Using ionic liquid cosolvents to improve enzymatic synthesis of arylalkyl β-d-glucopyranosides

Article abstract of DOI:10.1016/j.molcatb.2011.08.009

Enzymatic synthesis of various arylalkyl β-d-glucopyranosides catalyzed by prune (Prunus domestica) seed meal via reverse hydrolysis in the mixture of organic solvent, ionic liquid (IL) and phosphate buffer was described. Among four hydrophilic organic solvents tested, ethylene glycol diacetate (EGDA) was found to be the most suitable for enzymatic synthesis of salidroside, a bioactive compound of commercial interest, from d-glucose and tyrosol. The effects of the nature of ionic liquids and their contents on the enzymatic glucosylation were studied. The addition of a suitable amount of ILs including denaturing ones was favorable to shift the reaction equilibrium toward the synthesis, thus improving the yields. Among the examined ILs, the novel IL [BMIm]I proved to be the best. And this IL was applied as the solvent in biocatalysis for the first time. The yields were found to be enhanced between 0.2-fold and 0.5-fold after the addition of 10% (v/v) [BMIm]I. In 10% (v/v) [BMIm]I-containing system, the desired arylalkyl β-d-glucopyranosides were synthesized with 15-28% yields, among which salidroside was obtained with a yield of 22%. .

Full text of DOI:10.1016/j.molcatb.2011.08.009

Journal of Molecular Catalysis B: Enzymatic 74 (2012) 24–28
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Using ionic liquid cosolvents to improve enzymatic synthesis of arylalkyl
␤-d-glucopyranosides
Rong-Ling Yang, Ning Li∗, Min-Hua Zong∗
State Key Laboratory of Pulp and Paper Engineering, College of Light Industry and Food Sciences, South China University of Technology, Guangzhou 510640, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 13 June 2011
Received in revised form 22 August 2011
Accepted 22 August 2011
Available online 27 August 2011
Enzymatic synthesis of various arylalkyl ␤-d-glucopyranosides catalyzed by prune (Prunus domestica)
seed meal via reverse hydrolysis in the mixture of organic solvent, ionic liquid (IL) and phosphate buffer
was described. Among four hydrophilic organic solvents tested, ethylene glycol diacetate (EGDA) was
found to be the most suitable for enzymatic synthesis of salidroside, a bioactive compound of commercial
interest, from d-glucose and tyrosol. The effects of the nature of ionic liquids and their contents on the
enzymatic glucosylation were studied. The addition of a suitable amount of ILs including denaturing ones
was favorable to shift the reaction equilibrium toward the synthesis, thus improving the yields. Among
the examined ILs, the novel IL [BMIm]I proved to be the best. And this IL was applied as the solvent in
biocatalysis for the first time. The yields were found to be enhanced between 0.2-fold and 0.5-fold after
the addition of 10% (v/v) [BMIm]I. In 10% (v/v) [BMIm]I-containing system, the desired arylalkyl ␤-d-
glucopyranosides were synthesized with 15–28% yields, among which salidroside was obtained with a
yield of 22%.
Keywords:
Ethylene glycol diacetate
␤-Glucosidase
Glycosides
Ionic liquids
Reverse hydrolysis
Salidroside
© 2011 Elsevier B.V. All rights reserved.
1. Introduction
condensation reaction of monosaccharides with alcohols to fur-
nish alkyl glycosides as the desired products and water as the
Glycosides are of commercial interest in the food, cosmetic
and pharmaceutical industries. For example, alkyl glycosides are
an important group of nonionic surfactants due to the excellent
properties such as good surfactant performance, low toxicity and
high biodegradability [1]. Salidroside (4-hydroxyphenethyl ␤-d-
glucopyranoside) is known as the main active component of the
plant Rhodiola, which has been used as the traditional medicinal
herb in China for more than 1000 years. Recently, this compound
was found to have neuron- and cardioprotection [2,3], anticancer
and antiviral activities [4,5], etc. As compared to chemical methods
[6], enzymatic synthesis of these compounds attracted more atten-
tion owing to the simplicity, high stereo- and regioselectivity, mild
reaction conditions [7].
Compared to glycosyl transferase-catalyzed synthesis of gly-
cosides, the processes mediated by glycosidases display greater
application potential since these enzymes are inexpensive, avail-
able in large quantities, generally stable and have low specificity
toward aglycones [8]. In addition, non-activated sugars such as
monosaccharides or disaccharides can be used directly as gly-
cosyl donors to synthesize glycosides. Reverse hydrolysis is the
byproduct that would not interfere with the downstream process-
ing, whereas the detrimental byproducts (monosaccharides) would
form in the transglycosylation of alcohols and disaccharides [7].
