Highly specific sophorose β-glucosidase from Sphingomonas elodea ATCC 31461 for the efficient conversion of stevioside to rubusoside

Article abstract of DOI:10.1016/j.foodchem.2019.05.164

Enzyme specificity and particularity is needed not only in enzymatic separation methods, but also in enzymatic determination methods for plant compound extraction. Stevioside, rubusoside, and rebaudioside A are natural sweet compounds from plants. These compounds have the same skeleton and only contain different side-chain glucosyl groups, making them difficult to separate. However, enzymes that target diterpenoid compounds and show specific activity for side-chain glucosyl groups are rare. Herein, we report the identification and characterization of an enzyme that can target both diterpenoid compounds and sophorose, namely, β-glucosidase SPBGL1 from Sphingomonas elodea ATCC 31461. SPBGL1 displayed high specificity toward sophorose, and activity toward stevioside, but not rebaudioside A. The stevioside conversion rate was 98%. SPBGL1 also operated at high substrate concentrations, such as in 50% crude steviol glycoside extract. Glucose liberated from stevioside was easy to quantify using the glucose oxidase method, allowing the stevioside content to be determined.

Full text of DOI:10.1016/j.foodchem.2019.05.164

Accepted Manuscript
Highly specific sophorose β-glucosidase from Sphingomonas elodea ATCC
31461 for the efficient conversion of stevioside to rubusoside
Qing Lan, Tingting Tang, Yu Yin, XiaoYi Qu, Zilong Wang, Hao Pang, Ribo
Huang, Liqin Du
PII:
S0308-8146(19)30977-X
DOI:
https://doi.org/10.1016/j.foodchem.2019.05.164
FOCH 24890
Reference:
Food Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
9 February 2019
17 May 2019
23 May 2019
Please cite this article as: Lan, Q., Tang, T., Yin, Y., Qu, X., Wang, Z., Pang, H., Huang, R., Du, L., Highly specific
sophorose β-glucosidase from Sphingomonas elodea ATCC 31461 for the efficient conversion of stevioside to
rubusoside, Food Chemistry (2019), doi: https://doi.org/10.1016/j.foodchem.2019.05.164
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
3
Highly specific sophorose -glucosidase from Sphingomonas
elodea ATCC 31461 for the efficient conversion of stevioside to
rubusoside
4
5
Qing Lana,1, Tingting Tanga,1, Yu Yina, XiaoYi Qua, Zilong Wanga, Hao Pangb,*, Ribo Huanga, Liqin Dua,*
6
7
aState Key Laboratory for Conservation and Utilization of Subtropical Agro-bioresources, Guangxi Research
Center for Microbial and Enzymatic Technology, College of Life Science and Technology, Guangxi University,
Daxue Road No. 100, Nanning, Guangxi 530005, China
8
9
10
11
12
bNational Engineering Research Center for Non-Food Biorefinery, State Key Laboratory of Non-Food Biomass
and Enzyme Technology, Guangxi Key Laboratory of Bio-refinery, Guangxi Academy of Sciences, Daling Road
No. 98, Nanning, Guangxi 530007, China
13
14
15
16
17
1These authors contributed equally to this work.
*Corresponding authors:
Email: haopang@gxas.cn (Hao Pang), duliqin@gxu.edu.cn (Liqin Du)
Phone: +86-771-2503987(Hao Pang), +86-771-3235706(Liqin Du)
18
19
20
21
22
23
Abstract
Enzyme specificity and particularity is needed not only in enzymatic separation
methods, but also in enzymatic determination methods for plant compound extraction.
Stevioside, rubusoside, and rebaudioside A are natural sweet compounds from plants.
These compounds have the same skeleton and only contain different side-chain
1

24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
glucosyl groups, making them difficult to separate. However, enzymes that target
diterpenoid compounds and show specific activity for side-chain glucosyl groups are
rare. Herein, we report the identification and characterization of an enzyme that can
target both diterpenoid compounds and sophorose, namely, -glucosidase SPBGL1
from Sphingomonas elodea ATCC 31461. SPBGL1 displayed high specificity toward
sophorose, and activity toward stevioside, but not rebaudioside A. The stevioside
conversion rate was 98%. SPBGL1 also operated at high substrate concentrations,
such as in 50% crude steviol glycoside extract. Glucose liberated from stevioside was
easy to quantify using the glucose oxidase method, allowing the stevioside content to
be determined.
Keywords
-glucosidase; stevioside; plant content measurement; enzymatic determination
method
1. Introduction
Steviol glycosides are natural sweeteners isolated from the leaves of Stevia
rebaudiana. Steviol glycosides are a class of tetracyclic diterpenoids consisting of at
least nine sweet ingredients, including stevioside, rebaudiosides A–D and F,
rubusoside, dulcoside, and steviolbioside (Supplementary Fig. S1, Table S1)
(Carakostas, Curry, Boilea, & Brusick, 2008). These substances differ in the type and
number of glycosyl groups linked to positions C-13 and C-19 (Lemus-Mondaca,
2

