Gentiobiose, a β-1,6-linked glycosyl-disaccharide, accumulates abundantly in Gentianaceae and is involved in aspects of plant development, such as fruits ripening and release of bud dormancy. However, the mechanisms regulating the amount of gentio-oligosaccharide accumulation in plants remain obscure. The present study aimed to identify an enzyme that modulates gentio-oligosaccharide amount in gentian (Gentiana triflora). A protein responsible for gentiobiose hydrolysis, GtGen3A, was identified by partial purification and its peptide sequence analysis. The enzyme had a molecular mass of ∼67 kDa without a secretory signal peptide sequence. Sequence analysis revealed that GtGen3A could be a β-glucosidase member belonging to glycoside hydrolase family 3 (GH3). GtGen3A showed a homology to GH3 β-glucan exohydrolases, ExoI of Hordeum vulgare, and ExgI from Zea mays, which preferentially hydrolyzed β-1,3- and β-1,4-linked oligosaccharides. The purified recombinant GtGen3A (rGtGen3A) produced in Escherichia coli showed optimal reaction at pH 6.5 and 20°C. The rGtGen3A liberated glucose from β-1,2-, β-1,3-, β-1,4-, and β-1,6-linked oligosaccharides, and showed the highest activity toward gentiotriose among the substrates tested. Kinetic analysis also revealed that rGtGen3A preferentially hydrolyzed gentiotriose. Virus-induced gene silencing of Gtgen3A in gentian plantlets resulted in predominant accumulation of gentiotriose rather than gentiobiose. Furthermore, the expression level of Gtgen3A was almost similar to the amount of gentiobiose in field-grown gentians. These findings suggest that the main function of GtGen3A is the hydrolysis of gentiotriose to gentiobiose, and that GtGen3A plays a role in modulating gentiobiose amounts in gentian.

Introduction

Sugars serve as an energy source and act as signaling molecules regulating growth and development for plants. For example, glucose regulates photosynthesis, senescence, hormone signaling, and stress responses through a hexokinase-dependent pathway [13]. Oligosaccharides also act as signal molecules to regulate plant metabolism. Especially, non-reducing disaccharides, such as sucrose and trehalose, are extensively investigated for their roles in the regulation of plant metabolism. Sucrose is known as an inducer of anthocyanin accumulation and as a signal negatively regulating the expression and transport activity of the proton-sucrose symporter, which is involved in assimilate partitioning in higher plants [4,5]. Trehalose seems to be involved in plant–pathogen interactions and in responses against several environmental stresses [6,7]. Furthermore, glucose oligosaccharides derived from pathogens and host plant cell walls act as signals in triggering defense responses [8]. Studies on several oligosaccharide elicitors and host plants revealed that plants recognize the oligosaccharides on the basis of the type of linkage, degree of polymerization (DP), and side-chain modification [9]. An oligosaccharide elicitor from Pyricularia oryzae cell wall induced phytoalexin accumulation in rice cells, but not in soybean cells [10], indicating that plants have evolved individual oligosaccharide recognition systems and the functions of oligosaccharides may vary depending on plant species.

Several reports indicate that oligosaccharide signaling is significantly modulated by synthetic and degradative enzymes [1114]. Even though these reports provide evidence for the regulation of oligosaccharide concentrations using enzymatic reactions, the identification of the regulating enzymes is necessary to reveal the actual oligosaccharide signaling. In this study, we identified a gentio-oligosaccharide degradative enzyme and explored the function of the oligosaccharides as signal molecules in Gentiana. Gentio-oligosaccharides, β-1,6-linked oligosaccharides, possess unique properties, including a bitter taste. The oligosaccharides are present abundantly in some members of the Gentianaceae and are scarce in other plants [15]. Gentiobiose [β-d-Glcp-(1 → 6)-d-Glc] is reported to hasten fruit ripening in tomato and capsicum [16,17]. Our previous report showed that exogenously applied gentiobiose induced budbreak of gentian overwintering bud in vitro with increased concentration of sulfur-containing amino acids, which are known precursors of glutathione. On the other hand, depletion of glutathione and ascorbate repressed the budbreak, suggesting that gentiobiose serves as a signal molecule releasing dormancy in gentian through glutathione and ascorbate [15]. On the basis of these reports, gentiobiose appears to be related to plant development or maturity, but the detailed mechanisms of its action are still mostly unknown. Our study also revealed that gentiobiose accumulation was attributed to the degradation of gentianose [β-d-Glcp-(1 → 6)-d-Glc-(1 → 2)-d-Fru] mediated by invertase. However, since a decrease in gentiobiose amount did not mirror the level of gene expression and activity of invertase, and the total amount of gentiobiose and gentianose was not constant, it is suspected that some other pathways modulating gentiobiose level must exist. Gentio-oligosaccharide metabolism has been reported in some fungal species such as Aspergillus, and the enzymes involved in the process have been identified [18,19] (see also Figure 5). However, there is no information about other enzymes controlling the concentration of gentio-oligosaccharides in plants.

Here, we purified GtGen3A, a gentio-oligosaccharide-hydrolyzing enzyme, from the leaves of Gentiana triflora. Based on the amino acid sequence analysis, GtGen3A was identified as a β-glucosidase (E.C. 3.2.1.21) belonging to the glycoside hydrolase (GH) family 3 (GH3). Although plant GH3 β-glucosidases of Hordeum vulgare and Zea mays have been reported [20,21], there are only a few reports characterizing plant GH3 β-glucosidases. Here, we showed that GtGen3A possesses characteristics that are apparently different from other plant GH3 β-glucosidase, despite the homology at the amino acid level. Furthermore, our results also showed that the amount of gentiobiose was modified in parallel with the expression level of Gtgen3A in gentian. We also discuss the significance of GtGen3A in gentio-oligosaccharide metabolism in gentian.

Results

Identification of a gentio-oligosaccharide-hydrolyzing enzyme of gentian

To identify the protein responsible for gentio-oligosaccharide degradation in gentian leaves, hydrolytic activities against gentiobiose were determined in the fractions separated at various stages by methods, such as ammonium sulfate precipitation, hydrophobic interaction chromatography on a Phenyl-Sepharose column, ion-exchange chromatography on a Mono-Q and DEAE-Sepharose columns (Supplementary Figure S1), and gel filtration chromatography on a TSK-gel G3000SW column (Table 1). Finally, the enzyme was purified to ∼20-fold specific activity compared with the crude extract. Fractions 15−21 obtained by gel filtration chromatography were analyzed by SDS–PAGE analysis, which showed higher activities for gentiobiose hydrolysis in fractions 17−19, with several protein bands (Figure 1). Judging by the signal intensity of proteins and their hydrolytic activities, proteins corresponding to 28 and 67 kDa were expected to be candidate proteins participating in gentiobiose hydrolysis. These protein bands were trypsin-digested and the peptides generated were identified using a gentian EST database [22] after separation by LC–MS/MS on an LTQ Orbitrap mass analyzer. The 67 and 28 kDa proteins were annotated as a β-glucosidase and an esterase, respectively. The 67 kDa protein consisted of 612 amino acids (Supplementary Figure S2) and showed a calculated molecular mass of 67.14 kDa, which agrees with that determined by SDS–PAGE. Since the GtGen3A has a conserved domain similar to that of β-glucosidase-related glycosidases (BglX) and glycosyl hydrolase family 3 C-terminal domain (pfam01915), the 67 kDa protein, named GtGen3A, was considered to be a GH3 β-glucosidase. GtGen3A had high (69–75%) amino acid similarity to that of β-glucosidases such as BoGH3B-like of Glycine max, Daucus carota, and Solanum pennellii, which have not been enzymatically characterized yet. GtGen3A shared 59 and 57% amino acid identity with ExoI of H. vulgare [20] and ExgI of Z. mays [21], respectively, and was separated from the group of putative α-arabinofuranosidases and β-xylosidases (Figure 2). Key amino acid residues forming the active sites, aspartic acid at 291 and glutamic acid at 496, were conserved in GtGen3A (Supplementary Figure S2). To confirm that the GtGen3A is responsible for gentiobiose hydrolysis and to determine its enzymatic properties, the recombinant protein rGtGen3A was prepared in Escherichia coli. rGtGen3A was sequentially purified by a Ni-NTA affinity column, a Phenyl-Sepharose column, and a DEAE-Sepharose column. The purified rGtGen3A protein showed a single band with a molecular mass of ∼67 kDa on SDS–PAGE (Supplementary Figure S3A), identical with that estimated from the deduced amino acid sequence and that of native GtGen3A shown in Figure 1. Peptide sequences of trypsin-digested rGtGen3A matched native GtGen3A sequence (Supplementary Figure S4A,B). We determined that rGtGen3A had optimal activity at 20°C and at pH 6.5, and the activities were stable after incubation for 10 min at temperatures lower than 30°C, but completely lost over 55°C (Supplementary Figure S3B,C). The optimal pH and temperature were identical with the result of native GtGen3A in terms of peptide sequences (Supplementary Figure S5A,B).