However, thermodynamically controlled reverse hydrolysis gen-
erally affords lower yields than the transglycosylation controlled
kinetically. Considerable efforts have been made to improve the
equilibrium yields [9–11], among which reducing water activities
by adding organic cosolvents is one of the most efficient strategies
[12,13].
Ionic liquids (ILs) represent a class of promising and “green”
nonmolecular solvents [14,15]. Like conventional water-miscible
organic solvents, these ionic solvents enabled to work as the cosol-
vents to control the water activities, thus inhibiting the second
hydrolysis of the glycosylated products and improving the yields
[16,17]. In addition, this type of solvents has proven to be the
media with positive effects on the enzymes, such as improving
the activity, selectivity and/or the stability [18,19]. Recently, we
found that a variety of fruit seed meals, especially prune seed
meal, were excellent biocatalysts for the synthesis of various alkyl
and arylalkyl ␤-d-glucopyranosides from d-glucose and alcohols
(data unpublished). In the present work, we focused our interest
on the enhancement of the enzymatic synthesis of arylalkyl ␤-d-
glucopyranosides by using ILs as the cosolvents, which were able
to control water activities of the systems, thus improving the yields
(Scheme 1).
∗
Corresponding authors. Tel.: +86 20 8711 1452; fax: +86 20 2223 6669.
E-mail addresses: lining@scut.edu.cn (N. Li), btmhzong@scut.edu.cn
(M.-H. Zong).
1381-1177/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcatb.2011.08.009

R.-L. Yang et al. / Journal of Molecular Catalysis B: Enzymatic 74 (2012) 24–28
25
2.3. HPLC analysis
HPLC analysis was carried out on an Agilent 1100 chromato-
graph with a variable wavelength detector and a Zorbax Eclipse
Plus C18 column (4.6 mm × 250 mm, 5 ␮m, Agilent). The flow rate
and column temperature were 1.0 ml/min and 30 ◦C, respectively.
The absorption wavelength for HPLC analysis was 257 nm, with the
exception of the analysis of salidroside 3d at 275 nm. A gradient
elution with methanol/water being 60/40 (v/v) from 0 to 4 min,
and then methanol/water being 80/20 (v/v) at 6.0 min was used.
The retention times of the compounds 3a–h were 3.02, 3.47, 2.52,
2.58, 3.06, 3.20, 3.28 and 3.09 min, respectively. The peak of salidro-
side 3d was identified by comparison with the retention time of
authentic standard. Yields were calculated by the ratio of the actual
product concentration to the theoretic product concentration based
on d-glucose. The analytical errors were less than 1%.
2.4. Structure characterization
The structures of glycosides were determined by 1H and
13C NMR (Bruker AVANCE Digital 400 MHz NMR spectrometer,
Germany) at 400 and 100 MHz, respectively. NMR data of the com-
pound 3a is the same as reported previously [20].
Scheme 1. Enzymatic synthesis of various arylalkyl ␤-d-glucopyranosides in IL-
containing systems.
Phenethyl ˇ-d-glucopyranoside 3b. 1H NMR (DMSO-d6):
ı
ꢀ
2.80–2.90 (m, 2H, H7), 2.94–2.99 (m, 1H, H4 ), 3.01–3.17 (m, 3H,
ꢀ
ꢀ
ꢀ
ꢀ
H2 + H3 + H5 ), 3.41–3.46 (m, 1H, 10.45, H6 ), 3.62–3.68 (m, 2H,
2. Materials and methods
ꢀ
ꢀ
H8), 3.91–3.98 (m, 1H, H6 ), 4.19 (d, 1H, J = 7.6 Hz, H1 ), 4.47 (t, 1H,
ꢀ
ꢀ
J = 6.0 Hz, OH6 ), 4.88 (d, 1H, J = 4.8 Hz, OH3 ), 4.91 (d, 1H, J = 4.4 Hz,
2.1. Materials
ꢀ
ꢀ
OH4 ), 4.96 (d, 1H, J = 4.8 Hz, OH2 ), 7.19–7.21 (m, 1H, H4), 7.27–7.30
(m, 4H, H2 + H3 + H5 + H6). 13C NMR (DMSO-d6): ı 35.62 (C2), 61.10
1-Butyl-3-methylimidazole tetrafluoroborate ([BMIm]BF4,
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
(C6 ), 69.39 (C4 ), 70.12 (C2 ), 73.14 (C3 ), 76.77 (C1), 76.84 (C5 ),
102.80 (C1 ), 125.97 (C4), 128.14 (C3 + C5), 128.83 (C2 + C6), 138.69
≥97%),
([BMMIm]BF4,
([BMIm]I,
≥98%),
1-butyl-2,
3-dimethylimidazole
1-butyl-3-methylimidazole
1,3-dimethylimidazole methylsulfate
tetrafluoroborate
ꢀ
≥97%),
iodide
(C1).