46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
Vega-Gálvez, Zura-Bravo, & Ah-Hen, 2012). Steviol glycosides show various
properties and different sweetness levels (Lemus-Mondaca, et al., 2012; Xu, Li,
Wang, Yang, & Ning, 2009), with higher sweetness (200–350-fold) and fewer
calories (300-fold) compared with sucrose (Carakostas, et al., 2008; Singhania, Patel,
Sukumaran, Larroche, & Pandey, 2013). Since 2008, steviol glycoside has been
successively approved by the Food and Agriculture Organization of the United
Nations, the World Health Organization Joint Expert Committee on Food Additives,
the U. S. Food and Drug Administration, and other official organizations as a food
and drug additive (Tada, et al., 2013). Therefore, the use of steviol glycoside as a
natural sugar substitute is increasing.
Among steviol glycosides extracted from S. rebaudiana, rebaudioside A and
stevioside have the highest contents. Rebaudioside A produces a clean sweet taste
with no significant undesirable taste characteristics, while stevioside has a slightly
bitter taste and aftertaste (Prakash, DuBois, Clos, Wilkens, & Fosdick, 2008).
Therefore, the separation and purification of steviol glycoside components from the
extraction mixture is required to produce high-quality products (Ba, Zhang, Yao, Ma,
& Wang, 2014; Li, Chen, & Di, 2012).
Such separation processes and methods have been developed, as well as methods
for the rapid detection and quantitation of components from the mixture (Formigoni,
et al., 2018; Logue, Dowey, Strain, Verhagen, McClean, & Gallagher, 2017; Pavlíček
& Tůma, 2017). Common methods involve measuring the steviol glycoside content in
stevia samples or extraction mixtures using standard chemical analyses, such as high-
3

68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
performance liquid chromatography (HPLC) (Aranda-González, Moguel-Ordoñez, &
Betancur-Ancona, 2015; Lorenzo, Serrano-Díaz, Plaza, Quintanilla, & Alonso, 2014).
However, these methods are time consuming and expensive. Earlier, Mizukami et al.
reported an enzymatic method for the quantitative analysis of stevioside in stevia
extract (Mizukami, Shiiba, & Ohashi, 1982). As the stevioside molecule contains
three molecules of glucose, glucose liberated from stevioside can be used to quantify
the stevioside content using crude hesperidinase. Therefore, this method combined the
use of a stevioside-specific enzyme and a standard glucose detection method.
Furthermore, Udompaisarn et al. showed that recombinant -glucosidase from
Bacteroides thetaiotaomicron HB-13 can selectively hydrolyze stevioside to produce
rubusoside and glucose (Udompaisarn, Arthan, & Somana, 2017). Glucose liberation
during enzymatic hydrolysis of stevioside can be used to quantify the stevioside
content by coupling with a glucose oxidase method. Using this glucosidase, which
liberates only one glucose molecule, stevioside determination is more accurate and
can be completed in a shorter time. Compared with HPLC-based methods, enzymatic
methods are cheaper and faster.
Some enzymes for specific reactions with steviol glycosides have been
identified. For example, lactase from Thermus thermophilus and -galactosidase from
Aspergillus sp. can produce rubusoside from stevioside (Nguyen, et al., 2014; Wan,
Guan-junTao, Kim, & Xia, 2012). -Glucosidase from Streptomyces sp. GXT6 can
also convert stevioside to rubusoside (Wang, et al., 2015). -Glucosidase from
Penicillium decumbens can convert stevioside to steviol (Ko, et al., 2013), while a -
4

