Biotransformation of Ginsenosides Re and Rg1 into Rg2 and Rh1 by Thermostable β-Glucosidase from Thermotoga thermarum

Article abstract of DOI:10.1007/s10600-017-2025-0

The recombinant thermostable β-glucosidase from Thermotoga thermarum DSM 5069^T exhibited high selectivity to catalyze the conversion of ginsenoside Re and Rg₁ to the more pharmacologically active minor ginsenoside Rg₂ and Rh₁, respectively. At a concentration of 1.36 U/mL of the enzyme, a temperature of 85°C, and pH 5.5, 10 g/L ginsenoside Re was transformed into 8.02 g/L Rg₂ within 60 min, and 2 g/L ginsenoside Rg₁ was transformed into 1.56 g/L Rh₁ within 60 min. This paper provides the first report on the production of ginsenoside Rg₂ and Rh₁ by a highly thermostable β-glucosidase.

Full text of DOI:10.1007/s10600-017-2025-0

DOI 10.1007/s10600-017-2025-0
Chemistry of Natural Compounds, Vol. 53, No. 3, May, 2017
BIOTRANSFORMATION OF GINSENOSIDES Re AND Rg1
INTO Rg AND Rh BY THERMOSTABLE ꢀꢁ-GLUCOSIDASE
2
1
FROM Thermotoga thermarum
Jianjun Pei, Tao Wu,1,2 Tao Yao, Linguo Zhao,
1
,2
1,2
1,2*
3
,4
3,4
3,4*
Gang Ding, Zhenzhong Wang, and Wei Xiao
T
The recombinant thermostable ꢁ-glucosidase from Thermotoga thermarum DSM 5069 exhibited high
selectivity to catalyze the conversion of ginsenoside Re and Rg to the more pharmacologically active minor
1
ginsenoside Rg and Rh , respectively. At a concentration of 1.36 U/mL of the enzyme, a temperature
2
1
of 85ꢂC, and pH 5.5, 10 g/L ginsenoside Re was transformed into 8.02 g/L Rg within 60 min, and 2 g/L
2
ginsenoside Rg was transformed into 1.56 g/L Rh within 60 min. This paper provides the first report on the
1
1
production of ginsenoside Rg and Rh by a highly thermostable ꢁ-glucosidase.
2
1
Keywords: biotransformation, ginsenosides, ꢁ-glucosidase, Thermotoga thermarum.
Ginseng is generally known for its medical properties. The pharmaceutical activities of ginseng roots were recently
demonstrated by many studies, and ginseng has become a world-famous medicinal plant [1]. Ginseng roots contain various
pharmaceutical components, such as ginsenosides, polyacetylenes, polyphenolic compounds, and acidic polysaccharides.
Ginsenosides are the most pharmaceutically active components. Ginsenosides are classified into three groups, namely oleanane,
protopanaxadiol (PPD), and protopanaxatriol (PPT), according to the ginsenoside aglycone, and more than 180 ginsenosides
have been isolated and characterized. Among them, five major ginsenosides (ginsenosides Rb , Rb , Rc, Re, and Rg ) constitute
1
2
1
more than 80% of the total ginsenosides [2]. The various ginsenosides have different pharmaceutical activities because of the
different skeletons and sugar moieties.
HO
O
O
2
1
2
1
OH
OH
HO
OH
2
6
26
OH
20
25
27
20
25
27
2
3
23
OH
17
17
1
9
18
19
10
18
1
1
1
0
15
1
5
3
H
O
3
H
O
3
7
0
30
7
HO
5
HO
5
2
9
28
29
28
HO 6'
HO 6'
4' OH
OH
OH
O
O
O
OH
1'
1'
Re, Rg
2
: R =
4
'
OH
OR
Re, Rg1
OR
OH OH
Rg , Rh : R = H
Rg , Rh
1
1
2
1
1
) College of Chemical Engineering, Nanjing Forestry University, Nanjing, P. R. China, e-mail: lg.zhao@163.com;
2
) Jiangsu Key Laboratory of Biomass Based Green Fuels and Chemicals, Nanjing, P. R. China; 3) Jiangsu Kanion
Pharmaceutical Co., Ltd., Lianyungang, Jiangsu Province, P. R. China; 4) State Key Laboratory of New-tech for Chinese
Medicine Pharmaceutical Process, Lianyungang, Jiangsu Province, P. R. China, e-mail: xw@kanion.com. Published in Khimiya
Prirodnykh Soedinenii, No. 3, May–June, 2017, pp. 402–405. Original article submitted September 14, 2015.
