Brassica napus cv. ZhongShuang 6 (ZS6), which is an elite Chinese cultivar, and the B. napus transgenic plants were grown in Pindstrup soil mix (Denmark) or 1/2 MS medium in a greenhouse at 22°C with 16 h light/8 h dark cycle, 40–65% humidity, and light intensity of 8,000 lux. Mature seeds of these plants were collected for experimental use.

The full-length BnSIP1-1 cDNA was isolated by PCR amplification of cDNA prepared from the mRNA of ZS6 leaves. The primers were listed in Supplementary Table S1. The resulting sequences and the predicted BnSIP1-1 protein sequence were analyzed using BLAST software¹. The protein structure regions were predicted by PSIPRED 3.3² (Sadowski and Jones, 2009) and the multiple protein alignment was performed using CLUSTALW2³ (Li et al., 2013).

The coding sequence of BnSIP1-1 was amplified using primers BnSIP1-1(3; Supplementary Table S1) for transient expression in onion epidermal cells. The PCR products were subcloned into PUC-35S-GFP expression vector under the control of the CaMV 35S promoter. The PUC-35S-GFP vector was transiently expressed in onion epidermal cells using a gene gun (PDS-1000, Bio-Rad, USA). Fluorescence was observed using a fluorescence microscopy (DM2500, Leica, Germany) after incubation at 25°C for 18 h on MS medium. Fluorophores were excited using an argon laser at 488 nm (GFP), and bright-field images were collected using a transmitted light detector.

To generate the transgenic plants, the CDS of BnSIP1-1 was amplified from cDNA isolated from ZS6 leaves using BnSIP1-1(2) primers (Supplementary Table S1) by reverse transcription PCR, and then subcloned into the PBI121S vector under the CaMV 35S promoter and a terminal poly A sequence (Liu et al., 2014). The 35S:BnSIP1-1 recombinant vector was introduced into Agrobacterium tumefaciens GV3101 by electroporation, and positive clones were selected on LB agar plates at 28°C, supplemented with appropriate concentrations of antibiotics (gentamicin 50 mg L^-1, rifampicin 50 mg L^-1 and kanamycin 50 mg L^-1) and verified using PCR. A single positive colony was used to transform B. napus (ZS6). The transgenic B. napus plants were generated by agrobacterium-mediated method as previously reported (Liu et al., 2014). In total, there were 22 independent T₀ generation of transgenic lines generated, and 77% were positive transformants as confirmed by PCR with a forward primer designed from the CaMV 35S sequence and a reverse primer designed from the BnSIP1-1 sequence. Positive T₃ generation transgenic plants were grown for phenotypic identification.

For the salt, osmotic or ABA stress tolerance assay of the transgenic B. napus during germination period, the wild-type and T₃ transgenic B. napus seeds were planted on plates filled with three-layer filter paper containing 8 ml of mannitol (400 and 600 mM), NaCl (150 and 200 mM), ABA (50 μM ABA and 100 μM ABA) or distilled water without any additions for 4 days. The germination rate was calculated every 12 h. Observations were made and images were acquired using a camera.

For the salt and osmotic stress assay of the transgenic B. napus during seedling period, the wild-type and transgenic seeds were germinated on wet filter paper for 7 days, and then transferred to liquid 1/2 MS medium under a 16 h light/8 h dark cycle at 25°C for 8 days. Finally, the 15-day-old seedlings were transferred to liquid 1/2 MS supplemented medium with NaCl solution (150 or 200 mM) or mannitol solution (300 or 600 mM). The control seedlings were transferred to liquid 1/2 MS medium supplemented without any additions. Photographs were taken after 6 h of treatment with 600 mM mannitol and 9 days of treatment with 300 mM mannitol. The chlorophyll contents of the wild-type and transgenic seedlings were detected after 9 days of treatment with 300 mM mannitol. Total chlorophyll was isolated using an extraction solution (2:1 acetone:95% alcohol), and the chlorophyll a and b contents were measured at 646.6, 663.6, and 750 nm using a spectrophotometer (Lambda, PerkinElmer, USA) according to previously described methods (Porra et al., 1989; Mishra et al., 2008).

For the ABA sensitivity assay during seedling period, the wild-type and transgenic B. napus seeds were sown in liquid 1/2 MS medium or soil with or without 1 μM ABA or 10 μM ABA. Photographs were taken after 12 days of plating on liquid 1/2 MS medium or 6 days of sowing in soil (Travaglia et al., 2010). The length of stem and root, and fresh weight of aboveground part and root of the wild-type and transgenic plants were measured after 12 days of treatment.