90
91
galactosidase from Kluyveromyces lactis was found to specifically catalyze hydrolysis
of the glycosyl ester linkage in stevioside to yield steviolbioside (Chen, Ding, Sui,
Xia, Wan, & Lu, 2016).
92
93
In the present study, we show that a -glucosidase from Sphingomonas elodea
ATCC 31461 has high activity in the conversion of stevioside to rubusoside with no
activity toward rebaudioside. Therefore, this enzyme can be used for the hydrolysis of
stevioside to release glucose, allowing fast and sensitive stevioside determination.
94
95
96
97
98
2. Materials and methods
99
2.1 Strains and materials
100
101
102
103
104
105
106
107
108
109
110
111
S. elodea ATCC 31461 was purchased from the American Type Culture
Collection. Vector pSE380 was purchased from Invitrogen Inc. (San Diego, CA). All
restriction endonucleases, ligases, and DNA polymerases were from Takara (Dalian,
China). p-Nitrophenyl--D-glucopyranoside (pNPG), 4-nitrophenyl--D-cellobioside
(pNPC), 4-nitrophenyl--D-galactopyranoside (pNPGal), 4-methylumbelliferyl--D-
glucuronide (MU-Glc), 4-nitrophenyl-2-acetylamino-2-deoxy--D-glucopyranoside
(pNPNAG), o-nitrobenzene--D-galactopyranoside (oNPG), 4-nitrophenyl--D-
xylopyranoside (pNP-Xyl), 4-nitrophenyl--D-rhamnoside (pNPR), 4-nitrophenyl--
L-arabinoside (pNPA), and p-nitrophenyl--D-glucopyranoside (α-pNPG) were
purchased from Sigma (St. Louis, MO). Stevioside (98%, HPLC), rubusoside,
rebaudioside, and steviol were purchased from Sichuan Weikeqi Biological
Technology Co., Ltd. (Chengdu, China). Steviolbioside was purchased from Chengdu
5

112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
Herbpurify Co., Ltd. (Chengdu, China). Crude steviol glycoside was obtained from
Shandong Huaxian Stevia Co., Ltd. (Shandong, China).
2.2 Expression and purification of recombinant SPBGL1
First, the protein was analyzed using the Simple Modular Architecture Research
Tool on the SMART server (http://smart.embl-heidelberg.de/). Signal peptides were
predicted using the SignalP 4.1 server (http://www.cbs.dtu.dk/services/SignalP/).
The gene spbgl1 was amplified from genomic DNA of S. elodea ATCC 31461
using primers spbgl1-F, 5′-
ATTACCATGGATATGCTGCTCCAGGCGGGTGCGCGG-3′
(containing a NcoI site at the 5′- end) and spbgl1-R, 5′-
ATTAAGCTTTCAGTGGTGGTGATGATGATGGGCCAGTTCGAAGCTGCCGG
ACAC-3′ (containing a HindIII site and a His6 tag at the 5′- end). The PCR fragment
was digested with NcoI and HindIII and ligated into pSE380. The recombinant
plasmid was transformed into Escherichia coli XL1-Blue. Transformants were grown
in Luria–Bertani (LB) medium until the OD600 reached 0.4–0.6. Then, 0.5 mM IPTG
was added (final concentration) and incubation was continued at 20°C for 20 h to
induce protein expression. The protein was purified by nickel–nitrilotriacetic acid (Ni-
NTA) chromatography (Qiagen, Hilden, Germany) according to manufacturer
instructions. The purified protein was desalted into 100 mM NaH2PO4–Na2HPO4
buffer (pH 7.0) using a PD MiniTrap G-25 column (GE Healthcare, Freiburg,
Germany). The molecular mass of the recombinant protein was determined by SDS-
6

134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
PAGE. The purified protein concentration was measured using the Bradford method.
The protein sequence accession number of SPBGL1 is WP_029622673.1 at
Genbank.
2.3 Enzymatic activity and kinetic parameters using pNPG as substrate
The activity of -glucosidase SPBGL1 was assessed using pNPG as substrate. A
standard reaction comprised McIlvaine buffer (170 L, 0.2 M Na2HPO4–0.1 M citric
acid, pH 5.5), 20 mM pNPG (20 L), and appropriately diluted enzyme (10 L) at
45°C for 20 min. The reaction was terminated by adding 2 M Na2CO3 (50 L). The
amount of p-nitrophenol (pNP) released was measured in a 96-well plate at OD405 in a
Sunrise microplate reader (Tecan Group Ltd., Männedorf, Switzerland). A reaction
system without any enzyme was used as a blank control. All reactions were
determined simultaneously in three parallel experiments. One enzyme unit was
defined as the amount of enzyme that produced 1 mol of pNP per min under the
above conditions.
The effect of different pH levels on the enzyme activity (pH 4.5–8.0 at intervals
of 0.5) was determined at 37°C in McIlvaine buffer. The optimum reaction
temperature for SPBGL1 was assessed at the optimum pH over the temperature range
of 25–55°C (5°C intervals). Under the optimum reaction conditions, the enzyme
activity was determined at a series of substrate concentrations (0.35–10 mM) by
nonlinear regression analysis and double reciprocal mapping using GraphPad Prism 5
software (GraphPad Software, San Diego, CA). The SPBGL1 activity at glucose
7

