Improving Drought Tolerance in Tobacco By Application of Salicylic Acid

Drought causes not only the decrease of tobacco yield and quality, but also the lowering of net photosynthetic rate, leading to reactive oxygen species accumulation and even the death of plants. Salicylic Acid is involved in regulating many plant physiological processes and has increasingly been applied to improve tolerance in plants exposed to drought stress. To explore the regulating mechanism of SA, ue-cured tobacco K326 was used in the hydroponic experiments to design PEG drought stress. The photosynthetic characteristics, antioxidant enzymes activities and osmotic regulatory substances contents of tobacco seedlings under drought stress were investigated after 0.3 mmol L -1 SA treatment. Transcriptome sequencing and GO/KEGG analysis were also performed. The main results showed that SA-applied greatly increased the activities of SOD, POD, CAT activity, Pn, proline and soluble protein by 44.27%, 50.18%, 26.23%, 45.74%, 34.67% and 24.91% while reduced the MDA content by 23.89%. GO and KEGG analysis showed that SA treatment was able to up-regulate the genes involved in photosynthesis, carbon metabolism, porphyrin and chlorophyll metabolism, photosynthesis-antenna proteins. The conclusion is that SA application would effectively improve the ability of pigment biosynthesis and photosystem repair of tobacco under drought conditions, thus enhance the photosynthesis, reduce the accumulation of ROS and increase drought resistance, which would provide a measure for alleviating the damage of tobacco caused by drought stress.


Introduction
Drought is one of the most severe environmental factors that adversely affects plant growth and development, as well as crop productivity and quality. Under drought stress conditions, many physiological and metabolic processes such as photosynthesis, are negatively affected. Drought causes the accumulation of reactive oxygen species (ROS) in plants, resulting in membrane lipid peroxidation, photosynthetic rate reduction and even the death of plants. Plants have evolved multifaceted adaptation mechanisms at morphological, physiological, biochemical, and molecular levels, to rapidly recognize and adapt to drought stress condition [1][2] .These internal responses range from changes in photosynthetic activity to the development of antioxidant defenses to enhance drought tolerance [ 3] . Tobacco (Nicotiana tabacum L.) is a model organism as well as a valuable economic crop. Recent years, drought stress has become a main limiting factor for the productivity and quality formation of tobacco due to the dramatic climate changes [ 4] . Especially, drought occurred at the vigorous growing stage has the most impact on yield and quality of tobacco leaf. During this stage, drought causes the decrease of pigment contents, nitrate reductase activity and the net photosynthesis of tobacco leaves, thus leading to the decrease of yield and the increase of nitrate accumulation [5][6] . Therefore, enhancement of tobacco drought resistance and improvement of tobacco physiological traits have become one of the main targets of tobacco breeders and agronomists.
Adding exogenous substances is one of the most important solution to abiotic stresses on plant damage.
Melatonin, brassinolactone and methyl jasmonate have been widely applied to improve the plant stress resistance [ 7] . Salicylic acid (SA) is an essential plant growth regulator which plays an important role in regulating physiological processes such as growth, photosynthesis and some other metabolic processes[ 8-9 [i] 10] . Prior research generally con rmed that plants treated with SA exhibited better resistance to drought stress [ 11-12[ii] 13] . Several studies agreed that SA alleviated drought stress by an increase in antioxidant enzyme activities [13 -14] , proline and soluble carbohydrate content [ 15] . A study reported that exogenous application of SA effectively improved Pn, Gs and Tr of Atractylodes lancea (Thumb.) DC., and protected its photosynthetic system and improved photosynthetic e ciency [ 16] .
Another revealed that foliar application of SA increased proline and soluble carbohydrate contents [15]. However, there is currently no in-depth coverage about the drought tolerance mechanisms at the genomewide transcriptional level in SA-regulated tobacco seedlings.
Transcriptome sequencing technology has been used to comprehensively study the species gene function and speci c biological process at mRNA level. It has clear advantages over other existing methods and numerous studies have applied the RNA-seq method to assess the drought responses in plants and to explore the mechanism of plant response to drought. For example, the global transcriptomic pro les of drought responses have been surveyed in rice [ 17] , maize[ 18] , wheat [ 19] , and soybean [ 20] . The objective of the present research is to comprehensively examine the effects of drought stress and SA treatment on the photosynthesis, secondary metabolites and transcriptome of ue-cured tobacco, aiming to reveal the regulating mechanism of SA and to provide effective methods to increase drought resistance.

