Abstract

   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.

Introduction

   Burley tobacco is one type of chlorophyll deficiency, and is well known
   as yellow-green leaf tobacco with reduced chlorophyll content and lower
   carbohydrate accumulation^[32]1,[33]2. This mutations is correlated
   with one double homozygous recessive genotypes at Yellow Burley 1 (YB1)
   and Yellow Burley 2 (YB2) loci, and has been reported that the changes
   resulted in low nitrogen use efficiency and high levels of nitrate
   accumulation which is one of the important precursors of
   tobacco-specific nitrosamines (TSNAs)^[34]3,[35]4. TSNAs are easy to
   induce malignant tumors in animals and was classified as group 1
   carcinogens by International Agency for Research on Cancer^[36]5.
   Nitrate was greater in burley tobacco than in other tobacco
   types^[37]6. Nitrate is mainly accumulated during plant growth and
   development in the field^[38]7. Mostly, the low capacity of nitrogen
   assimilation and translocation lead to nitrate accumulation in burley
   tobacco^[39]8. Moreover, decreasing nitrate content in burley tobacco
   is always the important approach to lower the formation of TSNAs. There
   were limited reports related to strategies aiming at nitrate reduction
   during the process of tobacco growth and development.

   In general, nitrogen metabolism is highly correlated with carbon
   metabolism in plants. Photosynthesis carbohydrates provide carbon
   backbones for nitrogen assimilation^[40]9. Most enzymes involved in
   nitrogen assimilation require reducing powers for their reactions in
   chloroplasts, and their activities are highly regulated by
   photosynthesis and its products^[41]10. Glycerol, which can supply
   carbon atom in conditions of reduced photosynthesis carbon fixation in
   plant^[42]11, was involved in cell growth, maintenance and
   metabolism^[43]12. Glycerol can freely inter into cells through water
   channels^[44]13. It is reported that glycerol is transported across the
   plasma membrane via MIPs/aquaporins^[45]14. Glycerol played an
   important role in cellular osmoregulation, photosynthetic carbon
   metabolism and starch metabolism^[46]15,[47]16, and it can be converted
   to sucrose by glycolysis and gluconeogenesis^[48]17. There are two
   pathways reportedly related to glycerol metabolism in plant^[49]18. The
   first one, glycerol is phosphorylated to glycerol-3-phosphate (G-3-P)
   by glycerol kinase (EC 2.7.1.30) and then G-3-P is catalyzed to
   dihydroxyacetone phosphate (DHAP) by a G-3-P dehydrogenase
   (GPDH)^[50]19. In this pathway, the enzyme of glycerol kinase plays a
   rate-limiting role in glycerol metabolism^[51]20,[52]21. In the second
   pathway, glycerol is converted to dihydroxyacetone by NAD^+-glycerol
   dehydrogenase (EC 1.1.1.6) and is then phosphorylated to DHAP by
   dihydroxyacetone kinase (EC 2.7.1.29)^[53]22, and DHAP is involved in
   the process of glycolysis and gluconeogenesis for sugar production. In
   preliminary work, spraying glycerol had largely decreased nitrate
   accumulation in vegetables but effects and its mechanisms were not
   clearly demonstrated^[54]23. The hypotheses of reducing nitrate in
   burley tobacco by glycerol were conceived: (1) Burley tobacco with low
   chlorophyll content and carbohydrates formation is one chlorophyll
   deficiency mutant, and carbon skeleton and energy might be insufficient
   for nitrate reducing and assimilation; (2) Glycerol can be converted to
   sucrose through the glycolysis and gluconeogenesis, which might
   increase carbohydrates formation and provide more carbon skeleton and
   energy supply; (3) Nitrate transport might be induced by sufficient
   energy and nitrate assimilation be promoted, leading to decrease of
   nitrate storage in tobacco leaves.