156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
concentrations of 0, 20, and 40 mM was determined using different concentrations of
pNPG (0.35–10 mM) under the optimum reaction conditions. The Ki value of glucose
(inhibitor constant) was measured by simultaneous nonlinear regression and
calculated using GraphPad Prism 5 software.
2.4 Substrate specificity of recombinant SPBGL1
The substrate specificity of SPBGL1 was determined by incubating purified
enzyme with 2 mM -pNPG, pNPNAG, oNPG, pNPGal, pNP-Xyl, pNPC, pNPA,
pNPR, MU-Glc, and 1% (w/v) cellobiose, gentiobiose, lactose, sophorose, sucrose,
trehalose, maltose, and isomaltose, respectively, in McIlvaine buffer (pH 5.5) at 45°C
for 20 min. The amounts of pNP or oNP released were detected using the methods
referred to in section 2.3. MU-Glc hydrolysis was visualized under ultraviolet light
due to the release of fluorescent methyl umbelliferone. Glucose release was
determined using a Glucose Assay Kit (Applygen Technology Inc., Beijing, China).
The rates of hydrolysis of different substrates were calculated as relative values
compared with the rate of hydrolysis of pNPG.
2.5 Hydrolysis activity toward steviol glycosides
To determine the hydrolysis activity of SPBGL1 toward stevioside substrates,
the reaction system consisted of 1% (w/v) substrate (stevioside, rubusoside,
steviolbioside, and rebaudioside, respectively), buffer (pH 5.5) and appropriately
purified enzyme (total system was 50 L), incubated at 45°C for 20 min. The reaction
8

178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
was terminated by boiling for 5 min. The products were detected using the Glucose
Assay Kit or HPLC. For HPLC, a G1311C 1260 Quat Pump VL, G1329B 1260 ALS
automatic sampler, G1314F 1260 VWD detector, and CO-1000 column oven from
Agilent Technologies Inc. (Santa Clara, CA, United States) were used. The stationary
phase was a C18 column (Alltima, 250×4.6 mm, 5 m). The column temperature was
40 °C, with a flow rate of 1 mL/min, and detection at 210 nm. The mobile phase
consisted of 10 mM NaH2PO4 solution in water (adjusted to pH 2.6 with phosphoric
acid) and acetonitrile. The gradient elution program was as follows: 32% acetonitrile
(1–10 min), 32–80% acetonitrile (10–20 min), 80% acetonitrile (20–21 min), 80–32%
acetonitrile (21–28 min), and 32% acetonitrile (28–30 min).
2.6 Enzymatic properties with stevioside as substrate
The effect of different pH levels on the hydrolysis of stevioside by SPBGL1 was
determined at 37°C for 20 min. The pH range was 4.0 to 8.0 (intervals of 0.5). The
effect of temperature on the hydrolysis of stevioside by SPBGL1 was determined at
the optimum reaction pH (pH 5.5) for 20 min at 25–60°C (5°C intervals). Under the
optimal reaction conditions, the enzyme activity at a series of substrate concentrations
(0.5–25 mM) was determined by nonlinear regression analysis and double reciprocal
mapping using GraphPad Prism 5 software. A Glucose Assay Kit was used to
measure glucose released by the reaction. Stevioside (1% and 24%, w/v) was
incubated with different concentrations of purified enzyme in McIlvaine buffer (pH
6.5) at 45°C for 6 h, and the products were determined by HPLC. The optimum
9

