Overexpression and characterization of a glycoside hydrolase family 1 enzyme from Cellulosimicrobium cellulans sp. 21 and its application for minor ginsenosides production

Article abstract of DOI:10.1016/j.molcatb.2015.06.015

Abstract A novel β-glucosidase gene (ccbgl1a) was cloned from the ginsenosides-transforming strain Cellulosimicrobium cellulans sp. 21. This enzyme was overexpressed in Escherichia coli, the recombinant β-glucosidase (CcBgl1A) containing N-terminal His-tag was sufficiently purified by nickel metal affinity chromatography with purification factor of 1.9-fold and specific activity of 31.5 U/mg. The molecular mass of recombinant CcBgl1A was estimated to be approximately 46 kDa. CcBgl1A exhibited optimal activity at 35°C and pH 5.5. However, above 40°C, the enzyme stability significantly decreased. The enzyme showed high bioconversion ability on protopanaxadiol-type ginsenosides mixture (PPDGM), which could hydrolyze the outer C-3 glucose moieties of ginsenosides Rb1, Rb2, Rc and Rd into the rare ginsenosides Gypenoside XVII (Gyp XVII), compound O, ginsenoside Mb and ginsenoside F2. Scaled-up production using 1 g of the PPDGM resulted in 292 mg Gyp XVII, 134 mg CO, 184 mg Mb, and 62 mg F2, with chromatographic purities. These results suggest that CcBgl1A would be potentially useful in the preparation of pharmacologically active minor ginsenosides Gyp XVII, CO, Mb and F2.

Full text of DOI:10.1016/j.molcatb.2015.06.015

Accepted Manuscript
Title: Overexpression and characterization of a glycoside
hydrolase family 1 enzyme from Cellulosimicrobium cellulans
sp.21 and its application for minor ginsenosides production
Author: Ye Yuan Yanbo Hu Chenxing Hu Jiayi Leng Honglei
Chen Xuesong Zhao Juan Gao Yifa Zhou
PII:
S1381-1177(15)30021-7
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcatb.2015.06.015
MOLCAB 3189
To appear in:
Journal of Molecular Catalysis B: Enzymatic
Received date:
Revised date:
Accepted date:
2-2-2015
26-6-2015
26-6-2015
Please cite this article as: Y. Yuan, Y. Hu, C. Hu, J. Leng, H. Chen, X. Zhao,
J. Gao, Y. Zhou, Overexpression and characterization of a glycoside hydrolase
family 1 enzyme from Cellulosimicrobium cellulans sp.21 and its application for
minor ginsenosides production, Journal of Molecular Catalysis B: Enzymatic (2015),
http://dx.doi.org/10.1016/j.molcatb.2015.06.015
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

1
2
Overexpression and characterization of a glycoside hydrolase family 1 enzyme from
Cellulosimicrobium cellulans sp.21 and its application for minor ginsenosides production
3
4
Ye Yuan1, Yanbo Hu1, Chenxing Hu1, Jiayi Leng1, Honglei Chen1, Xuesong Zhao2, Juan Gao1, *
Yifa Zhou1, *
,
5
6
7
1Jilin Province Key Laboratory on Chemistry and Biology of Natural Drugs in Changbai
Mountain, School of Life Sciences, Northeast Normal University, Changchun 130024, PR China
2College of Mining Industry, Liaoning Technical University, Fuxin 123000, PR China
8
9
10
11
12
13
14
15
16
*Corresponding author, Juan Gao and Yifa Zhou. Jilin Province Key Laboratory on Chemistry and
Biology of Natural Drugs in Changbai Mountain, School of Life Sciences, Northeast Normal
University, Changchun 130024, PR China.
Tel/Fax: +86-431-85098212; E-mail: gaoj199@nenu.edu.cn (Juan Gao); zhouyf383@nenu.edu.cn
(Yifa Zhou).
Page 1 of 30

1
2
Abstract
A novel β-glucosidase gene (ccbgl1a) was cloned from the ginsenosides-transforming strain
3
Cellulosimicrobium cellulans sp.21. This enzyme was overexpressed in Escherichia coli, the
recombinant β-glucosidase (CcBgl1A) containing N-terminal His-tag was sufficiently purified by
nickel metal affinity chromatography with purification factor of 1.9-fold and specific activity of
31.5 U/mg. The molecular mass of recombinant CcBgl1A was estimated to be approximately 46
kDa. CcBgl1A exhibited optimal activity at 35oC and pH 5.5. However, above 40oC, the enzyme
stability significantly decreased. The enzyme showed high bioconversion ability on
protopanaxadiol-type ginsenosides mixture (PPDGM), which could hydrolyze the outer C-3
glucose moieties of ginsenosides Rb1, Rb2, Rc and Rd into the rare ginsenosides Gypenoside
XVII (Gyp XVII), compound O, ginsenoside Mb and ginsenoside F2. Scaled-up production using
1 g of the PPDGM resulted in 292 mg Gyp XVII, 134 mg CO, 184 mg Mb, and 62 mg F2, with
chromatographic purities. These results suggest that CcBgl1A would be potentially useful in the
preparation of pharmacologically active minor ginsenosides Gyp XVII, CO, Mb and F2.
Keywords: Biotransformation; Cellulosimicrobium cellulans sp.21; ginsenosides; β-glucosidase;
overexpression
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
1. Introduction
Ginseng, the root of Panax ginseng C.A. Meyer, has been used as a traditional medicine in
Asian counties to strengthen immunity, supply nutrition and decrease fatigue for over 2000 years
[1-3]. Ginsenosides are reported to be the major active components of ginseng [4, 5]. The major
ginsenosides Rb1, Rb2, Rc, Rd, Re, and Rg1 represents more than 80% of total ginseng
Page 2 of 30