Partial purification of gentiobiose-hydrolyzing enzyme from gentian leaves.

Figure 1.
Partial purification of gentiobiose-hydrolyzing enzyme from gentian leaves.

CBB-stained SDS–PAGE of partially purified proteins by TSK-gel G3000SW column. Lane M, protein marker; lanes 15 to 21, number of fractions obtained by gel filtration chromatography. Hydrolytic activities are indicated as a relative activity for gentiobiose hydrolysis. Arrowheads indicate candidate protein bands corresponding to the activities.

Figure 1.
Partial purification of gentiobiose-hydrolyzing enzyme from gentian leaves.

CBB-stained SDS–PAGE of partially purified proteins by TSK-gel G3000SW column. Lane M, protein marker; lanes 15 to 21, number of fractions obtained by gel filtration chromatography. Hydrolytic activities are indicated as a relative activity for gentiobiose hydrolysis. Arrowheads indicate candidate protein bands corresponding to the activities.

Phylogenetic tree of representative plant GH3 family enzymes from the CAZy database.

Figure 2.
Phylogenetic tree of representative plant GH3 family enzymes from the CAZy database.

Deduced amino acid sequences of the enzymes were used to construct phylogenetic tree by the neighbor-joining method using the MEGA7 software. Bootstrap values are indicated at the branch points. The scale bar represents the number of amino acid substitutions per site. At, Arabidopsis thaliana; Dc, Daucus carota; Gm, Glycine max; Gt, Gentiana triflora; Hv, Hordeum vulgare; Ll, Lilium longiflorum; Msv, Medicago sativa x varia; Pta, Populus tremula x alba; Sp, Solanum pennellii; Zm, Zea mays.

Figure 2.
Phylogenetic tree of representative plant GH3 family enzymes from the CAZy database.

Deduced amino acid sequences of the enzymes were used to construct phylogenetic tree by the neighbor-joining method using the MEGA7 software. Bootstrap values are indicated at the branch points. The scale bar represents the number of amino acid substitutions per site. At, Arabidopsis thaliana; Dc, Daucus carota; Gm, Glycine max; Gt, Gentiana triflora; Hv, Hordeum vulgare; Ll, Lilium longiflorum; Msv, Medicago sativa x varia; Pta, Populus tremula x alba; Sp, Solanum pennellii; Zm, Zea mays.

Table 1
Purification of gentiobiose-hydrolyzing enzyme1 from gentian leaves
Purification steps Total protein (mg) Total activity (nmol/h) Specific activity (nmol/h/mg) Yield (%)2 Purification (fold)3 
Crude extract 20.97 40.61 1.16 100.00 1.00 
20–40% (NH4)2SO4 10.20 23.39 1.56 57.60 1.34 
Phenyl-Sepharose 4.77 21.39 2.14 52.67 1.84 
Mono-Q 0.32 7.40 7.40 18.22 6.38 
DEAE-Sepharose 0.13 6.14 8.77 15.12 7.56 
TSK-gel G3000SW 0.01 1.17 23.31 2.88 20.09 
Purification steps Total protein (mg) Total activity (nmol/h) Specific activity (nmol/h/mg) Yield (%)2 Purification (fold)3 
Crude extract 20.97 40.61 1.16 100.00 1.00 
20–40% (NH4)2SO4 10.20 23.39 1.56 57.60 1.34 
Phenyl-Sepharose 4.77 21.39 2.14 52.67 1.84 
Mono-Q 0.32 7.40 7.40 18.22 6.38 
DEAE-Sepharose 0.13 6.14 8.77 15.12 7.56 
TSK-gel G3000SW 0.01 1.17 23.31 2.88 20.09 
1

Activities were determined by the release of glucose from gentiobiose.

2

Yield was expressed as the percentage relative to total activity in crude extract.

3

Purification was expressed as fold change relative to specific activity in crude extract.

Substrate specificity and kinetic analysis of rGtGen3A

For determining the substrate specificity, GtGen3A was incubated with various oligosaccharides and polysaccharides, and the amount of glucose released was measured (Table 2). GtGen3A hydrolyzed oligosaccharides with various linkages, such as β-1,2-linked disaccharide (sophorose), β-1,3-linked oligosaccharides with a DP of 2–5 (laminaribiose, laminaritriose, laminaritetraose, and laminaripentaose), β-1,4-linked oligosaccharides with DP of 2–5 (cellobiose, cellotriose, cellotetraose, and cellopentaose), and β-1,6-linked oligosaccharides with DP of 2–5 (gentiobiose, gentiotriose, gentiotetraose, and gentiopentaose). GtGen3A also displayed hydrolysis of polymeric laminarin with β-1,3-linkages. Although hydrolytic activity was not observed toward pustulan, a polymeric β-1,6-glucan, prolonged reaction allowed the release of glucose from pustulan (Supplementary Figure S6). In this experiment, only glucose was detected as a reaction product from laminarin and pustulan, indicating that GtGen3A acts as an exo-type β-glucanase. The activity showed a trend to be higher for β-1,3- and β-1,6-linked oligosaccharides compared with β-1,2- and β-1,4-linked oligosaccharides, and the highest activity was observed toward gentiotriose (1.67 µmol glucose min−1 mg−1 protein). The hydrolytic activity toward laminari-oligosaccharides and cello-oligosaccharides of DP 2–5 increased with the DP. No hydrolytic activity was detected for gentianose, polymeric phosphoric acid swollen cellulose (PSC), and barley β-1,3-1,4-glucan. Native GtGen3A, which was partially purified to homogeneity from gentian leaves, showed similar substrate specificity to that of rGtGen3A (Supplementary Figure S5C). The activities increased with the increase in the substrate concentration of glycosyl-disaccharides tested (Supplementary Figure S3D). The activities for laminaribiose were relatively high at low concentration (<2.5 mM), but still remained lower than that for gentiobiose at 10 mM concentration. Based on the hydrolytic activities at various substrate concentrations, kinetic parameters were determined (Table 3). The largest kcat value was observed for gentiotriose (4.01 s−1) and the smallest was for cellobiose (0.94 s−1). The Km values for laminaribiose and gentiotriose were 1.17 and 1.65 mM, respectively. The kcat/Km values obtained for gentiotriose and laminaribiose were 2.47 s−1 mM−1 (the largest among tested) and 1.77 s−1 mM−1, whereas the values for sophorose and cellobiose were much lower. These results suggest that GtGen3A preferentially hydrolyzes gentiotriose and laminaribiose.