200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
reaction time with 1% (w/v) stevioside was determined using the optimum
concentration of SPBGL1 in McIlvaine buffer (pH 6.5) incubated at 45°C for 0, 5, 10,
20, 40, 60, 120, 240, and 360 min. These products were also determined by HPLC.
2.7 Hydrolysis of crude steviol glycoside and NMR product analysis
The optimum enzyme amounts and reaction time for the hydrolysis of crude 50%
(w/v) steviol glycoside were determined using the method described above for
SPBGL1. For control enzyme BGL1, the reaction conditions were 50% stevioside (50
L, 0.1 M Na2HPO4–citric acid solution, pH 6.5) at 45°C for 6 h.
Rubusoside and rebaudioside A were isolated and purified from the hydrolysates
of crude steviol glycoside by preparative HPLC (Waters Corporation, Milford, MA).
A Waters 2535 quaternary gradient module, a 2707 automatic sampler, and a 2489
UV/visible detector were used for preparative HPLC. The stationary phase was a C18
column (SunFire, 25010 mm, 5 m). The flow rate was 2 mL/min and detection was
at 210 nm. The mobile phase consisted of acetonitrile and H2O. The gradient elution
program was as follows: 32% acetonitrile (0–13 min), 32–80% acetonitrile (13–25
min), 80% acetonitrile (25–30 min), 80–32% acetonitrile (30–40 min), and 32%
acetonitrile (40–55 min).
Purified rubusoside and rebaudioside A were lyophilized for NMR determination
as follows: Rubusoside (10 mg) dissolved in deuterated water (500 L) and
rebaudioside A (10 mg) dissolved in deuterated pyridine (500 L) were placed in
separate 5-mm tubes for NMR structural elucidation. Spectra were recorded using a
10

222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
Varian VNMRS 600 MHz spectrometer (Varian, Palo Alto, CA) operating at 600
1
MHz for H and 150 MHz for 13C NMR at 25°C for rubusoside and 40°C for
rebaudioside A.
The stevioside conversion rate and rubusoside yield were calculated using the
following equations:
Stevioside conversion rate = (initial stevioside amount in reaction  amount of
stevioside remaining at the end)/initial stevioside amount in reaction × 100.
Rubusoside yield = amount of rubusoside produced/theoretical maximum
amount of rubusoside produced × 100.
3. Results and discussion
3.1 Enzymatic properties of SPBGL1
A -glucosidase of glycoside hydrolase family 3 was cloned from S. elodea
ATCC 31461. The open reading frame, designated spbgl1, was 2289-bp long, which
encoded 763 amino acids. The purified protein was detected by SDS-PAGE, and the
molecular weight of SPBGL1 was about 85 kDa, which was consistent with the
theoretical molecular weight (Supplementary Fig. S2, lane 3).
The enzymatic properties of SPBGL1 were initially detected using pNPG as
substrate. The optimum reaction pH of SPBGL1 was 5.5. The enzyme showed high
activity (>80%) in the range pH 5.0–6.5 (Supplementary Fig. S3a). The optimum
reaction temperature of SPBGL1 was 45°C. Between 35 and 45°C, the activity of
SPBGL1 remained over 50% (Supplementary Fig. S3b). The Km, vmax, and Ki
11

244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
(glucose) values were 1.364±0.11 mM, 11.37±0.28 mol/min/mg and 16.74±1.29
mM, respectively (Supplementary Figs. S3 and S4).
3.2 Substrate specificity of recombinant SPBGL1
The specificity of SPBGL1 toward various substrates was determined
(Supplementary Table S2). In addition to the hydrolysis of -pNPG, SPBGL1
hydrolyzed MU-Glc and had weak hydrolytic activity toward cellobiose, gentiobiose,
salicin, and amygdalin. Notably, SPBGL1 had very high activity toward sophorose,
about 12.8 times that toward pNPG.
SPBGL1 also hydrolyzed daidzin and genistin, showing higher hydrolysis
activity toward daidzin than that toward genistin. The conversion rates of daidzin and
genistin were 76.6% in 15 min and 22.4% in 3 h, respectively (Fig. S5), meaning that
SPBGL1, even as a glucosidase, targeted large substrates, such as diterpenoid
compounds.
Compared with other -glucosidases (SSGase, BGL1, and SPGase) reported to
hydrolyze stevioside, SPBGL1 showed higher activity toward cellobiose, gentiobiose,
salicin, and amygdalin (Ko, et al., 2012; Ko, et al., 2013; Wang, et al., 2015).
SPBGL1 showed 12.8 times the activity toward sophorose that that toward pNPG,
while BGL1 showed the same activity toward sophorose and pNPG. This indicated
that sophorose was the optimum substrate for SPBGL1. As SPBGL1 showed the
highest activity toward sophorose and also targeted compounds with tetracyclic
diterpenoid skeletons, SPBGL1 might be very useful for the hydrolysis of stevioside
12