4-Hydroxybenzyl ˇ-d-glucopyranoside 3c. 1H NMR (DMSO-d6):
ı 2.97–3.15 (m, 4H, H2 + H3 + H4 + H5 ), 3.43–3.49 (m, 1H, H6 ),
3.68–3.72 (m, 1H, H6 ), 4.19 (d, 1H, J = 7.8 Hz, H1 ), 4.45 (d, 1H,
J = 11.4 Hz, H7), 4.53 (t, 1H, J = 5.8 Hz, OH6 ), 4.70 (d, 1H, J = 11.4 Hz,
([MMIm]MeSO4, ≥98%) and salidroside were purchased from
Sigma–Aldrich (USA). 1-Butyl-3-methylimidazole octylsulfate
([BMIm]C8SO4, ≥98%) was from Merck (Germany). 1-Butyl-
3-methylimidazole hexafluorophosphate ([BMIm]PF6, ≥96%),
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
H7), 4.91 (apparent dd, 2H, OH3 + OH4 ), 5.01 (d, 1H, J = 4.8 Hz, OH2 ),
1-hexyl-3-methylimidazole
tetrafluoroborate
([HMIm]BF4,
6.73 (apparent d, 2H, H3 + H5), 7.18 (apparent d, 2H, H2 + H6), 9.35 (s,
≥96%), 1-butyl-3-methylimidazole chloride ([BMIm]Cl, ≥96%),
1-ethoxyethyl-3-methylimidazole chloride ([EOEMIm]Cl, ≥96%),
and 1-acetoxyethyl-3-methylimidazole chloride ([AcOEMIm]Cl,
≥96%) were from Lanzhou Institute of Chemical Physics (China).
4-Nitrophenol, 4-nitrobenzyl alcohol and tyrosol were from
Linfeng Chemical Co., (Shanghai, China). 4-Hydroxybenzyl
alcohol, 4-methoxybenzyl alcohol, 3-methoxybenzyl alcohol, 2-
methoxybenzyl alcohol and phenethyl alcohol were obtained from
Haiqu Chemical Co., (Shanghai, China). Ethylene glycol diacetate
(EGDA) was from TCI (Japan). Prune seeds were from local food
processing company. The defatted prune seed meal (23.7 U/g) was
prepared and the enzyme activity was assayed as described by Yu
et al. [11]. One unit of hydrolytic activity was defined as the amount
of enzyme that released 1 ␮mol 4-nitrophenol per minute at pH
7.0 and 50 ◦C. All other chemicals are of high purity commercially
available.
1H, OH4). 13C NMR (DMSO-d6): ı 61.08 (C6 ), 69.31 (C4 ), 70.11 (C2 ),
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
73.38 (C3 ), 76.71 (C7), 76.78 (C5 ), 101.57 (C1 ), 114.74 (C3 + C5),
127.97 (C1), 129.37 (C2 + C6), 156.64 (C4).
4-Methoxylbenzyl ˇ-d-glucopyranoside 3e. 1H NMR (DMSO-d6):
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ı 2.99–3.14 (m, 4H, H2 + H3 + H4 + H5 ), 3.44–3.50 (m, 1H, H6 ),
3.68–3.73 (m, 1H, H6 ), 3.74 (s, 3H, OCH3), 4.20 (d, 1H, J = 7.7 Hz,
H1 ), 4.49–4.53 (m, 2H, H7 + OH6 ), 4.76 (d, 1H, J = 11.7 Hz, H7), 4.89
(d, 1H, J = 4.7 Hz, OH4 ), 4.92 (d, 1H, J = 4.6 Hz, OH3 ), 5.04 (d, 1H,
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
J = 4.9 Hz, OH2 ), 6.91 (apparent d, 2H, H3 + H5), 7.32 (apparent d, 2H,
H2 + H6). 13C NMR (DMSO-d6): ı 54.94 (OCH3), 61.07 (C6 ), 69.04
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
(C4 ), 70.08 (C7), 73.37 (C2 ), 76.68 (C3 ), 76.81 (C5 ), 101.65 (C1 ),
113.41 (C3 + C5), 129.22 (C2 + C6), 129.76 (C1), 158.56 (C4).
3-Methoxylbenzyl ˇ-d-glucopyranoside 3f. 1H NMR (DMSO-d6):
ı 3.01–3.17 (m, 4H, H2 + H3 + H4 + H5 ), 3.44–3.50 (m, 1H, H6 ),
3.67–3.72 (m, 1H, H6 ), 3.75 (s, 3H, OCH3), 4.22 (d, 1H, J = 7.7 Hz, H1 ),
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
4.53 (t, 1H, J = 5.9 Hz, OH6 ), 4.59 (d, 1H, J = 12.6 Hz, H7), 4.81 (d, 1H,
ꢀ
J = 12.6 Hz, H7), 4.89 (d, 1H, J = 4.6 Hz, OH4 ), 4.93 (d, 1H, J = 4.6 Hz,
ꢀ
ꢀ
OH3 ), 5.13 (d, 1H, J = 4.9 Hz, OH2 ), 6.84 (dd, 1H, J = 2.4, 8.2 Hz, H4),
6.95 (d, 1H, J = 7.6 Hz, H2), 7.01 (s, 1H, H6), 7.26 (t, 1H, J = 7.9 Hz, H5).
