Identification and transcriptomic profiling of genes involved in increasing sugar content during salt stress in sweet sorghum leaves

Background Sweet sorghum is an annual C4 crop considered to be one of the most promising bio-energy crops due to its high sugar content in stem, yet it is poorly understood how this plant increases its sugar content in response to salt stress. In response to high NaCl, many of its major processes, such as photosynthesis, protein synthesis, energy and lipid metabolism, are inhibited. Interestingly, sugar content in sweet sorghum stems remains constant or even increases in several salt-tolerant species. Results In this study, the transcript profiles of two sweet sorghum inbred lines (salt-tolerant M-81E and salt-sensitive Roma) were analyzed in the presence of 0 mM or 150 mM NaCl in order to elucidate the molecular mechanisms that lead to higher sugar content during salt stress. We identified 864 and 930 differentially expressed genes between control plants and those subjected to salt stress in both M-81E and Roma strains. We determined that the majority of these genes are involved in photosynthesis, carbon fixation, and starch and sucrose metabolism. Genes important for maintaining photosystem structure and for regulating electron transport were less affected by salt stress in the M-81E line compared to the salt-sensitive Roma line. In addition, expression of genes encoding NADP+-malate enzyme and sucrose synthetase was up-regulated and expression of genes encoding invertase was down-regulated under salt stress in M-81E. In contrast, the expression of these genes showed the opposite trend in Roma under salt stress. Conclusions The results we obtained revealed that the salt-tolerant genotype M-81E leads to increased sugar content under salt stress by protecting important structures of photosystems, by enhancing the accumulation of photosynthetic products, by increasing the production of sucrose synthetase and by inhibiting sucrose decomposition. Electronic supplementary material The online version of this article (doi:10.1186/s12864-015-1760-5) contains supplementary material, which is available to authorized users.


Background
Soil salinity is not only one of the major factors leading to deterioration of the ecological environment but also a major abiotic stress in plant agriculture worldwide [1]. Salt stress involves a combination of osmotic stress and ionic stress that greatly affects plant growth and crop production [2]. Upon salt treatment, lots of the major processes within plants, such as photosynthesis, protein synthesis, energy metabolism and lipid metabolism are affected [3]. Salt treatment also regulates the expression level of many genes involved either directly or indirectly in plant protection [2]. In the past few decades, many efforts have been made to understand the molecular mechanisms of salt tolerance. Utilization of genes related to compatible solutes [4], ion transporters [5] and transcription factors [6] is regarded as a way to improve the salt tolerance of plants.
Sweet sorghum [Sorghum bicolor (L.) Moench] which originates from Africa is an annual C4 crop [7]. Sweet sorghum has a fast growth rate and high efficiency of biomass accumulation. It is consumed as a food source for humans and as livestock feed. In addition, it has been considered to be one of the most promising bio-energy crops [8], as the stalks are rich in fermentable sugars. The tolerance of sweet sorghum to salinity is thought to be high. However, there are salt-tolerant and saltsensitive genotypes of sweet sorghum. Salt-tolerant genotypes have a greater ability to exclude toxic ions and to store the absorbed toxic ions in the root cell vacuoles while maintaining high levels of K + uptake. As a result, the accumulation of Na + and Cl − in actively growing shoots and leaves may be limited. This mechanism can effectively prevent the photosynthetic apparatus of sweet sorghum from being damaged by Na + and Cl − . On the other hand, the ability to compartmentalize Na + within root cell vacuoles is lower in salt-sensitive genotypes [9], which results in a higher level of Na + accumulated in leaves. Due to this accumulation, the photosynthetic apparatus may be damaged by Na + and the photosynthesis will significantly decrease.
Interestingly, it has been shown that the brix of saltsensitive sweet sorghum decreases under salt stress. The brix of salt-tolerant species, on the other hand, stays stable or is even increased by salt stress [10][11][12]. As we know, the main source of carbon and energy in the sink tissues of sweet sorghum is sucrose. Several physiological processes play important roles in maintaining the high sugar content in stems of sweet sorghum. A) CO 2 from the atmosphere is fixed in the mesophyll cells. Stalk sugar content accumulation of sweet sorghum depends on the synthesis and accumulation of photosynthetic products. The initial product of CO 2 fixation is oxaloacetate (OAA). OAA is converted into a transportable form (malate) and is then transported to the bundle sheath. After a series reactions through the C4 pathway, 3-phosphoglycerate is produced which is then converted to triose phosphate (TP) [13]. B) Once TP has been produced, it either leaves the chloroplast via the triose phosphate translocator (TPT) in exchange for orthophosphate or it remains in the chloroplast stroma for the completion of the Calvin cycle or to be converted to starch [14]. C) TP in the cytoplasm can be converted to fructose-6-phosphate (Fru-6-P). Then, Fru-6-P can be further converted to sucrose by sucrose phosphate phosphatase (SPS, EC3. 1.2.24) or to UDP-glucose (UDP-Glu). UDP-Glu is used as substrates in sucrose synthesis, a reaction catalyzed by sucrose synthase (SS, EC2.4.1.13). Sucrose can be decomposed into glucose and fructose by invertase (INV, EC3.2.1.26) in vacuole. D) Six sucrose transporters (SUT1-6) have been reported in monocots [15], which are located in the plasma membranes of sieve elements and companion cells, or in tonoplasts of storage cells. SUTs have been reported to play an important role in the re-distribution of sucrose [16].
Although there are numerous studies on the response mechanism of sweet sorghum to salt stress, most of which are restricted to the eco-physiological level or to the study of a single pathway. The physiological and molecular mechanisms of increasing sugar content in salt-tolerant sweet sorghum species under salt stress is remain unclear. In recent years, with the increasing availability of sequence data, expression profiling has been used to identify genes involved in the adaptive responses abiotic stresses. A common strategy to identify genes related to salt stress is using a comparative study of different genotypes or cultivars in the tolerance to the abiotic stress [17][18][19]. Comparisons between salt-sensitive and salt-tolerant genotypes of model and non-model plant species have been reported, including Arabidopsis [19], rice [18,20], olive [17], populus [21] and tomato [22]. In the present study, the transcriptomes of salt-sensitive and a salt-tolerant sweet sorghum inbred lines were analyzed by high-throughput Illumina RNA-sequencing (RNA-seq). By comparing the transcriptomes of a salt-sensitive and a salt-tolerant sweet sorghum inbred line under salt stress, we identified 864 and 930 differentially expressed genes (DEGs) between control plants and those subjected to salt in M-81E and Roma, respectively. Results of this study should provide further insight into the complex regulatory networks underlying the mechanism of higher sugar content under salt stress in sweet sorghum.