266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
at the C13 sophorose residue.
3.3 Hydrolysis of stevioside by SPBGL1
The hydrolyses of stevioside, rubusoside, steviolbioside, and rebaudioside A by
SPBGL1 were determined at 45°C and pH 5.5 (Fig. 1). SPBGL1 hydrolyzed
stevioside to produce rubusoside (Fig. 1c) and converted steviolbioside to
steviolmonoside (Fig. 1e). However, SPBGL1 could not hydrolyze rebaudioside A or
rubusoside. Therefore, SPBGL1 exhibited highly specific hydrolytic activity toward
the C-13 sophorosyl group of stevioside.
As SPBGL1 showed the ability to hydrolyze stevioside to produce rubusoside,
this activity was studied further. The optimum conditions for stevioside as substrate
were different from those for pNPG as substrate. For stevioside, the optimum pH was
6.5, with SPBGL1 retaining >80% activity between pH 6.0 and 7.0 (Fig. S6a), while
the optimum reaction temperature was 45°C, with SPBGL1 retaining >80% activity
between 35 and 45°C (Fig. S6b). Using stevioside as substrate, the values of Km, vmax
kcat, and kcat/Km for SPBGL1 were 3.097±0.1939 mM, 85.08±1.854 mol/min/mg
protein, 121.3 s1, and 39.1 s1mM1, respectively (Fig. S7a).
,
When the amount of enzyme was 20 g/mL at 6 h, the stevioside conversion rate
was 95%. When the amount of enzyme was increased to 30 g/mL, the stevioside
conversion rate reached 98% for 1 h, and the final stevioside conversion rate was
98.2% when the reaction time was extended to 6 h (Fig. 2).
13

288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
3.4 Activity toward crude steviol glycoside extract
We also determined the conversion rate when the stevioside concentration was
increased to 24%. The conversion rate increased to 98.6%, and the rubusoside yield
was 99% when using 600 g/mL of SPBGL1 for 6 h (Fig. 3a). Therefore, SPBGL1
retained high activity even at a high substrate concentration.
In further experiments, we used crude steviol glycoside extract, which contained
about 50% stevioside, as substrate instead of purified stevioside. Most stevioside was
converted to rubusoside at 4 h. The highest conversion rate of stevioside was >98%
and the highest rubusoside yield was >99% in the reaction with 600 g/mL of
SPBGL1 for 6 h (Fig. 3b). The ability of enzyme BGL to hydrolyze the crude steviol
glycoside extract was compared with that of SPBGL1, which resulted in stevioside
still being present, even when the amount of enzyme was increased to 800 g/mL (Fig
S8e).
The products from the reaction of steviol glycoside were separated by
preparative HPLC, and the separated rebaudioside A and rubusoside were lyophilized
1
13
for NMR analysis. H and C spectral analyses were performed simultaneously with
purchased standards of rebaudioside A and rubusoside, with the results shown in
Supplementary Table S3. By comparison with the reported NMR data for rubusoside
(Wang, et al., 2015) and rebaudioside A (Steinmetz & Lin, 2009), we concluded that
the products obtained by preparative HPLC in this study were indeed rubusoside and
rebaudioside A.
Only a few enzymes have been identified to hydrolyze stevioside substrates,
14

310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
including lactase (Nguyen, et al., 2014), -galactosidase (Chen, et al., 2016; Chen,
Xia, Wan, Wang, & Liu, 2014; Wan, et al., 2012; Wan & Xia, 2015), and -
glucosidase (Katsuyuki, Hirofumi, Tsunneya, Taro, & Sumio, 2000; Ko, et al., 2012;
Ko, et al., 2013; Nakano, Okamoto, Yatake, Kiso, & Kitahata, 1998; Wang, et al.,
2015; Xu, et al., 2009). However, not all produce rubusoside. The stevioside
conversion rate and rubusoside yield were both higher for SPBGL1 than for other
reported enzymes (Table 1). In particular, we compared the Km, kcat, and kcat/Km
values of SPBGL1 with those of three reported β-glucosidases that can hydrolyze
stevioside to produce either rubusoside or steviol (Table 2). Among them, SPBGL1
showed the lowest Km and highest kcat/Km values. With a high concentration of ST
(24% w/v), SPBGL1 (600 g/mL) still maintained a >98% stevioside conversion rate
and 99% rubusoside yield. A small amount of SPBGL1 (30 g/mL) also rapidly (1 h),
and almost completely, converted stevioside to rubusoside. Furthermore, SPBGL1
could not hydrolyze rebaudioside A, with only specific hydrolysis of the glycosidic
bond at the C-13 position of stevioside observed. Therefore, SPBGL1 would be
suitable for use in an enzymatic stevioside determination method, and may be
applicable to the high-throughput analysis of stevioside content. Furthermore, as this
enzyme works on crude steviol glycoside extracts, it might be used in method
development for the direct measurement of stevioside content in stevia crude extracts.
Acknowledgments
331
332
We thank the Natural Science Foundation of China (31360369), the Natural Science
Foundation of Guangxi (2014GXNSFAA118078), and the Guangxi Chairman
15

