Metabolome and molecular basis for carbohydrate increase and nitrate reduction in burley tobacco seedlings by glycerol through upregulating carbon and nitrogen metabolism

Burley tobacco (Nicotiana Tabacum) is a chlorophyll-deficiency mutant. Nitrate is one precursor of tobacco-specific nitrosamines (TSNAs) and is largely accumulated in burley tobacco. To decrease nitrate accumulation in burley tobacco, glycerol, a polyhydric alcohol compound and physiological regulating material, was sprayed and its effects were investigated based on metabolomic technology and molecular biology. The results showed that glucose, glutamine and glutamic acid increased by 2.6, 5.1 and 196, folds, respectively, in tobacco leaves after glycerol application. Nitrate content was significantly decreased by 12–16% and expression of eight genes responsible for carbon and nitrogen metabolism were up-regulated with glycerol applications under both normal and 20% reduced nitrogen levels (P < 0.01). Leaf biomass of plants sprayed with glycerol and 20% nitrogen reduction was equivalent to that of no glycerol control with normal nitrogen application. Carbohydrates biosynthesis, nitrate transport and nitrate assimilation were enhanced in glycerol sprayed burley tobacco seedlings which might contribute to reduced nitrate and increased carbohydrates contents. In conclusion, glyerol spray coupled with 20% nitrogen reduction would be an effective method to reduce nitrate accumulation in burley tobacco.


RNA-Seq Statistics, and analysis of DEGs induced by deficient nitrogen.
In this study, leaf biomass was significantly lowered for low nitrogen plants than that for normal nitrogen plants (P < 0.01, Fig. 1e). There were twelve cDNA libraries (2 varieties * 2 nitrogen application levels * 3 biological replicates) prepared to analyze the effects of nitrogen deficiency on metabolism in burley tobacco seedlings. After removing sequencing adaptors and low quality reads, 72.56 M reads were obtained in tobacco leaves under nitrogen sufficiency and nitrogen deficiency conditions. A total of 85.14% reads from TN90 and 84.82% reads from TN86 were mapped to the reference genome with almost 81% of them having unique location in that genome.
The DEGs induced by nitrogen deficiency both in TN90 and TN86 at the same time were highly obtained. In all, 428 of up-regulated common genes and 213 of down-regulated common genes were analyzed, respectively (Fig. 1a,c). There are 12.85% of up-regulated common genes involved in the biological process of defense response (GO:0050832, GO:0042742, GO:0031347) (Fig. 1d). There are almost 15.96% of down-regulated common genes correlated with biological process (GO-BP) of photosynthesis (GO:0009765, GO:0015979). We observed high values for categories involved in gene ontology cellular component (GO-CC), such as photosystem I (GO:0009522), photosystem II (GO:0009523), chloroplast thylakoid membrane (GO:0009535), cytosol (GO:0005829). Genes mostly involved in gene ontology molecular function (GO-MF), such as chlorophyll binding (GO:0016168), glutamate synthase (NADH) activity (GO:0016040).
qRT-PCR analysis of 8 DEGs was conducted to validate the accuracy of RNA-Seq date. Expression patterns of selected genes were consistent with those in RNA-seq assay, indicating that results of RNA-Seq were reliable.

Differences of metabolites determined by GC/MS. Fifteen metabolites determined by metabolome
between treatments were different, including amino acids and carbohydrates ( Table 1). The result of PCA was showed that differences between treatments were substantial (Fig. 2). Amino acids and carbohydrates content were increased by spraying glycerol, including glucose, sucrose, fructose-6-phosphate, succinic acid and fumaric acid, which were mainly correlated with metabolic pathway of glycolysis, TCA cycle and starch and sucrose metabolism. Glutamine and L-glutamic acid were also significantly increased by spraying glycerol (P < 0.01).