2.2. General procedure for enzymatic glucosylation
13C NMR (DMSO-d6): ı 55.63 (OCH3), 61.96 (C6 ), 69.91 (C4 ), 71.03
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
d-Glucose (0.5 mmol) and alcohol (5 mmol) were added to 2 ml
mixture of 10% (v/v) phosphate buffer (pH 6.0, 50 mM) and organic
solvent. The reaction was initiated by adding 5.5 U crude enzyme
(defatted prune seed meal) at 50 ◦C and 200 rpm. Aliquots were
withdrawn from the reaction mixture at specified time intervals,
and then diluted by 25-fold with the corresponding mobile phase
prior to HPLC analysis.
(C7), 74.19 (C2 ), 77.50 (C3 + C5 ), 102.74 (C1 ), 113.57 (C4), 113.67
(C2), 120.20 (C6), 129.59 (C5), 140.37 (C1), 159.94 (C3).
2-Methoxylbenzyl ˇ-d-glucopyranoside 3g. 1H NMR (DMSO-d6):
ı 3.03–3.18 (m, 4H, H2 + H3 + H4 + H5 ), 3.43–3.49 (m, 1H, H6 ),
3.67–3.72 (m, 1H, H6 ), 3.78 (s, 3H, OCH3), 4.25 (d, 1H, J = 7.7 Hz,
ꢀ ꢀ
1 ), 4.52 (t, 1H, J = 5.9 Hz, OH6 ), 4.60 (d, 1H, J = 13.5 Hz, H7), 4.83
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
H
ꢀ
(d, 1H, J = 13.5 Hz, H7), 4.89 (d, 1H, J = 4.8 Hz, OH4 ), 4.92 (d, 1H,

26
R.-L. Yang et al. / Journal of Molecular Catalysis B: Enzymatic 74 (2012) 24–28
ꢀ
ꢀ
J = 4.7 Hz, OH3 ), 5.08 (d, 1H, J = 4.9 Hz, OH2 ), 6.91–6.98 (m, 2H,
highest initial reaction rate (3.9 mM/h) was afforded in 80% (v/v)
EGDA-containing system.
H3 + H4), 7.23–7.28 (m, 1H, H5), 7.48–7.50 (m, 1H, H6). 13C NMR
ꢀ
ꢀ
ꢀ
(DMSO-d6): ı 55.05 (OCH3), 60.96 (C6 ), 64.25 (C4 ), 70.11 (C2 ),
On the contrary, the yields decreased significantly with the
decrement of the contents of organic solvents (Fig. 1b). For exam-
ple, a yield of 17% was achieved in 90% (v/v) t-butanol-containing
system, while the yield decreased to 12% in the presence of 70%
(v/v) t-butanol. And the best yield (18%) was furnished in 90%
(v/v) EGDA-containing system. The results showed that among four
organic solvents, EGDA was the best for the enzymatic synthesis of
salidroside.
ꢀ
ꢀ
ꢀ
70.06 (C3 ), 73.30 (C7), 76.54 (C5 ), 102.14 (C1 ), 110.28 (C3), 119.74
(C5), 126.05 (C4), 127.89 (C1 + C6), 156.12 (C2).
4-Nitrobenzyl ˇ-d-glucopyranoside 3h. H NMR (DMSO-d6): ı
3.05–3.20 (m, 4H, H2 + H3 + H4 + H5 ), 3.44–3.49 (m, 1H, H6 ),
3.68–3.72 (m, 1H, H6 ), 4.29 (d, 1H, J = 7.7 Hz, H1 ), 4.55 (t, 1H,
J = 5.8 Hz, OH6 ), 4.77 (d, 1H, J = 14.0 Hz, H7), 4.92–4.98 (m, 3H,
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
OH3 + OH4 + H7), 5.21 (d, 1H, J = 4.7 Hz, OH2 ), 7.69 (apparent d,
2H, H2 + H6), 8.22 (apparent d, 2H, H3 + H5). 13C NMR (DMSO-d6):
ꢀ
ꢀ
ꢀ
ꢀ
ı 61.04 (C6 ), 68.39 (C4 ), 70.02 (C2 ), 73.42 (C3 ), 76.59 (C7), 76.94
3.2. Effects of ILs and their contents on enzymatic synthesis of
glucosides
ꢀ
ꢀ
(C5 ), 102.36 (C1 ), 123.17 (C3 + C5), 127.96 (C2 + C6), 146.29 (C1),
146.65 (C4).