Effects of salt stress on growth parameters
After treated with 50 mM NaCl for 7 days, there was no significant difference in M-81E (Fig. 1a), while growth of Roma was significantly inhibited (Fig. 1b). In the presence of 150 mM NaCl, the growth of both genotypes was inhibited, but it was more severe in Roma. Leaf length of M-81E was not affected by 50 mM NaCl treatment, but slightly decreased 15.6 % at 150 mM NaCl treatment. Leaf length of Roma decreased 27.2 % at 50 mM NaCl treatment and 41.6 % at 150 mM NaCl treatment (Additional file 1: Figure S1). Leaf numbers of M-81E and Roma were not affected by 50 mM NaCl, but decreased 23.2 and 31.3 %, respectively, when treated with 150 mM NaCl (Additional file1: Figure S1). Fresh weight (FW) of leaves of both genotypes gradually decreased with an increase in NaCl concentration. The reductions were more severe at 150 mM, particularly for Roma (Additional file 2: Figure S2) in which values decreased 43.1 and 68.6 % for 50 and 150 mM NaCl concentrations, respectively. Dry weight (DW) of leaves also decreased with an increase in NaCl concentration. The highest reduction in Roma was 62.9 % at 150 mM NaCl (Additional file 2: Figure S2). There was no significant effect on water content during NaCl treatment. (Additional file 2: Figure S2).

Effects of salt stress on ion concentration
After treated with 50 mM NaCl for 7 days, there were no significant changes in Na + concentrations in leaves of both genotypes compared to control plants (Additional file 3: Figure S3). When a higher concentration of salt (150 mM) was applied, Na + concentration increased significantly, especially for Roma. The K + concentration in leaves gradually decreased in response to NaCl. At the 150 mM NaCl treatment, K + concentration of M-81E and Roma decreased 30.6 and 41.6 %, respectively (Additional file 3: Figure S3). The K + /Na + ratio in leaves of M-81E increased under 50 mM NaCl treatment and then decreased when treated by 150 mM NaCl. While the K + /Na + ratio in leaves of Roma decreased under the NaCl treatment. At 150 mM NaCl treatment, the K + /Na + ratio in Roma decreased by a factor of fourteen times (Additional file 3: Figure S3).

Effects of salt stress on PSII photochemical efficiency
In both genotypes the potential efficiency of PSII photochemistry (Fv/Fm) was reduced with increasing NaCl concentration (Fig. 2). After treated with 50 mM NaCl for 7 days, Fv/Fm of M-81E and Roma decreased 3.6 and 11.1 %, respectively. For 150 mM NaCl, Fv/Fm of M-81E and Roma decreased 4.2 and 20.8 %, respectively (Fig. 2). The actual PSII efficiency (ΦPSII) decreased in both genotypes after treated with NaCl. ΦPSII of M-81E treated with 50 and 150 mM NaCl decreased 10.7 and 14.4 %, respectively. In Roma, ΦPSII decreased 36.6 and 50.7 % for 50 and 150 mM NaCl treatment, respectively.

Effects of salt stress on chlorophyll content
The effects of increasing level of NaCl salinity on chlorophyll contents in the two genotypes were determined after 7 day exposure to salinity ( Fig. 3). Chlorophyll content in M-81E was not changed significantly by 50 mM NaCl but decreased 46.5 % under 150 mM NaCl treatment. On the other hand, in Roma, chlorophyll content decreased gradually with the increasing NaCl treatments. Chlorophyll content of Roma treated with 50 and 150 mM NaCl decreased 37.6 and 68.4 %, respectively.