   In burley tobacco, high nitrate accumulation is correlated with
   nitrogen amount applied and reducing nitrogen level was effective in
   decreasing nitrate content^[55]3, although there is currently no
   in-depth coverage at molecular level in burley tobacco. From our early
   study, we have found that nitrogen application was 5-folds higher in
   burley tobacco than in flue-cured tobacco to produce similar leaf
   biomass, and lower contents of pigment and energy substances and weaker
   nitrogen assimilation were the key factors contributing to nitrate
   accumulation in burley tobacco^[56]7,[57]8. In this study, glycerol was
   applied to regulate carbohydrate and nitrate accumulation in two burley
   tobacco varieties. RNA sequencing technology was used to analyze the
   effects of reducing nitrogen fertilizer level on metabolism in burley
   tobacco seedlings, and metabolome analysis was used to explore the
   regulating effects of exogenous glycerol on carbohydrate biosynthesis
   and nitrate reduction in burley tobacco seedlings. Pigment content,
   photosynthetic trait, nitrate and carbohydrates content, expression of
   genes related to carbon and nitrogen assimilation and glycerol
   metabolism were investigated at sufficient and 20% reducing nitrogen
   applying levels, aiming to provide effective methods to decrease
   nitrate accumulation and TSNAs formation in burley tobacco cultivation.

Results

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. [58]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.

Figure 1.

   [59]Figure 1
   [60]Open in a new tab

   Transcriptome analysis strategies and GO enrichment of target DEGs and
   leaf biomass (a) Venn diagram of down-regulated DEGs of deficient
   nitrogen condition vs sufficient nitrogen condition in TN90 and TN86
   (b) GO enrichment in common down-regulated DEGs between TN90 and TN86
   (c) Venn diagram of up-regulated DEGs of deficient nitrogen condition
   vs sufficient nitrogen condition in TN90 and TN86 (d) GO enrichment in
   common up-regulated DEGs between TN90 and TN86 (e) Leaf biomass (f)
   RNA-seq results confirmed by quantitative qRT-PCR. Nd-TN90: TN90
   cultivated under deficient nitrogen condition; Ns-TN90: TN90 cultivated
   under sufficient nitrogen condition; Nd-TN86: TN86 cultivated under
   deficient nitrogen condition; Ns-TN86: TN86 cultivated under sufficient
   nitrogen condition. Error bars of leaf biomass indicate standard error
   of the means (n = 3). Error bars of qRT-PCR indicate standard error of
   the means (n = 6).

   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. [61]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. [62]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 [63]1). The
   result of PCA was showed that differences between treatments were
   substantial (Fig. [64]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).

Table 1.

   Profiles of metabolites in response to glycerol spray on burley tobacco
   seedling leaf under sufficient nitrogen condition.
   Pathway Name Fold Change P-Value
   Nitrogen metabolism glutamine 3.6 0.003
   glutamic acid 6.0 0.002
   Porphyrin and chlorophyll metabolism glycine 1.6 0.011
   Carbon fixation in photosynthetic organisms sedoheptulose 1.5 0.017
   Glycolysis, TCA cycle fructose-6-phosphate 2.4 0.014
   succinic acid 1.6 0.010
   fumaric acid 1.5 0.031
   glycerol 1.84 0.010
   Starch and sucrose metabolism glucose 200 0.001
   glucuronic acid 36 0.010
   sucrose 1.4 0.009
   trehalose-6-phosphate 1.4 0.112
   cellobiose 1.3 0.160
   maltose 1.2 0.329
   D-galacturonic acid 1.2 0.341
   [65]Open in a new tab

   Fold Change: Ratio of glycerol treatment to the control.

Figure 2.

   Figure 2
   [66]Open in a new tab

   PCA of metabolome of spraying glycerol at normal nitrogen level. The
   blue boxes indicate that control group with no spraying glycerol. The
   red circles indicate that treatments with spaying glycerol.