Table 1 showed the effect of a variety of ILs on the enzymatic
synthesis of salidroside. It was found that the initial reaction rates
(3.0–3.3 mM/h) decreased slightly upon addition−of several widely
3. Results
3.1. Effects of organic solvents and their contents on enzymatic
synthesis of salidroside
−
used ILs such as those containing BF4 and PF6 anions, but the
improved yields (20–21%) were obtained except for that (17%)
in [BMIm]PF6-containing system (Table 1, entries 2–5). As shown
in Table 1 (entries 6–8), similarly, the addition of [BMIm]C8SO4,
[MMIm]MeSO4 and [BMIm]I resulted in slower reaction rates and
higher yields (20–22%). However, the enzyme displayed extremely
low activities and even no activity in the presence of 10% (v/v) dena-
turing ILs such as [BMIm]Cl, [EOEMIm]Cl and [AcOEMIm]Cl, and the
enzymatic reactions afforded low yields (Table 1, entries 9–11).
With the enzymatic synthesis of salidroside 3d and 4-
nitrobenzyl ␤-d-glucopyranoside 3h as model reactions, the
contents of ILs were optimized (Table 2). In the synthesis of 3d, the
optimum contents of [BMIm]BF4, [BMMIm]BF4, and [BMIm]C8SO4
were 10–15% (v/v) and the yields were increased to 21%. In the cases
of [BMIm]C8SO4- and [BMIm]I-containing systems, the optimum
contents of the two ILs were 10% (v/v). And among all the examined
IL-containing systems, the best result (a yield of 22%) was obtained
in 10% (v/v) [BMIm]I-containing system. However, the optimum
content of the hydrophobic IL [BMIm]PF6 was 5% (v/v), in which the
improvement of the yield was negligible. Interestingly, such “dena-
turing” ILs as [BMIm]Cl, [EOEMIm]Cl and [AcOEMIm]Cl were found
to enhance the enzymatic glucosylation at the optimum contents,
affording higher yields (20%).
Four water-miscible organic solvents including EGDA, t-butanol,
dioxane and acetone were tested as the cosolvents for the enzy-
matic synthesis of salidroside (Fig. 1). As shown in Fig. 1a, the
initial reaction rates were comparable in the media containing
organic solvents of the same content. In 90% (v/v) organic solvent-
containing systems, for example, the initial reaction rates ranged
from 3.1 to 3.5 mM/h. Besides, the initial reaction rates increased
with the decrease of the contents of organic solvents and to
the highest when the content of organic solvents decreased to
80% (v/v), with the exception of dioxane-containing systems. The
It was found that in the enzymatic synthesis of 3h, the optimum
contents of all the ILs were the same as those in the synthesis of 3d.
For example, the optimum content of [BMIm]BF4 and [EOEMIm]Cl
were 15% and 5% (v/v), respectively. It seemed not to be occa-
sional since similar phenomena occurred in the synthesis of 3e
(Table 1S, available as supplementary material). Also, the highest
yield (15%) was obtained in 10% (v/v) [BMIm]I-containing system.
Table 1
Effect of ILs on enzymatic synthesis of salidroside 3d.
Entry
IL
V0 (mM/h)
Yield (%)
1
2
3
4
5
6
7
8
9
–
3.5
3.0
3.0
3.3
3.2
2.9
3.0
3.1
0
18
21
21
20
17
20
21
22
0
[BMIm]BF4
[BMMIm]BF4
[HMIm]BF4
[BMIm]PF6
[BMIm]C8SO4
[MMIm]MeSO4
[BMIm]I
[BMIm]Cl
10
11
[EOEMIm]Cl
[AcOEMIm]Cl
0.5
0.4
9
2
Fig. 1. Effects of organic solvents and their contents on the initial reaction rate (a)
and yield (b) in the enzymatic synthesis of salidroside 3d. Conditions: 0.5 mmol
glucose, 5 mmol tyrosol, 5.5 U prune seed meal, organic solvent-phosphate buffer
(pH 6.0, 50 mM), total volume 2 ml, 50 ◦C, 200 rpm. 90%, 85%, 80% and 70% represent
volume percent of corresponding organic solvents in the reaction systems.
Conditions: 0.5 mmol glucose, 5 mmol tyrosol, 5.5 U prune seed meal, EGDA-10%
(v/v) IL-10% (v/v) phosphate buffer (pH 6.0, 50 mM), total volume 2 ml, 50 ◦C,
200 rpm. Reaction time: 72 h.

R.-L. Yang et al. / Journal of Molecular Catalysis B: Enzymatic 74 (2012) 24–28
27
Table 2
Effects of ILs and their contents on the enzymatic synthesis of the compounds 3d and 3 h.
