A novel EST clone (Sal-C-64; EB484712) was used to design primers (Table S1). The gene was converted to full length using RACE (rapid amplification of cDNA ends) and was sequenced and analyzed. The NCBI (National Center for Biotechnology Information) database was used to search for nucleotide and protein homologs. The secondary structure was predicted using ExPASy tools. Amino acid sequences deduced from nucleotide sequences were imported to the ProtParam tool of the ExPASy server (Artimo et al., 2012) for primary analysis and the PSIPRED server (Buchan et al., 2010) for the prediction and analysis of the secondary structure. The functional activity of SbSRP was predicted using TrSSP (Mishra et al., 2014). The amino acid sequences were subjected to BLAST (Basic Local Alignment Search Tool) and compared with the Protein Data Bank (PDB) and Conserved Domain Data bank (CDD). The phylogram study was accomplished using the Maximum Likelihood (ML) statistical method based on the JTT matrix-based model using Molecular Evolutionary Genetics Analyses version 6 [MEGA6] (Tamura et al., 2013).

For transcript profiling, 1-month-old seedlings were transferred to a hydroponic culture medium containing ¼ strength MS basal medium and grown in vitro with 8/16 h dark/light cycle at 25°C for 15 days. Different NaCl (0.05, 0.10, 0.25, 0.50, and 1.00 M) stress treatments were given to plants for 24 h by transferring plants grown under equivalent condition to new hydroponic culture media (¼ MS containing different NaCl concentrations). In the second set of experiments, different abiotic stresses (salt, 250 mM; desiccation; heat, 45°C; cold, 4°C) were applied for different time periods (2, 6, 12, 24 h) to plants. Total RNA was isolated from control and stressed plant samples using the GITC method (Chomczynski and Sacchi, 1987) and quantified with a Nanodrop spectrophotometer (NanoDrop, USA). cDNA was prepared using a ImpromII reverse transcriptase first strand cDNA synthesis kit (Promega, USA). The quantitative real-time PCR (qRT-PCR) reactions were carried out with gene-specific primers and β-tubulin was used as the internal reference gene (Table S1). At the end of the PCR cycles, products were retained through a melt curve analysis to check the specificity of PCR amplification in qRT-PCR. The amplified products were checked on 1% agarose gel to confirm the expected size. The qRT-PCR data were analyzed using the comparative C_T method, and the relative fold-gene expression (2^−ΔΔC_T) of the SbSRP gene in stressed plants in comparison to control plants (without treatment) was determined after normalizing with internal control β-tubulin C_T-values (Livak and Schmittgen, 2001; Schmittgen and Livak, 2008).

A translational fusion cassette of SbSRP along with RFP (red fluorescent protein) was made to study subcellular localization. The SbSRP open reading frame (ORF) was amplified, cloned into pENTER/D-TOPO (Invitrogen, USA) and sequenced (Table S1). An LR recombination reaction was performed between an attL-containing pENTER/D-TOPO-SbSRP recombinant vector and an attR-containing pSITE-4CA (RFP) destination vector using LR Clonase II enzyme mix (Invitrogen, USA). Positive clones were selected, confirmed by PCR and transferred to onion epidermal cells using a gene gun (PDS-1000/He Biolistic, Biorad, USA). Vector pSITE-4CA:RFP was also transformed and used as a control. Transformants were incubated in the dark for 12 h and the transient expression of RFP was observed using an epifluorescence microscope (Axio Imager, Carl Zeiss AG, Germany).

The SbSRP was amplified and the gene copy number was determined using DNA gel blot analysis (Joshi et al., 2011; Table S1). Genomic DNA of Salicornia was extracted using a modified CTAB (N-cetyl-N, N, Ntrimethyl ammonium bromide) method (Bubner et al., 2004) and was quantified (ND-1000, Nanodrop Technology, USA) and digested with different restriction enzymes (EcoRI, SacI, SmaI, and HindIII). Digested DNA was transferred to a Hybond N⁺ membrane (Amersham Pharmacia, UK) using an alkaline transfer buffer (0.4 N NaOH along with 1 M NaCl). The blot was hybridized with PCR-generated DIG-11-dUTP labeled SbSRP probe overnight at 42°C (Table S1). Signals were obtained using CDP-Star as a chemiluminescent substrate and detected on X-ray film.