333
334
Technology Fund (17290-03).
335
336
References
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
Aranda‐González, I., Moguel‐Ordoñez, Y., & Betancur‐Ancona, D. (2015). Validation of HPLC-UV
method for determination of minor glycosides contained in Stevia rebaudiana Bertoni leaves.
Biomedical Chromatography, 29(5), 733-738.
Ba, J., Zhang, N., Yao, L., Ma, N., & Wang, C. (2014). Separation of Rebaudiana A from Steviol glycoside
using a polymeric adsorbent with multi-hydrogen bonding in a non-aqueous system. Journal
of Chromatography B, 971, 141-149.
Carakostas, M. C., Curry, L. L., Boilea, A. C., & Brusick, D. (2008). Overview: The history, technical
function and safety of rebaudioside A, a naturally occurring steviol glycoside, for use in food
and beverages. Food and Chemical Toxicology, 46(7), S1-S10.
Chen, J.-m., Ding, L., Sui, X.-c., Xia, Y.-m., Wan, H.-d., & Lu, T. (2016). Production of a bioactive
sweetener steviolbioside via specific hydrolyzing ester linkage of stevioside with a β -
galactosidase. Food Chemistry, 196, 155-160.
Chen, J.-M., Xia, Y.-m., Wan, H.-d., Wang, H.-j., & Liu, X. (2014). A complete specific cleavage of
glucosyl and ester linkages of stevioside for preparing steviol with a β-galactosidase from
Sulfolobus solfataricus. Journal of Molecular Catalysis B: Enzymatic, 105, 126-131.
Formigoni, M., Milani, P. G., Avíncola, A. S., Santos, V. J., Benossi, L., Dacome, A. S., Pilau, E. J., &
Costa, S. C. (2018). Pretreatment with ethanol as an alternative to improve steviol glycosides
extraction and purification from a new variety of stevia. Food Chemistry, 241, 452-459.
Katsuyuki, O., Hirofumi, N., Tsunneya, Y., Taro, K., & Sumio, K. (2000). Purification and some
properties of a β-glucosidase from Flavobacterium johnsonae. Bioscience, Biotechnology, and
Biochemistry, 64(2), 333-340.
Ko, J.-A., Kim, Y.-M., Ryu, Y. B., Jeong, H. J., Park, T.-S., Park, S.-J., Wee, Y.-J., Kim, J.-S., Kim, D., & Lee,
W. S. (2012). Mass production of rubusoside using a novel stevioside-specific beta-
glucosidase from Aspergillus aculeatus. Journal of Agricultural and Food Chemistry, 60(24),
6210-6216.
Ko, J.-A., Ryu, Y. B., Kwon, H.-J., Jeong, H. J., Park, S.-J., Kim, C. Y., Wee, Y.-J., Kim, D., Lee, W. S., & Kim,
Y.-M. (2013). Characterization of a novel steviol-producing β-glucosidase from Penicillium
decumbens and optimal production of the steviol. Applied Microbiology and Biotechnology,
97(18), 8151-8161.
Lemus-Mondaca, R., Vega-Gálvez, A., Zura-Bravo, L., & Ah-Hen, K. (2012). Stevia rebaudiana Bertoni,
source of a high-potency natural sweetener: A comprehensive review on the biochemical,
nutritional and functional aspects. Food Chemistry, 132(3), 1121-1132.
Li, J., Chen, Z., & Di, D. (2012). Preparative separation and purification of Rebaudioside A from Stevia
rebaudiana Bertoni crude extracts by mixed bed of macroporous adsorption resins. Food
Chemistry, 132(1), 268-276.
Logue, C., Dowey, L. R. C., Strain, J. J., Verhagen, H., McClean, S., & Gallagher, A. M. (2017).
Application of Liquid Chromatography–Tandem Mass Spectrometry To Determine Urinary
16