IL
3d
3h
Optimum content (%, v/v)
Yield (%)/relative yield (%)
Optimum content (%, v/v)
–
Yield (%)/relative yield (%)
The control
[BMIm]BF4
[BMMIm]BF4
[BMIm]C8SO4
[BMIm]I
[HMIm]BF4
[BMIm]PF6
[BMIm]Cl
–
18/100
21/117
21/117
21/117
22/122
20/111
18/100
20/111
20/111
20/111
10/100
14/140
13/130
13/130
15/150
12/120
11/110
12/120
12/120
12/120
10–15
10–15
15
10
10
5
2.5
5
15
15
15
10
10
5
2.5
5
5
[EOEMIm]Cl
[AcOEMIm]Cl
5
Conditions: 0.5 mmol glucose, 5 mmol alcohol, 5.5 U prune seed meal, EGDA-various content of IL-10% (v/v) phosphate buffer (pH 6.0, 50 mM), total volume 2 ml, 50 ◦C,
200 rpm.
3.3. Enzymatic synthesis of arylalkyl glycosides in
[BMIm]I-containing system
with decreasing contents of the organic solvents, while the yields
deceased significantly. The possible reason might be that reverse
hydrolysis is thermodynamically controlled, and a low water activ-
ity is desirable to shift the reaction equilibrium toward synthesis,
thus leading to a higher yield [7,8]. On the other hand, the gly-
cosidases are amenable to denaturation at a low water activity
[13,22].
Zhao proposed that like the ions, the effects of ILs on the enzyme
activity and stability followed the Hofmeister series in aqueous
solutions of hydrophilic ILs: kosmotropic anions and chaotropic
cations stabilize enzymes while chaotropic anions and kosmotropic
cations destabilize them [23]. The kosmotropicity of the anions
follow a decreasing order: Cl− > I− > BF4− > PF6−; according to the
abovementioned principle, the enzyme activity should be the high-
est in Cl−-containing system, and lower in I−-containing system,
the lowest in PF6−-containing system. Nevertheless, indeed, no
activity was observed in [BMIm]Cl-containing system (Table 1,
entry 9), and the enzyme activity in [BMIm]PF6-containing system
was comparable to those in [BMIm]BF4- and [BMIm]I-containing
systems (Table 1, entries 2, 5 and 8). Obviously, the effects of ILs
on the enzyme activity did not follow the Hofmeister series in the
present work, which might be ascribed to the presence of low con-
tent of water (10%, v/v) in the systems, thus preventing ILs from
dissociating into individual ions.
As compared to that (3.5 mM/h) in the control, the initial reac-
tion rates (2.9–3.3 mM/h) were slightly lower in IL-containing
systems (Table 1, entries 2–8), which might stem from the high
visc−osities of ILs. The most extensively used ILs such as those with
BF4 were able to promote the enzymatic reactions, thus lead-
ing to higher yields (Table 1, entries 2–4), while a slightly lower
yield (17%) was obtained in [BMIm]PF6-containing system (Table 1,
entry 5). The possible reason is that [BMIm]PF6 is a hydrophobic IL,
which has more poor ability of lowering water activity than those
hydrophilic ILs. Previously, [MMIm]MeSO4 was demonstrated to
enable to improve the yields and/or the enzyme stability in the
enzymatic glycosylation reactions [16,17]. Also, a higher yield (21%)
was achieved with this IL as the cosolvent (Table 1, entry 7). Inter-
estingly, unlike its counterparts containing Cl− and Br− anions, the
novel IL [BMIm]I did not denature the enzyme. In addition, the
enzymatic glycosylation of tyrosol was enhanced by adding this IL,
affording a higher yield (Table 1, entry 8). To the best of our knowl-
edge, the biocatalytic application of this IL was reported for the first
time. The significant denaturation of the enzyme in the systems
containing Cl− anion (Table 1, entries 9–11) might be attributed to
high hydrogen bond basicity of this anion [14,24–26].
The synthesis of a group of arylalkyl ␤-d-glucopyranosides
catalyzed by prune seed meal was performed in 10% (v/v) [BMIm]I-
containing system (Table 3). The enzymatic glucosylation of benzyl
alcohol afforded the product 3a with a 21% yield (Table 3,
entry 1). And the yield decreased to 17% with the elongation
of alkyl chain (Table 3, entry 2). The highest yield (28%) was
obtained after 68 h in the synthesis of 4-hydroxybenzyl ␤-D-
glucopyranoside 3c (Table 3, entry 3). As shown in Table 3 (entries
5–7), in the enzymatic glycosylation of a series of methoxy-
substituted benzyl alcohol, the highest yield (26%) was furnished
for 4-methoxybenzyl ␤-d-glucopyranoside 3e, and lower (22%) for
3-methoxybenzyl ␤-D-glucopyranoside 3f, and the lowest (15%) for
2-methoxybenzyl ␤-d-glucopyranoside 3g. Besides, 4-nitrobenzyl
␤-d-glucopyranoside 3h was synthesized with a yield of 15% after
50 h (Table 3, entry 8).