1
2
ginsenosides [6, 7]. Some ginsenosides, such as Gypenoside XVII (Gyp XVII or GXVII), F2,
Compound O (CO) and Mb, that are deglycosylated from the major ginsenosides, exist in smaller
amounts or are absent in ginseng. Deglycosylated ginsenosides are more pharmaceutically active
than major glycosylated ginsenosides, because of their small size, high bioavailability and good
permeability across the cell membrane [8, 9]. Thus, the minor ginsenosides have attracted more
attention in recent years. These minor ginsenosides are difficult to obtain from a direct extraction
of ginseng. However, the minor ginsenosides Gyp XVII, CO, Mb and F2 can be produced by
hydrolyzing the β-(1→2)-glucosidic linkage at the C-3 position of the major ginsenosides Rb1,
Rb2, Rc and Rd, respectively. Many methods including heating, acidic and alkali treatments, and
enzymatic conversion, have been developed for the production of the minor ginsenosides [10-12].
Among these, the enzymatic conversion is the most used because of its high specificity, yield, and
productivity [13-16].
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Some bacteria and fungi, and the ginsenoside-hydrolyzing enzymes produced by
microorganisms are reported for ginsenoside conversion [17, 18]. Compared to native
ginsenoside-hydrolyzing enzymes or gensinosidase producing microorganisms, recombinant
enzyme are better in ginsenoside biotransformation due to its high yield, easy preparation
procedure and more stable biotransformation process [9]. Several reports focused on
biotransformation of ginsenosides by recombinant enzymes [9, 14, 16, 19-23]. In the present study,
a ginsenoside-hydrolyzing β-glucosidase gene ccbgl1a was cloned from Cellulosimicrobium
cellulans sp. 21, the recombinant protein, CcBgl1A, was overexpressed, purified and the
enzymatic properties were investigated. The biotransformation of the major ginsenosides Rb1,
Rb2, Rc, and Rd by CcBgl1A, to the rare ginsenosides Gyp XVII, compound O, compound Mb,
Page 3 of 30

1
2
and F2 was also investigated.
3
2. Materials and Methods
4
2.1 Chemicals and Reagents
5
Substrates p-nitrophenyl (pNP)-β-glucopyranoside (pNPβGlc), pNP-α-glucopyranoside
(pNPαGlc), pNP-β-galactopyranoside (pNPβGal), pNP-α-galactopyranoside (pNPαGal),
6
7
pNP-β-mannopyranoside
(pNPβMan),
pNP-α-mannopyranoside
(pNPαMan),
8
pNP-β-xylopyranoside (pNPβXyl), pNP-α-L-arabinofuranoside (pNPαAraf), 4-methylumbelliferyl
β-D-glucopyranoside (MUG) were purchased from Sigma (St. Louis, MO, USA). All other
chemicals and reagents were of analytical grade. Protopanaxadiol-type ginsenosides mixture
(PPDGM), single ginsenosides Rb1, Rb2, Rc, Rd, F2, Gyp XVII, compound O and compound Mb,
were prepared in our lab as described before [12].
9
10
11
12
13
14
15
16
17
18
19
20
21
22
2.2 Bacterial Strains, Plasmids and Media
C. cellulans sp. 21 with the collection number CGMCC 7587 in China General Microbiological
Culture Collection Center was grown on Luria Bertani (LB) agar at 30°C for 24 h. The stain was
identified by 16S rDNA sequencing. The 16S rDNA amplification was performed using 27F
(AGAGTTTGATCCTGGCTCAG) and 1492R (TACGGCTACCTTGTTACGACTT) as primers.
The program for PCR was 94°C for 5 min, 30 cycles of three steps (94°C for 1 min, 55°C for 30 s,
and 72°C for 1 min), and followed by 72°C for 10 min additionally. Obtained 1409-bp amplicon
was analyzed by Sangon Biotech. Co. (Shanghai, China) with 27F and 1492R as sequencing
primers.
E. coli BL21 (DE3) and pET-28a (+) (Novagen, Madison, WI, USA) were used as host and
Page 4 of 30

1
2
expression vector sources, respectively. Recombinant E. coli strains were grown in LB broth with
50 μg/ml kanamycin at 37°C.
3
2.3 Basic Recombinant DNA Methods
4
Genomic DNA of C. cellulans sp. 21 was isolated using DNeasy Tissue kit (Qiagen, Hilden,
Germany). Primers ccbgl1a-F (5’-GGAATTCCATATGGTGTCGATCACCTTCCCCGAGTC-3’)
and ccbgl1a-R (5’-CGGGATCCTTATCAGTCGTCGAGCGAGACGGTC-3’) were used for gene
ccbgl1a amplification to introduce NdeI and BamHI restriction sites (underlined) for cloning to the
pET-28a (+). PCR was performed using Phusion polymerase and the following program: 98°C for
30 s, 30 successive cycles of 98°C for 10 s, 68°C for 45 s, and 72°C for 1 min, and final 72°C for
10 min. The PCR product and pET28a (+) were digested with NdeI and BamHI and ligated to
insert the ccbgl1a gene into pET28a (+) for expression. The resulting recombinant
pET28a-ccbgl1a was used to transform E. coli BL21 (DE3) cells. The positive transformants were
verified by DNA sequencing. All enzymes used were from New England Biolab (Beverly, MA).
The DNA isolations, restriction enzyme digestions, ligations and transformations were performed
as the suppliers’ recommendations.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
2.4 Expression and Purification of Enzyme CcBgl1A
Recombinant E. coli BL21 (DE3) harboring the recombinant plasmid was grown in 200 mL of
LB broth with 50 μg/ml kanamycin at 37°C. When the OD 600nm reached 0.5, the culture was
induced with 0.5 mM isopropyl-β-D-thiogalactopyranoside (IPTG) and then grown for another 20
h at 25°C. Subsequently, the cells were harvested and disrupted by sonication in binding buffer (10
mM imidazole, 0.1 M NaCl, 20 mM Tris-HCl, pH 7.5). Cell debris were removed by
centrifugation (14,000 × g, 4°C, 30 min) to obtain the crude extract.
Page 5 of 30