Expression levels of genes induced by glycerol

   Two days after spraying glycerol, expression of genes involved into
   response to light stimulus (EGY1, CP12-2), carbon fixation pathways
   (PGK, PPC16, GLPK), sucrose biosynthesis (SPS, SUS2-2), nitrate
   transport and nitrate assimilation (NIA1, NLP7, NPF7.3) were all
   up-regulated (Fig. [67]3). Compared to CK, expression of genes CP12-2,
   PPC16, NPF7.3 increased more than 0.5-fold after spraying glycerol,
   indicating that glycerol might enhance carbon fixation, photosynthesis
   and nitrate transport in tobacco leaves.

Figure 3.

   [68]Figure 3
   [69]Open in a new tab

   Light stimulus, carbon fixation, sucrose biosynthesis and nitrate
   assimilation induced by spraying glycerol at 2 days after spray.
   Expression levels of ten genes were obtained by qRT-PCR. (a) Genes in
   response to light stimulus. (b) Genes in carbon fixation pathways (c)
   Genes in sucrose biosynthesis pathways. (d) Genes in nitrate
   assimilation pathways. CK: control; Gl: glycerol. EGY1 (Probable zinc
   metalloprotease, Gene ID 833476), CP12-2 (Calvin cycle protein, Gene ID
   825414), PGK (Phosphoglycerate kinase, Gene ID 107787830), PPC16
   (Phosphoenolpyruvate carboxylase, Gene ID 547769), GLPK (glycerol
   kinase [Nicotiana attenuata], Gene ID 109221820), SPS
   (sucrose-phosphate synthase, Gene ID 107766133), SUS2_2 (sucrose
   synthase [Nicotiana attenuata], Gene ID 109214879), NIA1 (nitrate
   reductase [NADH] 1 [Nicotiana attenuata], Gene ID 107823732), NLP7
   (Protein NLP7, Gene ID 828502), NPF7.3 (Protein NRT1/ PTR family 7.3,
   Gene ID 840139). Error bars of qRT-PCR indicate standard error of the
   means (n = 3). Symbols **and * indicate that the significant
   differences between treatment of control and treatment of spraying
   glycerol are at 0.01 and 0.05, respectively.

   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. [70]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 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.

Figure 4.

   Figure 4
   [71]Open in a new tab

   Glycerol enhanced carbon and nitrogen metabolism in burley tobacco 7
   days after spray under different nitrogen levels. Expression levels of
   nine genes were quantified by qRT-PCR. Carbon metabolism: GLPK
   (glycerol kinase [Nicotiana attenuata], Gene ID 109221820), gpmA
   (2,3-bisphosphoglycerate-dependent phosphoglycerate mutase, accession
   number [72]BX908798), PGK (Phosphoglycerate kinase, Gene ID 107787830),
   SPS (sucrose-phosphate synthase, Gene ID 107766133), SUS2_2 (sucrose
   synthase [Nicotiana attenuata], Gene ID 109214879). Nitrogen
   metabolism: NLP7 (Protein NLP7, Gene ID 828502), NPF3.1 (Protein NRT1/
   PTR family 3.1, Gene ID 843186), NPF7.3 (Protein NRT1/ PTR family 7.3,
   Gene ID 840139), NIA1 (nitrate reductase [NADH] 1 [Nicotiana
   attenuata], Gene ID 107823732).

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 [73]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 [74]5 and
   [75]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. [76]6).

Table 2.

   ANOVA analysis of differences between treatments, varieties and their
   interactions on LB, TN, SP, NO[3]-N, Pigment, Pn, TS and RS content in
   burley tobacco seedlings.
   Parameters LB TN SP NO[3]-N Chlorophyll a Pigment Pn TS RS
   Varieties (V) 2.8 3.5 6.8 18.6 1.5 1.0 0.8 2.0 1.0
   ns ** ** ** ** ** * ** **
   Treatments (T) 25.1 8.0 4.5 1.9 1.7 1.6 2.3 9.7 10.5
   ** ** * ns ns ns ns ** **
   Varieties (V)*Treatments (T) 67.8 201.9 40.7 1272.0 70.6 96.3 18.5
   1244.1 3486.5
   ** ** ** ** ** ** ** ** **
   [77]Open in a new tab

   Note: LB, Leaf biomass; TN, Total nitrogen content; SP, Soluble protein
   content; NO[3]-N, NO[3]-N content; TS, Total soluble sugar content; RS,
   Reducing soluble sugar content. Symbols ∗∗ and ∗ indicate significant
   difference at 0.01 or 0.05, respectively, and ns indicate that there
   was no significant difference between treatments.