4. Discussion
t-Butanol and dioxane were widely used as organic cosolvents
in glycosides synthesis [11,20]. However, little attention was paid
to the enzymatic glycosylation in EGDA-containing systems [21].
The results showed that EGDA as the cosolvent was slightly bet-
ter than other three organic solvents at the same content. Besides,
compared to dioxane that is classified by the international agency
for research on cancer (IARC) as a Group 2B carcinogen, EGDA is
a green and safe solvent. And it has many good features such as
slow evaporating rate, good flow properties and biodegradability.
More importantly, unlike t-butanol, EGDA is partially miscible with
water. So upon completion of the reaction, two phases would be
formed by adding water, thus remarkably simplifying the down-
stream processing. In addition, the initial reaction rates increased,
Table 3
Enzymatic synthesis of various glucosides in [BMIm]I-containing system.
Entry
Glucoside
Time (h)
Yield (%)
1
2
3
4
5
6
7
8
3a
3b
3c
3d
3e
3f
42
42
68
72
50
50
50
50
21
17
28
22
26
22
15
15
3g
3h
Table 2 showed that the optimization led to significant incre-
ment of the yields in the systems containing the denaturing ILs
[BMIm]Cl, [EOEIm]Cl and [AcOEMIm]Cl. In the enzymatic synthesis
of salidroside 3d, the denaturation of the enzyme was remarkably
Conditions: 0.5 mmol glucose, 5 mmol alcohol, 5.5 U prune seed meal, EGDA-10%
(v/v) [BMIm]I-10% (v/v) phosphate buffer (pH 6.0, 50 mM), total volume 2 ml, 50 ◦C,
200 rpm.

28
R.-L. Yang et al. / Journal of Molecular Catalysis B: Enzymatic 74 (2012) 24–28
relieved by reducing the contents of these ILs, thus giving satis-
factory yields (20%). Likewise, slightly higher yields (12%) were
achieved in the synthesis of 3h under the optimum contents of
these ILs. To date, few enzymes were reported to be active in the
denaturing ILs [27]. The striking results revealed that this biocata-
lyst (the prune seed meal) was strongly resistant to nonaqueous
media including the denaturing ILs, displaying great synthetic
application potential. Although the yield was only increased by
4–5% in the absolute value, which was not as significant as we
expected, the addition of IL cosolvents led to significant improve-
ments of the relative yields within the range from 0.2- to 0.5-fold.
It was expected that the yields would be improved further by
coupling the “solvent engineering” with in situ product removal
[10,11,28].
Acknowledgments
This work was financially supported by the National Natu-
ral Science Foundation of China (Grant no. 20906032, 20876059
and 21072065), the Fundamental Research Funds for the Cen-
tral Universities, SCUT (Grant no. 2009zz0018, 2009zz0026 and
2009zm0199), and the Open Project Program of State Key Labo-
ratory of Pulp and Paper Engineering, SCUT (Grant no. 200913).
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.molcatb.2011.08.009.
As shown in Table 3, the effect of the structures of the gly-
cosyl acceptors, especially the chain length of the alkyl, on the
enzymatic reaction is remarkable. For example, the enzymatic glu-
cosylation of benzyl alcohol 2a afforded the product 3a with a yield
of 21%, whereas the yield (17%) was lower with phenethyl alco-
hol 2b as the glycosyl acceptor. Similarly, the glucosylation yield
(28%) of 4-hydroxybenzyl alcohol 2c was higher than that (22%)
of 4-hydroxyphenethyl alcohol 2d. Sheldon and coworkers pro-
posed that the equilibrium yield was reduced by about 40% for
each extra carbon atom [7]. In addition, the enzyme preferred cat-
alyzing the glucosylation of 4-hydroxy substituted aryl alcohols to
that of no substituted analogs (Table 3, entries 1 and 3, 28% vs.
21%; entries 2 and 4, 22% vs. 17%). Besides, 4-methoxy substitution
resulted in a higher yield (26%, Table 3, entry 5). The substitution
pattern of the methoxyl on phenyl ring exerted a great effect on
the yields (Table 3, entries 5–7). The yield was the highest for the
para-substituted benzyl alcohol, and lower for the meta-substituted
counterpart, and the lowest for the ortho-substituted counterpart.
It might derive from the steric hindrance. The lowest yield (15%)
was obtained in the glucosylation of 4-nitrobenzyl alcohol, which
might be attributed to the electronic effect and/or steric strain of
the strong electron-withdrawing NO2 group [29]. Similarly, Xu and
coworkers reported that the yield was the lowest for the glucosyla-
tion of 4-nitrobenzyl alcohol catalyzed by apple seed meal among
a series of 4-substituted benzyl alcohol [20].