Seven days after spraying glycerol, differences in genes expression levels related to carbohydrates biosynthesis and nitrate assimilation between treatments were all substantial under both sufficient and 20% nitrogen reduction conditions. And expression of genes (SUS2-2, SPS) involved in glucose and starch synthesis were all up-regulated (Fig. 4), which was consist with the increase of carbohydrate in burley tobacco seedlings. Expression of genes GLPK, gpmA and PGK correlated with glycerol metabolism were up-regulated by spraying glycerol under SCientifiC RepoRtS | (2018) 8:13300 | DOI:10.1038/s41598-018-31432-3 different nitrogen conditions. After spraying glycerol, changes of genes expression levels were greater at 7 days after than at 2 days after. Glycerol induced the biosynthesis of carbohydrates and chlorophyll, which would lead to increase of carbohydrate contents and biomass accumulation in burley tobacco seedlings. Moreover, expression of genes (NLP7, NIA1, NPF3.1 and NPF7.3) correlated with nitrate transport and assimilation were all up-regulated by glycerol.
Increase of chlorophyll and carbohydrate formation by glycerol. Differences in leaf biomass, parameters of carbon and nitrogen metabolism and its products between varieties, treatments and their interactions were substantial in tobacco leaves (Table 2). Leaf biomass, chlorophyll a contents and photosynthetic rates in both TN90 and TN86 significantly increased by spraying glycerol under both normal and reduced nitrogen conditions, respectively (P < 0.05, Figs 5 and 6). It is noteworthy that the photosynthetic rate and contents of chlorophyll a and other pigments, leaf biomass in glycerol-sprayed plants with 20% less nitrogen application were all equivalent to those in no glycerol-sprayed plants with normal nitrogen application. The total sugar, soluble reducing sugar contents and biomass accumulation (root, stem, leaf) in both TN90 and TN86 significantly increased by glycerol treatment under two different nitrogen conditions, respectively (P < 0.01, Fig. 6).

Reduction of nitrate and increase of soluble protein by glycerol application. Total nitrogen and
NO 3 -N content were significantly reduced by spaying glycerol under different nitrogen conditions in burley tobacco seedlings (P < 0.01, Fig. 7). In addition, soluble protein content in burley tobacco seedlings significantly increased after spaying glycerol at different nitrogen levels, indicating that the ability of nitrate reduction and assimilation in burley tobacco seedlings were enhanced by glycerol application (P < 0.05). Moreover, soluble protein content in glycerol-sprayed plants under 20% less nitrogen application condition was similar to the normal levels, while the NO 3 -N content was low.

Discussion
Burley tobacco is one typical yellow-green leaf tobacco due to chlorophyll mutation, and chlorophyll and carbohydrate content are always lower than other tobacco types during the whole period of growth 24 . Nitrogen fertilizer application in burley tobacco is normally 4-5 fold higher than other tobacco types to produce the same levels of biomass, which resulted in low nitrogen use efficiency and more than 100 times higher of nitrate content in burley tobacco, further contributing to much more TSNA formation during both leaf curing and storage 2,3 . Transcriptomic analysis showed that expression of genes involved in photosynthesis were decreased by reducing nitrogen application, resulting in decreased leaf biomass accumulation and carbohydrates formation in burley  tobacco 25 . Photosynthesis rate and leaf biomass accumulation for the treatment of 20% nitrogen reduction coupled with glycerol spray were similar to those in the control with sufficient nitrogen application, representing that glycerol can compensate one part of nitrogen fertilizer reduction in burley tobacco cultivation.
In burley tobacco, we found that low pigment and carbohydrates and weak ability of nitrogen assimilation were the key factors contributing to nitrate accumulation 7,8 . And nitrate concentration can be diluted with increased biomass in plant at the same nitrogen level 7 . In this work, spraying glycerol decreased nitrate accumulation in burley tobacco seedlings. For a better understanding of glycerol application in reducing nitrate and rising carbohydrates, technologies of molecular biology combined with metabolomes were used to explore the molecular mechanisms. Spraying glycerol has increased expression of genes involving into response to light stimulus, carbon fixation, nitrate transport and assimilation, and photosynthesis rate, nitrate assimilation and transportation abilities, and carbohydrates biosynthesis (glucose, starch and sucrose) were greatly induced in two burley tobacco varieties under different nitrogen conditions, indicating that spraying glycerol might enhance the ability of carbon and nitrogen assimilation in burley tobacco seedlings.