0
472
0009-3130/17/5303-0472 2017 Springer Science+Business Media New York

TABLE 1. 13C NMR Spectra of Ginsenosides 1, 20(S)-Rg [11], and Re [12] (ꢃ, ppm)
2
C atom
Re
1
20(S)-Rg
2
C atom
Re
1
20(S)-Rg
2
1
2
3
4
5
6
7
8
9
39.4
27.8
78.4
40.0
60.9
74.6
46.0
41.2
49.6
39.6
31.0
70.2
49.1
51.5
30.8
26.7
51.7
17.4
17.6
83.3
22.4
36.1
23.3
125.9
130.9
25.9
39.4
27.4
78.1
40.9
60.9
73.9
45.8
40.9
51.4
39.7
31.9
69.1
48.2
51.4
31.9
26.8
54.4
17.4
17.4
72.8
26.9
36.2
22.6
126.5
131.3
25.7
39.5
27.7
78.3
41.1
60.7
74.2
46.0
41.1
49.7
39.9
32.0
71.0
48.1
51.6
31.2
26.8
54.6
17.6
17.5
72.9
27.0
35.7
22.9
126.3
130.7
25.8
27
28
29
30
17.9
32.3
17.8
17.3
17.4
32.0
16.9
17.3
17.6
32.1
16.8
17.1
6-O-Outer-Rha
1ꢄꢄ
2ꢄꢄ
3ꢄꢄ
4ꢄꢄ
5ꢄꢄ
6ꢄꢄ
101.8
72.5
72.4
74.2
69.5
18.8
101.5
72.2
72.3
73.9
69.2
18.5
101.7
72.2
72.6
74.1
69.4
18.7
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
6-O-Inter-Glc
1ꢄ
2ꢄ
3ꢄ
4ꢄ
5ꢄ
6ꢄ
101.8
79.4
78.6
72.6
78.3
63.1
101.7
79.2
78.3
72.3
78.1
62.9
101.9
79.4
78.5
72.4
78.3
63.1
20-O-Glc
1ꢄꢄꢄ
2ꢄꢄꢄ
3ꢄꢄꢄ
4ꢄꢄꢄ
5ꢄꢄꢄ
6ꢄꢄꢄ
98.3
75.2
79.2
71.6
78.4
62.9
In recent decades, many studies have focused on the pharmaceutical activities of the minor ginsenosides, such as
ginsenosides Rg , Rh , F , ginsenoside K (compound K), Rg , F , and Rh , because their activities are found to be superior to
3
2
2
2
1
1
those of the major ginsenosides [3]. The ginsenosides Rg and Rh have pharmaceutical effects such as anti-allergy and
2
1
anti-inflammatory activities, and they inhibit glutamate-induced neurotoxicity in PC12 cells [4]. However, the Rg and Rh
2
1
contents of ginseng roots are less than those of major ginsenosides. By analyzing their structures, we see that Re and Rg have
1
structural similarities to Rg and Rh , respectively. Recently, some enzymes have been applied to the production of ginsenosides
2
1
Rg and Rh from ginsenosides Re and Rg . However, most of the enzymes lacked high catalytic efficiency, thermostability,
2
1
1
or high specificity [5–7]. To date, studies on production of ginsenosides Rg and Rh using the highly thermostable extremozyme
2
1
are not available. In this paper, the induction conditions of expression of recombinant thermostable ꢁ-glucosidase from
T
Thermotoga thermarum DSM 5069 are discussed, and the ꢁ-glucosidase is shown to catalyz the conversion of ginsenosides
Re and Rg to the more pharmacologically active minor ginsenosides Rg and Rh , respectively.