Effects of salt stress on photosynthesis
There were no significant changes in photosynthetic rate, stomatal conductance and intercellular CO 2 concentration in M-81E under salt stress. However, photosynthesis in Roma was significantly influenced by salt stress (Fig. 4). The photosynthetic rate of Roma was inhibited after treated with NaCl for 7 days. The reduction percentage of photosynthetic rate of Roma was 45.1 and 67.5 % for 50 mM and 150 mM NaCl treatment, respectively. Stomatal conductance of Roma decreased 35.5 and 60.9 % after treated with 50 mM and 150 mM NaCl, respectively. Intercellular CO 2 concentration of

Effects of salt stress on sugar content
The effects of increasing level of NaCl salinity on sugar contents in the two genotypes were determined after 7 days exposure to salinity. After treated for 7 days, the sugar content of M-81E increased 15.6 and 99.7 % under 50 mM and 150 mM NaCl, respectively. While, there was no significant change in sugar content of Roma under 50 mM NaCl. Under 150 mM NaCl, the sugar content of Roma decreased 30.5 % (Fig. 5).

Sequencing output and assembly
In order to investigate the molecular mechanisms of high sugar content under salt stress in sweet sorghum, libraries (MC, MS, RC and RS) were designed for RNAseq. MC and MS libraries were used for leaves of M-81E treated with 0 mM and 150 mM NaCl, respectively. RC and RS libraries were used for leaves of Roma treated with 0 mM and 150 mM NaCl, respectively. In total, 78.41 million reads were generated. After trimming adapters and filtering out low quality reads, more than 67.08 million clean reads were retained for assembly and further analysis. Among all the reads, more than 94 % had Phred-like quality scores at the Q30 level (an error   Table S1). All these data showed that the throughput and sequencing quality were high enough for further analysis. The reads produced in this study have been deposited in the National Center for Biotechnology Information (NCBI) SRA database and accession number was shown in "Availability of supporting data".

Exploration of DEGs in response to salt stress
In the absence of salt, 3342 genes showed differential expression levels when comparing M-81E vs. Roma. While in the presence of salt, the DEGs between them were 2265. For M-81E, 864 genes were differentially expressed between control plants and those subjected to salt. Among these DEGs, 236 genes were up-regulated in leaves under salt stress. For Roma, 930 genes were differentially expressed between control plants and those subjected to salt. Among these DEGs, 442 genes were up-regulated in leaves under salt stress (Fig. 6). All of these DEGs were selected for further analysis.

Functional categorization of stress-regulated genes Functional classification by GO
In order to assign functional information to the DEGs between control plants and those treated with NaCl, Gene Ontology (GO) [23] analysis was carried out. This analysis provides a dynamic, controlled vocabulary and also hierarchical relationships for the representation of information on biological processes, molecular function, and cellular components, forming a coherent annotation of various gene products [23]. In M-81E, there were 812 unique transcripts assigned to 48 level-2 GO terms, which were summarized under three main GO categories, including 13 for cellular component, 12 for molecular function and 23 for biological process, respectively. In Roma, there were 878 unique transcripts assigned to 47 level-2 GO terms including 13 for cellular component, 12 for molecular function and 22 for biological process, respectively. For the cellular group, in both M-81E and Roma, the most represented category was cell part, cell and organelle. For molecular function, the category of binding was the most represented GO term, followed second by the category of catalytic activity. Regarding biological process, NCBI UniGene for cellular process and metabolic process were highly represented (Fig. 7).

Functional classification by COG
In addition, all the DEGs were subjected to a search against the Clusters of Orthologous Groups (COG) [24] classification. Among the 864 DEGs, 349 sequences showed a COG classification in M-81E (Additional file 5: Figure S4A). Among the 25 COG categories, the cluster for "general function prediction only" was the largest group, followed by "secondary metabolites biosynthesis, transport and catabolism", "amino acid transport and metabolism", "carbohydrate transport and metabolism" and "transcription". The categories "chromatin structure and dynamics", "extracellular structure" and "nuclear structure" had no corresponding genes. The 360 sequences of the 930 sequences could be assigned to COG classifications in Roma (Additional file 5: Figure S4B). The cluster for "general function prediction only" represented the largest group, followed by "signal transduction mechanisms", "transcription", "replication, recombination and repair" and "carbohydrate transport and metabolism". Whereas no unigenes were assigned to "extracellular structure", "nuclear structure", "cell motility" and "intracellular trafficking, secretion, and vesicular transport".

Functional classification by KEGG
Kyoto Encyclopedia of Genes and Genomes database (KEGG) [25] was used to identify potential biological pathways represented in the sweet sorghum transcriptome. There were 150 DEGs of M-81E and 174 DEGs of Roma assigned to 70 and 63 KEGG pathways, respectively. The majority of these DEGs mapped to "photosynthesis", "photosynthesis-antenna proteins", "carbon fixation in photosynthetic organisms" and "starch and sucrose metabolism" categories ( Fig. 8, Table 1), which indicated that salt stress mainly affected photosynthesis and carbohydrate metabolism in leaves of sweet sorghum.

