Enzymatic resolution of (RS)-1-phenylalkyl β-D-glucosides to (R)-1-phenylalkyl β-primeverosides and (S)-1-phenylalkyl β-D-glucosides via plant xylosyltransferase

Article abstract of DOI:10.1016/j.tetasy.2010.07.023

The stereoselective xylosylation of (RS)-1-phenylalkyl β-D-glucosides was investigated using plant xylosyltransferase isolated from cultured Catharanthus roseus cells. Enzymatic xylosylation of (RS)-1-phenylethyl β-D-glucoside afforded (R)-1-phenylethyl β

Full text of DOI:10.1016/j.tetasy.2010.07.023

Tetrahedron: Asymmetry 21 (2010) 2060–2065
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Enzymatic resolution of (RS)-1-phenylalkyl b-D-glucosides to (R)-1-phenylalkyl
b-primeverosides and (S)-1-phenylalkyl b-D-glucosides via plant
xylosyltransferase
b,
Kei Shimoda a, , Hisashi Katsuragi
*
*
a Department of Chemistry, Faculty of Medicine, Oita University, 1-1 Hasama-machi, Oita 879-5593, Japan
b Sunny Health Co., Ltd, Nakajima Bidg., 8-8 Kabuto-cho, Nihonbashi, Chuo-ku, Tokyo 103-0026, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
The stereoselective xylosylation of (RS)-1-phenylalkyl b-D-glucosides was investigated using plant xylo-
Received 30 June 2010
Accepted 20 July 2010
Available online 13 August 2010
syltransferase isolated from cultured Catharanthus roseus cells. Enzymatic xylosylation of (RS)-1-phenyl-
ethyl b- -glucoside afforded (R)-1-phenylethyl b-primeveroside and (S)-1-phenylethyl b- -glucoside. The
(R)-selective xylosylation of (RS)-1-phenylbutyl b- -glucoside also occurred to give (R)-1-phenylbutyl
b-primeveroside and recovered (S)-1-phenylbutyl b- -glucoside.
Ó 2010 Elsevier Ltd. All rights reserved.
D
D
D
D
1. Introduction
To examine the substrate specificity of the xylosyltransferase
isolated from C. roseus cell cultures, 2-hydroxybenzyl b- -glucoside
1, 2-hydroxymethylphenyl b- -glucoside 2, 3-hydroxybenzyl b-
glucoside 3, 3-hydroxymethylphenyl b- -glucoside 4, 4-hydroxy-
benzyl b- -glucoside 5, and 4-hydroxymethylphenyl b- -glucoside
6, p-nitrophenyl b- -glucoside 7, p-nitrophenyl b- -galactoside 8,
p-nitrophenyl b- -xyloside 9, and p-nitrophenyl -arabinoside
10 were used as substrates (Fig. 1). As shown in Table 1, only ben-
zyl b- -glucoside derivatives 1, 3, and 5 were xylosylated to the
D
Enzymatic glycosylation of organic compounds with glycosi-
dases and glycosyltransferases has been widely studied, since
one-step enzymatic glycosylation is more attractive than chemical
glycosylation which requires tedious protection–deprotection pro-
cedures.1–6 However, little attention has been paid to the stereose-
lectivity of the enzymatic glycosylations.7–9 Recently, it has been
reported that (RS)-1-phenylethanol was transformed to (R)-1-
phenylethyl b-primeveroside, (R)-1-phenylethyl b-gentiobioside,
and (RS)-1-phenylethyl b-vicianoside by plant cell cultures of Cath-
aranthus roseus.10 The (R)-selective disaccharide formation was
also found in the biotransformation of (RS)-tropic acid to give an
isotrehalose ester of its (R)-isomer by cultured plant cells of Coffea
arabica.11 Herein we report the stereoselective xylosylation of (RS)-
D
D-
D
D
D
D
D
D
a-L
D
corresponding b-primeverosides 13–15 by the enzyme, whereas
no xylosylated products were obtained in the cases of phenyl
b-
D-glucoside derivatives 2, 4, 6, and 7 when used as the substrates.
The yields of primeveroside products 13–15 were 52% (70%
OR2
2
3
1-phenylalkyl b-
sides by plant xylosyltransferase from C. roseus.