Table 2
Specific activity of rGtGen3A
Substrate Relative activity (%) 
Sophorose 14.0 ± 0.9 
Laminaribiose 45.7 ± 2.4 
Laminaritriose 82.0 ± 2.6 
Laminaritetraose 86.3 ± 6.1 
Laminaripentaose 90.7 ± 3.1 
Laminarin 46.7 ± 6.4 
Cellobiose 7.3 ± 0.3 
Cellotriose 35.7 ± 1.1 
Cellotetraose 37.9 ± 1.6 
Cellopentaose 38.2 ± 0.7 
PSC UD 
Gentiobiose 36.9 ± 0.4 
Gentiotriose 100.0 ± 3.4 
Gentiotetraose 73.4 ± 4.4 
Gentiopentaose 55.6 ± 1.8 
Gentianose UD 
Pustulan UD 
Barley glucan UD 
Substrate Relative activity (%) 
Sophorose 14.0 ± 0.9 
Laminaribiose 45.7 ± 2.4 
Laminaritriose 82.0 ± 2.6 
Laminaritetraose 86.3 ± 6.1 
Laminaripentaose 90.7 ± 3.1 
Laminarin 46.7 ± 6.4 
Cellobiose 7.3 ± 0.3 
Cellotriose 35.7 ± 1.1 
Cellotetraose 37.9 ± 1.6 
Cellopentaose 38.2 ± 0.7 
PSC UD 
Gentiobiose 36.9 ± 0.4 
Gentiotriose 100.0 ± 3.4 
Gentiotetraose 73.4 ± 4.4 
Gentiopentaose 55.6 ± 1.8 
Gentianose UD 
Pustulan UD 
Barley glucan UD 

Values shown are means ± SD from triplicate measurements. In each experiment, 300 ng of rGtGen3A and 2 mM substrates were used. Specific activity at 100% was 1.67 μmol Glc min−1 mg−1 protein for gentiotriose. Abbreviations: PSC, phosphoric acid swollen cellulose; Barley glucan, barley β-1,3-1,4-glucan; UD, under detectable levels.

Table 3
Kinetic parameters of GtGen3A for hydrolysis of glucose disaccharides
Substrates kcat (s−1Km (mM) kcat/Km (s−1 mM−1
Sophorose 2.04 ± 0.16 4.98 ± 0.74 0.41 ± 0.03 
Laminaribiose 2.06 ± 0.06 1.17 ± 0.12 1.77 ± 0.14 
Cellobiose 0.94 ± 0.21 21.88 ± 6.15 0.04 ± 0.00 
Gentiobiose 2.86 ± 0.19 4.40 ± 0.75 0.66 ± 0.07 
Gentiotriose 4.01 ± 0.79 1.65 ± 0.42 2.47 ± 0.27 
Substrates kcat (s−1Km (mM) kcat/Km (s−1 mM−1
Sophorose 2.04 ± 0.16 4.98 ± 0.74 0.41 ± 0.03 
Laminaribiose 2.06 ± 0.06 1.17 ± 0.12 1.77 ± 0.14 
Cellobiose 0.94 ± 0.21 21.88 ± 6.15 0.04 ± 0.00 
Gentiobiose 2.86 ± 0.19 4.40 ± 0.75 0.66 ± 0.07 
Gentiotriose 4.01 ± 0.79 1.65 ± 0.42 2.47 ± 0.27 

Substrates were used at a concentration of 1–50 mM for sophorose, 0.1–2 mM gentiotriose, and 0.25–50 mM for other substrates. Values shown are means ± SD from triplicate measurements. In each experiment, 200 ng of rGtGen3A was used.

Knockdown of Gtgen3A modified the oligosaccharide composition in gentian plantlets

To investigate the effects of GtGen3A on sugar accumulation in gentian plantlets, we generated Gtgen3A knockdown plantlets using a virus-induced gene silencing (VIGS) technique. Three VIGS-Gtgen3A lines with a significant decrease in Gtgen3A expression level (VIGS-Gtgen3A-1, -2, and -3; Figure 3) were used for further analysis. There was almost no difference in appearance and growth between vector control (VC) and VIGS-Gtgen3A lines (Figure 3A). The Gtgen3A expression levels of VIGS-Gtgen3A plantlet leaves were approximately one-third of that in VC plantlet (Figure 3B and Supplementary Figure S7). Gentiobiose hydrolytic activities in VIGS-Gtgen3A plantlets decreased to about one-third of VC plantlet (Figure 3C). On the other hand, no significant difference in the GtINV expression levels was observed between VC and VIGS-Gtgen3A plantlets (Figure 3D).

Figure 3.

Effects of VIGS of Gtgen3A on gene expression levels and gentiobiose hydrolytic activities.(A) Photograph of plantlets of VC and VIGS lines of Gtgen3A cultured for 1 month. Plantlets were cultured on MS medium at 22°C with photoperiods of 16 h at 50 µmol m−2 s−1. Bar = 1 cm. (B) Expression of Gtgen3A in VC and VIGS-Gtgen3A plantlets. (C) Enzyme activities using gentiobiose as a substrate. (D) Expression of GtINV in VC and VIGS-Gtgen3A plantlets. Gene expression levels and enzyme activities are represented as means ± SD calculated from five and three independent experiments, respectively. All data are represented as the percentage of the level in VC (taken as 100%) and significant differences from VC are shown (**P <0.01; Student's t-test).

Figure 3.

Effects of VIGS of Gtgen3A on gene expression levels and gentiobiose hydrolytic activities.(A) Photograph of plantlets of VC and VIGS lines of Gtgen3A cultured for 1 month. Plantlets were cultured on MS medium at 22°C with photoperiods of 16 h at 50 µmol m−2 s−1. Bar = 1 cm. (B) Expression of Gtgen3A in VC and VIGS-Gtgen3A plantlets. (C) Enzyme activities using gentiobiose as a substrate. (D) Expression of GtINV in VC and VIGS-Gtgen3A plantlets. Gene expression levels and enzyme activities are represented as means ± SD calculated from five and three independent experiments, respectively. All data are represented as the percentage of the level in VC (taken as 100%) and significant differences from VC are shown (**P <0.01; Student's t-test).

The concentrations of sugar alcohols, monosaccharides, disaccharides, and trisaccharides in leaves of the plantlets were quantified by liquid chromatography–mass spectrometry (LC–MS) (Table 4). The concentrations of glucosyl-disaccharides were apparently higher in VIGS-Gtgen3A plantlets than those in VC plantlet. Laminaribiose and cellobiose concentrations in VIGS-Gtgen3A plantlets were at least 2.8- and 1.7-fold higher than in VC plantlets, respectively. The concentration of gentiobiose, the most abundant glycosyl-disaccharide, was 2.6- to 3.7-fold higher in VIGS-Gtgen3A plantlets than in VC plantlet. Gentiotriose was observed in VIGS-Gtgen3A plantlets, whereas it was below detectable levels in the VC plantlet. Conversely, fructose and glucose concentrations tended to be lower in VIGS-Gtgen3A plantlets compared with VC plantlet. The difference in the concentrations of inositol, arabinose, xylose, sucrose, trehalose, and raffinose was not significant between VC and VIGS-Gtgen3A plantlets. Gentianose concentration was slightly different in VIGS-Gtgen3A plantlets, but no discernible pattern was observed. These results imply that the action of GtGen3A remarkably affects the accumulation of gentiotriose and gentiobiose.