The SbSRP ORF was amplified (Table S1) and cloned in pCAMBIA1301 under the control of a 35S promoter via an intermediate pRT101 plant expression vector and the recombinant pCAMBIA1301:35S:SbSRP vector was mobilized into Agrobacterium tumefaciens (strain LBA4404). Tobacco (Nicotiana tabacum cv. Petit Havana) plants were transformed by the leaf disc method (Horsch et al., 1985) and transgenic plants (T0) were obtained using standard tissue culture methods. The T1 transgenic lines, obtained from T0 plants, were subjected to PCR and southern analysis for the confirmation of trans-gene (SbSRP) integration.

The total RNA was extracted using the GITC method from six transgenic lines (L2, L7, L10, L16, and L17), vector control (VC) and wild-type (WT) plants (abiotic stressed and unstressed) and quantified (ND-1000). The mRNA was converted to cDNA using Superscript II RT (Invitrogen, USA). The β-tubulin gene was used as an internal reference gene and resulting cDNA of both (SbSRP and β-tubulin) were amplified using PCR (Table S1). The semi-quantitative RT-PCR was performed three times independently and amplicons were observed through agarose gel electrophoresis. Three transgenic lines (L2, L7, and L16) showed better over-expression of the SbSRP gene and so were selected for further analysis.

The copy number of the SbSRP gene in transgenic lines was determined by DNA gel blotting (Joshi et al., 2011) as described above using gene-specific probes (Table S1). The T1 seeds of SbSRP over-expressing transgenic lines (L2, L7, and L16) along with VC and WT plants were germinated on MS media that contained NaCl (200 mM) or Polyethylene glycol (10% PEG) under controlled conditions. The seedling growth percentage was calculated using the following equation:- Percent seedling growth=(Number of seeds grown/                                                              Total number of seeds)×100

T1 seedlings that were uniform in size and germinated on MSB containing hygromycin (20 mg L⁻¹) were transferred to a hydroponic system with ½ strength MS media, incubated for 45 days, and then subjected to stress treatment (for 24 h under 200 mM NaCl or 10% PEG) for biochemical and physiological analysis.

Leaf segments of 45 day old treated and untreated plants were collected and incubated overnight at −20°C. The next day, plant samples were thawed and centrifuged at 13,000 rpm for 10 min at 4°C. The cell sap extract was used to determine the cellular osmotic potential of the sample using a vapor pressure osmometer (VAPRO, Wescor Inc., USA).

Leaf fresh weight was measured from treated and untreated, control (WT and VC), and transgenic plants, following overnight incubation of leaf samples in deionized water. The next day, turgid weight, and dry weight were determined after drying at a constant temperature (40°C). Relative water content (RWC, %) was calculated using the following equation: RWC (%)=100×[(Fresh weight−Dry weight)/                            (Turgid weight−Dry weight)]

Similar sized leaf discs (n = 10) from 45 day old control and transgenic plants were floated in ½ strength MS (control) media supplemented with NaCl (100, 150, or 200 mM) or PEG (10 or 20%) and incubated for 8 days under controlled conditions. The effects of salt and osmotic stress were examined by visually observing phenotypic alteration amongst different leaf segments. In addition, total chlorophyll content was calculated for leaf discs per gram of tissue fresh weight (Inskeep and Bloom, 1985).