References
[1] V.M. Dembitsky, Lipids 39 (2004) 933–953.
[2] S. Yu, M. Liu, X. Gu, F. Ding, Cell. Mol. Neurobiol. 28 (2008) 1067–1078.
[3] J.P. Zhang, A.H. Liu, R.R. Hou, J. Zhang, X. Jia, W.F. Jiang, J.Z. Chen, Eur. J. Phar-
macol. 607 (2009) 6–14.
[4] H.B. Wang, Y.Y. Ding, J. Zhou, X.L. Sun, S.W. Wang, Phytomedicine 16 (2009)
146–155.
[5] X. Hu, S. Lin, D. Yu, S. Qiu, X. Zhang, R. Mei, Cell Biol. Toxicol. 26 (2010)
499–507.
[6] X. Zhu, R.R. Schmidt, Angew. Chem. Int. Ed. 48 (2009) 1900–1934.
[7] F. van Rantwijk, M.W.V. Oosterom, R.A. Sheldon, J. Mol. Catal. B: Enzym. 6 (1999)
511–532.
[8] B.M. de Roode, M.C.R. Franssen, A. van der Padt, R.M. Boom, Biotechnol. Prog.
19 (2003) 1391–1402.
[9] A. Millqvist-Fureby, I.S. Gill, E.N. Vulfson, Biotechnol. Bioeng. 60 (1998)
190–196.
[10] B.M. de Roode, J. van Beek, A. van der Padt, M.C.R. Franssen, R.M. Boom, Enzyme
Microb. Technol. 29 (2001) 513–520.
[11] H.L. Yu, J.H. Xu, W.Y. Lu, G.Q. Lin, J. Biotechnol. 133 (2008) 469–477.
[12] G. Vic, D. Thomas, D.H.G. Crout, Enzyme Microb. Technol. 20 (1997)
597–603.
[13] A. Ismail, S. Soultani, M. Ghoul, J. Biotechnol. 69 (1999) 145–149.
[14] S. Park, R.J. Kazlauskas, Curr. Opin. Biotechnol. 14 (2003) 432–437.
[15] Z. Yang, W.B. Pan, Enzyme Microb. Technol. 37 (2005) 19–28.
[16] M. Lang, T. Kamrat, B. Nidetzky, Biotechnol. Bioeng. 95 (2006) 1093–1100.
[17] N. Kaftzik, P. Wasserscheid, U. Kragl, Org. Process Res. Dev.
553–557.
6 (2002)
[18] F. van Rantwijk, R.A. Sheldon, Chem. Rev. 107 (2007) 2757–2785.
[19] U. Kragl, M. Eckstein, N. Kaftzik, Curr. Opin. Biotechnol. 13 (2002) 565–571.
[20] A.M. Tong, W.Y. Lu, J.H. Xu, G.Q. Lin, Bioorg. Med. Chem. Lett. 14 (2004)
2095–2097.
[21] T. Balogh, L. Boross, J. Kosary, Tetrahedron 60 (2004) 679–682.
[22] G. Ljunger, P. Adlercreutz, B. Mattiasson, Enzyme Microb. Technol. 16 (1994)
751–755.
[23] H. Zhao, J. Mol. Catal. B: Enzym. 37 (2005) 16–25.
[24] N. Li, D. Ma, M.H. Zong, J. Biotechnol. 133 (2008) 103–109.
[25] S. Lee, S. Ha, S. Lee, Y.-M. Koo, Biotechnol. Lett. 28 (2006) 1335–1339.
[26] M.B. Turner, S.K. Spear, J.G. Huddleston, J.D. Holbrey, R.D. Rogers, Green Chem.
5 (2003) 443–447.
[27] A.R. Toral, A.P. de los Rios, F.J. Hernandez, M.H.A. Janssen, R. Schoevaart, F. van
Rantwijk, R.A. Sheldon, Enzyme Microb. Technol. 40 (2007) 1095–1099.
[28] F. Ahmed, A. Stein, G.J. Lye, J. Chem. Technol. Biotechnol. 76 (2001)
971–977.
5. Conclusions
In conclusion, the ILs proved to be excellent cosolvents to low-
ering the water activity of the reaction system, thus improving the
reaction equilibrium yields. Interestingly, the IL with halo anion
[BMIm]I not only did not inactivate the enzyme, but also pro-
moted the enzymatic glucosylation. Besides, the crude enzyme
from prune seeds was highly active in the systems containing lower
contents (2.5–5%) of denaturing ILs such as [BMIm]Cl, [EOEIm]Cl
and [AcOEMIm]Cl, displaying great application potential in
synthetic organic chemistry.
[29] M.B. Smith, J. March, March’s Advanced Organic Chemistry: Reactions, Mech-
anisms and Structure, 6th ed., John Wiley & Sons, Inc, Hoboken, New Jersey,
2007.