Photosynthesis-antenna proteins
In the first steps of photosynthesis, light energy is captured and converted into chemical energy. A large part of the light is absorbed by the outer light-harvesting complexes (LHCs), which contain most of the chlorophyll and carotenoid pigments and are peripherally associated with PSI and PSII [26,27]. These LHC proteins are encoded by nuclear genes of the LHC multi-gene family coding for proteins that contain one to four trans-membrane helices and share a number of conserved chlorophyll-and xanthophyll-binding motifs [28]. In higher plants, 14 different types of LHC proteins (Lhca1-Lhca6 and Lhcb1-Lhcb8) are expressed [29]. Lhca-type proteins are organized into two heterodimeric domains (Lhca1/Lhca4 and Lhca2/Lhca3) as an external antenna with the PSI core. The reaction center of PSII is surrounded by Lhcb-type proteins. In the present study, 8 DEGs of M-81E and 14 DEGs of Roma were mapped to the antenna proteins, respectively. In comparison with the untreated control, the expression of DEGs encoding Lhca1 and Lhcb1-5 were down-regulated in both of the two genotypes under salt stress. However, the expression level of DEGs encoding Lcha2-4 and Lchb6 dropped under salt stress in Roma but did not change in M-81E (Additional file 6: Figure S5).

Photosynthesis
Photosynthesis is one of the most important metabolic processes in plants. Salt stress significantly impacts the photosynthetic rate [30,31]. The four protein components of the photosynthetic electron transport chain responsible for the electron transfer from water to NADP + are Photosystem II (PSII), Photosystem I (PSI), cytochrome (Cytb6f ) complex, and ATP synthase. There were 11 and 20 DEGs of M-81E and Roma, respectively, that mapped to the photosynthesis pathway, which led to changes in the structure and function of the four protein components (Fig. 9, Table 1).
Photosystem II is a protein complex consisting of several different types of chlorophyll binding components. The function of these components is to organize chlorophylls for light harvesting and to harbor the electron transport intermediates as well as cofactors needed for   the oxidation of water [32]. After treated with NaCl, DEGs encoding PsbQ, which is necessary for regulation of activity and assembly [33,34] of PSII in both M-81E and Roma, were down-regulated. PsbR has been proved to be an important link in the PSII core complex to permit stable assembly of the oxygen-evolving complex proteins PsbP and PsbQ [35]. DEGs encoding PsbR were up-regulated in M-81E after treated with 150 mM NaCl for 48 h. DEGs encoding PsbW, which stabilize the supramolecular organization of photosystem II, were down-regulated only in Roma. These results suggested that salt stress reduced the binding stability of several subunits of PSII. However, we predict that M-81E may protect important connective structures from being destroyed by increasing expression of specific genes. Photosystem I (PSI) from higher plants is a supramolecular complex which catalyzes the light-driven electron transfer from plastocyanin to ferredoxin and is composed of a chlorophyll binding core complex and a chlorophyll a/b binding peripheral antenna called LHCI [36]. After treated with NaCl for 48 h, the expression of DEGs encoding PsaK, PsaH and PsaO decreased in both genotypes. All of these three subunits are involved in the interaction between the light-harvesting complex (LHC) and Photosystem I [37][38][39], suggesting that salt stress weakened the connection between LHCs and PSI and reduced the conversion of light energy to chemical energy. PsaD, PsaE, PsaF, PsaG, PsaL and PsaN encoding genes were down-regulated only in Roma. Among them, four subunits (PsaD, PsaE, PsaF, PsaN) are considered to be important for the interaction with ferredoxin or plastocyanin [40,41], indicating that the electron transport mechanism was inhibited by salt stress in Roma. These observations agreed fairly well with the down-regulation of petE and petF in Roma after treated with salt. Moreover, expression of Sb04g027810, a gene encoding the ATP synthase delta chain, decreased in both genotypes when treated with 150 mM NaCl, while the gene encoding the ATP synthase gamma chain was only down-regulated in Roma.

Carbon fixation in photosynthetic organisms
There were 6 and 8 DEGs of M-81E and Roma, respectively, mapped to the carbon fixation in photosynthetic organisms pathway. Ribulose-bisphosphate carboxylase (rubisco, EC: 4.1.1.39), phosphoenolpyruvate carboxylase (PEPC, EC:4.1.1.31) and pyruvate orthophosphate dikinase (PPDK, EC:2.7.9.1) are considered as key enzymes in the process of carbon fixation. Rubisco catalyzes the incorporation of CO 2 into ribulose 1,5-bisphosphate [42]. Under salt stress for 48 h, the expression of DEGs encoding rubisco decreased while the PPDK and PEPC encoding genes remained unchanged in both genotypes based on our RNA-seq data. Surprisingly, the expression  Figure S6). ) are considered to be key enzymes in sucrose metabolism. SS is known to play a role in sucrose synthesis using uridine diphosphate (UDP)-glucose and fructose as substrates and its activity is high in source tissues such as leaves [43]. After a 48 h treatment with NaCl, the expression of DEGs encoding SS were enhanced in M-81E but unchanged in Roma. INV plays the most important role in the decomposition of sucrose. In the present study, the expression of DEGs encoding INV decreased in M-81E but increased in Roma during salt stress (Additional file 8: Figure S7).