Plant Material and Drought Treatment
A hydroponic experiment was carried out on substrate culture in the greenhouse located in National Tobacco Cultivation & Physiology & Biochemistry Research Center of Henan Agricultural University. Seeds of ue-cured tobacco variety K326 were sown in a oating system in green house that maintained 12/12 h light/dark cycle, 28/19 • C light/dark temperature regime, average photosynthetic photon ux density of 400 μmol m −2 s −1 and relative humidity 65%~80%. Six weeks later, well-rooted and same size tobacco seedlings (with 5-6 leaves) were selected and cultivated in 25 cm×30 cm (diameter ×depth) plastic pots (plant/pot). Seedlings were cultivated with Hoagland solution. Nutrient solutions were refreshed every three days. The K326 seeds used in this study were provided by Yunnan Tobacco Company. The collection of the plant material complied with relevant institutional, national and international guidelines and legislation.
Flue-cured tobacco variety K326 seedlings were subjected to treatment after seven days (for recovery). One third of the tobacco seedlings were treated with SA. SA solutions were sprayed on leaves every day for 3 days. Then polyethylene glycol (PEG) simulated water stress method was applied. Moderate drought stress was simulated using 15% (−0.388 MPa) PEG6000 treatment [45] . The seedlings were grouped into three experimental test sets with 15 plants per treatment: CK (water + Hoagland nutrient solution), D (water +15% PEG6000-treated nutrient solutions), D+SA (0.3 mmol/L SA + 15% PEG6000). Sampling and analysis were carried out at 14h after the treatment. Fully expanded leaves (length > 5 cm, up to down, the fourth leaf from top) from the same position in three pots of each treatment were sampled and immediately frozen in liquid nitrogen and then stored at -80℃ for RNA-sequencing.

Measurement of Enzymatic Antioxidant Activities and osmotic regulating substance
Leave lamina between the sixth and eighth side vein were sliced into small sections for determination of the physiological and biochemical indices. Every treatment had three biological replicates. Fresh lamina tissues were cut into 2 mm × 5 mm pieces. Superoxide dismutase (SOD), peroxidase (POD), catalase (CAT) enzyme activity was respectably determined by NBT, Guaiacol and Hydrogen peroxide method [46] .
Proline and MDA (malondialdehyde) content were determined as described by Fan [47] . About 1.0g of each samples was frozen in liquid N 2 and used to assay the soluble protein content according to Li [48] .

RNA Extraction, Preparation of cDNA Library, and Sequencing
The transcriptome sequencing and analysis were conducted by OE biotech Co., Ltd. (Shanghai, China). Total RNA was extracted using the mirVana miRNA Isolation Kit (Ambion, Waltham, MA, USA) following the manufacturer's proltocol. RNA integrity was evaluated using the Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). The samples with RNA Integrity Number (RIN) ≥7 were subjected to the subsequent analysis. The libraries were constructed using TruSeq Stranded mRNA LTSample Prep Kit (Illumina, San Diego, CA, USA) according to the manufacturer's instructions. These libraries were then sequenced on the Illumina sequencing platform (HiSeqTM2500 or Illumina HiSeq X Ten) and 125 bp/150 bp paired-end reads were generated. Quality control was assessed on the remaining reads using NGS QC Toolkit [49] . Raw data (raw reads) were processed using Trimmomatic [50] . The reads containing ploy-N and the low quality reads were removed to obtain the clean reads. Then the clean reads were mapped to reference genome using hisat2 [51] . After removing low quality date, the clean reads were mapped to reference P. trichocarpa genome (ftp://ftp.solgenomics.net/genomes/Nicotiana_tabacum/assembly/Ntab-K326_AWOJ-SS.fa.gz)using bowtie2 or Tophat (http://tophat.cbcb.umd.edu/) [52][53] .
RNA-Seq Analysis, GO and KEGG Pathway Enrichment Analysis of Differentially Expressed Genes (DEGs) Transcript pro les of RNA-seq date were analyzed by calculating the read fragments per kilo base per million mapped reads (FPKM). FPKM value of each gene was calculated using cu inks [54] , and the read counts of each gene was obtained by htseq-count [55] . DEGs were identi ed using the DESeq (2012) functions estimate Size Factors and nbinomTest [56] . p-value < 0.05 and fold change > 2 or fold change < 0.5 was set as the threshold for signi cantly differential expression. GO enrichment and KEGG pathway enrichment analysis of DEGs were respectively performed using R based on the hypergeometric distribution [57] .