Table 4
Amounts of sugar in leaves of VC and VIGS lines of Gtgen3A plantlets
 Amounts (nmol mg−1 DW) 
VC VIGS-Gtgen3A 
Inositol 10.0 ± 3.4 8.2 ± 3.9 13.6 ± 1.7 7.8 ± 1.8 
Arabinose 1.6 ± 0.5 0.9 ± 0.5 1.3 ± 0.6 1.2 ± 0.2 
Xylose 1.4 ± 0.3 1.1 ± 0.2 1.5 ± 0.2 1.3 ± 0.1 
Fructose 17.2 ± 6.2 8.3 ± 1.9 13.5 ± 2.7 10.7 ± 1.6 
Glucose 29.6 ± 9.2 15.9 ± 2.6* 23.8 ± 2.5 22.8 ± 2.4 
Sucrose 37.7 ± 4.9 31.9 ± 4.4 37.0 ± 3.4 33.9 ± 1.6 
Trehalose 1.5 ± 0.5 1.2 ± 0.1 1.2 ± 0.2 1.6 ± 0.7 
Laminaribiose 0.1 ± 0.0 0.4 ± 0.1** 0.4 ± 0.1** 0.5 ± 0.0** 
Cellobiose 0.1 ± 0.0 0.1 ± 0.0** 0.2 ± 0.1* 0.1 ± 0.1** 
Gentiobiose 8.3 ± 2.2 21.7 ± 7.3** 31.0 ± 17.3* 22.9 ± 11.0* 
Gentiotriose UD 6.9 ± 1.9 12.8 ± 8.5 11.4 ± 7.7 
Raffinose 0.03 ± 0.01 0.02 ± 0.01 0.04 ± 0.01 0.02 ± 0.01 
Gentianose 4.3 ± 1.2 2.9 ± 1.0 6.5 ± 2.2 3.4 ± 0.9 
 Amounts (nmol mg−1 DW) 
VC VIGS-Gtgen3A 
Inositol 10.0 ± 3.4 8.2 ± 3.9 13.6 ± 1.7 7.8 ± 1.8 
Arabinose 1.6 ± 0.5 0.9 ± 0.5 1.3 ± 0.6 1.2 ± 0.2 
Xylose 1.4 ± 0.3 1.1 ± 0.2 1.5 ± 0.2 1.3 ± 0.1 
Fructose 17.2 ± 6.2 8.3 ± 1.9 13.5 ± 2.7 10.7 ± 1.6 
Glucose 29.6 ± 9.2 15.9 ± 2.6* 23.8 ± 2.5 22.8 ± 2.4 
Sucrose 37.7 ± 4.9 31.9 ± 4.4 37.0 ± 3.4 33.9 ± 1.6 
Trehalose 1.5 ± 0.5 1.2 ± 0.1 1.2 ± 0.2 1.6 ± 0.7 
Laminaribiose 0.1 ± 0.0 0.4 ± 0.1** 0.4 ± 0.1** 0.5 ± 0.0** 
Cellobiose 0.1 ± 0.0 0.1 ± 0.0** 0.2 ± 0.1* 0.1 ± 0.1** 
Gentiobiose 8.3 ± 2.2 21.7 ± 7.3** 31.0 ± 17.3* 22.9 ± 11.0* 
Gentiotriose UD 6.9 ± 1.9 12.8 ± 8.5 11.4 ± 7.7 
Raffinose 0.03 ± 0.01 0.02 ± 0.01 0.04 ± 0.01 0.02 ± 0.01 
Gentianose 4.3 ± 1.2 2.9 ± 1.0 6.5 ± 2.2 3.4 ± 0.9 

Each value represents the means ± SD of five independent experiments. Significant differences from VC are shown (*P < 0.05; **P < 0.01; Student's t-test). Abbreviations: UD, under detectable levels.

Seasonal fluctuations in Gtgen3A expression and gentio-oligosaccharide accumulation in the leaves of field-grown gentian

To further investigate the role of GtGen3A in the mechanism of gentio-oligosaccharide accumulation, we measured the seasonal changes in the Gtgen3A expression levels in leaves throughout the year (Figure 4A). Gentian plants used in this experiment pass the winter (dormancy phase) from October to March and develop vegetative organs (vegetative growth phase: April and May) followed by reproductive organs (reproductive growth phase: June to August). The level of Gtgen3A transcript decreased in the dormancy phase and retained at a low level until the end of that phase. However, the level increased gradually during the vegetative growth phase and decreased during the reproductive growth phase. We also measured the amounts of sugar in the leaves (Figure 4B,C). Similar to that in plants, the amounts of glucose, gentiobiose, and gentiotriose in the leaves increased during vegetative growth phase and decreased during reproductive growth phase. The levels of gentiotriose were far low compared with those of glucose and gentiobiose, suggesting that the generated gentiotriose may be rapidly converted into gentiobiose by the actions of hydrolytic enzymes such as GtGen3A. The amount of laminaribiose was detectable but was very low compared with gentiobiose. These results suggest that GtGen3A is involved in the hydrolysis of gentiotriose to gentiobiose.

Seasonal changes of Gtgen3A levels and sugar amounts in leaves of field-grown gentian.

Figure 4.
Seasonal changes of Gtgen3A levels and sugar amounts in leaves of field-grown gentian.

(A) The expression levels of Gtgen3A and GtINV. (B) The amounts of glucose, gentiobiose, and laminaribiose. The amounts during dormancy were quoted from our previous results (Takahashi et al. [15]). (C) The amounts of gentiotriose. Values are means ± SD calculated from five independent experiments.

Figure 4.
Seasonal changes of Gtgen3A levels and sugar amounts in leaves of field-grown gentian.

(A) The expression levels of Gtgen3A and GtINV. (B) The amounts of glucose, gentiobiose, and laminaribiose. The amounts during dormancy were quoted from our previous results (Takahashi et al. [15]). (C) The amounts of gentiotriose. Values are means ± SD calculated from five independent experiments.

Discussion

Enzymatic properties of GtGen3A

β-Glucosidases are thought to play key roles in the metabolism of oligo- and polysaccharides. Here, we identified GtGen3A, a gentio-oligosaccharide-hydrolyzing enzyme from G. triflora, to investigate the mechanism for modulating the amounts of gentio-oligosaccharides. We used a recombinant protein rGtGen3A to this purpose. Plant β-glucosidases are mainly classified into GH1, GH3, GH5, GH9, and GH116 (http://www.cazy.org) based on amino acid similarity. GtGen3A has two GH3-conserved domains and showed 59 and 57% amino acid identity with H. vulgare ExoI [20] and membrane-associated ExgI from Z. mays [21], both of which are reported to belong to the GH3 family. Furthermore, we searched other GH3 glucosidases in a G. triflora EST database [22], and five candidate genes possessing GH3-conserved domain were identified (Supplementary Figure S8A). The expression levels of these candidate genes were apparently low compared with that of Gtgen3A (Supplementary Figure S8B), indicating that Gtgen3A is the most expressed GH3 glucosidase in G. triflora. Most GH3 enzymes are thought to be secreted extracellularly and are involved in the degradation of cell wall polysaccharides derived from plants and plant–pathogens. For example, Ustilago esculenta glucosidase, which possesses a secretory signal peptide, is a glycosylphosphatidylinositol-anchored plasma membrane protein involved in the degradation of laminarin [23]. The fact that GtGen3A does not possess a secretory signal peptide strongly suggests that GtGen3A has different properties and roles compared with other GH3 enzymes reported previously. As expected, GtGen3A possesses unique biological properties, including substrate specificity and optimal pH and temperature. The release of glucose from the substrates indicates that GtGen3A is an exo-β-glucanase. The recombinant protein rGtGen3A showed the maximal activity for gentiobiose hydrolysis at pH 6.5 and at 20°C, which were comparable with those of native GtGen3A partially purified from gentian leaves (Supplementary Figures S3 and S5). This optimum pH value was higher than other GH3 β-glucosidases or β-glucanases from barley, nasturtium, maize, and tobacco [21,2426]. The optimal temperature of the hydrolytic activity was comparable to the temperature (20−23°C) favorable for gentian growth (Supplementary Figure S3C). Thus, GtGen3A could have evolved to metabolize oligosaccharides during the growth and development of gentians. The kinetic analyses for disaccharides revealed high kcat values for sophorose, laminaribiose, gentiobiose, and gentiotriose as substrates, whereas the Km value for sophorose and gentiobiose was far higher than those for laminaribiose and gentiotriose (Table 3). Therefore, the kcat/Km values for laminaribiose and gentiotriose were higher than others. However, at high concentration of substrates (10 mM), hydrolytic activity toward gentiobiose was higher than that of laminaribiose (Supplementary Figure S3D). This may be due to the inhibitory effect of a high concentration of laminaribiose, which might bind to the subsites, but not to the catalytic center of GtGen3A. Notably, gentiotriose was found to be the best substrate among tested as judged by the results from substrate specificity and kinetic experiments (Tables 2 and 3), implying that the main function of GtGen3A is to hydrolyze gentiotriose.