Verification of RNA-seq data
We performed quantitative real-time PCR on 14 randomly selected DEGs to validate the RNA-seq gene expression analysis. As shown in Fig. 10, a high correlation (R 2 = 0.93) between RNA-seq and qRT-PCR was observed. Also, three genes (Sb03g034280, Sb06g023760 and Sb01g035890) which may play important roles in improving sugar content in sweet sorghum were confirmed by qPCR, too. As shown in Fig. 11, a high correlation (R 2 = 0.92) was observed, confirming the reliability of the RNA-seq data.

Discussion
Sweet sorghum has been considered to be a plant with relatively high salt tolerance [9,12,44]. In our work, however, NaCl caused a dramatic decrease in leaf length, leaf number, FW and DW in Roma, while growth parameters in M-81E were less affected (Fig. 1, Additional file 1: Figure S1 and Additional file 2: Figure S2). These results were consistent with previous studies that M-81E was observed to be relatively salt-tolerant but Roma was salt-sensitive. Generally, the effects of salt stress include ion toxicity and osmotic stress. Ion toxicity is mainly caused by Na + . Increases in Na + concentration during salt stress have been well established [45,46]. Since sweet sorghum has the ability to exclude toxic ions and store the absorbed toxic ions in the root cell vacuoles while maintaining higher levels of K + uptake [9], the accumulation of Na + in leaves can be limited. In the present study, the accumulation of Na + in leaves was not significantly affected in both our tested genotypes under 50 mM NaCl stress condition for 7 days. At 150 mM NaCl, however, Na + concentration was significantly increased in both genotypes, particularly in Roma by a factor of six (Additional file 3: Figure S3). High concentration of Na + impairs the ability of plants to accumulate essential nutrients [47], such as K + , which are required to maintain the stability and functioning of cell membranes and associated enzymes. Maintenance of adequate K + levels in plant tissues under salt stress has been reported to be dependent on selective cellular K + and Na + distribution [48]. K + content decreased with raised NaCl concentration and as a result, the K + /Na + ratios decreased. These results indicated that sweet sorghum limited accumulation of Na + under 50 mM NaCl treatment, while at a NaCl concentration increased of 150 mM, Roma lost the ability to control the absorption of Na + .
In order to compare the salt response of sugar content in leaves of M-81E and Roma at the transcriptome level, RNA-seq was performed using leaves treated with 0 and 150 mM NaCl for 48 h. In response to salt stress, a larger number of DEGs were observed in Roma compared to M-81E (Fig. 6). In a previous study, the results showed that the salt tolerant plants had a smaller number of salt-regulated genes in salt cress [49]. However, another study indicated that a salt-sensitive tomato PI365967 showed a relatively smaller amount of saltregulated genes than Moneymaker which is more tolerant to salt [22]. These results suggested that the number of salt-regulated genes may not be directly linked to the degree of salt tolerance. Fig. 10 Validation of RNA-seq results by RT-qPCR. Expression levels of 14 randomly selected genes in the four samples used in this study were detected by RT-qPCR. R 2 represents the correlation coefficient value between the two platforms. The numbers in the scale bar stand for RPKM values in RNA-seq and ΔΔCt in qRT-PCR, which were used to evaluate the correlation (R 2 ). Primers are listed in (Additional file 9: Table S2) Pathway analysis is an effective way to characterize "gene networks" under salt stress. We observed that genes related with photosynthesis and light-harvesting proteins were mainly repressed by salinity. There were 20 DEGs mapped to the photosynthesis pathway in Roma and all of them were down-regulated under salt stress. Most of these genes were related to the structure of the photosystem complex, the electron transport chain and the connection between photosystem complexes and light-harvesting proteins (Table 1). Only 11 DEGs in M-81E mapped to the photosynthesis pathway with 9 down-regulated genes and 2 up-regulated genes, which were related to the stable assembly of the oxygenevolving complex and ATP synthase. These results suggest that salt stress could damages the structure of the photosystem and reduces the efficiency of electron transportation, which may result in decreased ATP and NADPH levels in plants under salt stress. This damage is particularly severe in salt-sensitive species, while salttolerant species can protect important connective structures from being destroyed by keeping low concentration of Na + in leaves and increasing the expression of particular genes. As a result, there was no significant change in Fv/Fm of M-81E, while Fv/Fm of Roma decreased gradually with the increasing NaCl treatments. Furthermore, the decrease in ΦPSII was more significant in Roma (Fig. 2). There were 7 and 13 DEGs mapped to the photosynthesis-antenna proteins pathway in M-81E and Roma, respectively. All of these genes were downregulated under salt stress ( Table 1), suggesting that the ability to capture and convert light energy of both genotypes was affected by salt stress. As light-harvesting complexes contain most of the chlorophyll and carotenoid pigments, the down-regulation of those genes resulted in decreases in chlorophyll content (Fig. 3), particularly in the salt-sensitive genotype Roma.