D
-glucosides to (R)-1-phenylalkyl b-primevero-
7
RO
NO2
1
R1O H2C
4
6
5
7: R=Glc
8: R=Gal
9: R=Xyl
10: R=Ara
2. Results and discussion
1: R1 = Glc, R2 = H (o-OH)
2: R1 = H, R2 = Glc (o-OGlc)
3: R1 = Glc, R2 = H (m-OH)
4: R1 = H, R2 = Glc (m-OGlc)
5: R1 = Glc, R2 = H (p-OH)
6: R1 = H, R2 = Glc (p-OGlc)
13: R1 = GlcXyl, R2 = H (o-OH)
14: R1 = GlcXyl, R2 = H (m-OH)
15: R1 = GlcXyl, R2 = H (p-OH)
A xylosyltransferase was isolated from cultured plant cells from
C. roseus with three-step column chromatographies on DEAE-Toyo-
pearl column, Sephadex G-150 column, and REACTIVE GREEN 19
agarose-gel column. The purified xylosyltransferase was judged
to be homogeneous by SDS–PAGE, and has a monomeric form with
a molecular mass of ca. 55 kDa by analyses of both SDS–PAGE and
size exclusion chromatography on a Sephadex G-150 column.
H
R2O
R1
11: R1 = Me, R2 = Glc
12: R1 = n-Pr, R2 = Glc
16: R1 = Me, R2 = GlcXyl
17: R1 = n-Pr, R2 = GlcXyl
* Corresponding authors. Tel.: +81 97 586 5606; fax: +81 97 586 5619.
E-mail address: shimoda@med.oita-u.ac.jp (K. Shimoda).
Figure 1. Chemical structures of substrates 1–12 and products 13–17.
0957-4166/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2010.07.023

K. Shimoda, H. Katsuragi / Tetrahedron: Asymmetry 21 (2010) 2060–2065
2061
intensities of the H-1000 anomeric proton signals [d 4.27 (1H, d,
J = 7.6 Hz) for (R)-1-phenylethyl b-primeveroside 16a and 4.31
(1H, d, J = 7.6 Hz) for (S)-1-phenylethyl b-primeveroside 16b] in
the 1H NMR spectrum of product 16 (Fig. 2). The enantiomeric ex-
cess of the aglycone moiety of product 16 was confirmed to be 92%
ee by chiral GLC analysis of the aglycone alcohols, which had been
prepared by the hydrolysis of 16 with 4.0 M HCl, on the basis of the
peak areas of each enantiomer. Thus, it was demonstrated that the
xylosyltransferase from C. roseus can xylosylate (RS)-1-phenylethyl
Table 1
Substrate specificity of the xylosyltransferase from C. roseus
Substrate
Product
Yield (%)
Relative activity (%)
1
2
3
4
5
13
—
14
—
15
—
—
52
0
66
Tr.
29
0
79
—
100
—
44
—
—
6
7
0
b-D-glucoside to (R)-1-phenylethyl b-primeveroside with a high
8
9
10
—
—
—
0
0
0
—
—
—
diastereomeric excess. According to the previously reported meth-
od,13 the E parameter was calculated to be 65.
The absolute configuration and enantiomeric composition of
the aglycone moiety of the recovered substrate 11 (33% yield;
52% conversion yield) were determined on the basis of the 1H
NMR analysis of H-100 anomeric proton signals of unreacted 11.
Figure 3(A) shows the 1H NMR spectrum of the anomeric proton
conversion yield), 66% (85% conversion yield), and 29% (39%
conversion yield), respectively. In addition, no xylosylation of
b-D-galactoside 8, b-D-xyloside 9, and a-L-arabinoside 10 occurred.
These findings indicate that the xylosyltransferase from C. roseus
cells had high substrate specificity for the benzyl hydroxyl group.
Next, the diastereoselectivity of the C. roseus xylosyltransferase
pairs of the substrate (RS)-1-phenylethyl b-
included two H-100 anomeric proton signals at d 4.41 (1H, d,
J = 7.9 Hz) for (R)-1-phenylethyl b- -glucoside 11a and 4.11 (1H,
d, J = 8.0 Hz) for (S)-1-phenylethyl b-
-glucoside 11b. The 1H
D-glucoside 11, which
D
was examined. The enzyme converted (RS)-1-phenylethyl b-D-glu-
D
coside 11 into product 16 in 30% yield (48% conversion yield). The
structure of product 16 was determined as 1-phenylethyl b-prime-
veroside by HRFABMS, 1H and 13C NMR, H–H COSY, C–H COSY, and
HMBC analyses. The 1H NMR spectrum of the authentic (R)-1-
phenylethyl b-primeveroside 16a, which had been prepared by a
chemosynthetic method,12 showed proton signals at d 3.67 (1H,
dd, J = 11.6, 5.8 Hz, H-6a00), 3.79 (1H, dd, J = 11.6, 3.4 Hz, H-5b000),
3.97 (1H, dd, J = 11.6, 1.8 Hz, H-6b00), 4.26 (1H, d, J = 7.6 Hz,
H-1000), and 4.46 ppm (1H, d, J = 8.2 Hz, H-100), whereas that of (S)-
1-phenylethyl b-primeveroside 16b included proton signals at d
3.73 (1H, dd, J = 11.4, 5.9 Hz, H-6a00), 3.85 (1H, dd, J = 11.6, 3.6 Hz,
H-5b000), 4.08 (1H, dd, J = 11.6, 2.0 Hz, H-6b00), 4.30 (1H, d,
J = 7.6 Hz, H-1000), and 4.43 ppm (1H, d, J = 8.0 Hz, H-100). The abso-
lute configuration of the aglycone moiety of product 16 obtained
by xylosylation of 11 with C. roseus xylosyltransferase was deter-
mined to be (R) by direct comparison of its 1H NMR spectrum,
which included proton signals at d 3.68 (1H, dd, J = 11.6, 5.8 Hz,
H-6a00), 3.79 (1H, dd, J = 11.6, 4.1 Hz, H-5b000), 3.97 (1H, dd,
J = 11.6, 1.9 Hz, H-6b00), 4.27 (1H, d, J = 7.6 Hz, H-1000), and
4.46 ppm (1H, d, J = 8.2 Hz, H-100) (Fig. 2), with those of authentic
(R)- and (S)-1-phenylethyl b-primeverosides. The diastereomeric
excess of product 16 was calculated to be 92% de based on the
NMR analysis of the recovered substrate 11 [Fig. 3(B)], which
showed strong H-100 anomeric proton signal at d 4.11 (1H, d,
J = 8.0 Hz) rather than that at d 4.41 (1H, d, J = 7.9 Hz), revealed that
the absolute configuration of the recovered substrate 11 was (S).