1
2
Protein CcBgl1A carrying His-tag was loaded onto Ni sepharose fastflow column (GE
healthcare). The column of 5 mL was pre-equilibrated by the binding buffer, washed with 20 mL
washing buffer (20 mM imidazole, 0.1 M NaCl, 20 mM Tris-HCl, pH 7.5) and then eluted with 20
mL eluting buffer (250 mM imidazole, 0.1 M NaCl, 20 mM Tris-HCl, pH 7.5). The active
fractions were pooled, desalted by membrane ultrafiltration with Amicon YM10 centriprep device
(Millipore, Bedford, Massachusetts, USA) and used for further characterizations.
3
4
5
6
7
The purified CcBgl1A was analyzed by SDS-PAGE on 10% separating gel. Densitometry of the
stained gels was used to estimate the quantity of enzyme in the preparation. Protein concentrations
were determined by Bradford method using bovine serum albumin (BSA) as a standard. For
detection of in-gel β-glucosidase activity, the purified CcBgl1A was analyzed by native PAGE
using 8% separation gel under non-reducing conditions. After electrophoresis, the gel was washed
twice with 50 mM sodium acetate buffer (pH 5.5) for 10 min and then overlaid with 2 mM MUG
in 50 mM sodium acetate buffer (pH 5.5) at 25oC for 5 min. The fluorescent band was visualized
under UV light and proteins were stained with Coomassie brilliant blue G-250.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
2.5 Enzymatic Activity Analysis of CcBgl1A on Different Substrates
Enzymatic activities of CcBgl1A on various substrates including pNP-linked glycosides,
disaccharides and ginsenosides were assayed. The reaction mixture of 200 μl composed of 50 mM
sodium acetate buffer (pH 6.0), 5 mM substrate and 2 μg β-glucosidase, was incubated at 30°C for
10 min. The reaction mixture without enzyme was used as a blank. The reaction was stopped by
the addition of 1 M NaOH (100 μl). For pNP-linked glycosides, the absorbance of the mixture was
measured at 405 nm by BioTek ELx808 microplate reader (Winooski, VT, USA). The absorbance
was compared to a standard curve of p-nitrophenol in a solution of the same composition (0.1-5
Page 6 of 30

1
2
mM). The enzymatic activities of CcBgl1A on sophorose, laminaribiose, cellobiose and
gentiobiose were measured with dinitrosalicylic acid reagent at 520 nm by determining the
reducing sugars, with glucose as a standard (0.1-5 mM) [24]. The hydrolysis activities on
ginsenosides by CcBgl1A were determined by HPLC (see analytical methods). The amount of
product was calculated according to its peak area. The standard curve of Gyp XVII, CO, Mb and
F2 was used at a concentration of 0.1-5 mM. One unit (U) of β-glucosidase activity was defined as
the amount of enzyme releasing 1 µmol/min of p-nitrophenyl/reducing sugar/ginsenoside under
assay conditions.
3
4
5
6
7
8
9
2.6 Effect of pH and Temperature on Enzyme Activity
10
11
12
13
14
15
16
17
18
19
20
21
22
The dependence of the purified CcBgl1A on pH was determined in different buffers (50 mM
final concentration) at different pH (pH 2.0-6.0, sodium acetate buffer; pH 6.0-8.0,
Na2HPO4-NaH2PO4 buffer; pH 8.0-11.0, Glycine-NaOH buffer) using pNPβGlc as a substrate.
The pH stability was investigated under standard assay conditions after incubation of the purified
enzyme for 24 h at 4°C in the buffer without substrate. The optimum temperature was determined
by measuring enzymatic activity at pH 5.5 in the temperature range of 20-80°C. Temperature
stability was measured by analyzing residual activity after incubation of aliquots of enzyme at
different time and different temperatures.
2.7 Effect of Metal Ions and Chemical Reagents on Enzyme Activity
The effects of metals and other chemicals on CcBgl1A activity were determined. CcBgl1A
activity was tested in the presence of 5 or 50 mM (final concentration) of metals or other
chemicals for 1 h at 30°C. The remaining activity was determined using pNPβGlc as a substrate
(final concentration 5 mM) as described before, and activities are expressed as a percentage of the
Page 7 of 30

1
2
activity obtained in the absence of the compound.
2.8 Determination of Kinetic Parameters
3
To determine the kinetic parameters on various substrates, substrate concentration was varied
(0.05 mM – 2 mM) and reaction time of 15 min was chosen to ensure initial rates. Buffer and
enzyme concentrations used are identical to that stated above. The Km and Vmax values were
calculated from GraphPad Prism V5.
4
5
6
7
2.9 Scaled-up Production of Gyp XVII, CO, Mb and F2 by Crude Recombinant CcBgl1A
Scaled-up production of minor ginsenosides Gyp XVII, CO, Mb and F2 was performed in a 1-L
stirred tank reactor. 200-ml reaction mixture contained 5 mg/ml of substrate PPDGM and crude
recombinant CcBgl1A (final protein concentration 1 mg/ml) in 50 mM phosphate buffer (pH 5.5).
After reaction at 30 °C for 2 h at 100 rpm, the mixture was extracted with one volume of n-butanol
twice. The n-butanol extracts were pooled and evaporated in vacuo, then applied onto preparative
Shim-pack PREP-ODS (H) column (20 mm × 250 mm, 5 μm) connected to an HPLC system
(Shimadzu, Japan). The column was eluted at the flow rate of 5.0 ml/min with the following
gradient program: 0–10 min, 32% acetonitrile (in distilled water, v/v); 10–40 min, 32%–60%
acetonitrile; and 40–50 min, 60% acetonitrile, monitored by the absorbance at 203 nm. Fractions
were taken every 5-ml elution. Each product was pooled and evaporated in vacuo.
2.10 Analytical Methods
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
TLC was carried out using a silica gel G60 plate and a developing solvent consisting of
chloroform, methanol and water (65:35:10, v/v/v, lower phase). The ginsenosides developed on
the plate were stained by spraying with 5% sulphuric acid in ethanol (v/v), and then heating at
110°C for 5 min. HPLC analysis was performed using analytical Shim-pack PREP-ODS (H)
Page 8 of 30