Rubisco, PEPC and PPDK are well known as the key enzymes in the dark reaction of photosynthesis. Rubisco plays an important role in CO 2 assimilation. Salt stress led to a reduced expression of Rubisco encoding gene in both M-81E and Roma, suggesting that salt stress reduced the efficiency of CO 2 assimilation in sweet sorghum as has been shown in previous studies [50][51][52][53]. To our surprise, the expression of the gene encoding NADP-ME was extremely enhanced by salt stress in M-81E. NADP-ME is important for the carbon fixation pathway because it catalyzes the reversible oxidative decarboxylation of L-malate to produce CO 2 , pyruvate and NADPH [54,55]. It has been shown that the expression of the gene encoding NADP-ME can be activated by salt stress [56,57]. NADP-MEs are not only an important for photosynthesis, but are also involved in plant defense reactions and environmental stress responses [58]. In our study, the gene encoding NADP-ME was up-regulated by salt stress in M-81E which could increase the content of CO 2 , pyruvate and NADPH. The increasing CO 2 and pyruvate levels enhanced the efficiency of the dark reaction of photosynthesis. In our research, photosynthetic rate of M-81E was not significantly affected by salt stress (Fig. 4), which might be related to the up-regulation of NADP-ME. After treated with 150 mM NaCl, in Roma, photosynthetic rate and stomatal conductance decreased, while the intracellular CO 2 concentration increased (Fig. 4). This showed that the decrease of photosynthetic rate was attributed to the non-stomatal factors, which might be related to the down expression of Rubisco. Recent studies showed that NADP-ME plays a role in enhancing tolerance of plants to salt stress [57,59]. Salt stress can produce superabundant reactive oxygen species (ROS) causing oxidative stress in plants [60][61][62]. Additionally, NADPH provides the reducing power required for ROS metabolism [63]. Møller and Rasmusson (1998) reported that NADPH can be used by the NADPH-specific glutathione reductase (GR) to catalyze the reduction of glutathione for scavenging ROS by an ascorbate coupled system [64]. In our study, the upregulation of NADP-ME encoding gene may play a role in the stress response and in the dark reaction of Fig. 11 Validation of RNA-seq results by RT-qPCR. Expression levels of 3 genes involved in sucrose synthesis and metabolism pathways were detected by RT-qPCR. R 2 represents the correlation coefficient value between the two platforms. The numbers in the scale bar stand for RPKM values in RNA-seq and ΔΔCt in qRT-PCR, which were used to evaluate the correlation (R 2 ). Primers are listed in (Additional file 9: Table S2) photosynthesis in salt-tolerant species of sweet sorghum. The increase of NADP-ME content enhanced the recycling of CO 2 in the C4 pathway. Furthermore, it may reduce the damage caused by ROS.
After treated with NaCl for 7 days, there was a significant difference in sugar content between M-81E and Roma. The sugar content increased 99.7 % in M-81E and decreased 30.5 % in Roma under 150 mM NaCl, suggesting that salt stress strongly induced the accumulation of sugar in salt-tolerant genotype of sweet sorghum (Fig. 5). Sucrose is the main source of carbon and of energy the sink tissues of sweet sorghum [8]. The cytoplasm of leaves is the site for sucrose synthesis. After synthesized, sucrose will be loaded into phloem and transported to sink tissues (stem and/or panicle). Various enzymes involved in sugar metabolism are required to ensure that sucrose is synthesized efficiently and the flow of sucrose is unidirectional (from source to the sink) [65]. SS is known to play a role in sucrose synthesis and its activity is high in source tissues such as leaves [43]. SPS can synthesize sucrose phosphate, which is converted to sucrose by sucrose phosphate phosphatase (SPP) in source tissues and then loaded into phloem. INV plays the most important role in the decomposition of sucrose. The vacuolar invertase activity is high in rapidly growing tissues [66]. Sucrose transporters (SUTs) are important exporters of photosynthetically-produced sugar, principally sucrose, from leaves to sink tissues [67]. It has been reported that the expression of sucrosemetabolizing enzymes play an important role in the accumulation of sucrose. In our study, the SS gene was up-regulated only in M-81E by salt treatment. However, the gene encoding INV was found to be up-regulated in Roma but down-regulated in M-81E (Additional file 8: Figure S7, Table 1). These findings suggest that salttolerant species of sweet sorghum accumulate more sucrose by enhancing the synthesis and reducing the decomposition of sucrose under salt stress. While, saltsensitive species enhance decomposition of sucrose under salt stress in order to meet the energy demand of growth. Furthermore, genes encoding SUTs showed no differential expression after treated with 150 mM NaCl in either genotype, suggesting that the transportation of sucrose from leaves to stem is not affected by salt stress treatment for 48 h at seedling stages.

Conclusions
In conclusion, we report here that the salt-tolerant genotype M-81E can increase sugar content under salt stress. This may be caused by the changes in expression level of genes related to important structures of photosystems and LHCs and genes encoding key enzymes of sucrose synthetase and sucrose decomposition under salt stress (Fig. 12). This RNA-seq dataset is an important resource for future studies aimed at improving sugar content of sweet sorghum under salt stress. Further genetic and biochemical analysis will be critical to understanding the detailed gene function and the relationship between salt tolerance and sugar content in sweet sorghum.