Previous reports showed that some fungi, such as Penicillium brefeldianum, U. esculenta, Neurospora crassa, and Aspergillus oryzae, possess GH3 β-glucosidases or β-glucanases preferentially to hydrolyze β-1,6-linked oligosaccharides or polysaccharides [18,23,27,28]. Furthermore, the fungal β-1,6-glucanase categorized in the GH30 family also reported to hydrolyze β-1,6-glucans [29]. On the other hand, no plant enzymes, except for GtGen3A, have been demonstrated to exhibit a preferential cleavage of β-1,6-oligosaccharides. Our results show that the purified rGtGen3A exhibited hydrolytic activities toward β-1,2-, β-1,3-, β-1,4-, and β-1,6-linked oligosaccharides as well as some polysaccharides, and the highest activity was observed against β-1,6-linked trisaccharide, gentiotriose (Table 2). Although plant GH3 enzymes showed broad substrate specificities, and hydrolyzed β-glucan with various linkages, β-1,6-oligosaccharides are minor substrates among them. For example, ExoI and ExgI preferentially hydrolyze β-1,3 glucan, and wall-bound exo-1,3-β-glucanase from H. vulgare showed hydrolytic activities toward various oligosaccharides, among which laminaritetraose is the best substrate [24]. Furthermore, ARA-I from H. vulgare exhibited a bifunctional hydrolysis as α-l-arabinofuranosidase and β-d-xylosidase, but showed no activity against 4NP-β-d-glucopyranoside [30]. Substrate specificity of the recombinant GtBglA implies that not all GH3 glucosidases possess high hydrolytic activity against gentio-oligosaccharides (Supplementary Figure S8C,D). Thus, to our knowledge, GtGen3A is the first enzyme to preferentially hydrolyze β-1,6-oligosaccharides in plants.

Involvement of GtGen3A in the gentio-oligosaccharide metabolism

Gentio-oligosaccharides are rare sugars found in some plants and thought to play key roles in plant developments, such as fruit ripening and dormancy release. Therefore, the amounts of gentio-oligosaccharides are predicted to be modulated by the actions of enzymes involved in biosynthesis and degradation. We found that the recombinant protein rGtGen3A preferentially hydrolyzed gentiotriose and gentiobiose into gentiobiose and glucose, respectively. Although rGtGen3A also hydrolyzed β-1,3-oligosaccharides comparable with gentiobiose, the amount of laminaribiose was far less compared with that of gentiobiose, suggesting that the main target of GtGen3A may be gentio-oligosaccharides in gentians. Our results highlighted the importance of gentiotriose, which markedly accumulated in VIGS-Gtgen3A plantlets, in which GtGen3A activity decreased by ∼70% compared with control plants (Table 4 and Figure 3C). So far, we have not detected gentio-oligosaccharides with more than four DP in gentian plantlets. These facts strongly suggest that gentiotriose could be the largest gentio-oligosaccharide synthesized in gentian, and that it is converted to gentiobiose by the enzymatic action of GtGen3A. There was considerable evidence that gentiobiose accumulation was also observed in VIGS-Gtgen3A plantlets, whereas gentiotriose hydrolysis was repressed. This finding suggests the presence of a gentiobiose-producing pathway separated from the GtGen3A-related gentiobiose production in gentian. Since the knockdown of Gtgen3A repressed not only gentiotriose hydrolysis but also gentiobiose hydrolysis, the decreased supply of gentiobiose from gentiotriose might be offset by the flow from other gentiobiose-producing pathways. We previously demonstrated that gentiobiose was produced by gentianose degradation mediated by the invertase, GtINV [15]. Thus, both GtGen3A and GtINV seem to be involved in the production of gentiobiose in gentian, at least in part. Seasonal changes in gene expression of Gtgen3A and GtINV, as well as the accumulation of gentiobiose in field-grown gentian, may support the contribution of these enzymes in controlling the amount of gentiobiose (Figure 4). The expression levels of Gtgen3A and GtINV increased during vegetative growth phase, accompanied by an increase in gentiobiose amount, suggesting the possibility that gentiobiose produced from gentiotriose and gentianose is used as an energy source or as a signal for growth and development. Although no direct evidence for the contribution rate of GtGen3A to the gentiobiose accumulation is currently available, it may be that gentiotriose degradation by GtGen3A is also involved in controlling gentiobiose amount in field-grown gentian.

Role of gentiotriose in the gentio-oligosaccharide metabolic pathway

In the present study, we purified GtGen3A, a GH3 β-glucosidase hydrolyzing gentio-oligosaccharide in gentian. The evidence for the involvement of GtGen3A in the gentio-oligosaccharide hydrolysis in vivo was obtained by the analyses of Gtgen3A gene-silenced gentian plantlets. It is noteworthy that the most pronounced effect of the silencing was the accumulation of gentiotriose; i.e. gentiotriose was detected only in VIGS-Gtgen3A plantlets, suggesting that the main function of GtGen3A is the hydrolysis of gentiotriose to gentiobiose. There is a paucity of information about the gentio-oligosaccharide metabolism in whole organisms, whereas in vitro synthesis of gentiobiose has shown that fungal β-glucosidases produced gentio-oligosaccharides from condensate glucose or gentiobiose by the transglycosylation reaction [3133]. Especially in plants, the hydrolysis of gentianose mediated by invertase is the only known pathway producing gentiobiose [15] (Figure 5). Here, we found a novel pathway of gentiobiose production from gentiotriose by the hydrolytic activity of GtGen3A in gentian plants (Figure 5). Gentiotriose is a rare sugar contained in gentian root [34] and is also reported to be one of the constitutive carbohydrate residues of crocin contained in Crocus sativus [35]. However, cellular functions and synthetic and degradative pathways of gentiotriose in plants have not yet been demonstrated. Our results highlight that gentiotriose is an actual precursor of gentiobiose, and that GtGen3A serves as one of the modulators of gentio-oligosaccharides in gentian.

Schematic metabolic pathway involved in gentio-oligosaccharides.

Figure 5.
Schematic metabolic pathway involved in gentio-oligosaccharides.

The dotted arrows indicate possible pathways producing gentio-oligosaccharides. G1P, glucose-1-phosphate; G6P, glucose-6-phosphate; INV, invertase; HXK, hexokinase; PGM, phosphoglucomutase; UGP, UDPG pyrophosphorylase; SPS, sucrose phosphate synthase; SUS, sucrose synthase.

Figure 5.
Schematic metabolic pathway involved in gentio-oligosaccharides.

The dotted arrows indicate possible pathways producing gentio-oligosaccharides. G1P, glucose-1-phosphate; G6P, glucose-6-phosphate; INV, invertase; HXK, hexokinase; PGM, phosphoglucomutase; UGP, UDPG pyrophosphorylase; SPS, sucrose phosphate synthase; SUS, sucrose synthase.

Experimental procedures

Plant materials

Leaves of gentian (G. triflora cv. SpB) were harvested from 4-year-old plants grown in an agricultural field at Iwate Prefecture in Japan on August 2013. The leaves were immediately frozen in liquid nitrogen, freeze-dried by a freeze dryer FDU-2210 (EYELA), and stored at −20°C until use. Plantlets of G. triflora cv. ‘Polano white’ were cultured in vitro on solid Murashige and Skoog (MS) medium containing 3% (w/v) sucrose and 0.2% (w/v) gellan gum at 22°C under a 16/8 h light/dark photoperiod at light intensity of 50 μmol m−2 s−1.