After spraying glycerol, glycerol metabolism was induced, which increased carbohydrates formation in burley tobacco seedlings. Glycerol is catalyzed to glycerol-3-phosphate (G-3P) by glycerol kinase (encoded by GLPK gene) and gene GLPK was significantly induced by spraying glycerol (P < 0.01). Glycerol-3-phosphate (G-3P) is one key crossroads of glucose, lipid and energy metabolism 26   nitrogen applying conditions, indicating that glycerol metabolism was greatly induced in burley tobacco seedlings (P < 0.05). In addition, genes related to starch and sucrose biosynthesis were also up-regulated after spraying glycerol, including SPS, which catalyzes the limiting steps in sucrose synthesis 28 , SUS2-2, which plays a dominant role in generating precursors for starch biosynthesis 29,30 . In analysis of metabolome, glucose, glucuronic acid and fructose-6-phosphate were markedly increased, which was consistent with increased expression of genes involved in carbohydrates biosynthesis.
Carbon metabolism supplies carbon skeleton and energy for nitrogen metabolism 31,32 . Nitrate is largely accumulated and carbohydrates are extremely low in burley tobacco, which are associated with one double recessive genes mutation 2 . In this study, nitrate content was significantly decreased and leaf biomass was significantly increased by spraying glycerol (P < 0.01), and the effect of 20% nitrogen reduction with spraying glycerol was the most effective in decreasing nitrate content in burley tobacco seedlings. Transcription factor NLP7 is a positive regulator of nitrate-induced expression of N-related genes through post-translation regulation, such as genes NRT2.1, NITR2:1, NIA1, NIR1, which plays an important role in nitrate assimilation pathway [33][34][35] . NLP7 was up-regulated by spraying glycerol, which was conducive to decreasing nitrate accumulation in burley tobacco leaves. Nitrate reductase (EC 1.6.6.1, encoded by NIA genes) catalyzes the limiting-rate step of nitrate reduction and assimilation in most organisms 36 . Expression of gene NIA1 was up-regulated and the products of nitrogen  assimilation (glutamic acid, glutamine and soluble protein) were increased by spraying glycerol in burley tobacco seedlings, indicating that glycerol application promoted nitrogen assimilation. Nitrate will be difficult to utilization once it is stored 37 . Nitrate transport was lower in burley tobacco than in other tobacco types 8 . NPF3.1 and NPF7.3 are low-affinity proton-dependent bidirectional nitrate transporter that are involved in regulation of nitrite uptake into chloroplasts and nitrate loading into xylem in higher plant 38 . Expression of genes NPF3.1 and NPF7.3 were up-regulated by spraying glycerol in burley tobacco seedlings, indicating that glycerol enhanced nitrate transport. These results showed that glycerol could increase the ability of nitrate transport and nitrate assimilation in burley tobacco seedlings and decrease nitrate content under 20% reducing nitrogen application conditions. And 20% nitrogen reduction with spraying glycerol would be an effective approach of decreasing nitrate accumulation in burley tobacco. In conclusion, spraying glycerol decreased nitrate accumulation and promoted carbohydrates formation in burley tobacco seedlings at different nitrogen levels. Reducing nitrogen application inhibited the expression of genes involved in photosynthesis and reduced leaf biomass. By spraying glycerol, chlorophyll content, photosynthesis rate, total soluble sugar, reducing soluble sugar content and leaf biomass were all increased. Total nitrogen and nitrate contents were significantly decreased by glycerol application in burley tobacco seedlings under different nitrogen conditions (P < 0.01). Spraying glycerol under 20% less nitrogen application condition was an effective method to reduce nitrate accumulation in burley tobacco seedlings while maintaining same levels of leaf biomass. Glycerol triggered the biosynthesis of carbohydrates and promoted the abilities of nitrate reductase, nitrate assimilation and nitrate transport, leading to the increase of carbohydrates and to the reduction of nitrate in burley tobacco seedlings.