1
2
column (4.6 mm × 250 mm, 5 μm) connected to an HPLC system (Shimadzu, Japan), eluted at the
flow rate of 1.0 ml/min with the following gradient program: 0–10 min, 32% acetonitrile (in
distilled water, v/v); 10–40 min, 32%–60% acetonitrile; and 40–50 min, 60% acetonitrile,
monitored by the absorbance at 203 nm. 13C-NMR spectra were carried out on a Bruker Av 600
NMR spectrometer at 150 MHz with CD3OD as the solvent and TMS as the internal standard.
2.11 Nucleotide Sequence Accession Numbers
3
4
5
6
7
The sequences for the 16S rRNA and CcBgl1A genes from C. cellulans sp. 21 were deposited in
GenBank under accession numbers KR349463 and KR185312, respectively.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
3. Results and Discussion
3.1 Gene Cloning and Analysis of CcBgl1A Gene
Biotransformation is considered to be a desirable method for the preparation of bioactive minor
ginsenosides because of its high specificity, yield, and productivity. Previously, we isolated a
ginsenoside-hydrolyzing strain from soil, which was identified by 16S rDNA sequencing to be C.
cellulans. In the present study, a ginsenoside-hydrolyzing β-glucosidase gene belonging to GH
family 1 was cloned from the genomic DNA of C. cellulans sp. 21. Although some
ginsenoside-hydrolyzing β-glucosidases from bacteria and fungi have been studied extensively [9,
11, 14, 19-21, 23], this represents the first report on the overexpression and characterization of C.
cellulans ginsenoside-hydrolyzing enzyme.
The β-glucosidase gene consisting of 1,173 bp encoding 390 amino acids with a theoretical
molecular
mass
of
43.33
kDa
and
a
theoretical
pI
value
of
4.85
(http://web.expasy.org/compute_pi/). Signal sequence of CcBgl1A was analyzed by the SignalP4.1
Page 9 of 30

1
2
server (http://www.cbs.dtu.dk/services/SignalP/), the lack of a signal sequence suggests its
intracellular location in C. cellulans. The multiple amino acid sequence alignment indicated
CcBgl1A exhibited identities with some characterized GH1 ginsenoside-hydrolyzing enzymes (Fig.
S1), such as β-glucosidases from Arthrobacter chlorophenolicus (43%), Sphingopyxis alaskensis
(33%), Sphingomonas sp. 2F2 (30%), Pyrococcus furiosus (29%), and β-galactosidase from
Sulfolobus acidocaldarius (25%) [9, 14, 19-20, 23]. Amino acid sequence comparisons revealed
that CcBgl1A shared the same conserved catalytic residue sequences with other GH1
ginsenoside-hydrolyzing enzymes, the Glu158 in conserved region Asn-Glu-Pro (NEP) was
identified as the acid/base catalyst, and Glu304 in conserved region Glu-Asn-Gly (ENG) was
identified as the catalytically nucleophile.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
3.2 Overexpression, Purification, and Molecular Mass Determination of CcBgl1A from C.
cellulans sp. 21
The recombinant enzyme was purified as a soluble protein from crude E. coli extracts. In order
to maximize the yield of the fusion protein in a soluble form, different induction conditions were
tested. It was found that 0.5 mM IPTG at 25oC for 20 h cultivation after induction produced the
maximum amount of soluble active fusion enzyme (data not shown). Under the optimal induction
condition, the concentration of total protein was 0.5 g/l culture and the specific activity of
recombinant β-glucosidase was 8350 U/l culture. This high productivity makes the recombinant
CcBgl1A possible for industrial application.
The fusion protein was purified by Ni sepharose fastflow column, β-glucosidase from C.
cellulans sp. 21 was purified with a final purification of 1.9-fold, with a yield of 72%, and a
specific activity of 31.5 U/mg (Table 1). The expressed protein analyzed by SDS-PAGE gel
Page 10 of 30

1
2
showed a single band with a molecular mass of approximately 46 kDa (Fig. 1a), consistent with
the calculated value of 45,495 Da based on the 410 amino acids, which includes the 6 histidines of
the affinity tag. The molecular weight of CcBgl1A is similar to that of some GH1
ginsenoside-hydrolyzing β-glucosidases, such as A. chlorophenolicus (43.7 kDa) [23], P.
mucilaginosus (~48 kDa) [21], Sphingomonas sp. 2F2 (50 kDa) [14], S. alaskensis (51 kDa) [9],
and P. furiosus (55.5 kDa) [20].
3
4
5
6
7
With native PAGE and MUG zymogram analysis (Fig. 1b), the single band stained by
Coomassie Brilliant Blue G-250 was at the same location as the β-glucosidase activity band,
which proved the recombinant CcBgl1A is indeed a β-glucosidase.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
3.3 Characterization of Recombinant β-Glucosidase CcBgl1A from C. cellulans sp. 21
The hydrolytic activity of CcBgl1A was examined over a pH range from 2.0 to 11.0 with
pNPβGlc as a substrate (Fig. 2). A maximum activity was observed at pH 5.5. CcBgl1A was stable
within the pH range from 6.0 to 10.0. After pre-incubated at 4°C for 24 h, over 60% of the activity
was retrieved at pH 6.0-10.0 (Fig. 2). The effect of temperature on the enzyme activity was
investigated at pH 5.5, and a maximum activity was recorded at 35°C (Fig. 3a). The effect of
temperature on enzyme stability was investigated by varying the temperature from 30 to 50°C at a
constant pH of 5.5. The results showed that CcBgl1A exhibited a poor thermal stability. More than
80% of its activity remained after treatment at temperature below 30°C for 1 h, however, the
activity significantly decreased above 35°C and was completely lost at 50°C for 5 min incubation
(Fig. 3b). The half-life values were 13.5 min, 7.6 min, 5.1 min and 3.4 min for 35, 40, 45 and
50°C. Therefore, for the efficiency of biotransformation, the reaction temperature was determined
to be 30°C.
Page 11 of 30