Leaf segments (of a uniform size and weight) of 45 day old plants (treated and untreated, control and transgenic) were used to quantify the accumulation of superoxide radicals (after stress) using a 2,3-bis-(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide (XTT) assay (Hema et al., 2014). Leaf segments were kept in a potassium phosphate buffer (20 mM, pH 6.0) containing 500 μM XTT for 5 h and the increase in absorbance was measured at 470 nm. Likewise, leaf segments (of equal weight and size) of 45 day old treated and untreated, control and transgenic, plants were used to evaluate cell viability using a 2,3,5-triphenyltetrazolium chloride (TTC) assay (Hema et al., 2014). Leaf fragments were carefully rinsed in sterile water and kept in TTC solution in the dark for 6 h at room temperature (RT). Leaf samples were then boiled in 5 ml of 2-methoxy ethanol until dry, to release bound formazan. Then, 5 ml of 2-methoxy ethanol was added for the second time, and the change in absorbance was measured at 485 nm.

Healthy fresh leaves were harvested (from treated and untreated; control and SbSRP transgenic plants) and MSI was determined by the process explained by Sairam (1994). Leaf segments were kept in sterile water, and 1st set (L1) was stored at 40°C for 30 min where 2nd set (L2) was stored at 100°C for 10 min. Electrical conductivity (EC) of samples from both sets was measured, and MSI was calculated as MSI = 100 × 1−L1/L2.

For analyzing the electrolyte leakage, leaf samples were thoroughly rinsed with sterile water to remove surface bound electrolytes. Fresh leaf segments were stored in the sterile water in sealed tubes and incubated, overnight, on a rotating shaker. The next day, EC was determined (Lt) by a conductivity meter (SevenEasy, Mettler Toledo AG 8603, Switzerland). Then, leaf samples from the each treatment were autoclaved at 120°C for 15 min, subsequently, they were chilled to RT and EC (L0) was determined again (Lutts et al., 1996). The percentage EL was determined as EL(%) = 100 × Lt/L0.

The free proline and H₂O₂ content in the treated and untreated, control, and transgenic plants were calculated by the methods described by Bates et al. (1973) and He et al. (2005), respectively. A standard curve was generated with the known concentrations of proline and H₂O₂, an increase in absorbance was recorded at 520 and 560 nm, respectively. Lipid peroxidation was calculated by quantifying the level of MDA (malondialdehyde) produced by TBA (thiobarbituric acid) reaction (Draper and Hadley, 1990).

The total quantities of oligo and reducing sugars were determined by reference to a standard curve that was generated using glucose (Sigma-Aldrich, USA). The treated and untreated, control and transgenic leaf segments were crushed in 85% ethanol, following incubation with an anthrone reagent. All samples were boiled then cooled to room temperature, and absorbance was recorded at 620 and 540 nm. The level of reducing sugar was computed using a colorimetric test and the DNS method (Yemm and Willis, 1954). Total amino acids and polyphenol content were calculated by reference to a standard curve, which was generated by glycine and catechol solution, respectively (Pandey et al., 2015; Patel et al., 2016b). Fresh leaf samples were crushed in 85% ethanol, evaporated to dryness following re-dissolution in ninhydrin or the Folin-Ciocalteu reagent. Variation in color intensity was then analyzed at 570 or 650 nm.

The hydrogen peroxide, superoxide-radicals, and starch under salt (200 mM NaCl) or osmotic (10% PEG) stress were determined in vivo using a histochemical stain of 3,3-diaminobenzidine (DAB), nitro-blue tetrazolium (NBT), or potassium iodide (KI), respectively (Shi et al., 2010). DAB and NBT were prepared as 1 mg ml⁻¹ solutions in 10 mM phosphate buffer (pH 7.8), whereas KI (1 mg ml⁻¹) solution made in distilled water. Fresh leaf segments (of treated and untreated, control, and transgenic plants) were dipped in the freshly prepared stain solutions, kept in the dark for 2 h followed by exposure to white light (40 μmol m⁻² s⁻¹ spectral flux photon) for DAB (8 h) and NBT (1 h). For in vivo localization of starch, leaf segments were immersed in KI solution for 1 min. The blue or brown spots that appeared on the leaves indicated in vivo localization.