1
2
1
To improve the expression of recombinant thermostable ꢁ-glucosidase (BGL3T) from Thermotoga thermarum DSM
069 , the induction conditions such as induction temperature and concentration of IPTG were determined. The temperature
25, 30, 37, 42ꢂC) and concentration of IPTG (0, 0.2, 0.3, 0.5, 0.8 mM) have an influence on the protein translation in E. coli,
T
5
(
which could sometimes lead to the formation of inclusion bodies [8]. The result showed that the optimal induction temperature
was 37ꢂC and the optimal IPTG concentration was 0.8 mM. The activity of the recombinant ꢁ-glucosidase was increased from
2
6 to 45 U/mL under the optimal conditions. The protein in the cell-free extract was purified to gel electrohomogeneity by heat
treatment. The final preparation produced a single band on SDS-PAGE gel, and the molecular mass of the enzyme was
estimated to be 82 kDa (Fig. 1). The recombinant BGL3T was purified 2-fold with 85% recovery.
473

TABLE 2. 13C NMR Spectra of Ginsenosides 2, 20(S)-Rh [10], and Rg [12] (ꢃ, ppm)
1
1
C atom
Rg
1
2
20(S)-Rh
1
C atom
Rg
1
2
20(S)-Rh
1
1
2
3
4
5
6
7
8
9
39.4
27.9
78.6
40.3
61.3
80.0
45.1
41.1
50.0
39.7
30.9
70.2
49.1
51.3
30.7
26.6
51.6
17.5
17.5
83.2
22.4
36.1
39.4
27.9
78.4
40.2
61.2
79.8
45.0
40.9
50.0
39.4
32.0
70.7
48.0
51.4
31.0
26.7
54.5
17.4
17.4
72.7
26.8
35.6
39.5
28.0
78.7
40.4
61.5
80.1
45.3
41.2
50.3
39.7
32.1
71.1
48.3
51.7
31.3
26.9
54.8
17.4
17.7
73.1
27.1
35.9
23
24
25
26
27
28
29
30
23.2
125.8
130.8
25.7
17.7
31.7
16.3
17.2
22.8
126.1
130.5
25.6
17.4
31.8
16.1
17.1
23.0
126.4
130.8
25.9
17.7
31.8
16.2
16.9
6-O-Glc
1
1
1
1
1
1
1
1
1
1
2
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
1ꢄ
2ꢄ
3ꢄ
4ꢄ
105.8
75.3
79.1
71.8
79.4
62.8
105.9
75.2
79.8
71.6
78.3
62.8
106.1
75.5
79.7
71.9
78.2
63.2
5ꢄ
6ꢄ
20-O-Glc
1ꢄꢄ
98.1
75.0
78.0
71.6
77.9
63.1
2ꢄꢄ
3ꢄꢄ
4ꢄꢄ
5ꢄꢄ
6ꢄꢄ
kDa
30
00
M
1
2
1
1
8
0
5
4
3
5
0
5
Fig. 1. SDS-PAGE analysis of recombinant
BGL3T in E. coli JM109(DE3). M – protein
marker; cell-free extract of JM109 (DE3)
harboring pET-20-BGL3T (1), purified
BGL3T (2).
The effects of temperature and pH on the enzymatic transformation of ginsenoside Re were investigated using the
purified recombinant BGL3T. The optimal pH on BGL3T-transformed ginsenoside Re was determined to be 5.5, whereas the
degradation rate was higher than 60% of the maximum degradation rate at the pH range from 5.0 to 6.0. ꢁ-glucosidases from
Microbacterium esteraromaticum, Bacterodies sp., and Aspergillus niger had optimal activity at pH 7.0. 7.0 and 7.5, respectively
[
5–7]. The optimal temperature of the enzymatic transformation of ginsenoside Re was 85ꢂC, while the activity was higher
than 90% of the maximum activity at the temperature range from 75 to 85ꢂC. The optimal activity of BGL3T was higher than
that of ꢁ-glucosidases from M. esteraromaticum, Bacterodies sp., and Aspergillus niger [7]. The dependence of the rate of the
enzymatic reaction on the substrate concentration followed Michaelis–Menten kinetics; the Km and Vmax for ginsenoside Re
were 0.19 mM and 49 U/mg, and for ginsenoside Rg , 0.29 mM and 7.3 U/mg, under optimal conditions.