1
2
The effects of metal ions and reagents on CcBgl1A activity were also investigated (Table 2).
CcBgl1A activity was not affected by most metal ions and reagents we detected at the
concentration of 5 mM, only Hg2+ and SDS significantly inhibited the enzymatic activity of
CcBgl1A. In contrast, EDTA and Mg2+ slightly activated the enzymatic activity of CcBgl1A. At
the concentration of 50 mM, most metal ions and reagents inhibited the activity of CcBgl1A.
3.4 Substrate Specificity of CcBgl1A on pNP Glycosides and Disaccharides
3
4
5
6
7
The substrate specificity of β-glucosidase CcBgl1A from C. cellulans sp. 21 was investigated
using aryl-glycosides, disaccharides, and ginsenosides (Table 3). When using aryl-glycosides as
substrates, the hydrolysis order was pNPβGlc >> pNPβXyl > pNPαAraf ≈ pNPβMan > pNPβGal.
The pNP-α-glycosides we tested were not substrates for CcBgl1A. The GH1 enzyme CcBgl1A
had good activity on sophorose (glucose-β-(1→2)-glucose), medium activity on laminaribiose
(glucose-β-(1→3)-glucose), but very weak activity on cellobiose (glucose-β-(1→4)-glucose), and
no activity on gentiobiose (glucose-β-(1→6)-glucose). The linkage preference for disaccharides
was β-1, 2 > β-1, 3 >>> β-1, 4.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Based on substrate specificity, β-glucosidases are classified as aryl β-glucosidases, that act on
aryl-glucosides; true cellobiases, that catalyze the hydrolysis of cellobiose to release glucose; and
broad substrate specificity β-glucosidases, that act on a variety of substrates [25]. CcBgl1A
showed significant activity for both aryl-glucosides pNPβGlc and glucose disaccharides,
indicating that it belongs to broad substrate specificity β-glucosidase. Enzymes from GH family 1
are reported to be good candidates for true cellobiase in a variety of microorganisms [26-28].
However, our results showed that CcBgl1A is a good β-glucosidase, but not an efficient cellobiase,
as it showed much better activity on β-1, 2 linked glucose (sophorose) and β-1, 3 linked glucose
Page 12 of 30

1
2
(laminaribiose) than β-1, 4 (cellobiose).
3.5 Substrate Specificity of β-Glucosidase CcBgl1A on PPD-type Ginsenosides
When using a PPD-type ginsenosides mixture PPDGM as the substrate, the enzyme showed
considerable activity on protopanaxadiol-type ginsenosides Rb1, Rb2, Rc and Rd. TLC and HPLC
analysis revealed that in the presence of CcBgl1A, PPDGM including ginsenosides Rb1, Rb2, Rc
3
4
5
6
and Rd were converted into different metabolites, suggesting that CcBgl1A is a
7
ginsenoside-hydrolyzing enzyme (Fig. 4). To verify the bioconversion pathway and efficiency of
CcBgl1A, each individual ginsenoside was used as substrates for the biotransformation, and
HPLC analyses were carried out for products determination. As shown in Fig. 5, the products were
Gyp XVII for Rb1, compound O for Rb2, compound Mb for Rc, and F2 for Rd. The results
indicated that the enzyme only removed the outer glucose linked to the β-1,2 linkage of C-3
position in ginsenosides Rb1, Rb2, Rc, and Rd. The inner glucose linked to the C-3 position and
the disaccharides at the C-20 position with a β-1,6 linkage were not attacked by CcBgl1A (Fig. 6).
It is noteworthy that although CcBgl1A exhibited some activity on pNPαAraf, it showed no
hydrolysis activity on the α-Araf (1→6) Glc- in the C-20 position of ginsenoside Rc, therefore, no
metabolite Rd or F2 was observed even by prolonging the reaction time of CcBgl1A with
ginsenoside Rc to 12 h (data not shown). The reason for this might be that the spatial
conformation of ginsenoside molecule blocked the attack of enzyme to the α-Araf (1→6) Glc-
linkage.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
In previous reports, some bacterial GH1 ginsenoside-hydrolyzing β-glucosidases were
heterologous expressed in E. coli (Table S1). The hydrolysis behavior of CcBgl1A on
ginsenosides was similar to the GH1 β-glucosidase from S. alaskensis [9], although they shared
Page 13 of 30