374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
Concentrations of Five Commonly Used Low-Calorie Sweeteners: A Novel Biomarker
Approach for Assessing Recent Intakes? Journal of Agricultural and Food Chemistry, 65(22),
4516-4525.
Lorenzo, C. n., Serrano-Díaz, J. s., Plaza, M., Quintanilla, C., & Alonso, G. L. (2014). Fast methodology
of analysing major steviol glycosides from Stevia rebaudiana leaves. Food Chemistry, 157,
518-523.
Mizukami, H., Shiiba, K., & Ohashi, H. (1982). Enzymatic determination of stevioside in Stevia
rebaudiana. Phytochemistry, 21(8), 1927-1930.
Nakano, H., Okamoto, K., Yatake, T., Kiso, T., & Kitahata, S. (1998). Purification and characterization of
a novel β-glucosidase from Clavibacter michiganense that hydrolyzes glucosyl ester linkage in
steviol glycosides. Journal of Fermentation and Bioengineering, 85(2), 162-168.
Nguyen, T. T. H., Jung, S.-J., Kang, H.-K., Kim, Y.-M., Moon, Y.-H., Kim, M., & Kim, D. (2014). Production
of rubusoside from stevioside by using a thermostable lactase from Thermus thermophilus
and solubility enhancement of liquiritin and teniposide. Enzyme and Microbial Technology,
64-65, 38-43.
Pavlíček, V., & Tůma, P. (2017). The use of capillary electrophoresis with contactless conductivity
detection for sensitive determination of stevioside and rebaudioside A in foods and
beverages. Food Chemistry, 219, 193-198.
Prakash, I., DuBois, G. E., Clos, J. F., Wilkens, K. L., & Fosdick, L. E. (2008). Development of rebiana, a
natural, non-caloric sweetener. Food and Chemical Toxicology, 46 Suppl 7, S75-82.
Singhania, R. R., Patel, A. K., Sukumaran, R. K., Larroche, C., & Pandey, A. (2013). Role and significance
of beta-glucosidases in the hydrolysis of cellulose for bioethanol production. Bioresource
technology, 127, 500-507.
Steinmetz, W. E., & Lin, A. (2009). NMR studies of the conformation of the natural sweetener
rebaudioside A. Carbohydrate research, 344(18), 2533-2538.
Tada, A., Takahashi, K., Ishizuki, K., Sugimoto, N., Suematsu, T., Arifuku, K., Tahara, M., Akiyama, T.,
Ito, Y., Yamazaki, T., Akiyama, H., & Kawamura, Y. (2013). Absolute Quantitation of Stevioside
and Rebaudioside A in Commercial Standards by Quantitative NMR. Chemical and
Pharmaceutical Bulletin, 61(1), 33-38.
Udompaisarn, S., Arthan, D., & Somana, J. (2017). Development and Validation of an Enzymatic
Method To Determine Stevioside Content from Stevia rebaudiana. Journal of Agricultural
and Food Chemistry, 65(15), 3223-3229.
Wan, H.-d., Guan-junTao, Kim, D., & Xia, Y.-m. (2012). Enzymatic preparation of a natural sweetener
rubusoside from specific hydrolysis of stevioside with β-galactosidase from Aspergillus sp.
Journal of Molecular Catalysis B: Enzymatic, 82, 12-17.
Wan, H.-d., & Xia, Y.-m. (2015). Enzymatic transformation of stevioside using a β-galactosidase from
Sulfolobus sp. Food Function, 6(10), 3291-3295.
Wang, Z., Wang, J., Jiang, M., Wei, Y., Pang, H., Wei, H., Huang, R., & Du, L. (2015). Selective
production of rubusoside from stevioside by using the sophorose activity of beta-glucosidase
from Streptomyces sp. GXT6. Applied Microbiology and Biotechnology, 99(22), 9663-9674.
Xu, Z., Li, Y., Wang, Y., Yang, B., & Ning, Z. (2009). Production of β-fructofuranosidase by Arthrobacter
sp. and its application in the modification of stevioside and rebaudioside A. Food Technology
and Biotechnology, 47(2), 137-143.
417
17