Statistical Analysis
Figures were processed using GraphPad Prism (v. 8.0.1, GraphPad Software Inc., CA, USA) and correlation analysis and variance between treatments were all processed using SPSS 20.0 (IBM, Palo Alto, CA, USA).
Treatments were compared by LSD multiple range test. All presented data is mean of three biological replicates (n = 3).

Phenotypic observation under PEG drought and natural drought stress
The growth and morphological changes of tobacco seedlings from different treatments were observed (Fig.1). Flue-cured tobacco seedlings were signi cant different between drought-only (D) and SA (D+S) treatments. It was worth noting that under drought stress (D), the bottom leaves of ue-cured tobacco turned yellow and all the leaves began wilting. In contrast, the leaves of ue-cured tobacco only had slighter symptoms of wilting under SA treatment (D+S).

Effect of SA on photosynthesis parameters
To evaluate the effects of SA on the photosynthesis of ue-cured tobacco seedlings, the net photosynthesis rate (Pn), stomatal conductance (Gs), transpiration rate (Tr) and intercellular CO 2 concentration (Ci) were measured (Fig.2a-d). Pn, Gs, Tr and Ci were in uenced signi cantly by imposed drought stress. In comparison with control, Pn, Gs and Tr markedly decreased by 45.23%, 40.31% and 59.67% in 15% PEG-treated seedlings, while the Ci increased by 9.59%. Whereas, SA treated leaves had higher Pn, Gs, Tr of 45.74% 26.82% 61.37%, and lower Ci of 3.29%, compared with the seedlings of PEG treatment alone.
Reduction of MDA and increase of antioxidant enzyme activity and osmotic regulating substance by SA Antioxidant enzymes play an important role in scavenging ROS which has been proven to hinder the synthesis of PSII core D1 [ 21] . An increase in POD, SOD, CAT activity, proline and soluble protein content was obtained in 15% PEG-treated tobacco seedlings (Fig.3a-c, e-f). In contrast, exogenous SA exerted a remarkable increase of POD, SOD and CAT by 50.18%, 44.27% and 26.23%, respectively compared with PEG treatment alone (Fig.3a-c, e-f). Moreover, a sharp increase in MDA (the product of lipid peroxidation) was observed under drought stress (Fig.3d). The seedlings under 15% PEG drought stress caused 87.85% increase in MDA content compared to control (Fig.3d). However, a signi cant reduction of MDA and increase of proline and soluble protein were observed in SA supplemented drought-stressed seedlings compared to the seedlings exposed to drought only, indicating that SA could improve the cell osmotic function and alleviate the damage of drought stress.

Data processing
Nine cDNA libraries were established for RNA-seq analysis. A total of 85.88M, 88.96M, and 89.26M raw reads were identi ed in the CK, D and D+S samples, respectively. After data processing, 83.57M (97.31%), 86.43M (97.16%) and 86.91M (97.37%) million clean reads were obtained for further analysis (Table 1). In all, the mapping rates of all the samples to the reference genome were above 95%, the GC content of all samples was stable with the distribution ranging from 43.28~44.13% and the QC30 value of all sample was above 95%, implying successful library construction and RNA sequencing.

Gene expression and correlation analysis between samples
After ltering the raw reads, a high rate of clean reads from each sample was achieved. The FPKM expression levels for each sample were calculated (Fig.4a). In addition, Pearson's correlation coe cient and principle component analysis (PCA) of the data pro les from all 9 samples revealed high correlation among all the samples (Fig.4b-c). These analysis demonstrated that the sequencing data in this study was adequately representative and valid.
Differentially expressed gene selection, Go enrichment and KEGG pathway analysis of differentially expressed genes Fold change (FC) > 2 or FC < 0.5, p-value < 0.05, was used as threshold to achieve the differentially expressed genes (DEGs).. A total of 6229 DEGs (2914 up-regulated and 3315 down-regulated) were identi ed between D and control groups, and 4840 DEGs (1944 up-regulated and 2896 down-regulated) were identi ed between D+S and D groups (Fig.5e).
Gene Ontology is a standardized system used for the functional classi cation of genes according to the 3 domains of biological process, cellular components, and molecular functions. The DEGs in D-vs-CK were further analyzed using GO and KEGG analyses (Figu.5a-d). We observed almost 5.97% of down-regulated genes correlated with biological process (GO-BP) of photosynthesis (GO:0015979), chlorophyll biosynthetic process (GO:0015995) and chloroplast organization (GO:0009658). In cellular component (GO-CC), the down-regulated genes were signi cantly enriched in thylakoid (GO:0009579), chloroplast (GO:0009507) and photosystem II (GO:0009523). The genes related to molecular function (GO-MF), such as pigment binding (GO:0031409), rRNA binding (GO:0019843), chlorophyll binding (GO:0016168) were also down-regulated. KEGG pathway analysis also showed that the down-regulated DEGs were mostly involved in photosynthesis (ko00195) and porphyrin and chlorophyll metabolism (ko00860). However, in D+S-vs-D, these DEGs in the categories of molecular function, biological process, and cellular component were almost up-regulated, which were consistent with KEGG pathway analysis.