1
2
low amino acid sequence similarity (33%). However, CcBgl1A and the S. alaskensis
β-glucosidase possessed different physicochemical properties, especially the molecular weight,
optimal temperature, and substrate specificity on pNP compounds. GH1 enzyme BglPm from P.
mucilaginosus also hydrolysed ginsenoside Rb1 to Gyp XVII, and hydrolysed ginsenoside Rd to
F2. However, Gyp XVII was not the final metabolite, which would be further converted to F2 [21].
GH1 enzyme BglSp from Sphingomonas sp. 2F2 also converted Rb1, Rb2, and Rc to Gyp XVII,
compound O, and compound Mb, respectively, which was further converted to F2 by hydrolyzing
the outer glycoside linkage at the C-20 position [14]. Other GH family 1 β-glucosidases from A.
chlorophenolicus, P. furiosus, S. solfataricus, and S. acidocaldarius hydrolysed the sugar moieties
at both the C-3 and C-20 positions in PPD-type ginsenosides Rb1, Rb2, Rc [11, 19, 20, 23].
3.6 Kinetic Parameters of CcBgl1A
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
The kinetic parameters of CcBgl1A on different substrates were determined, as shown in Table
4. The catalytic efficiencies (kcat/Km) for the substrates decreased in the following order:
pNPβGlc > Rb1 > Rb2 > Rc > sophorose > Rd > laminaribiose > pNPβXyl. CcBgl1A had a
comparable Km value for pNPβGlc (0.53 mM), lower than that of GH1 β-glucosidases from
bacteria, such as P. mucilaginosus (3.24 mM) [21], Sphingomonas sp. 2F2 (2.9 mM) [14],
Sphingomonas paucimobilis (1.3 mM) [29], and fungi including Neosartorya fischeri (2.8 mM)
[30], Piromyces sp. E2 (1 mM) [31] and Stereum hirsutum (2.5 mM) [32]. The catalytic
efficiencies for hydrolysis ginsenosides by CcBgl1A were higher than those of β-glycosidase from
S. acidocaldarius for ginsenosides Rd (4.8 mM-1min-1) and Rb1 (4.8 mM-1min-1) [19], and
β-glucosidase from P. mucilaginosus for Rb1 (1.23 mM-1s-1), Rd (1.78 mM-1s-1), and Gyp XVII
(1.17 mM-1s-1) [21].
Page 14 of 30

1
2
3.7 Scaled-up Production of Minor Ginsenosides by Crude Recombinant CcBgl1A
PPDGM was used as substrate for the mass production of minor ginsenosides Gyp XVII,
compound O, compound Mb and F2. After biotransformation, 1 g PPDGM yielding 292 ± 12 mg
Gyp XVII, 134 ± 5.8 mg compound O, 184 ± 7 mg compound Mb, and 62 ± 3.5 mg F2 with
chromatographic purities. The transformation was completely converted in 2 h, which is faster
than the microbial transformations (usually 24 h or even longer). To confirm the structure, each
product was identified by 13C NMR. The 13C NMR spectra of the obtained compounds (Table S2)
were consistent with Gyp XVII, compound O, compound Mb and F2 reported in the literatures.
Comparing to the other enzymes, this biotransformation would be potentially useful because of
following advantages: (i) The rare ginsenosides were produced using one recombinant enzyme,
which was easier to prepare than native enzymes from natural sources, and more stable than the
microbial biotransformation. (ii) The substrate PPDGM for biotransformation, which is relatively
abundant in total ginsenosides, did not need to be further separated for the minor ginsenosides
preparation. (iii) The recombinant CcBgl1A did not need to be purified, the crude recombinant
CcBgl1A was used for biotransformation. This high yields and simple procedure indicated that the
biotransformation has potential for industrial application. Further work are still needed for the
enlargement of the recombinant CcBgl1A transformation system to adapt it to the industrial
production.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
4. Conclusions
In conclusion, we describe here the cloning of a ginsenoside-hydrolyzing β-glucosidase gene
belonging to glycoside hydrolase family 1 from C. cellulans sp.21. This enzyme was successfully
Page 15 of 30

1
2
overexpressed in E. coli and the recombinant enzyme was found to be specific to β-1, 2-
glucosidic linkage in protopanaxadiol type ginsenosides, which transformed ginsenoside Rb1,
Rb2, Rc and Rd to minor ginsenosides Gyp XVII, compound O, compound Mb and F2. Amounts
of minor glycosides in the range of 62-292 mg were produced with the simple recombinant
enzyme biotransformation.
3
4
5
6
7
Acknowledgements
8
This work was supported by the National Natural Science Foundation of China (Nos. 31400299
and 31270479), and the Scientific and Technologic Foundation of Jilin Province (20110242).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
References
[1] D. Kiefer, T. Pantuso, Am. Fam. Physician 68 (2003) 1539-1542.
[2] T.K. Lee, R.M. Johnke, R.R. Allison, K.F. O’Brien, LJ.Jr. Dobbs, Mutagenesis 20 (2005)
237-243.
[3] C.L. Saw, A.Y. Yang, D.C. Cheng, S.S. Boyanapalli, Z.Y. Su, T.O. Khor, S. Gao, J. Wang, Z.H.
Jiang, A.N. Kong, Chem. Res. Toxicol. 25 (2012) 1574-1580.
[4] J.D. Park, D.K. Rhee, Y.H. Lee, Phytochem. Rev. 4 (2005) 159-175.
[5] K.W. Leung, A.S. Wong, Chin. Med. 5 (2010) 20-26.
[6] C.S. Yuan, J.A. Wu, J. Osinski, Am. J. Clin. Nutr. 75 (2002) 600-601.
[7] L.P. Christensen, Adv. Food Nutr. Res. 55 (2009) 1-99.
[8] Q.F. Xu, X.L. Fang, D.F. Chen, J. Ethnopharmacol. 84 (2003) 187-192.
[9] K.C. Shin, D.K. Oh, J. Biotechnol. 172 (2014) 30-37.
Page 16 of 30