1
The ginsenoside Re was transformed into metabolite 1 (ginsenoside Rg ), and the ginsenoside Rg was transformed
2
1
into metabolite 2 (ginsenoside Rh ) (Fig. 2). The substrate specificity of BGL3T showed that it has high selectivity to cleave
1
the outer and inner glucopyranosyl moieties at the C-20 carbon of protopanaxadiol (PPD) and protopanaxatriol (PPT) [9].
474

2
.0
.5
.0
1
0
8
a
b
2
4
1
6
4
2
1
0
.5
0
1
60
3
0
0
20
40
0
20
40
60
Time, min
Time, min
Fig. 2. Ginsenosides Rg (a) and Rh (b) production from ginsenosides Re (1) and Rg (3) by BGL3T. Re (1),
2
1
1
Rg (2), Rg (3), Rh (4).
2
1
1
+
A comparison of the m/z of molecular ions [M + Na] of Re (970) and Rg (824) with metabolite 1 (808) and metabolite 2
1
(
662) showed that the differences were due to a D-glucose residue, respectively. These indicated that metabolites 1 and 2 were
1
3
ginsenosides Rg and Rh , respectively. Furthermore, this was confirmed by comparing the PMR and C NMR spectra of
these compounds (Table 1 and 2). The C NMR spectrum of metabolite 1 contained resonances for only two anomeric C
2
1
1
3
atoms, the chemical shifts of which were similar to those of the C-6 disaccharide of Re. Moreover, the chemical shifts of the
resonances for metabolite 1 were similar to those of 20(S)-Rg but different from those of 20(R)-Rg [10]. Thus, metabolite 1
2
2
1
3
was identical to 20(S)-Rg . Like metabolite 1, the C NMR spectrum of metabolite 2 showed a strong-field shift for the C-20
2
resonance compared with that for Rg with the same C-6 resonance. This indicated that C-20 in metabolite 2 lacked a glucose
1
1
3
residue. Thus, it was identical to Rh . On the other hand, the C NMR spectra of metabolite 2 were similar to those of
1
2
0(S)-Rh but different from those of 20(R)-Rh [10]. This indicated unambiguously that metabolite 1 was identical to 20(S)-Rh .
1
1
1
To verify the biotransformation pathways of the ginsenosides Re and Rg , a time-course experiment was performed,
1
and the hydrolyzed products were analyzed by HPLC. As shown in Fig. 2, at a concentration of 1.36 U/mL of the enzyme, a
temperature of 85ꢂC, and pH 5.5, 10 g/L ginsenoside Re was transformed into 8.12 g/L Rg (96.3% molar conversion) within
2
6
0 min, and 2 g/L ginsenoside Rg was transformed into 1.56 g/L Rh (95.6% molar conversion) also within 60 min. Thus,
1
1
BGL3T shows high selectivity to cleave the glucopyranosyl moiety at the C-20 carbon of ginsenosides Re and Rg . It has been
1
reported that some ꢁ-glucosidases can cleave the glucopyranosyl moiety at the C-20 carbon of ginsenosides Re and Rg to
1
produce ginsenosides Rg and Rh [6, 7]. However, these enzymes exhibited low thermostability, low catalytic efficiency, and
2
1
poor productivity.
In this study, the BGL3T expression level was increased from 26 to 45 U/mL in Luria–Bertani (LB) medium by
optimizing the induction conditions. The BGL3T converted almost completely ginsenosides Re and Rg to ginsenosides Rg
1
2
and Rh with high selectivity and high catalytic efficiency. Thus, this study provides a useful ꢁ-glucosidase that may be used
1
to improve the enzymatic conversion of the ginsenosides Re and Rg to the ginsenosides Rg and Rh .
1
2
1
EXPERIMENTAL
Bacterial Strains, Plasmids, Growth Media, Materials. Ginsenosides Re, Rg , Rg , and Rh were purchased from
1
2
1
Chendu Must Bio-Technology (Chendu, China). p-Nitrophenyl-ꢁ-D-glucopyranoside was purchased from Sigma-Aldrich
St Louis, MO, USA).