Plant materials and growth conditions
Seeds of two sweet sorghum genotypes M-81E and Roma were used as the experimental materials in this study. M-81E is considered to be tolerant to salt stress [9], while Roma is sensitive to salt stress. After being washed with tap water for 8 h, plump seeds were selected and sowed in plastic pots filled with river sand and irrigated with tap water. After germination, they were irrigated with 1/2 Hoagland solution in controlled growth chambers. The seedlings were cultured at 28 ± 3/23 ± 3°C (day/night) at a light intensity of 600 μmol m −2 s −1 (15 h photoperiod) and 70 % relative humidity. Salt treatment was performed at the four-leaf stage. The treated plants were irrigated with nutrient solution supplemented with 0, 50 and 150 mM NaCl. The NaCl concentrations were increased stepwise towards the final concentrations by 50 mM each day.

Measurement of fresh weight and dry weight
After exposure to salt treatments for 7 days, 15 plants from each treatment (5 per replicate) were sampled to determine leaf length and leaf number. Leaves were then separated and their FW were directly determined. For DW determination, the leaves were weighed after being dried at 150°C for 15 min and 70°C for 72 h. Water content (WC) was defined as follows: WC (%) = [(FW − DW)/ FW × 100.

Analysis of inorganic ions
For Na + and K + analysis, samples of dried leaves were ashed in a furnace for 6 h at 500°C. The ash was dissolved in 20 % nitric acid, diluted in distilled water and filtered through a sheet of filter paper. Na + and K + contents were determined by flame emission photometry (Flame Photometer 410,UK). Inorganic ion contents were expressed in mg g −1 DW.

Measurements of chlorophyll fluorescence
After exposure to salt treatments for 7 days, chlorophyll fluorescence was measured. For each treatment, the parameters of Chl fluorescence were measured independently on five plants. Measurements were taken on the mature leaves of each of the chosen plants. Chl fluorescence was measured using a portable fluorometer (FMS2, Hansatech, King's Lynn, UK) following the protocol described by Kooten and Snel [68]. Leaves had been pre-darkened for at least 1 h in order to determine the minimal and the maximal fluorescence. Minimal fluorescence (Fo) with all PSII reaction centers open was determined by modulated light which was low enough not to induce any significant variable fluorescence (Fv). Maximal fluorescence (Fm) [43] with all reaction centers closed was determined by 0.8 s saturating light of 8000 μmol m −2 s −1 on a dark-adapted leaf. Then the leaf was illuminated by an actinic light of 500 μmol m −2 s −1 . Steady-state fluorescence (Fs) was recorded when the leaf reached steady-state photosynthesis. A second 0.8 s saturating light of 8000 μmol m −2 s −1 was given to determine maximal fluorescence in the light-adapted state (Fm'). Maximal photochemical efficiency (Fv/Fm) of PSII was expressed as: Fv/Fm = (Fm-Fo)/Fm. Quantum yield of PSII electron transport was: ΦPSII = (Fm'-Fs)/Fm'.

Measurement of chlorophyll content
Leaves (0.2 g FW) were washed in distilled water and extracted in 5 ml 80 % acetone and 5 ml dimethyl sulfoxide at 65°C in darkness for 24 h. The extract was adjusted to a total volume of 25 ml with 80 % acetone. The absorbance of the extract was determined at 663 and 645 nm using a spectrophotometer of BECKMAN DU2600. The amount of total chlorophyll was calculated using the Arnon [69] formulae as follows: Chla(mg/g) = (12.

Measurement of photosynthesis
photosynthetic rate, stomatal conductance and intercellular CO 2 concentration were measured using Li-6400 photosynthesis measurement system. Measurements were taken on the mature leaves of each of the chosen plants.

Measurement of sugar content
After exposure to salt treatments for 7 days, 15 plants from each treatment (5 per replicate) were sampled to determine sugar content. Sugar content were determined by the anthrone method described by Spiro [70]. 100 μl leaf extract were added to 3 ml (final volume) assay media containing 1.08 M H 2 SO 4 , 1.09 mM thiourea and 2.1 mM anthrone. The mixture was heated at 100°C for 10 min and absorbance read at 620 nm. A calibration curve with D-glucose was done as a standard.

Total RNA extraction
Total RNA was isolated from the leaves of sweet sorghum of each genotype treated with 0 and 150 mM NaCl for 48 h using a Total Plant RNA Extraction Kit (Karroten, Beijing, China) following the manufacturer's protocols. The RNA was quantified using a Nanodrop ND-1000 spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA). A 1 % agarose gel buffered by Tris-acetate-EDTA was also run to determine the integrity of the RNA.