RNA extraction and the first-strand cDNA synthesis were performed using an RNAprep Pure Plant Kit (TIANGEN BIOTECH, Beijing, China) and FastQuant RT Kit (with gDNase; TIANGEN BIOTECH, Beijing, China) according to the manufacturers’ instructions. Real-time PCR was performed using fivefold diluted cDNA and a SYBR Green I kit (Bio-Rad, Hercules, CA, USA) on the CFX96 real-time PCR platform (Bio-Rad, Hercules, CA, USA). Three independent biological replicates and three technical replicates for each biological replicate were run and the significance was determined through t-test of SPSS statistical software (p < 0.05). The B. napus β-actin gene (accession No. AF111812.1) was used as an internal reference control. Quantitative analysis was performed using the ΔΔCt comparative method (Livak and Schmittgen, 2001).

For the tissue specific analysis, the roots, leaves, and stems were harvested from randomly selected 2-month-old plants; and the pollens, stigmas, petals and immature seeds were collected from plants during reproduction period. For exploring the response of BnSIP1-1 to different abiotic stresses, the leaves and roots of 15-day-old seedlings (under 50 μM ABA, 200 mM NaCl, or 300 mM mannitol for 0, 1, 2, 4, 8, 12, and 24 h of treatments) were harvested for QPCR analysis. For detection of BnSIP1-1 expression of transgenic B. napus, the leaves of different independent transgenic lines were harvested. The leaves of 15-day-old transgenic and wild-type seedlings (under normal condition or treated with 50 μM ABA, 200 mM NaCl, or 300 mM mannitol for 0, 1, 6, 48, and 72h) were harvested for detection of the expression of stress-related genes (BnRD29A, BnERD15, BnLEA1, BnNAC485, BnCIPK6, BnSOS1, BnNHX1, BnKIN1, BnHKT, and BnABI5). The corresponding specific primers were listed in Supplementary Table S1.

For ABA content measurement, the leaves of 15-day-old transgenic and wild-type seedlings (under normal condition or treated with 50 μM ABA, 200 mM NaCl, or 300 mM mannitol for 0, 1, 6, and 24 h) were harvested. Quantification of endogenous ABA was performed as described (Chen M. et al., 2012).

All of the expression, phenotypic and quantitative experiments have been replicated three times and data have been presented as mean ± SD (standard deviation). Significant variations between genotypes or treatments were evaluated statistically by Student’s t-test or one-way ANOVA (Sokal and Rohlf, 1995). Mean values that were significantly different (p < 0.05) from each other are marked with asterisks or different lowercase letters inside the figures.

The full-length gene sequence of BnSIP1-1 was amplified from B. napus to analyze the structural properties. This analysis suggested that BnSIP1-1 has only one exon. The ORF of BnSIP1-1 is 924 bp long and encodes a predicted protein of 307 amino acids. Its predicted molecular mass and isoelectric point were 34.6 kDa and 9.30, respectively. Similarity searches revealed that BnSIP1-1 has 100% overall amino acid identity with a predicted gene product having an unknown function (accession No. CDX73485) in B. napus, 96% identity with an uncharacterized protein (accession No. XP_009116124) from B. rapa, 83% identity with an unknown protein (accession No. XP_006403581.1) from Eutrema salsugineum, and 26 and 19% identity with Arabidopsis ASIL1 (accession No. NP_564648.1) and ASIL2 (accession No. OAP02717.1), respectively. Analyses of the helix structure of the predicted protein suggested BnSIP1-1 protein had the classical conserved secondary structure of the SIP1 clade of the trihelix family, including the N-terminal trihelix domain, the fourth amphipathic α-helix, and the long, uninterrupted α-helical (Figure 1). Moreover, the three conserved amino acids in the three individual amphipathic α-helices of the N-terminal trihelix were also observed in BnSIP1-1, including two tryptophans (W) and one isoleucine (I). Overall, these results suggested that the BnSIP1-1 gene isolated in this study was a member of the SIP1 clade of the trihelix family.

To further confirm the localization of the BnSIP1-1 gene product, the BnSIP1-1 gene was cloned the downstream of the CaMV (constitutive Cauliflower mosaic virus) 35S promoter of pCAMBIA1301-GFP vector and upstream of the GFP gene to create the 35S:BnSIP1-1-GFP fusion construct. The positive colonies were identified by PCR (Supplementary Figure S1). The BnSIP1-1-GFP fusion protein was transiently expressed in onion epidermal cells. Confocal microscopy analysis revealed that the BnSIP1-1 gene product was localized exclusively in the nucleus of cells, whereas the GFP control was found throughout the entire cell (Figure 2).