Purification of the gentiobiose-hydrolyzing enzyme

Freeze-dried gentian leaves (1 gDW) were powdered in liquid nitrogen using a pestle and mortar and homogenized in 35 ml of 10 mM sodium phosphate buffer (pH 7.0; Buffer A). After filtration through cheesecloth, the homogenate was centrifuged at 20 000×g for 10 min at 4°C and the supernatant was used as a crude extract. Chromatography was performed using a fast protein FPLC system (GE Healthcare). Thirty milliliters of the crude extract was subjected to ammonium sulfate precipitation at different saturation levels (20−80%), and the most active fraction (40−60% saturation) was dissolved in 10 ml of 10 mM phosphate buffer (pH 7.0) containing 15% ammonium sulfate (Buffer B). The solution was loaded onto a Phenyl-Sepharose 6 fast flow (high sub; GE Healthcare; 0.7 × 2.5 cm) column equilibrated with Buffer B. After washing the column with 10 volumes of Buffer B, proteins were eluted with three volumes of Buffer A at a flow rate of 1 ml/min. The eluent was concentrated and desalted using an Amicon Ultra-15 centrifugal filter (Millipore), and 1 ml of the solution was subject to ion-exchange chromatography on a Mono-Q 5/50 GL column (GE Healthcare; 0.5 × 5 cm) equilibrated with Buffer A. After applying the enzyme solution, proteins were eluted with a linear gradient of 0–0.5 M NaCl in Buffer A at a flow rate of 0.5 ml/min. Active fractions were collected, desalted using an Amicon Ultra-15 centrifugal filter (Millipore), and further purified by a DEAE-Sepharose column (GE Healthcare). One milliliter of the solution was loaded onto the column and eluted with a linear gradient of 0–0.5 M NaCl in Buffer A at a flow rate of 0.5 ml/min. The fractions with the highest activity from the ion-exchange chromatography steps were subjected to gel permeation chromatography on a TSK-gel G3000SW column (Toso-Haas) and eluted with Buffer A at a flow rate of 0.5 ml/min. SDS–PAGE was performed according to the method of Laemmli [36] using 10% (w/v) running gels with 5.1% (w/v) stacking gels. Proteins were visualized with Coomassie Brilliant Blue (CBB) (Nacalai).

Identification of the trypsin-digested peptides by Orbitrap mass spectrometry

Protein digestion and identification were performed according to the method of Imamura et al. [37]. One microgram of the TSK-gel G3000SW-purified protein or the rGtGen3A was subjected to SDS–PAGE (10% gel). The 67-kDa band was excised and in-gel protein digestion was performed using the In-Gel Tryptic Digestion Kit (Thermo Fisher Scientific) according to the manufacturer's instructions. Briefly, excised gels were destained, reduced by Tris[2-carboxyethyl]phosphine, alkylated by iodoacetamide, and dehydrated by acetonitrile. The proteins were digested with 10 ng/µl trypsin overnight. The solution was purified using Pierce C-18 Spin Columns (Thermo Fisher Scientific) and subject to nanoLC–MS/MS analysis. Digested peptides were separated by the nanoLC system and analyzed on an LTQ Orbitrap XL mass spectrometer (Thermo Fisher Scientific) operated with the Xcalibur software (version 2.0.7, Thermo Fisher Scientific). Peptides were identified using a MASCOT MS/MS ion search (http://www.matrixscience.com/home.html) and partial ORF (open reading frame) sequence was obtained from our G. triflora EST library [22].

Heterologous expression and purification of GtGen3A

RNA extraction and cDNA synthesis were conducted according to the method of Takahashi et al. [15]. Total RNA from leaves of G. triflora was isolated using an RNAs-ici!-P Kit (Rizo Inc.) and incubated with RNase-free DNase I (Wako) to remove DNA contamination. cDNA was generated from 500 ng of total RNA using an RNA PCR Kit (AMV) version 3.0 (Takara) with oligo(dT) primer following the manufacturer's instructions and used as a template. For cloning of GH3 glucosidases, sequences were obtained from our G. triflora EST library [22], cDNA was isolated by PCR amplification with PrimeSTAR Max DNA Polymerase (Takara) and subcloned into a pCR-Blunt II TOPO vector (Thermo Fisher Scientific). The rapid amplification of DNA end (RACE) was performed to determine the complete nucleotide sequences of Gtgen3A using a Gene Racer (Invitrogen). The ORF cDNA for Gtgen3A was obtained by PCR amplification with AccuPrime Taq DNA Polymerase, high fidelity (Thermo Fisher Scientific) using primers: ORF-Fw and ORF-Rv, and cloned into a pTrcHis2 TA cloning vector (Invitrogen), resulting in the recombinant plasmid pTrcHis2-Gtgen3A. The plasmid was sequenced with the ABI PRISM Big Dye Terminator Cycle Sequencing Kit using an ABI 3100 DNA sequencer (Applied Biosystems). The primers used are shown in Supplementary Table S1. For expression of GtGen3A, pTrcHis2-GtGen3A was transformed into E. coli NiCo21 (DE3; New England BioLabs). The rGtGen3A was induced with 0.1 mM IPTG overnight at 20°C, purified using HisTalon gravity column, and then further purified through Phenyl-Sepharose and DEAE-Sepharose columns as described above. Purified protein was evaluated by SDS–PAGE and LC–MS/MS analysis.

Assay for hydrolytic activity

For the purification of gentiobiose-hydrolyzing enzyme in gentian leaves, the release of glucose from gentiobiose was monitored. The reaction mixture was incubated at 30°C for 1 h in a total volume of 10 µl containing 10 mM sodium phosphate, pH 7.0, 10 mM gentiobiose, and protein preparation. The reaction was stopped by incubation at 95°C for 2 min and the amount of released glucose was determined. For the characterization of the rGtGen3A, the activity was evaluated in a total volume of 10 µl containing 10 mM gentiobiose and 200 ng of purified protein. The optimum pH was evaluated in the range of pH 5.5–9.0 at 30°C for 30 min. Buffers used were 50 mM MES (pH 5.5–6.5) and Tris–HCl (pH 6.5–9.0) buffer. The pH stability was determined by measuring residual activity after incubation at 4°C for 1 h at each pH. The optimal temperature was measured from 4 to 60°C after 30 min of incubation in MES buffer (pH 6.5). The stability was also evaluated by pre-incubating at 4–60°C for 10 min and residual activity was measured under the optimal condition for 30 min. The reaction was stopped by incubating at 95°C for 2 min and released glucose was measured. The amount of glucose was measured using a glucose CII-Test Wako Kit (Wako Pure Chemicals). When 10 mM p-nitrophenyl-(pNP) sugars were used as a substrate, the amount of p-nitrophenol liberated was measured colorimetrically at 410 nm. The release of glucose from polysaccharides was also determined by thin-layer chromatography according to the method of Takahashi et al. [15]. The reacted mixtures were spotted onto a silica gel 60 F254 plate (Merck), chromatographed with ethyl acetate–acetic acid–water (3 : 2 : 1), and stained with thymol in H2SO4/EtOH (5 : 95, by vol). Substrates tested were sophorose (Serva Electrophoresis GmbH), laminaribiose (Megazyme), laminaritriose (Megazyme), laminaritetraose (Megazyme), laminaripentaose (Megazyme), laminarin (Sigma), cellobiose (Megazyme), cellotriose (Megazyme), cellotetraose (Megazyme), cellopentaose (Megazyme), PSC made from cellulose (Sigmacell20; Sigma), gentiobiose (Sigma), gentianose (LGC Standards), pustulan (Calbiochem), and barley β-1,3-1,4-glucan (Megazyme). Gentiotriose, gentiotetraose, gentiopentaose, and gentiohexaose were kindly provided by Dr Taichi Usui, Dr Takeshi Hattori, and Dr Makoto Ogata (Shizuoka University, Shizuoka, Japan). The protein concentration was determined by the Bradford assay or A280 assay using bovine serum albumin as a standard. The specific activity was represented as μmol glucose min−1 mg−1 protein. Kinetics parameters were calculated by regression analysis using Kaleidagraph (Synergy Software, version 3.51) based on the Michaelis–Menten equation.