GO enrichment and KEGG pathway analysis of DEGs Response to SA
To explore the DEGs in tobacco seedlings improved by spraying SA under drought stress, venn digram analysis between D-vs-CK down and D+S-vs-D up was conducted (Fig.6a). GO and KEGG were then conducted to examine the potential gene function and metabolism pathway of DEGs (Fig.6b-c) Genes involved in molecular function mostly enriched in chlorophyll binding (GO:0016168). KEGG pathways analysis presented that the genes were signi cantly enriched in the pathway of photosynthesis-antenna proteins (ko00196), photosynthesis (ko00195), carbon xation in photosynthetic organisms (ko00710), carbon metabolism (ko01200), starch and sucrose metabolism (ko00500), porphyrin and chlorophyll metabolism (ko00860) and carotenoid biosynthesis (ko00906).

DEGs Involved in photosynthesis and carbon metabolism
Transcriptome sequencing technology provides a large amount of information regarding the DEGs that are involved in speci c biological responses. The results showed that carbon metabolism, porphyrin and chlorophyll metabolism, photosynthesis were lower in PEG-treated tobacco than in well-watered tobacco (Fig.7a-e), which was consistent with the low Pn in PEG-treated tobacco seedlings. The expression levels of genes involved in carbon metabolism, such as FBP (Nitab4.5_0002152g0090), photosynthesis, such as PSBO (Nitab4.5_0000108g0110), PSBR (Nitab4.5_0002475g0030) and PSAO (Nitab4.5_0003068g0060), photosynthesis-antenna proteins, such as lhcA-P4 (Nitab4.5_0001128g0010), LHC4.2 (Nitab4.5_0000476g0110), and CAB50 (Nitab4.5_0000592g0320) were consistently suppressed at drought stress in ue-cured tobacco while up-regulated after SA spraying. To explore the reason why SA can alleviate the damage of drought stress in ue-cured tobacco, we also analyzed the differences in gene expression levels of porphyrin and chlorophyll metabolism. The results showed that genes of POR1 (Nitab4.5_0002526g0100), hemC (Nitab4.5_0001791g0020), HEMA1 (Nitab4.5_0000377g0190) and CHLM (Nitab4.5_0000976g0010) were consistently and signi cantly down-regulated under drought stress while up-regulated in SA-treated tobacco (Fig.7d) , which might also be the cause for higher photosynthesis in SA-treated tobacco.