The pET-20-BGL3T (recombinant plasmid carrying the thermostable ꢁ-glucosidase gene) was constructed by our
(
laboratory [9]. Escherichia coli JM109 and JM109(DE3) were grown at 37ꢂC in LB and supplemented with ampicillin when
T
required. Thermotoga thermarum DSM 5069 was purchased from DSMZ (www.dsmz.de). It was grown anaerobically at
8
0ꢂC as described previously [13].
Expression and Purification of BGL3T. The pET-20-BGL3T plasmid was transformed into E. coli JM109 (DE3)
and induced to express the recombinant BGL3T by adding isopropyl-ꢁ-D-thiogalactopyranoside (IPTG) to final concentration
0
to 1 mM at an OD600 of approximately 0.6, and incubated further at 25ꢂC or 30ꢂC or 37ꢂC or 42ꢂC for about 12 h.
475

The recombinant cells (200 mL) carrying pET-20-BGL3T were harvested by centrifugation at 5,000g for 10 min at
4
ꢂC, washed twice with distilled water, resuspended in 50 mL of 20 mM Tris-HCl buffer (pH 7.5), and French-pressured three
times. The cell extracts were heat treated (80ꢂC, 40 min), cooled in an ice bath, and centrifuged (20,000g, 4ꢂC, 30 min). The
protein was examined by SDS-PAGE, and the protein bands were analyzed by density scanning with an image analysis system
(
Bio-Rad, USA). The protein concentration was determined by the Bradford method using BSA as a standard.
Determination of EnzymeActivities and Properties. The reaction mixture, which contained 50 mM imidazole-potassium
buffer (pH 5.0), 1 mM p-nitrophenyl-ꢁ-D-glucopyranoside, or 1 mM ginsenosides Re or Rg as substrates, and a certain
1
amount of ꢁ-glucosidase in 0.2 mL, was incubated for 5 min at 90ꢂC. The reaction was stopped by the addition of 1 mL of
1
M Na CO . For p-nitrophenyl-ꢁ-D-glucopyranoside, the absorbance of the mixture was measured at 405 nm. For the
2 3
ginsenosides Re or Rg , the residue was assayed by HPLC. One unit of enzyme activity was defined as the amount of enzyme
1
necessary to liberate 1 ꢅmol of pNP (or ginsenosides Rg or Rh ) per min under the assay conditions.
2
1
The effects of pH and temperature on the activity of recombinant BGL3T were determined using ginsenoside Re as a
substrate. The optimum pH for BGL3T activity was determined by incubation at 85ꢂC for 20 min in 50 mM imidazole-potassium
buffer at a pH in the range of 4.5 to 6.5. The optimum temperature for BGL3T activity was determined by a standard assay with
a temperature ranging from 75 to 95ꢂC in 50 mM imidazole-potassium buffer at pH 5.5. The results are expressed as percentages
of the activity obtained at either the optimum pH or the optimum temperature.
The kinetic constant of BGL3T was determined by measuring the initial rates at various ginsenoside (Re or Rg1)
concentrations under standard reaction conditions.
Analysis of Ginsenosides Re and Rg Degradation. The ginsenoside Re or Rg was treated with purified BGL3T,
1
1
and the degradation was subjected to HPLC analysis. The reaction mixture (50 ꢅL) contained 50 mM imidazole-potassium
buffer (pH 5.5), 10 g/L ginsenoside Re or 2 g/L ginsenoside Rg , and 1.36 U/mL BGL3T. The reaction was performed for
1
various times at 85ꢂC and stopped by introduction into an ice bath. The reaction was analyzed by HPLC at various times.
HPLC and LC/MS Analysis. The HPLC analysis of ginsenosides was performed as described previously [9].
LC/MS for ginsenosides was analyzed by Agilent 6310 Ion Trap LC/MS with positive polarity and an ion trap analyzer.
Ion spray was operated under 5 L N /min, 3.5 kV, 25 psi, and 300ꢂC.
2
Structural Identification. The reaction metabolite 1 and metabolite 2 by BGL3T were separated on a silica gel
column using choloform–methanol–water (8:3:1). Their structures were determined using the proton and carbon nuclear
1
13
magnetic resonance ( H NMR, C NMR) spectrum method (Bruker Avance III 400) with pyridine-d as the solvent.