Figure 5.

   [78]Figure 5
   [79]Open in a new tab

   Effect of pigment content and photosynthesis rate by spraying glycerol
   under different nitrogen conditions (a,d) Pn (b,e) Chlorophyll a (c,f):
   Pigment. CK: control; Gl: glycerol. Pn: photosynthesis rate. Error bars
   of photosynthesis rate indicate standard error of the means (N = 6, “N”
   means the number of individuals), and error bars of chlorophyll a
   content and pigment content indicate standard error of the means
   (n = 3, three biological replicates). Symbols ** ^1and * ^1indicate
   that the significant differences between treatment of control and
   treatment of spraying glycerol under sufficient nitrogen condition
   (N100) are at 0.01 and 0.05, respectively. Symbols ** ^2and *
   ^2indicate that the significant differences between treatment of
   control and treatment of spraying glycerol under 20% less nitrogen
   application condition (N80) are at 0.01and 0.05, respectively (a–c)
   TN90 (d–f), TN86.

Figure 6.

   [80]Figure 6
   [81]Open in a new tab

   Effect of spraying glycerol on total soluble sugar (a,d), reducing
   sugar content (b,e) and biomass accumulation (c,f) in burley tobacco
   seedlings under different nitrogen conditions. CK: control; Gl:
   glycerol. TS: total soluble sugar content. RS: reducing sugar content.
   Error bars indicate standard error of the means (n = 3). Symbols **
   ^1indicate that the significant differences between treatment of
   control and treatment of spraying glycerol under sufficient nitrogen
   application (N100) are at 0.01. Symbols ** ^2indicate that the
   significant differences between treatment of control and treatment of
   spraying glycerol under 20% less nitrogen application condition (N80)
   are at 0.01 (a–c) TN90 (d–f) TN86.

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. [82]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.

Figure 7.

   [83]Figure 7
   [84]Open in a new tab

   Effect of spraying glycerol on NO[3]-N (a,d), total nitrogen content
   (b,e) and soluble protein content (c,f) in burley tobacco seedlings
   under different nitrogen conditions. CK: control; Gl: glycerol. TN:
   total nitrogen content. Error bars indicate standard error of the means
   (n = 3). Symbols ** ^1indicate that the significant differences between
   treatment of control and treatment of spraying glycerol under
   sufficient nitrogen condition (N100) are at 0.01. Symbols ** ^2and *
   ^2indicate that the significant differences between treatment of
   control and treatment of spraying glycerol under 20% less nitrogen
   application condition (N80) are at 0.01 and 0.05, respectively. (a–c),
   TN90; (d–f), TN86.

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^[85]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^[86]2,[87]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^[88]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^[89]7,[90]8. And nitrate concentration can be
   diluted with increased biomass in plant at the same nitrogen
   level^[91]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^[92]26. The 2,
   3-bisphosphoglycerate-dependent phosphoglycerate mutase (EC 5.4.2.1,
   encoded by gpmA gene) and phosphoglycerate kinase (EC 2.7.2.3, encoded
   by PGK gene) can catalyze the interconversion of 3-phosphoglycerate
   (3-PGA) and 2-phosphoglycerate (2-PGA) during glycolysis and
   gluconeogenesis and are involved in photosynthetic carbon
   fixation^[93]27. Expression of genes GLPK, PGK and gpmA were
   significantly up-regulated by spraying glycerol under sufficient and
   20% reducing 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^[94]28, SUS2-2, which plays a
   dominant role in generating precursors for starch
   biosynthesis^[95]29,[96]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^[97]31,[98]32. Nitrate is largely accumulated and
   carbohydrates are extremely low in burley tobacco, which are associated
   with one double recessive genes mutation^[99]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^[100]33–[101]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^[102]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^[103]37. Nitrate transport was lower in burley tobacco than
   in other tobacco types^[104]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^[105]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^[106]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).