The plant samples (treated and untreated, control, and transgenic plants) were dried at 65°C for 72 h and their dry weights were determined. Plant samples were acid digested overnight in perchloric acid-nitric acid solution (3:1), heated to dryness following re-dissolution in sterile water. Ion content was determined using an inductively coupled plasma optical emission spectrometer (Optima2000DV, PerkinElmer, Germany).

Total RNA was extracted from control and treated plant samples (WT, VC, and transgenic lines) and cDNA was synthesized using a Superscript RT III cDNA synthesis kit (Invitrogen, USA). A real-time qRT-PCR analysis was performed for antioxidant and transcription factor genes of the host plant (NtSOD, NtAPX, NtCAT, NtDREB, and NtAP2). A melt curve analysis was also implemented to validate the specificity of the qRT-PCR reaction (Table S1). The housekeeping gene β-tubulin was used as a reference and the relative expression change (compared to WT and control condition) was calculated using the CT method (Livak and Schmittgen, 2001; Schmittgen and Livak, 2008).

Photosynthesis was computed by determining the rate of change of CO₂ over time in a leaf covered with a moderately big chamber, following the manufacturer's instructions (LI-6400XTv6, LI-COR Biosciences Inc., Nebraska). The treated (NaCl 200 mM and PEG 10%) and untreated plants of control and transgenic lines were kept under controlled conditions for 24 h and then chlorophyll fluorescence was computed. Photosynthesis, photochemical efficiency, and water use efficiency were investigated. All measurements were taken on four fully expanded leaves. The ratio of variable to maximum fluorescence was then determined as Fv/Fm = (Fm − F₀/Fm). The computed light was switched on and actinic light was given for 30 s at an intensity of ~1,000 μmol photons m⁻² s⁻¹ PAR (photosynthetically active radiation) and Fm′ was calculated.

Data from 3 replicates, each containing 15 plants, were documented for each set of experiments. Results are expressed as mean ± standard error. Statistical significance was determined by ANOVA (analysis of variance) amongst the means of WT, VC, and transgenic plants. A Tukey HSD test was used for comparisons and the differences were considered to significant when p < 0.05 and these differences are designated by different letters.

The full-length SbSRP gene (1262 bp; gene accession no. KX242114) contained a 5′ untranslated leader sequence (5′-UTR: 1–294 bases), an open reading frame (ORF: 753 bp; 295–1047), 3′-UTR (215 bp; 1048–1262) and a poly A tail of 19 base pairs (Figure S1). The SbSRP ORF (753 bp) encoded a protein of 250 amino acid residues (Figure S1) with an estimated molecular mass of 80–85 KDa and theoretical isoelectric point (pI) of 8.34. The total number of negatively charged residues (aspartate and glutamate) is 28 and there are 30 positively charged residues (arginine and lysine). The N-terminal of SbSRP is M (methionine) and its estimated half-life is 30 h for mammalian reticulocytes (in vitro), >20 h for yeast (in vivo), and >10 h for E. coli (in vivo). The instability index of the SbSRP protein was computed to be 34.52; this classifies the protein as stable according to ExPasy's ProtParam Tool (SIB, Swiss Institute of Bioinformatics). The TMpred program predicts membrane-spanning regions and their respective orientation (Figure S2). The prediction was made using an amalgamation of several weight matrices for the scoring of TMbase, a database of naturally occurring transmembrane proteins. The prediction of the topology of the membrane spanning domains revealed the presence of one transmembrane domain score from 25 to 44 at the N-terminal region (Figure S2). The predicted secondary structure of the deduced SbSRP protein contained topological arrangements of elements (coil, α-helices, and β-strand) with classical folding (Figure S2). The TrSSP program predicts that SbSRP is a transporter. The conserved domain database (CDD) categorized the deduced protein as part of the rubber elongation factor/stress related protein particle like superfamily (REF/SRPP). Phylogenetic analyses revealed that SbSRP showed proximity with the Beta vulgaris and Vitis vinifera hypothetical or predicted REF/SRPP like particle. The SbSRP gene sequence showed 78% identity with Beta vulgaris subsp. vulgaris predicted REF/SRPP like protein At3g05500 mRNA, 71% with Populus trichocarpa REF family protein, 71% with Populus trichocarpa unknown clone mRNA, 71% with Populus euphratica REF/SRPP like proteins At3g05500 mRNA, 70% with Ricinus communis REF like protein, 69% with Citrus sinensis predicted REF/SRPP like protein At3g05500 mRNA, 69% with Citrus clementina hypothetical protein mRNA, and 69% with Vitis vinifera predicted SRP like mRNA hence was considered uncharacterized.