The trihelix family genes play important roles in different development processes and growth conditions (Kaplan-Levy et al., 2012). To determine the expression pattern of BnSIP1-1 in different B. napus tissues, QPCR was conducted using mRNA which was isolated from different tissues of the wild-type. The results showed that BnSIP1-1 was expressed in all of the examined tissues, including the roots, stems, leaves, pollens, stigmas, petals and immature seeds, with high expression levels in the roots, leaves, pollen and petals (Figure 3).

In order to explore the response of the BnSIP1-1 gene to various abiotic stresses, QPCR was used to analyze the BnSIP1-1 gene expression pattern in leaves and roots of 15-day-old seedlings under ABA, NaCl, and mannitol treatments (Figure 4). As shown in Figure 4A, the BnSIP1-1 transcript was remarkably enhanced in the leaves by ABA treatment, which amounted to an approximate threefold increase at the 1 h time point relative to the untreated control plants and reached the highest expression (an approximately fourfold increase) after 12 h of ABA treatment. Under NaCl treatment, BnSIP1-1 gene expression in the leaves increased gradually and also showed the highest level of expression (an approximately threefold increase) at the 12 h time point. By contrast, BnSIP1-1 gene expression in the leaves under mannitol treatment showed a rapid and transient approximately fourfold increase at the 1 h time point and then declined over time. The BnSIP1-1 gene expression eventually was 40% lower than the control plants after 24 h of osmotic stress treatment. The expression pattern of BnSIP1-1 was similar in the roots as in the leaves under salt stress. However, the expression pattern of BnSIP1-1 varied in roots under ABA and mannitol stress. BnSIP1-1 expression in the roots was up-regulated and reached its highest level (an approximately threefold increase) after 2 h of ABA treatment. Under mannitol treatment, the expression of BnSIP1-1 in the roots was comparatively higher in the first 12 h of treatment and then decreased to a similar level as the control plants after 24 h of treatment.

To determine whether BnSIP1-1 played a role in the abiotic stress response, 17 independent transformants of the B. napus that OE BnSIP1-1 under the control of the CaMV 35S promoter were generated. Total RNA from young leaves of independent transgenic lines was extracted for QPCR. The results showed that the BnSIP1-1 gene was highly expressed in all of the positive lines (Figure 5A), and two lines (OE-7, OE-27) were chosen for further analysis. Fifteen-day-old seedlings were treated with salt (150 mM or 200 mM NaCl) or osmotic (300 mM or 600 mM mannitol) stress to analyze their tolerance to abiotic stresses. Compared to the wild-type seedlings, the transgenic lines showed enhanced tolerance to 600 mM mannitol after 6 h of treatment (Figure 5B), but no obvious differences were observed after salinity stress (data not shown). The leaves of 15-day-old transgenic seedlings treated with 300 mM mannitol for 9 days looked greener than the wild-type seedlings of the same age (Figure 5C). Therefore, we further determined the chlorophyll content of the leaves of the mannitol-treated seedlings. The chlorophyll content of the OE-7 and OE-27 transgenic B. napus leaves were 49.6 and 55.6% higher than that of the wild-type leaves, respectively (Figure 5D), which indicated that chlorophyll was reduced at a slower rate in the transgenic lines than in leaves of the wild-type seedlings after osmotic stress.

The expression pattern of BnSIP1-1 under ABA and abiotic stresses suggested that it had a significant role in ABA signaling, which prompted us to determine its in planta functional characterization. When plated and grown on MS medium, the BnSIP1-1 transgenic plants grew almost as same as the wild-type plants (Figures 6A,B). However, when cultured in MS medium or soil with 1 or 10 μM ABA for several days, the BnSIP1-1 OE seeds were less sensitive to ABA than the wild-type seeds (Figures 6A,B). In the absence of ABA, the root length and weight were similar in the wild-type and transgenic plants. However, under 10 μM ABA treatment, the stems of the transgenic plants were 35–45% longer than those of the wild-type plants (Figure 6C). There was no significant difference in the stem length between the wild-type and transgenic seedlings when under 1 μM ABA treatment. The root of the transgenic plants was 22–38% and 160–224% longer than the wild-type plants under 1 and 10 μM ABA, respectively (Figure 6C). At the same time, the fresh weight of the transgenic roots was 21–37% and 116–163% higher than the wild-type plants under 1 and 10 μM ABA treatments, respectively (Figure 6D). The fresh weight of the aboveground part of the transgenic plants was 24–28% and 142–185% higher than the wild-type plants under 1 and 10 μM ABA, respectively (Figure 6D). These data suggested that overexpression of the BnSIP1-1 gene caused reduced sensitivity to ABA in B. napus seedlings.