The enantiomeric excess of the aglycone moiety of the recovered
substrate 11 was calculated to be 88% ee based on the intensities
of H-100 anomeric proton signals at d 4.41 (1H, d, J = 7.9 Hz) for
(R)-1-phenylethyl b-
for (S)-1-phenylethyl b-
of unreacted 11 [Fig. 3(B)]. It was demonstrated that (S)-1-phenyl-
ethyl b- -glucoside can be recovered with high diastereomeric ex-
cess after xylosylation of (RS)-1-phenylethyl b- -glucoside by C.
D
-glucoside 11a and 4.11 (1H, d, J = 8.0 Hz)
D
-glucoside 11b in the 1H NMR spectrum
D
D
roseus xylosyltransferase. The results obtained here revealed that
the xylosyltransferase from C. roseus discriminated the absolute
configuration of the aglycones of substrate (RS)-1-phenylethyl b-
D
-glucoside and that the enzyme catalyzed the (R)-preferential for-
mation of primeveroside from (RS)-1-phenylethyl b- -glucoside.
Scheme 1 shows the stereoselective xylosylation of (RS)-1-phenyl-
ethyl b- -glucoside 11 to give (R)-1-phenylethyl b-primeveroside
D
D
and unreacted (S)-1-phenylethyl b-
D
-glucoside by C. roseus xylosyl-
6''
HO
HO
4''
5''
O
O
HO
O
HO
2''
O
1''
HO
HO
4'''
3'
6'
5'''
3''
2'''O
O
OH
OH
HO
HO
HO
6''
O
O
1'''
2'
1'
4'
5'
HO
4''
3'''
5''
OH
OH
O
O
2''
HO
HO
O
O
11a
11b
1''
1
HO
HO
3''
OH
OH
2
4.41
1''-H (11a)
4.11
1''-H (11b)
16a
16b
4.46
1''-H (16a)
4.27
(A)
(B)
1'''-H (16a)
3.97
6b''-H (16a)
3.79
5b'''-H (16a)
3.68
6a''-H (16a)
δ (ppm)
4.43
4.31
4.07
6b''-H (16b)
3.85
3.73
1''-H (16b) 1'''-H (16b)
5b'''-H (16b) 6a''-H (16b)
δ (ppm)
Figure 2. Proton signals in the 1H NMR of the product 16 obtained by xylosylation
of 11 with xylosyltransferase from C. roseus.
Figure 3. Anomeric proton signals in the 1H NMR of (A) substrate 11 and (B)
recovered substrate 11 after xylosylation by xylosyltransferase from C. roseus.

2062
K. Shimoda, H. Katsuragi / Tetrahedron: Asymmetry 21 (2010) 2060–2065
O
HO
HO
HO
HO
HO
HO
O
O
O
O
HO HO
HO
O
O
O
HO
HO
Xylosyltransferase from
C. roseus
OH
R
OH
OH
R
R
+
R=Me, n-Pr
Scheme 1. Stereoselective xylosylation of (RS)-1-phenylalkyl b-
xylosyltransferase from C. roseus.