The transient expression of RFP only and SbSRP:RFP (Figure 1A) in onion epidermal cells revealed that SbSRP is expressed in the plasma membrane whereas RFP is expressed uniformly throughout the cells (Figure 1B). The sequence analysis of the cDNA clone compared to the genomic clone revealed a single exon structure of SbSRP gene, confirming the intronless genomic organization of the gene. In the DNA gel blot analysis, EcoRI, HindIII, SacI, and SmaI-digested genomic DNA produced two intense bands; hence, the existence of at least two homologs of the SbSRP gene was depicted (Figure 1C).

The transcript profile of SbSRP was studied in Salicornia under salt, desiccation, heat, and cold stress using qRT-PCR (Figure S3). A 3.7-fold exponential increase was observed in the SbSRP transcript under 250 mM salt stress; when the salt concentrations was increased further to 500 and 1,000 mM, there was a reduction in the transcript level, and about 1.7 and 1.5-fold expression were detected respectively. Therefore, 250 mM NaCl was found to be the optimum concontration for SbSRP transcript expression. Consequently, plants were subjected to 250 mM salt stress for a range of durations from 2 to 24 h, and an exponential increase in the mRNA level of SbSRP from 2.5 to 6.5-fold was observed up to 12 h, thereafter expression decreaded and reached to 3.5-fold at 24 h compared to control condition. Under cold and desiccation stresses, relative expression of SbSRP transcript was maximum 8.6 and 3.8-fold, respectively at 12 h. Under heat, the SbSRP transcript reached a maximum of 25.63-fold at 24 h (Figure S3).

Putative transgenic lines (T1) were screened by germination on hygromycin (20 mg L⁻¹) and the presence of transgenes was confirmed by PCR amplification of uidA and hptII genes (Figures S4A,B). Acclimatized putative SbSRP over-expressing transgenic lines were confirmed for the stable integration of the transgene into the tobacco genome by DNA hybridization (Figure S4C). Amplification of the uidA and hptII genes was observed in all the transgenic lines (L1–L17) and the positive control, whereas no amplification was observed in WT (wild type, non-transgenic) plants. DNA gel blot analysis revealed that transgenic lines L3, L6, and L12 had two copies of the transgene, whereas L2, L7, L10, L16, and L17 clearly showed a single transgenic event, validating the single-copy integration of T-DNA into the tobacco genome. All other lines showed faint or unclear transgenic events while WT and VC plants did not illustrate any detectable hybridization signal. Therefore, the comparative ectopic expression analysis of lines L2, L7, L10, L16, and L17 was carried out by semi-quantitative RT-PCR and real-time (qRT) PCR (Figures S5A,B). Transgenic lines L2, L7, and L16 showed 4.6, 5.9, and 5.6-fold increases whereas L10, L12, and L17 showed about 3–4-fold increases under salinity stress. The SbSRP over-expressing transgenic lines L2, L7, and L16 exhibited higher expression than the others lines, and no amplification was observed from the cDNA of WT and VC plants. The stringency of the PCR was checked by amplifying the β-tubulin as the internal reference gene using the same cDNA and PCR conditions. An amplicon of ~300 bp was obtained in all transgenic lines and the WT and VC plants. Semi-quantitative RT-PCR and qRT-PCR additionally confirmed the stable integration and expression of the transgene in putative transgenic lines. Transgenic lines (L2, L7, and L16) and control plants (WT and VC) were selected for further analysis. The activity of the reporter gene, β-glucuronidase, in the leaf segment of transgenic, VC and WT plants was visualized, and efficient GUS activity was observed (Figure S5C).