1
2
[10] E.A. Bae, M.J. Han, E.J. Kim, D.H. Kim, Arch. Pharm. Res. 27 (2004) 61-67.
[11] K.H. Noh, J.W. Son, H.J. Kim, D.K. Oh, Biosci. Biotechnol. Biochem. 73 (2009) 316-321.
[12] J. Gao, W.J. Xu, F. Liang, M. Hao, X.Y. Liu, L. Fu, R.T. Jin, N. Wang, G.H. Tai, Y.F. Zhou,
Ann. Microbiol. 63 (2013) 139-149.
3
4
5
[13] D.S. An, C.H. Cui, H.G. Lee, L. Wang, S.C. Kim, S.T. Lee, F. Jin, H. Yu, Y.W. Chin, H.K. Lee,
W.T. Im, S.G. Kim, Appl. Environ. Microbiol. 76 (2010) 5827-5836.
[14] L. Wang, Q.M. Liu, B.H. Sung, D.S. An, H.G. Lee, S.G. Kim, S.C. Kim, S.T. Lee, W.T. Im, J.
Biotechnol. 156 (2011) 125-133.
6
7
8
9
[15] J.K. Kim, C.H. Cui, M.H. Yoon, S.C. Kim, W.T. Im, J. Biotechnol. 161 (2012) 294-301.
[16] C.H. Cui, S.C. Kim, W.T. Im, Appl. Microbiol. Biotechnol. 97 (2013) 649-659.
[17] C.S. Park, M.H. Yoo, K.H. Noh, D.K. Oh, Appl. Microbiol. Biotechnol. 87 (2010) 9-19.
[18] X.D. Yang, Y.Y. Yang, D.S. Ouyang, G.P. Yang, Fitoterapia 100C (2015) 208-220.
[19] K.H. Noh, D.K. Oh, Biol. Pharm. Bull. 32 (2009) 1830-1835.
[20] M.H. Yoo, S.J. Yeom, C.S. Park, K.W. Lee, D.K. Oh, Appl. Microbiol. Biotechnol. 89 (2011)
1019-1028.
10
11
12
13
14
15
16
17
18
19
20
21
22
[21] C.H. Cui, J.K. Kim, S.C. Kim, W.T. Im, PLoS One 9 (2014) e85727.
[22] J. Du, C.H. Cui, S.C. Park, J.K. Kim, H.S. Yu, F.X. Jin, C. Sun, S.C. Kim, W.T. Im, PLoS
One 9 (2014) e96914.
[23] M.K. Park, C.H. Cui, S.C. Park, S.K. Park, J.K. Kim, M.S. Jung, S.C. Jung, S.C. Kim, W.T.
Im, J. Microbiol. 52 (2014) 399-406.
[24] G.L. Miller, Anal. Chem. 31 (1959) 426-428.
[25] Y. Bhatia, S. Mishra, V.S. Bisaria, Crit. Rev. Biotechnol. 22 (2002) 375-407.
Page 17 of 30

1
2
[26] D.R. Love, R. Fisher, P.L. Bergquist, Mol. Genet. Genomics 213 (1988) 84-92.
[27] F. Grabnitz, M. Seiss, K.P. Rucknagel, W.L. Staudenbauer, Eur. J. Biochem. 200 (1991)
301-309.
3
4
[28] J. Gao, W. Wakarchuk, J. Bacteriol. 196 (2014) 4103-4110.
[29] A.R. Marques, P.M. Coutinho, P. Videira, A.M. Fialho, I. Sá-Correia, Biochem. J. 370 (2003)
793-804.
5
6
7
[30] P. Ramachandran, M.K. Tiwari, R.K. Singh, J.R. Haw, M. Jeya, J.K. Lee, Process Biochem.
47 (2012) 99-105.
8
9
[31] H.R. Harhangi, P.J. Steenbakkers, A. Akhmanova, MS.M. Jetten, C. van der Drift, H.J. Op
den Camp, Biochim. Biophys. Acta 1574 (2002) 293-303.
10
11
12
13
[32] N.P. Nguyen, K.M. Lee, K.M. Lee, I.W. Kim, Y.S. Kim, M. Jeya, J.K. Lee, Appl. Microbiol.
Biotechnol. 87 (2010) 2107-2116.
Page 18 of 30

1
2
Figure captions
Fig. 1 Electrophoresis analysis of CcBgl1A. (a) SDS-PAGE analysis on 10% resolving gel. Lane 1,
culture lysate of E. coli BL21-pET28a-ccbgl1a before IPTG induction; lane 2, culture lysate of E.
coli BL21-pET28a-ccbgl1a after IPTG induction; lane 3, CcBgl1A purified from Ni sepharose
fastflow column; M, molecular weight marker (PageRuler Prestained Protein Ladder, Thermo
Scientific); (b) native PAGE of CcBgl1A with Coomassie blue G-250 staining (1) and
β-glucosidase activity staining with MUG (2).
3
4
5
6
7
8
Fig. 2 Effect of pH on activity (solid symbols) and stability (hollow symbols) of CcBgl1A. The
optimal pH of CcBgl1A was determined by ranging pH from 2.0 to 11.0 using following buffers:
sodium acetate buffer, pH 2.0-6.0; Na2HPO4-NaH2PO4 buffer, pH 6.0-8.0; Glycine-NaOH buffer,
pH 8.0-11.0. The maximum activity obtained was defined as 100%. The pH stability of CcBgl1A
was determined by pre-incubating CcBgl1A in different pH for 24 h at 4°C and then determining
the percentage of residual activity under standard assay conditions. The activity of CcBgl1A
without pre-incubating was defined as 100%. Results are presented as means ± standard deviations
(n = 3).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Fig. 3 Effect of temperature on activity (a) and stability (b) of CcBgl1A using pNPβGlc as
substrate. The optimal temperature (a) was determined at different temperatures from 20 - 80°C.
The maximum activity obtained was defined as 100% activity. Thermal stability was determined
by incubating the enzyme for 1 h at different temperatures. The activity of the enzyme before
incubation was defined as 100%. Results are presented as means ± standard deviations (n = 3).
Fig. 4 Analysis of ginsenosidase activity of CcBgl1A on PPDGM by TLC (a) and HPLC (b). (a): 1,
biotransformation substrate PPDGM; 2, transformed products of PPDGM; S, standards; (b): 1,
Page 19 of 30