The soil culture experiments of above mentioned dropped a hint that the germination rate of transgenic B. napus seeds was higher than that of the wild-type seeds under ABA treatment. To systematically study the effect of BnSIP1-1 on the germination process, we sowed the wild-type and transgenic T₃ seeds on plates containing distilled water, mannitol, NaCl, and ABA for 4 days. Without any addition, there were no differences between the wild-type and transgenic seeds. However, when mannitol, NaCl or ABA were applied, the transgenic seeds germinated much better than the wild-type seeds (Figure 7). Under lower concentrations of mannitol (400 mM) or NaCl (150 mM), the germination rate of the transgenic seeds could reach higher than 80%, while the germination rate of the wild-type seeds was lower than 40% (Figures 7A-b/c,B-b/c) after 96 h of treatment. Under higher concentrations of mannitol (600 mM) or NaCl (200 mM), the germination rate of the transgenic seeds was approximately 30%, while the germination rate of the wild-type seeds was only about 10% (Figures 7A-e/f,B-e/f) after 96 h of treatment. In term of the ABA treatment, the transgenic seeds germinated faster than wild-type under both 50 μM ABA and 100 μM ABA treatment (Figures 7A-d/g,B-d/g). The increase of ABA concentration had a greater impact on the germination of the wild-type seeds than the transgenic seeds.

ABA is an extensively studied hormone that plays essential roles in stress responses of plants. Many kinds of abiotic stress promote ABA synthesis, which further induce stomatal closure and up-regulation of the expression of ABA-responsive genes, thereby contributing to plant stress resistance (Finkelstein et al., 2002).

To study whether OE BnSIP1-1 affect ABA accumulation, we detected endogenous ABA content in plants (Figure 8). All of the stress treatments, including ABA, mannitol and NaCl, can induce significant accumulation of ABA both in the wild-type and the transgenic plants. In addition, endogenous ABA level was 51% higher in transgenic plants than the wild-type during normal growth condition. Under exogenous ABA and mannitol treatment, the accumulation of ABA in the transgenic plants was higher than that in the control plants (Figures 8A,B). Under salt stress, the difference of ABA content before treatment was gradually decreased with the prolongation of salt treatment time, then after 24 h of treatment the ABA level in the transgenic plants became similar to the wild-type plants, which showing opposite change trend with ABA or mannitol treatments (Figure 8C).

The BnSIP1-1 may function in response to stress and ABA through the regulation of downstream genes. To broaden our knowledge about the molecular mechanism of ABA-insensitivity and stress-tolerance in BnSIP1-1-OE plants, we first analyzed the expression level of related genes, including BnRD29A, BnERD15, BnLEA1, BnNAC485, BnCIPK6, BnSOS1, BnNHX1, BnKIN1, and BnHKT by QPCR in the wild-type and transgenic plants under normal conditions (Figures 9A,B). These genes were categorized into two groups: the expression-increased group (Figure 9A) and expression-unchanged/decreased group (Figure 9B). As shown in Figure 9A, the expression of the transcription factor BnNAC485 had the most significant increase in BnSIP1-1 transgenic plants, being approximately 14-fold higher than that in the wild-type plants. Similarly, the expression of BnRD29A was increased about 11-fold higher level in the transgenic lines than in the wild-type plants. Furthermore, the expression levels of BnERD15, BnLEA1 and BnCIPK6 were also up-regulated 3- to 5-fold in the BnSIP1-1-OE plants. However, there was no significant difference in the expression levels of BnSOS1, BnNHX1, BnABI5 and BnKIN1 between the transgenic and wild-type plants, as shown in Figure 9B. In addition, the expression level of BnHKT declined by 68% in the transgenic plants compared with the wild-type under normal growth conditions.

BnNAC485 and BnCIPK6 are two rare rapeseed genes which were experimentally identified to play significant roles in ABA signaling and stress responses (Chen L. et al., 2012; Ying et al., 2014). ABI5 is a basic leucine zipper protein which has already been identified to have important function in the ABA signaling pathway (Piskurewicz et al., 2008). Therefore, to further expand our understanding of regulation mechanism of BnSIP1-1, we compared the expression level of BnNAC485, BnCIPK6 and BnABI5 in the wild-type and transgenic plants treated by ABA, mannitol and NaCl stresses for different time periods. After 1 h of ABA treatment, the expression levels of BnNAC485 and BnCIPK6 were about 40-fold and 16-fold higher in the transgenic plants than in the wild-type plants, respectively. As the ABA-treatment time increased to 6 h, the high-expression level of BnNAC485 remained steady, while the expression level of BnCIPK6 was further increased to 27-fold in the transgenic plants compared to the wild-type plants. There were no significant differences in the expression levels of BnNAC485 and BnCIPK6 under osmotic or salt stresses between the transgenic and wild-type plants. Notably, the expression level of BnABI5 did not present any difference between the transgenic and wild-type plants under all three of the stress conditions.