The seeds from WT, VC, and transgenic lines (L2, L7, and L16) were harvested from the greenhouse, surface-sterilized, and germinated under different conditions. The SbSRP over-expressing transgenic lines L2, L7, and L16 showed ~75% seedling growth under NaCl and osmotic stress conditions (Figure 2). This seedling growth percent was significantly different (p < 0.05) to that observed in WT and VC, which exhibited ~54% seedling growth. After 3 weeks of salinity (NaCl 200 mM) and osmotic stress (mannitol 300 mM) treatment, transgenic lines illustrated phenotypically better tolerance than WT and VC plants. Likewise, growth parameters were significantly (p < 0.05) enhanced in SbSRP over-expressing transgenic lines compared to WT and VC plants under stress conditions (Figure 2).

After 7 days of incubation in control conditions, NaCl (100, 150, 200 mM) or PEG (10 and 20%), leaf discs of SbSRP over-expressing transgenic lines showed lower degradation of chloroplasts and stayed greener than WT and VC leaf discs (Figure S6A). The total chlorophyll (Chl; about 29.5 mg g⁻¹ Fw) and carotenoid (about 6.8 mg g⁻¹ Fw) content were almost comparable between control (WT and VC) and transgenic plants under normal (unstressed) conditions (Figure 3). After NaCl and osmotic stress, the total Chl content was reduced ~80%. About 5.7 and 2.6 mg g⁻¹ Fw Chl contents were observed in WT under NaCl and osmotic stress, respectively. Similarly, about 3.8 and 7.3 mg g⁻¹ Fw Chl were detected in VC plant under NaCl and osmotic stress, respectively. The degree of reduction was less in transgenic lines and about 13.8 and 3.6 mg g⁻¹ Fw total Chl and carotenoid contents were quantified, respectively. Transgenic lines showed better protection of Chl from salt and osmotic stress conditions than WT and VC plants. Individually, transgenic lines, L2, L7, and L16, showed about 50% (30.94–15.35 = 15.59 mg g⁻¹ Fw), 42% (29.88–17.21 = 12.67 mg g⁻¹ Fw) and 48% (26.92–14.01 = 12.91 mg g⁻¹ Fw) reduction under salinity (NaCl) stress. Similarly, a 61% (30.94–12.03 = 18.91 mg g⁻¹ Fw), 63% (29.88–11.12 = 18.76 mg g⁻¹ Fw) and 52% (26.92–13.06 = 13.86 mg g⁻¹ Fw) reduction was found in transgenic lines (L2, L7, and L16) under osmotic stress, respectively. Total carotenoid content also exhibited a similar pattern as chlorophyll content (Figure 3). About an average of 47% (6.8–3.6 = 3.2 mg g⁻¹ Fw) decrease was observed in transgenic lines compared to 79.5% (6.8–1.4 = 5.4 mg g⁻¹ Fw) reduction in control plants under stress conditions. Furthermore, transgenic lines showed significantly (p < 0.05) improved cell viability in contrast to WT and VC plants under salt and osmotic stress conditions (Figure S6B).