Experiment 1

   Deficient nitrogen level: (1) Nd-TN90, 4 mM, TN90; (2) Nd-TN86, 4 mM,
   TN86. Sufficient nitrogen level: (1) Ns-TN90, 24 mM, TN90; (2) Ns-TN86,
   24 mM, TN86. Sampling and determining were carried out 10 days after
   seedlings being transplanted (around 10: 00 am). Leaves of five plants
   from each treatment were mixed and frozen in liquid nitrogen
   immediately, and then kept at −80 °C. Leaves of fifteen seedlings per
   treatment were deactivated at 105 °C for 20 min and then dried at 60 °C
   for 48 h and weighed, finally. Every treatment had three biological
   replicates.

Experiment 2

   Varieties of TN90 and TN86: Nitrogen reduction level, 20% less nitrogen
   application condition: (1) N80-CK, 19.2 mM nitrogen level, spraying
   pure water; (2) N80-Gl, 19.2 mM nitrogen level, spraying 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)^[107]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^[108]39. Photosynthetic
   rate (Pn) was measured with a portable photosynthesis system (LI-COR
   Biotechnology, 6400XT, Lincoln, NE, USA)^[109]40.

Measurement of carbonitride content

   Nitrate content was determined by the method described by
   Cataldo^[110]41. Soluble protein content was assayed according to
   Li^[111]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)^[112]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^[113]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
   ([114]http://tophat.cbcb.umd.edu/)^[115]44,[116]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^[117]46,[118]47. DEGs were identified
   using the DESeq (2012) functions estimate Size Factors and nbinom
   Test^[119]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)^[120]49, Swiss-Prot (A
   manually annotated and reviewed protein sequence database)^[121]50, KO
   (KEGG Ortholog database)^[122]51, GO (Gene Ontology)^[123]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^[124]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 [125]S1, [126]S2). The expression levels of mRNAs were
   normalized and calculated using the 2^−ΔΔCt method^[127]54.

GC-MS analysis of metabolites

Metabolites Extraction

   Five plants per treatment were determined by GC-MS. Take 60 ± 1 mg
   sample into the 2 mL EP tubes, extracted with 0.48 mL extraction liquid
   (V[Methanol]: V[H2O = ]3: 1), add 20 μL of adonitol (0.5 mg mL^−1 stock
   in dH[2]O) as internal standard, vortex mixing for 30 s; Homogenized in
   ball mill for 4 min at 45 Hz, then ultrasound treated for 5 min
   (incubated in ice water); Centrifuge for 15 min at 12000 rpm, 4 °C;
   Transfer the supernatant (0.4 mL) into a fresh 2 mL GC/MS glass vial.
   Dry completely in a vacuum concentrator without heating; Add 40 μL
   Methoxyamination hydrochloride (20 mg mL^−1 in pyridine) incubated for
   30 min at 80 °C; Add 60 μL of the BSTFA regent (1% TMCS, v/v) to the
   sample aliquots, incubated for 1.5 h at 70 °C; All samples were
   analyzed by gas chromatograph system coupled with a Pegasus HT
   time-of-flight mass spectrometer (GC-TOF-MS).

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.

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^[128]55.
   Both of mass spectrum match and retention index match were considered
   in metabolites identification.

Statistical analysis

   Figures were applied with Origin Pro 9.0 (Origin Lab Corporation,
   Northampton, MA, USA). Heatmaps were generated with R (3.4.1 version)
   pheatmap package. Statistical analysis was performed with SPSS 20
   software (IBM, Palo Alto, CA, USA). 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.

Electronic supplementary material

   [129]Supporting tables^ (140.4KB, pdf)

Acknowledgements