5
1
Metabolite 1 [20(S)-Rg ]. White powder. H NMR spectrum (400 MHz, Py-d , ꢃ, ppm, J/Hz): 0.92 (3H, s, CH -30),
2
5
3
0
.94 (3H, s, CH -19), 1.17 (3H, s, CH -18), 1.33 (3H, s, CH -29), 1.37 (3H, s, CH -21), 1.61 (3H, s, CH -27), 1.65 (3H, s,
3 3 3 3 3
CH -26), 2.11 (3H, s, CH -28), 5.24 (1H, d, J = 8, H-1ꢄ), 6.49 (1H, br.s, H-1ꢄꢄ) (only characteristic proton resonances are
3
3
+
given). Mass spectrum, FAB m/z 808 [M + Na] .
Metabolite 2 [20(S)-Rh ]. Whiter powder. H NMR spectrum (400 MHz, Py-d , ꢃ, ppm, J/Hz): 0.80 (3H, s, CH -30),
1
1
5
3
1
.01 (3H, s, CH -19), 1.17 (3H, s, CH -18), 1.38 (3H, s, CH -29), 1.39 (3H, s, CH -21), 1.61 (3H, d, CH -27), 1.64 (3H, s,
3 3 3 3 3
+
CH -26), 2.06 (3H, s, CH -28), 5.01 (1H, d, J = 8, H-1ꢄ) (only characteristic proton resonances are given). FAB m/z 662 [M + Na] .
3
3
ACKNOWLEDGMENT
This work was supported by the Natural Science Foundation of Jiangsu Province of China (Grant No. BK20131423),
the Open Fund of Jiangsu Key Laboratory of Biomass Based Green Fuels and Chemicals (JSBGFC12003), Postdoctoral
Science Foundation of Jiangsu Province (1302022B), and a project funded by the Priority Academic Program Development of
Jiangsu Higher Education Institutions (PAPD).
REFERENCES
1
2
.
.
L. P. Christensen, Adv. Food Nutr. Res., 55, 1 (2008).
B. S. Sun, L. J. Gu, Z. M. Fang, C. Y. Wang, Z. Wang, M. R. Lee, Z. Li, J. J. Li, and C. K. Sung, J. Pharm. Biomed.
Anal., 50, 15 (2009).
476

3
.
D. S. An, C. H. Cui, B. H. Sung, H. C. Yang, S. C. Kim, S. T. Lee, W. T. Im, and S. G. Kim, Appl. Microbiol.
Biotechnol., 94, 673 (2012).
4
5
6
7
8
9
0.
1.
2.
3.
.
.
.
.
.
.
N. Li, B. Liu, D. E. Dluzen, and Y. Jin, J. Ethnopharmacol., 111, 458 (2007).
S. R. Ko, K. J. Choi, K. Suzuki, and Y. Suzuki, Chem. Pharm. Bull., 51, 404 (2003).
E. A. Bae, J. E. Shin, and D. H. Kim, Pharm. Bull., 28, 1903 (2005).
L. H. Quan, J. W. Min, S. Sathiyamoorthy, D. U. Yang, Y. J. Kim, and D. C. Yang, Biotechnol. Lett., 34, 913 (2012).
K. L. Pan, H. C. Hsiao, C. L. Weng, M. S. Wu, and C. P. Chou, J. Bacteriol., 185, 3020 (2003).
J. J. Pei, J. C. Xie, R. Yin, L. G. Zhao, G. Ding, Z. Z. Wang, and W. Xiao, J. Mol. Catal. B Enzym., 113, 104 (2015).
L. N. Ten, S. M. Chae, and S. A. Yoo, Chem. Nat. Compd., 51, 81 (2015).
H. Yang, J. Y. Kim, S. O. Kim, Y. H. Yoo, and S. H. Sung, J. Ginseng Res., 38, 194 (2014).
R. Teng, H. Li, J. Chen, D. Wang, Y. He, and C. Yang, Magn. Reson Chem., 40, 483 (2002).
E. Windberger, R. Huber, A. Trincone, H. Fricke, and K. O. Stetter, Arch. Microbiol., 151, 506 (1989).
1
1
1
1
477