Osmotic potential and water retention ability of the WT, VC, and transgenic plant leaves were evaluated under control, salinity, and osmotic stress conditions (Figure 4). All transgenic lines were better able to adjust osmotically (−1.7 to −1.9 MPa) under both salinity and osmotic stress conditions compared to WT and VC plant (−0.6 to −0.8 MPa) leaves. The relative water content (RWC) of all transgenic lines was significantly higher under stress conditions than that of WT and VC plants during salinity and osmotic stress. At control condition, an average of 55% RWC was observed in all plants. About 42–50% RWC was detected in transgenic lines under stress conditions compared to control plants (28–33% RWC in WT and VC). Lipid peroxidation of WT, VC, and transgenic lines was compared in leaf segments by estimating the MDA content (about 0.24 mmol g⁻¹ Fw), which is generated after lipid peroxidation and is accumulated in leaves. The MDA content abruptly increased (about 6-fold) in the control (WT and VC) plants under stress, and 1.38 and 1.45 mmol g⁻¹ Fw MDA contents were detected in WT and VC, respectively. The MDA content also increased in SbSRP over-expressing transgenic lines by 1.8–2.7-fold, and 0.43, 0.66, and 0.58 mmol g⁻¹ Fw MDA contents were detected in transgenic lines, L2, L7, and L16, respectively, under salinity stress. Similarly, 2.2–4.7-fold increase was observed under osmotic stress, and 1.15, 0.89, and 0.54 mmol g⁻¹ Fw MDA contents were measured. Electrolyte leakage (%) from the leaves of WT, VC, and transgenic lines were not significantly different and was approximately an average of 8% in control conditions. Under stress conditions, the leakage in WT and VC plants increased 2.9-fold, and approximately an average of 23% EL was observed in comparison to that of the control. The electrolyte leakage in SbSRP over-expressing transgenic lines (13.5%) increased about 1.7-fold upon stress treatment, but it was significantly lower (23.0–13.5 = 9.5%) than that of WT and VC plants (i.e., 23%). In control conditions, the membrane stability index (MSI) was comparable (about 0.92) in all plants (WT, VC, and transgenic lines). Transgenic lines showed a significantly (p < 0.05) higher MSI (about 0.83) under stress conditions compared to control plants (0.59; WT and VC). Results suggest that transgenic plants maintained membrane stability about 1.4-fold better under stress condition than that of WT and VC plants.

The Na⁺ content under control conditions was almost comparable (0.55 mg g⁻¹ Dw) in WT, VC and transgenic lines but found lower under in transgenic lines (compared to WT and VC) under stress conditions. During salinity stress, Na⁺ content in WT and VC plants increased about 1.1-fold; transgenic lines also accumulated Na⁺, but not to the same extent as WT and VC plants (Figure S7). The K⁺ content under control conditions was similar (about 0.4 mg g⁻¹ Dw) in all plant types, but WT and VC showed more than a 60% reduction in K⁺ content (0.4–0.15 = 0.25 mg g⁻¹ Dw) under salinity stress, in comparison to transgenic lines (0.4–0.26 = 0.14 mg g⁻¹ Dw) where the reduction was only 35%. Likewise, the K⁺/Na⁺ ratio was found to be significantly higher in transgenic lines over-expressing the SbSRP gene (Figure S7).

No significant difference was observed in the proline (~3.05 mg g⁻¹ Fw), free amino acid (~4.32 mg g⁻¹ Fw), and polyphenol (0.12 mg g⁻¹ Fw) contents between control (WT and VC) and transgenic lines, over-expressing SbSRP gene, under control condition (Figure 5). Proline content increased about 1.5–2-fold under salinity stress (5.46, 6.61, and 7.99 mg g⁻¹ Fw) and osmotic stress (5.2, 3.19, and 2.85 mg g⁻¹ Fw) conditions in transgenic lines L2, L7, and L16, respectively, while WT and VC plants showed a non-significant increase (about 0.03 mg) in proline content (Figure 5A). Transgenic plants maintained the free amino acid content (about an average of 3.6 mg g⁻¹ Fw) under stress conditions in contrast to WT and VC (2.1 mg g⁻¹ Fw) plants (Figure 5B). The polyphenol content was increased about 2.6-fold (~ 0.32 mg g⁻¹ Fw) in all transgenic lines under stresses, except L16, which showed a different result and accumulated only 0.04 mg (about 1.3 times) polyphenol than control plants (WT and VC) under osmotic stress (Figure 5C).