Gene expression assay

Quantification of Gtgen3A transcript level by quantitative RT-PCR was conducted in accordance with the method of Takahashi et al. [15]. Synthesis of first-strand cDNA was carried out using a High-Capacity cDNA Reverse transcription Kit (Applied Biosystems) with 500 ng of total RNA according to the manufacturer's instructions. Quantitative RT-PCR was carried out using the StepOnePlus Real-Time PCR System (Applied Biosystems) with the KAPA SYBR FAST ABI Prism qPCR Kit (Kapa Biosystems, Woburn, MA). For the semi-quantitative RT-PCR analysis, cDNA undiluted or diluted to 1 : 5 was used as a template. Actin gene (GtACT) was used as an internal control. Primers used are shown in Supplementary Table S1.

VIGS of Gtgen3A in gentian plants

The expression of Gtgen3A was suppressed by VIGS using the apple latent spherical virus (ALSV) vector according to recent report [38]. Trigger fragment was amplified by PCR using cDNA of G. triflora × G. scabra cv. ‘Polano White’ as a template using primers Target-Fw and Target-Rv as noted in Supplementary Table S1. The amplified fragment was subcloned into a pCR-Blunt II TOPO vector (Thermo Fisher Scientific) and sequenced using an ABI 3100 DNA sequencer as described above. The fragment was digested with XhoI and BamHI, and then ligated into pEALSR2L5R5 vector. The resulting vector was named pEALSR2L5R5-Gtgen3A and used for VIGS. The ALSV vectors, pEALSR1, pE ALSR2L5R5, and pEALSR2L5R5-Gtgen3A, were prepared using the NucleoBond Xtra Midi plus Kit (Macherey-Nagel, Takara-bio) according to the manufacturer's instructions and inoculated into G. triflora × G. scabra cv. ‘Polano White’ plantlet by the PDS-1000/He particle Delivery system (Bio-Rad Laboratories). A 0.5-mg aliquot of gold particles (1.0 μm diameter; Bio-Rad Laboratories) was mixed with 100 µl of plasmid solution containing 5 µg of pEALSR1 and 5 μg of pEALSR2L5R5-Gtgen3A, 10 μl of 10 M ammonium acetate, and 220 μl of isopropanol. After washing with ethanol, particles were suspended in 10 µl of ethanol and bombarded with 1100 psi pressure at a distance of 10 cm from the microcarrier holder. Virus-infected plantlets were selected by PCR for use with the primers: Check-Fw and Check-Rv as noted in Supplementary Table S1. Plantlets carrying pEALSR2L5R5 instead of pEALSR2L5R5-Gtgen3A were used as VC.

Abbreviations

     
  • ALSV

    apple latent spherical virus

  •  
  • CBB

    Coomassie Brilliant Blue

  •  
  • DP

    degree of polymerization

  •  
  • GH3

    glycoside hydrolase family 3

  •  
  • LC–MS

    liquid chromatography–mass spectrometry

  •  
  • MS medium

    Murashige and Skoog medium

  •  
  • ORF

    open reading frame

  •  
  • PSC

    phosphoric acid swollen cellulose

  •  
  • rGtGen3A

    recombinant GtGen3A

  •  
  • VC

    vector control

  •  
  • VIGS

    virus-induced gene silencing

Author Contribution

H.T. and T.T. designed the research and wrote the paper. H.T., S.K.-F., H.Y., C.Y., T.Y., N.K., and T.T. performed the experiments. H.T. analyzed the data.

Funding

This work was supported by a Grant-in-Aid for Young Scientists (B) (No. 25850023 to H.T.) and a Grant-in-Aid for Scientific Research (B) (No. 15H04454 to H.T.) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan.

Acknowledgments

The authors thank Dr Masahiro Nishihara for providing helpful advice and Mr Kohei Fujita, Ms Ayumi Obara, and Ms Yuko Kanno for their technical assistance. The authors also thank Dr Taichi Usui, Dr Takeshi Hattori, and Dr Makoto Ogata (Shizuoka University, Shizuoka, Japan) for providing gentiotriose, gentiotetraose, gentiopentaose, and gentiohexaose.

Competing Interests

The Authors declare that there are no competing interests associated with the manuscript.