Materials and Methods
Plant materials. Seeds of burley tobacco varieties TN90 and TN86 were sown in a floating system in greenhouse that maintained a temperature ranging from 19 °C (night) to 28 °C (day), average photosynthetic photon flux density of 600 μmol m −2 s −1 and relative humidity 80%. Seedlings (sown 40 days later) were transplanted in 25 cm × 30 cm (diameter × depth) plastic pots (plant/pot) while seedlings had four to five permanent leaves. Seedlings were cultivated with Hoagland solution containing either 4 mM, 24 mM and 19.2 mM nitrogen levels 8 . All nutrient solutions were continuously aerated with an air pump. After germination, nutrient solution was replaced every six days. Nutrient solutions were refreshed every two days when seedlings were transplanted (sown 30 days later) in plastic pots. Seedlings were treated after three days (for recovery). 0.1% glycerol. Nitrogen sufficiency level: (1) N100-CK, 24 mM nitrogen level, spraying pure water; (2) N100-Gl, 24 mM nitrogen level, spraying 0.1% glycerol. Every treatment had three biological replicates. In our preliminary research, seven glycerol concentrations of 0.00%, 0.025%, 0.05%, 0.075%, 0.1%, 0.15% and 0.2% were compared to test their effects on reducing nitrate content in burley tobacco, and 0.1% was proved to be the most effective concentration.
The two sides of all leaves were sprayed by glycerol with a manual pump at different nitrogen levels (around 8: 00-9: 00 am). Control seedlings were sprayed with same volume of pure water. Sampling and determination were carried out after seedlings being treated 2 days and 7 days (around 10: 00 am), respectively.
Leaves of five plants from each treatment were mixed and frozen in liquid nitrogen immediately, then kept at −80 °C. Fifteen plants each treatment were separated into root, stem and leaf and then deactivated at 105 °C for 20 min and dried at 60 °C for 48 h. Tissues (root, stem and leaf) were weighed and ground to pass through a screen with 60 meshes, and the final powder mixture was used to determine nitrate, total nitrogen, total soluble sugar and soluble reducing sugar content in plant.
Note: It was reported that glycerol inhibited cotyledon greening and development of true foliar leaves at 100 mM (0.728%, V/V) 11 . And the dose of glycerol (0.1%, V/V) in this work was much lower than the concentration of 100 mM.
Measurement of pigment content and photosynthetic rate. Pigment content was determined by 95% ethanol 39 . Photosynthetic rate (Pn) was measured with a portable photosynthesis system (LI-COR Biotechnology, 6400XT, Lincoln, NE, USA) 40 .
Measurement of carbonitride content. Nitrate content was determined by the method described by Cataldo 41 . Soluble protein content was assayed according to Li 42 . Total nitrogen, total soluble sugar and reducing sugar content were determined according to methods modified from the Chinese Tobacco Industry standard (YC/T 161, 159-2002) 8 .