1
2
biotransformation substrate PPDGM; 2, transformed products of PPDGM; 3, standards.
Fig. 5 HPLC analysis of the hydrolysis products of single ginsenosides by CcBgl1A. Purified
CcBgl1A was incubated with single Rb1, Rb2, Rc and Rd, respectively. The reaction was
terminated by adding equal volume of n-butanol. The n-butanol fraction was subjected to HPLC
analysis as described in Methods section. Peak 1-8 represents ginsenosides Rb1, Rb2, Rc, Rd, Gyp
XVII, compound O, compound Mb and F2, respectively, the retention times of which were
determined to be 15.2 min, 22.1 min, 18.4 min, 26.1 min, 28.1 min, 30.9 min, 30.2 min and 35.2
min, respectively.
3
4
5
6
7
8
9
Fig. 6 Biotransformation pathways of ginsenosides Rb1, Rb2, Rc and Rd by recombinant
CcBgl1A.
10
11
12
13
∙ A new GH1 gene ccbgl1a from C. cellulans sp. 21 was cloned and expressed in E. coli.
∙ Enzyme CcBgl1A was specific to β-1,2- glycosidic bond at C-3 position of ginsenosides.
∙ CcBgl1A was applied for one-pot preparation of ginsenoside GXVII, CO, Mb and F2.
14
15
16
Table 1 Summary of purification of recombinant CcBgl1A
Purification step
Volume
(l)
Total Protein
(g)a
Total activity
(kU)b
Sp act
Recovery
(%)
Purification
(fold)
(U/mg)
Crude enzyme extract
1.2
0.60
0.23
10.0
7.2
16.7
31.5
100
72
1.0
1.9
Ni sepharose fastflow column
0.02
17
aProtein was quantified according to the Bradford method using bovine serum albumin (BSA) as standard.
18
19
bThe activity was reported as activity on pNPβGlc.
Page 20 of 30

1
2
Table 2 Effects of metal ions and chemical agents on the activity of CcBgl1A
Relative activity (%)a
Metal ions or
reagents
5 mM
50 mM
K+
Mg2+
Ca2+
Na+
105.5 ± 5.2
110.6 ± 2.8
108.4 ± 2.4
109.0 ± 2.9
91.1 ± 1.5
91.3 ± 2.1
15.5 ± 8.4
106.7 ± 1.4
87.5 ± 2.9
104.6 ± 2.1
121.2 ± 1.6
_
79.4 ± 2.3
38.2 ± 1.5
11.3 ± 0.5
96.0 ± 0.5
2.1 ± 0.7
Cu2+
Fe3+
b
_
Hg2+
Ba2+
Mn2+
DTT
EDTA
SDS
10.8 ± 0.8
7.8 ± 0.3
10.4 ± 0.2
91.9 ± 8.6
7.7 ± 0.4
_
3
4
aThe activity assayed in the absence of cations or reagents was taken as 100%. Results are presented as means ±
standard deviations (n = 3).
b-: Not detected.
5
6
Page 21 of 30

1
2
Table 3 Relative activity of CcBgl1A on different substratesa
Substrateb
pNPβGlc
pNPβXyl
pNPβMan
pNPβGal
pNPαAraf
pNPαGlc
pNPαGal
pNPαMan
Relative activity (%)c
100 ± 0.0
Substrate
sophorose
laminaribiose
cellobiose
gentiobiose
Rb1
Relative activity (%)
100 ± 0.0
51.5 ± 0.9
5.9 ± 0.1
30.8 ± 0.6
7.2 ± 0.1
1.4 ± 0.1
0
9.5 ± 0.5
100 ± 0.0
100 ± 0.0
87.5 ± 0.1
66.8 ± 0.2
0.7 ± 0.2
Rb2
0.6 ± 0.1
Rc
0.4 ± 0.1
Rd
3
4
5
6
7
8
a Reactions were performed with 5 mM substrate, pH 6.0, at 30°C for 10 min.
bAbsorption caused by released p-nitrophenol or reducing sugars was measured at 405 or 520 nm. The hydrolysis
product of ginsenosides was measured by HPLC. The relative activity on pNPβGlc (31.5 μmol/min/mg), sophorose
(19.3 μmol/min/mg) and ginsenoside Rb1 (7.5 μmol/min/mg) was taken as 100%, respectively.
c The data are reported as means ± standard errors from the mean for three independent experiments.
Page 22 of 30

1
2
Table 4 Kinetic parameters of CcBgl1A on different compounds
Substratea
Km
Vmax
kcat
kcat/Km
(mM)b
(μmol/min/mg)
39.71 ± 0.37
0.34 ± 0.05
26.57 ± 0.84
38.12 ± 2.72
11.10 ± 0.42
6.10 ± 0.05
6.93 ± 0.13
13.91 ± 0.11
(s-1)
(s-1 mM-1)
0.53 ± 0.04
1.34 ± 0.03
1.89 ± 0.15
4.89 ± 0.69
0.26 ± 0.03
0.15 ± 0.01
0.22 ± 0.01
1.24 ± 0.03
30.08 ± 0.28
0.26 ± 0.04
20.37 ± 0.64
29.23 ± 2.09
8.52 ± 0.33
4.67 ± 0.04
5.31 ± 0.10
10.64 ± 0.08
57.16 ± 3.37
0.19 ± 0.02
10.84 ± 0.49
6.04 ± 0.43
32.77 ± 1.29
31.44 ± 1.63
24.16 ± 0.64
8.64 ± 0.11
pNPβGlc
pNPβXyl
Sophorose
Laminaribiose
Ginsenoside Rb1
Ginsenoside Rb2
Ginsenoside Rc
Ginsenoside Rd
3
4
aThe hydrolysis rates for CcBgl1A with pNPβMan and pNPβGal were too low to calculate the kinetics parameters,
even when increasing the enzyme concentration to 1 mg/ml in the reaction mixture.
5
6
7
bThe data are reported as means ± standard errors from the mean for three independent experiments.
Page 23 of 30

Page 24 of 30

Page 25 of 30

Page 26 of 30

Page 27 of 30

Page 28 of 30

Page 29 of 30