References

References
1
Granot
,
D.
,
Kelly
,
G.
,
Stein
,
O.
and
David-Schwartz
,
R
. (
2014
).
Substantial roles of hexokinase and fructokinase in the effects of sugars on plant physiology and development
.
J. Exp. Bot.
65
,
809
819
2
Dai
,
N.
,
Schaffer
,
A.
,
Petreikov
,
M.
,
Shahak
,
Y.
,
Giller
,
Y.
,
Ratner
,
K.
et al. 
(
1999
).
Overexpression of Arabidopsis hexokinase in tomato plants inhibits growth, reduces photosynthesis, and induces rapid senescence
.
Plant Cell
11
,
1253
1266
3
Eveland
,
A.L.
and
Jackson
,
D.P
. (
2012
).
Sugars, signalling, and plant development
.
J. Exp. Bot.
63
,
3367
3377
4
Li
,
Y.
,
Van den Ende
,
W.
and
Rolland
,
F.
(
2014
).
Sucrose induction of anthocyanin biosynthesis is mediated by DELLA
.
Mol. Plant
7
,
570
572
5
Chiou
,
T.J.
and
Bush
,
D.R
. (
1998
).
Sucrose is a signal molecule in assimilate partitioning
.
Proc. Natl Acad. Sci. U.S.A.
95
,
4784
4788
6
Lunn
,
J.E.
,
Delorge
,
I.
,
Figueroa
,
C.M.
,
Van Dijck
,
P.
and
Stitt
,
M
. (
2014
).
Trehalose metabolism in plants
.
Plant J.
79
,
544
567
7
Garg
,
A.K.
,
Kim
,
J.-K.
,
Owens
,
T.G.
,
Ranwala
,
A.P.
,
Choi
,
Y.D.
,
Kochian
,
L.V
et al.  et al.  (
2002
).
Trehalose accumulation in rice plants confers high tolerance levels to different abiotic stresses
.
Proc. Natl Acad. Sci. U.S.A.
99
,
15898
15903
8
Shibuya
,
N.
and
Minami
,
E
. (
2001
).
Oligosaccharide signalling for defence responses in plant
.
Physiol. Mol. Plant Pathol.
59
,
223
233
9
Trouvelot
,
S.
,
Héloir
,
M.C.
,
Poinssot
,
B.
,
Gauthier
,
A.
,
Paris
,
F.
,
Guillier
,
C.
et al. 
 (
2014
).
Carbohydrates in plant immunity and plant protection: roles and potential application as foliar sprays
.
Front. Plant Sci.
5
,
592
10
Yamaguchi
,
T.
,
Yamada
,
A.
,
Hong
,
N.
,
Ogawa
,
T.
,
Ishii
,
T.
and
Shibuya
,
N.
(
2000
).
Differences in the recognition of glucan elicitor signals between rice and soybean: β-glucan fragments from the rice blast disease fungus Pyricularia oryzae that elicit phytoalexin biosynthesis in suspension-cultured rice cells
.
Plant Cell
12
,
817
826
PMID:
[PubMed]
11
Jia
,
H.
,
Jiu
,
S.
,
Zhang
,
C.
,
Wang
,
C.
,
Tariq
,
P.
,
Liu
,
Z.
et al. 
 (
2016
).
Abscisic acid and sucrose regulate tomato and strawberry fruit ripening through the abscisic acid-stress-ripening transcription factor
.
Plant Biotechnol. J.
14
,
2045
2065
12
van Dijken
,
A.J.
,
Schluepmann
,
H.
and
Smeekens
,
S.C
. (
2004
).
Arabidopsis trehalose-6-phosphate synthase 1 is essential for normal vegetative growth and transition to flowering
.
Plant Physiol.
135
,
969
977
13
Schluepmann
,
H.
,
Pellny
,
T.
,
van Dijken
,
A.
,
Smeekens
,
S.
and
Paul
,
M
. (
2003
).
Trehalose 6-phosphate is indispensable for carbohydrate utilization and growth in Arabidopsis thaliana
.
Proc. Natl Acad. Sci. U.S.A.
100
,
6849
6854
14
Wahl
,
V.
,
Ponnu
,
J.
,
Schlereth
,
A.
,
Arrivault
,
S.
,
Langenecker
,
T.
,
Franke
,
A.
et al. 
(
2013
).
Regulation of flowering by trehalose-6-phosphate signaling in Arabidopsis thaliana
.
Science
339
,
704
707
15
Takahashi
,
H.
,
Imamura
,
T.
,
Konno
,
N.
,
Takeda
,
T.
,
Fujita
,
K.
,
Konishi
,
T.
et al. 
 (
2014
).
The gentio-oligosaccharide gentiobiose functions in the modulation of bud dormancy in the herbaceous perennial Gentiana
.
Plant Cell
26
,
3949
3963
16
Dumville
,
J.C.
and
Fry
,
S.C
. (
2003
).
Gentiobiose: a novel oligosaccharin in ripening tomato fruit
.
Planta
216
,
484
495
17
Aizat
,
W.M.
,
Dias
,
D.A.
,
Stangoulis
,
J.C.R.
,
Able
,
J.A.
,
Roessner
,
U.
and
Able
,
A.J
. (
2014
).
Metabolomics of capsicum ripening reveals modification of the ethylene related-pathway and carbon metabolism
.
Postharvest. Biol. Technol.
89
,
19
31
18
Watanabe
,
A.
,
Suzuki
,
M.
,
Ujiie
,
S.
and
Gomi
,
K.
(
2016
).
Purification and enzymatic characterization of a novel β-1,6-glucosidase from Aspergillus oryzae
.
J. Biosci. Bioeng.
121
,
259
264
19
Smaali
,
M.I.
,
Michaud
,
N.
,
Marzouki
,
N.
,
Legoy
,
M.D.
and
Maugard
,
T
. (
2004
).
Comparison of two β-glucosidases for the enzymatic synthesis of β-(1–6)-β-(1–3)-gluco-oligosaccharides
.
Biotechnol. Lett.
26
,
675
679
20
Hrmova
,
M.
,
Harvey
,
A.J.
,
Wang
,
J.
,
Shirley
,
N.J.
,
Jones
,
G.P.
,
Stone
,
B.A.
et al. 
 (
1996
).
Barley β-d-glucan exohydrolases with β-d-glucosidase activity. Purification, characterization, and determination of primary structure from a cDNA clone
.
J. Biol. Chem.
271
,
5277
5286
21
Kim
,
J.-B.
,
Olek
,
A.T.
and
Carpita
,
N.C
. (
2000
).
Cell wall and membrane-associated exo-β-d-glucanases from developing maize seedlings
.
Plant Physiol.
123
,
471
486
22
Nakatsuka
,
T.
,
Yamada
,
E.
,
Saito
,
M.
,
Fujita
,
K.
and
Nishihara
,
M
. (
2013
).
Heterologous expression of gentian MYB1R transcription factors suppresses anthocyanin pigmentation in tobacco flowers
.
Plant Cell Rep.
32
,
1925
1937
23
Nakajima
,
M.
,
Yamashita
,
T.
,
Takahashi
,
M.
,
Nakano
,
Y.
and
Takeda
,
T.
(
2012
).
Identification, cloning, and characterization of β-glucosidase from Ustilago esculenta
.
Appl. Microbiol. Biotechnol.
93
,
1989
1998
24
Kotake
,
T.
,
Nakagawa
,
N.
,
Takeda
,
K.
and
Sakurai
,
N
. (
1997
).
Purification and characterization of wall-bound exo-1,3-β-d-glucanase from barley (Hordeum vulgare L.) seedlings
.
Plant Cell Physiol.
38
,
194
200
25
Crombie
,
H.J.
,
Chengappa
,
S.
,
Hellyer
,
A.
and
Reid
,
J.S.
(
1998
).
A xyloglucan oligosaccharide-active, transglycosylating β-d-glucosidase from the cotyledons of nasturtium (Tropaeolum majus L) seedlings — purification, properties and characterization of a cDNA clone
.
Plant J.
15
,
27
38
26
Koizumi
,
N.
,
Okushima
,
Y.
and
Sano
,
H
. (
2000
).
Isolation, characterization and molecular cloning of β-d-glucan exohydrolase from cultured tobacco cells
.
J. Plant Physiol.
157
,
691
698
27
Pei
,
X.
,
Zhao
,
J.
,
Cai
,
P.
,
Sun
,
W.
,
Ren
,
J.
,
Wu
,
Q.
et al. 
(
2016
).
Heterologous expression of a GH3 β-glucosidase from Neurospora crassa in Pichia pastoris with high purity and its application in the hydrolysis of soybean isoflavone glycosides
.
Protein Expr. Purif.
119
,
75
84
28
Schep
,
G.P.
,
Shepherd
,
M.G.
and
Sullivan
,
P.A
. (
1984
).
Purification and properties of a β-1,6-glucanase from Penicillium brefeldianum
.
Biochem. J.
223
,
707
714
29
Konno
,
N.
and
Sakamoto
,
Y
. (
2011
).
An endo-β-1,6-glucanase involved in Lentinula edodes fruiting body autolysis
.
Appl. Microbiol. Biotechnol.
91
,
1365
1373
30
Lee
,
R.C.
,
Hrmova
,
M.
,
Burton
,
R.A.
,
Lahnstein
,
J.
and
Fincher
,
G.B
. (
2003
).
Bifunctional family 3 glycoside hydrolases from barley with α-l-arabinofuranosidase and β-d-xylosidase activity. Characterization, primary structures, and COOH-terminal processing
.
J. Biol. Chem.
278
,
5377
5387
31
Semenova
,
M.V.
,
Okunev
,
O.N.
,
Gusakov
,
A.V.
and
Sinitsyn
,
A.P
. (
2009
).
Disaccharide synthesis by enzymatic condensation of glucose: glycoside linkage patterns for different fungal species
.
Open Glycosci.
2
,
20
24
32
Kim
,
T.-Y.
,
Lee
,
D.-S.
and
Shin
,
H.-J
. (
2003
).
Gentiobiose synthesis from glucose using recombinant β-glucosidase from Thermus caldophilus GK24
.
Biotechnol. Bioprocess. Eng.
8
,
210
212
33
Fujimoto
,
Y.
,
Hattori
,
T.
,
Uno
,
S.
,
Murata
,
T.
and
Usui
,
T.
(
2009
).
Enzymatic synthesis of gentiooligosaccharides by transglycosylation with β-glycosidases from Penicillium multicolor
.
Carbohydr. Res.
344
,
972
978
34
Kenner
,
D.
and
Requena
,
Y.
(
2001
).
Botanical Medicine: A European Professional Perspective
,
Paradigm Publications
,
Brookline, MA
. p.
146
35
Pittenauer
,
E.
,
Koulakiotis
,
N.S.
,
Tsarbopoulos
,
A.
and
Allmaier
,
G
. (
2013
).
In-chain neutral hydrocarbon loss from crocin apocarotenoid ester glycosides and the crocetin aglycon (Crocus sativus L.) by ESI-MSn (n=2,3)
.
J. Mass. Spectrom.
48
,
1299
1307
36
Laemmli
,
U.K
. (
1970
).
Cleavage of structural proteins during the assembly of the head of bacteriophage T4
.
Nature
227
,
680
685
37
Imamura
,
T.
,
Higuchi
,
A.
,
Sekine
,
K.-T.
,
Yamashita
,
T.
and
Takahashi
,
H
. (
2013
).
High concentrations of sucrose induce overwintering bud formation in gentian plantlets cultured in vitro
.
Plant Biotechnol.
31
,
97
104
38
Nakatsuka
,
T.
,
Saito
,
M.
,
Yamada
,
E.
,
Fujita
,
K.
,
Yamagishi
,
N.
,
Yoshikawa
,
N
et al.  et al.  (
2015
).
Isolation and characterization of the C-class MADS-box gene involved in the formation of double flowers in Japanese gentian
.
BMC Plant Biol.
15
,
182