RNA Extraction, Preparation of cDNA Library, and Sequencing. Total RNA was extracted using the mirVana miRNA Isolation Kit (Ambion) following the manufacturer's protocol. RNA integrity was evaluated with the Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). Samples with RNA Integrity Number (RIN)≥7 were subjected to the subsequent analysis. TruSeq Stranded mRNA LT Sample Prep Kit (Illumina, San Diego, CA, USA) was applied to conduct the libraries by the manufacturer's instructions. Then the libraries were built on the Illumina sequencing platform (HiSeqTM 2500 or Illumina HiSeq X Ten) and 125 bp/150 bp paired-end reads were generated. The quality control was evaluated on the remaining reads according to NGS QC Toolkit 43 . Low quality date was removed, and 92.86% of Q20 percentage from clean reads was 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/) 44,45 . RNA-Seq analysis. Transcript profiles of RNA-seq were analyzed by calculating the read fragments per kilo base per million mapped reads (FPKM). FPKM value of each gene was calculated using cufflinks, and the read counts of each gene were obtained by htseq-count 46,47 . DEGs were identified using the DESeq (2012) functions estimate Size Factors and nbinom Test 48 . In the process of DEGs screening, fold change (FC)>2 or FC <0.5, p-value < 0.05, was used as threshold to determine the significance of gene expression differences between nitrogen deficiency condition and nitrogen sufficiency condition; moreover, genes of DEGs, both nitrogen sufficiency and nitrogen deficiency condition, with FPKM < 1 were removed. FC is radio of FPKM between nitrogen sufficiency and nitrogen deficiency conditions. Gene function was annotated based on databases of NR (NCBI non-redundant protein sequences), KOG (Clusters of Orthologous Groups of proteins) 49 , Swiss-Prot (A manually annotated and reviewed protein sequence database) 50 , KO (KEGG Ortholog database) 51 , GO (Gene Ontology) 52 . GO enrichment and KEGG pathway enrichment analysis of DEGs were respectively achieved using R based on the hypergeometric distribution. Heatmaps analysis of genes expression was generated with R (3.4.1 version) pheatmap package 53 . Gene expression analysis by qRT-PCR. Eight genes determined by qRT-PCR were randomly selected to validate the transcript levels obtained by RNA-Seq (Experiment 1). RT reactions were performed in a Gene Amp ® PCR System 9700 (Applied Biosystems, Foster City, USA) and Gene Amp ® PCR System 9700 (Applied Biosystems, Foster City, USA). Expression of twelve genes correlated with glycerol metabolism, carbohydrates biosynthesis and nitrogen metabolism were observed (Experiment 2). Real-time PCR was performed using Light Cycler ® 480 II Real-time PCR Instrument (Roche, Basel, Swiss). Reactions were incubated in a 384-well optical plate (Roche, Basel, Swiss) at 95 °C for 5 min, followed by 40 cycles of 95 °C for 10 s, 60 °C for 30 s. L25 was used as the endogenous control (Supplementary Tables S1, S2). The expression levels of mRNAs were normalized and calculated using the 2 −ΔΔCt method 54 . GC-TOF-MS Analysis. GC-TOF-MS analysis was performed using an Agilent 7890 gas chromatograph system coupled with a Pegasus HT time-of-flight mass spectrometer. The system utilized a DB-5MS capillary column coated with 5% diphenyl cross-linked with 95% dimethylpolysiloxane (30 m × 250 μm inner diameter, 0.25 μm film thickness; J&W Scientific, Folsom, CA, USA). A 1 μL aliquot of the analyte was injected in split less mode. Helium was used as the carrier gas, the front inlet purge flow was 3 mL min −1 , and the gas flow rate through the column was 1 mL min −1 . The initial temperature was kept at 50 °C for 1 min, then raised to 310 °C at a rate of 10 °C min −1 , then kept for 8 min at 310 °C. The injection, transfer line, and ion source temperatures were 280, 280, and 250 °C, respectively. The energy was −70 eV in electron impact mode. The mass spectrometry data were acquired in full-scan mode with the m/z range of 50-500 at a rate of 20 spectra per second after a solvent delay of 6.27 min.

GC-MS analysis of metabolites. Metabolites
Data preprocessing and annotation. Chroma TOF 4.3 X software of LECO Corporation and LECO-Fiehn Rt x 5 database were used for raw peaks exacting, the data baselines filtering and calibration of the baseline, peak alignment, deconvolution analysis, peak identification and integration of the peak area 55 . Both of mass spectrum match and retention index match were considered in metabolites identification. . For comparison between two data sets, a Student's t test was used. For analysis of three or more sets of data, individual comparisons between mean values performed by using the least significant differences (LSD) test. *P < 0.05, **P < 0.01 were considered statistically significant.

Data Availability
All of the materials, data and associated protocols will be made available upon request without preconditions. All data generated or analyzed during this study are included in this published article (and its Supplemental Information files).