C to detect the activity of different promoter fragments according to the method described by Jefferson et al. [53]. After staining, the tobacco leaves were bleached three times with 75 % (v/v) ethanol in a water bath at 95 °C to remove the chlorophyll. Then, we photographed the tobacco leaves with a Leica DM2500 fluorescence microscope. The fluorometric GUS assays were measured according to the method described by Wang et al. [11].

Total RNA extraction and qPCR analysis
Total RNA was extracted from the leaves of the cucumber seedlings using the RNA  Table S2. The gene expression fold analysis was performed using the 2 − ΔΔCt method [54].

Quantification of endogenous GA 3 , SA and MeJA
For determination of the contents of GA 3 , SA and MeJA, approximately 1.0 g of leaves was ground in liquid nitrogen and extracted three times with 4 ml precooled 80% (v/v) chromatographic methanol. The homogenate was placed on ice for 12 h in the dark and then centrifuged at 10,000 g for 15 min at 4 °C, and the supernatant was stored in a refrigerator at 4 °C in the dark. The supernatant was added to 1.0 g PVPP and then shaken in a shaker at 4 °C for 1 h in the dark. After centrifugation at 10,000 g for 15 min at 4 °C, the supernatant was passed through a C18 extraction cartridge (Waters, Ireland). The filtered liquid was stored in a 50 ml centrifuge tube and dried in a vacuum for 3 d until there was no liquid at the bottom. Then, we added 1 ml precooled chromatographic methanol to the centrifuge tube under low light conditions to completely dissolve the phytohormone as much as possible. The solution was filtered into a brown sample vial with a 0.45 μm organic ultrafiltration membrane, and the sample was stored at 4 °C in the dark. The high-performance liquid chromatography (HPLC) instrument used in the experiment was the Waters 1525 system (UPLC H-Class, USA). GA 3 , SA and MeJA standards were purchased from Sigma, and chromatographic methanol was obtained from Tedia.

Measurement of GA oxidase activity
For enzymatic activity assays, 0.2 g of the cucumber leaves were ground in phosphate buffer (pH 7.4) and the homogenates were centrifuged at 4 °C 4,000 g for 15 min. The GA3, GA2 and GA20 oxidase activity were measured with the plant GA3-oxidase, GA2-oxidase and GA20-oxidase ELISA Assay Kit (Sino Best Biology Technology, China) according to the manufacturer's instructions.

Determination of the malondialdehyde and proline contents
To determine the content of malondialdehyde (MDA), 0.2 g of cucumber leaves were ground in 0.1% trichloroacetic acid (TCA), and the homogenates were centrifuged at 6,000 g for 10 min. Then, 2 mL of 0.6% thiobarbituric acid (TBA) was added into the 2 mL supernatant. The mixture was heated in a water bath at 100 °C for 10 min and then recentrifuged at 6,000 g for 10 min after cooling. The absorbance of the supernatant was read at 532 nm and 600 nm, and the MDA concentration was calculated as previously described [55].
To determine the content of proline, 0.2 g of cucumber leaves were ground and 20 placed in a 15 ml centrifuge tube, to which was added 5 ml 3% sulfosalicylic acid.
The mixture was heated in a water bath at 100 °C for 10 min and then recentrifuged at 3,000 g for 10 min after cooling. Then, 1 ml of supernatant, 1 ml water, 1 ml of glacial acetic acid and 2 ml of ninhydrin acid solution was added into a 15 ml centrifuge tube and heated in a boiling water bath for 60 min. When the mixture was cooled, we added 4 ml toluene into the centrifuge tube and vortexed for 30 s.
The absorbance of the supernatant was read at 520 nm, and the proline concentration was calculated using the method described by Vicente et al. [56].

Statistical analysis
All of the data are presented as the mean ± SE. Significant differences (P < 0.05) between treatments were determined using Duncan's multiple range tests in SPSS   Table S1. Putative cis elements in the promoter of CsGT-3b. Additional file 4: Table S2. Sequence of primers used in this study.