Discussion
Recent years, drought stress has become one of the most serious environmental factors for the productivity and quality formation of tobacco. Our investigation study revealed that the application of SA was able to protect the tobacco seedlings from damage of drought stress and the corresponding mechanism, which would have signi cant practical application. Under drought stress condition, a large amount of ROS would accumulate inside the plant, resulting in lipid peroxidation, MDA production, cell membrane damage, cell aging and death [22][23] . Therefore, MDA levels was used as an indicator to report the damage of cellular components. Plants have evolved multifaceted adaptation mechanisms to rapidly recognize and adapt to drought stress conditions such as the increase of antioxidant enzymes activity and osmotic regulating substance [ 24] . In wheat, exogenous SA had been reported to play an important role in improving the antioxidant enzymes activity and decreasing MDA content [25][26] . As the important osmotic regulators in plants, proline and soluble protein have the ability to improve the cell osmotic function and alleviate the damage of drought stress [ 27] . Positive effects of SA in triggering the synthesis of soluble protein and proline to improve the drought resistance of plants have been reported by Lu et al[ 28] . In our study, an increase in the MDA, soluble protein, proline contents and antioxidant enzyme activities were obtained in PEG drought stress and natural drought-stressed tobacco treatments. However, less MDA content and higher soluble protein and proline contents were observed in SA-treated plants compared to no SA-treated drought stressed seedlings (Fig.3a-f). The decrease in MDA content in SAtreated plants may be associated with the induction of antioxidant responses which would protect the plant from damage. Zhang et al. have suggested a similar mechanism to be responsible for SA induced drought stress tolerance in eggplant plant [ 29] . In ROS-scavenging systems, antioxidant enzymes have key roles to play, with SOD catalyzing the dismutation of O 2 to oxygen and H 2 O 2 , which is then scavenged by CAT and POD [ 30] . The results proved that SA was able to enhance antioxidant defense systems and increase osmotic regulating substance level to eliminate ROS, which is in accordance with the fact that the ROS and MDA levels were markedly reduced in plants.
Leaf chlorophyll content is used as an indicator of photosynthetic capacity and plant vitality [31] . Photosynthesis plays an important role in crop growth, and is the ultimate source of yield formation.
Besides, net photosynthesis rate (Pn) is regarded as an effective indicator to assess plants drought resistance [32] . Previous studies reported that the decrease in photosynthesis was usually caused by stomatal limitation when both Gs and Ci decline while non-stomatal limitation was the main reason for photosynthesis when Ci increased and Gs decreased [33][34] . In the present study, the data clearly showed a reduction in Pn, Tr, Gs accompanied with an increase in Ci under PEG-drought stress (Fig.2a-d). The results indicated that non-stomatal limitation primarily led to the drop of Pn in PEG-drought stress. Under PEG-drought stress, SA application signi cantly reversed the trend compared with the water stressed control. The results of this study were in the harmony with previous studies, demonstrating that drought stress inhibits carbon assimilation, chlorophyll biosynthesis and damages photosynthetic apparatus The decrease in photosynthesis could make plants absorb more light energy than the level necessary for photosynthetic carbon xation. This excess energy has the potential to trigger an increase in reactive oxygen species (ROS) production, which has been proven to be capable of damaging the photosystem, decreasing the activity of photosynthetic electron transport chain and hindering pigment synthesis [ 38] . For a better understanding of the mechanism of SA induced increase of photosynthesis rate, technologies of molecular biology were used in this study. Under drought stress, the down-regulated DEGs were most correlated with chlorophyll biosynthetic process, photosynthesis, light harvesting, chloroplast and photosystem II (Fig.5a). KEGG pathway analysis also showed that the down-regulated DEGs were mainly enriched in starch and sucrose metabolism, photosynthesis-antenna proteins and porphyrin and chlorophyll metabolism (Fig.5c). However, spraying SA has increased expression of genes involved in photosystem II, photosystem I, chloroplast thylakoid, photosynthetic electron transport chain, chloroplast thylakoid membrane, carbon xation and synthesis of carotenoids (Fig.5b, d), indicating that spraying SA was effective in protecting the cell membrane and photosystem from damage, which then enhanced photosynthesis rate and pigment synthesis. Further investigation of down-regulated DEGs between watered-plants and drought treated plants and up-regulated DEGs between drought treated plants and SA-sprayed plants suggested SA promoted pigment biosynthesis, photosynthesis, carbon xation and protect the photosystem from damage (Fig.6a-c). Besides, some key genes related to photosynthesis, photosynthesis-antenna proteins, porphyrin and chlorophyll metabolism and carbon metabolism were upregulated after spraying SA (Fig.7a-e) , including the RCA1 genes, which plays an important role in activation of the rubisco enzyme [39] , lhcA-P4, which encode key subunit of light-harvesting complex and could convert light energy into chemical energy [40] , AGPS1 and SS1, which catalyzes the limiting steps in starch synthesis [41] , photosystem II genes (PSBO, PSBR, PSBS), photosystem I genes (PSAF, PSAG, PSAK, PSAO), POR1 and CHLP, which play crucial roles in the chlorophyll biosynthesis [42] . Rubisco, a key enzyme in the calvin cycle, assimilates atmospheric CO 2 into the biosphere [ 43] . Previous study has showed that overexpression of ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) activase gene (RCA) was able to maintain photosynthesis under drought stress, which contributed to the higher drought tolerance in transgenic plants [ 44] . Together, these gene products might contribute to the adaptation or resistance of tobacco to drought stress.
In conclusion, spraying SA decreased MDA accumulation while promoted photosynthesis, antioxidant enzyme activity and osmotic regulators in ue-cured tobacco seedlings under PEG drought stress. Drought stress inhibited the expression of genes involved in photosynthesis, carbon xation and pigment biosynthesis. Spraying SA up-regulated the expression of genes involved in photosynthesis, photosynthesis-antenna proteins, photosystem II, photosystem I, carbon metabolism, porphyrin and chlorophyll metabolism. Spraying SA under drought stress was an effective method to alleviate the damage caused by drought stress.  represents three biological replications. Figure 1 The phenotype of ue-cured tobacco seedlings under PEG drought stress