Library construction and Illumina sequencing
Libraries were constructed following a High Throughput Illumina Strand-Specific RNA Sequencing Library protocol [71]. Briefly, mRNA was purified from 5 μg of total Fig. 12 visualization of DEGs involved in pathways related with the accumulation of sugar. A square block represents a gene assigned to our RNA-seq data. Blue represents the gene was down-regulated in salt-treated samples compared to the control samples. Red represents the gene was up-regulated. For each gene, the square block on the left stand for M-81E and the right one stand for Roma RNA using oligo (dT) magnetic beads. The purified mRNA was fragmented into small pieces using fragmentation buffer. Taking these short fragments as templates, first-strand cDNA was synthesized using reverse transcriptase and random hexamer primers. Second-strand cDNA synthesis was followed using DNA polymerase I and RNase H. Sequencing adapters were ligated to short fragments after purification with the QiaQuick PCR extraction kit, and which were used to distinguish different sequencing samples. Fragments with different lengths were then separated by agarose gel electrophoresis and selected for PCR amplification as sequencing templates. The final cDNA library was sequenced using Illumina HiSeq™ 2500 at BioMarker Technologies Co Ltd, Beijing. RNA-seq data of the untreated control and salttreated samples were obtained from two and three biological replicates, respectively. The raw reads were cleaned by removing adaptor sequences, empty reads and low quality sequences. Then, clean reads were generated.

Mapping and detection of DEGs
Clean reads were mapped to the sorghum genome [72,73] using TopHat version 2.0.10 [74]. Mapping results generated by TopHat were filtered to retain only unique mapped reads before being piped into Cuffdiff (http://cole-trapnell-lab.github.io/cufflinks/) to estimate read counts for each gene. Reads per KB per million (RPKM) values were calculated by an in-house script based on the count table of Cuffdiffs output. The RPKM measure of read density reflects the molar concentration of a transcript in the starting sample by normalizing for RNA length and for the total read number in the measurement. A RPKM threshold value of 0.1was set to detect the presence of a transcript for a particular gene. DEGs were defined using DESeq [75] as fold changes≧2 with a false discovery rate (FDR) adjusted p value ≤0.01.

Gene annotation and classifications
The optimal assembly results were chosen according to the assembly evaluation. The assembled sequences were compared against the NCBI non-redundant (nr) database [76], Swiss-Prot [77], GO [23], COG [24] and KEGG [25] database using BLAST [78] with E-value ≦ 1e-10 as the cutoff. To annotate the assembled sequences with GO terms, the Swiss-Prot BLAST results were imported into Blast2GO [79]. These GO terms were assigned to query sequences, producing a broad overview of groups of genes catalogued in the transcriptome for each of three ontology vocabularies, biological processes, molecular functions [50] and cellular components [23]. The unigenes sequences were also aligned to the COG database to predict and classify functions. KEGG pathways were assigned to the assembled sequences using the online KEGG web server (http://www.genome.jp/ kegg/). The output of KEGG analysis includes KO assignments and KEGG pathways that are populated with the KO assignments.

Quantitative real-time PCR analysis
Fourteen DEGs were randomly selected for quantitative real-time PCR to verify the RNA-seq results. Also, three genes which may play important roles in improving sugar content in sweet sorghum were confirmed by qPCR, too. Primers for these 17 genes were designed using the Beacon Designer software (version 7.0) (Additional file 9: Table S2). S. bicolor's housekeeping gene β-actin (Gen-Bank ID: X79378) was used as an internal standard. 1 μg of total RNA was used per 20 μl reaction for reverse transcription. Polymerase chain reaction was performed in a 20 μl reaction mixture with 10 μl SYBR Premix Ex Taq (Bio-RAD, California, USA), 0.5 μl of both forward and reverse primers, 7 μl of double distilled H 2 O and 2 μl (40 ng/μl) of the cDNA. Real-time PCR was performed on a real-time quantitative PCR instrument (Bio-RAD, California, USA). 2 -△△Ct method was used to calculate the relative expression of each gene [80].

Statistical analysis
Multiple comparisons were performed between different samples using Duncan's test at the 0.05 significance level. All tests were performed with SPSS Version 16.0 for Windows (SPSS, Chicago. IL, USA).

Availability of supporting data
The data sets supporting the results of this article are included within the article and its additional files. The reads produced in this study have been deposited in the National Center for Biotechnology Information (NCBI) SRA database with accession number of SRX1048181, SRX1050048 for M-81E control, SRX1050049, SRX1050050 for M-81E salt-treated, SRX1050054, SRX1050055 for Roma control, and SRX1050056, SRX1050057 for Roma salt-treated. Access to the data is available upon publication at http:// www.ncbi.nlm.nih.gov/sra/.

Additional files
Additional file 1: Figure S1. Effect of increasing NaCl concentration on leaf length and leaf number of Roma and M-81E under 3 salt treatments (0, 50 and 150 mM) for 7 days. Values are means ± SD of five replicates. Bars with the different letters are significantly different at p = 0.05 according to Duncan's multiple range test.
Additional file 2: Figure S2. Fresh weight, dry weight and water content of M-81E and Roma treated with different concentrations of NaCl (0, 50 and 150 mM) for 7 days. Values are means ± SD of five replicates. Bars with the different letters are significantly different at p = 0.05 according to Duncan's multiple range test. Bars with same letter are not significantly different.