Total soluble sugars (about an average of 1.2 mg g⁻¹ Fw) increased about 3.5–4.5-fold under salinity and osmotic stress conditions in transgenic lines (about an average of 4.1 mg g⁻¹ Fw) as compared to the control conditions, but only a slight change (1–2-fold) was observed in control (WT and VC; about an average of 2.2 mg g⁻¹ Fw) plants (Figure 5D). The reducing sugar level (about an average of 0.47 mg g⁻¹ Fw in control condition) in transgenic plants increased 2.2-fold (1.03 mg g⁻¹ Fw) under salinity and drought stress conditions whereas a slight increase (about 1.2-fold) was observed in WT and VC (about an average of 0.55 mg g⁻¹ Fw) plants (Figure 5E). The oligo-sugar content (about an average of 0.74 mg g⁻¹ Fw) was comparable under control conditions, whereas transgenic lines L7 and L16 exhibited about 4.4-times higher oligo-sugar content (about an average of 3.23 mg g⁻¹ Fw) under stress conditions (Figure 5F).

Fresh leaf segments (of treated and untreated, control and transgenic plants) were dipped in the freshly prepared DAB and NBT stain solutions, kept in the dark for 2 h followed by exposure to white light for in-vivo localization study. There was no variation in peroxide localization or free radicals among leaves of WT, VC, and SbSRP over-expressing transgenic lines under control conditions following staining with DAB and NBT (Figure 6A). In contrast, WT and VC plants leaves displayed elevated levels of brown and blue insoluble precipitate formation in comparison to transgenic lines under salt and osmotic stress. These results demonstrate that WT and VC leaves build up more superoxide-radicals and H₂O₂ than those of SbSRP over-expressing transgenic lines under stress, confirming that SbSRP mitigates abiotic oxidative stress in situ. Similarly, the cellular starch content was equivalent under control conditions, but transgenic lines showed a deep purple stain under salt and osmotic stress conditions, showing that SbSRP over-expressing transgenic lines maintained cellular starch content, even under stress conditions.

Lower accumulation of ROS was observed in transgenic lines (an average of 0.063 μmol g⁻¹ Fw), whereas WT and VC plants exhibited the higher accumulation of H₂O₂ (an average of 0.25 μmol g⁻¹ Fw) under salt and osmotic stress conditions (Figure 6B). The superoxide content was measured by the reduction of XTT tetrazolium salt, where the transgenic lines showed a significant (p < 0.05) increase in absorbance (470 nm) in contrast to WT and VC plants under stress conditions (Figure 6C).

The mRNA level of genes encoding antioxidative enzymes and transcription factors (ascorbate peroxidase [Nt-APX], catalase [Nt-CAT], superoxide dismutase [Nt-SOD], DREB [Nt-DREB], and AP2 [Nt-AP2]) were studied in WT, VC, and the transgenic lines (L2, L7, and L16) under control and stress conditions (salinity, osmotic, heat, cold, ABA, and SA; Figure 7). The expression of Nt-APX was upregulated by more than 8-fold under salt, drought, ABA, SA, heat, and cold stress in L2, whereas an ~5–7-fold increase was observed in L7 under stress compared to WT, VC plants, and control condition. However, L16 showed lower accumulation of the transcript of Nt-APX under stress conditions. The Nt-CAT gene transcript also showed a similar pattern, although L16 accumulated about 8 times more transcript under salt and drought stress than WT and VC plants. The transcript of Nt-SOD increased by 8-fold in L2, and by 7-fold in L7 and L16 under stress conditions. Likewise, the transcript of Nt-DREB and Nt-AP2 accumulated 4–6 times more in transgenic lines compared to WT and VC plants under stress conditions.

Net assimilation rate was ~12 μmol CO₂ m⁻² s⁻¹ in WT, VC, and transgenic plants under control conditions. Under stress conditions, it reduced to 4 μmol CO₂ m⁻² s⁻¹ in WT and VC plants, but transgenic plants (L2, L7, and L16) maintained values of 6.5, 8.0, and 5.0 μmol CO₂ m⁻² s⁻¹, while transpiration rate was 2.0, 1.5, and 2.8 mmol H₂O m⁻² s⁻¹. These results confirmed the role of SbSRP in providing abiotic stress tolerance to the transgenic tobacco plants by maintaining photosynthesis under stress condition (Figure 8).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.