D-glucosides to give (R)-1-phenylalkyl b-primeverosides and (S)-1-phenylalkyl b-D-glucosides by
6''
HO
HO
transferase. In our recent paper, we reported that (RS)-1-phenyl-
ethanol was transformed to (R)-1-phenylethyl b-primeveroside
16a (100% de) by C. roseus cell cultures.10 Compared with the re-
sults in the present study, the excellent diastereomeric excess
(100% de) of 16a found in the biotransformation of (RS)-1-phenyl-
ethanol with C. roseus cell cultures may be explained by the (R)-
selective glucosylation of (RS)-1-phenylethanol by C. roseus cells
HO
4''
5''
O
O
HO
2''
O
O
1''
HO
HO
3''
OH
OH
12b
12a
4.49
1''-H (12a)
4.15
1''-H (12b)
to give (R)-1-phenylethyl b-
D
-glucoside 11a with 63% de.10
-glucoside 12 was subjected to
Substrate (RS)-1-phenylbutyl b-
D
the same enzymatic xylosylation system as described above to give
product 17 (21% yield; 32% conversion yield). Product 17 was iden-
tified as 1-phenylbutyl b-primeveroside by HRFABMS, 1H, and 13C
NMR, H–H COSY, C–H COSY, and HMBC analyses. The 1H NMR
spectrum of the chemically synthesized authentic (R)-1-phenylbu-
tyl b-primeveroside 17a included proton signals at d 3.67 (1H, dd,
J = 11.6, 5.9 Hz, H-6a00), 3.78 (1H, dd, J = 11.6, 3.7 Hz, H-5b000), 4.00
(1H, dd, J = 11.6, 1.8 Hz, H-6b00), 4.30 (1H, d, J = 7.8 Hz, H-1000), and
4.48 ppm (1H, d, J = 7.9 Hz, H-100), while that of (S)-1-phenylbutyl
b-primeveroside 17b showed proton signals at d 3.72 (1H, dd,
J = 11.6, 5.8 Hz, H-6a00), 3.86 (1H, dd, J = 11.6, 3.6 Hz, H-5b000), 4.10
(1H, dd, J = 11.6, 2.0 Hz, H-6b00), 4.34 (1H, d, J = 7.9 Hz, H-1000), and
4.45 ppm (1H, d, J = 8.0 Hz, H-100). The absolute configuration of
the aglycone moiety of product 17 obtained by enzymatic xylosy-
lation of 12 with C. roseus xylosyltransferase was determined to be
(R) by direct comparison of its 1H NMR spectrum, which included
proton signals at d 3.68 (1H, dd, J = 11.6, 5.9 Hz, H-6a00), 3.78 (1H,
dd, J = 11.5, 3.9 Hz, H-5b000), 4.01 (1H, dd, J = 11.6, 2.0 Hz, H-6b00),
4.30 (1H, d, J = 7.8 Hz, H-1000), and 4.48 ppm (1H, d, J = 7.9 Hz, H-
100) (Fig. 4), with those of authentic (R)- and (S)-1-phenylbutyl b-
primeverosides. The diastereomeric excess of product 17 was
determined to be 100% de based on the 1H NMR analysis of product
17, which showed not H-1000 anomeric proton signal at d 4.34 for
(S)-1-phenylbutyl b-primeveroside 17b but one signal at d 4.30
(1H, d, J = 7.8 Hz) for (R)-1-phenylbutyl b-primeveroside 17a
(A)
(B)
δ (ppm)
Figure 5. Anomeric proton signals in the 1H NMR of (A) substrate 12 and (B)
recovered substrate 12 after xylosylation by xylosyltransferase from C. roseus.
(Fig. 4). The enantiomeric composition of the aglycone moiety of
product 17 was confirmed to be 99% ee by chiral GLC analysis of
the aglycone alcohol obtained by the hydrolysis of 17 with 4.0 M
HCl. It was demonstrated that C. roseus xylosyltransferase can
xylosylate (RS)-1-phenylbutyl b-D-glucoside to (R)-1-phenylbutyl
b-primeveroside with excellent diastereomeric excess. The E value
was calculated to be 317.
The absolute configuration and enantiomeric excess of the agly-
cone moiety of the recovered substrate 12 (45% yield; 68% conver-
sion yield) were determined on the basis of 1H NMR analysis of the
H-100 anomeric proton signals of unreacted 12 (Fig. 5) [d 4.49 (1H, d,
J = 8.0 Hz) for (R)-1-phenylbutyl b-
J = 8.0 Hz) for (S)-1-phenylbutyl b-
an (S)-configuration with 44% ee. The resolution of (RS)-1-phenyl-
butyl b- -glucoside to give (R)-1-phenylbutyl b-primeveroside and
(S)-1-phenylbutyl b- -glucoside by xylosylation with C. roseus
4'''
3'
6'
5'''
O
O
HO
HO
HO
6''
D
-glucoside 12a and 4.15 (1H, d,
-glucoside 12b] and indicated
2'''
O
O
1'''
2'
1'
4'
5'
HO
4''
3'''
5''
2''
OH
OH
HO
HO
O
O
D
HO
O
O
1''
1
HO
3''
OH
OH
2
D
3
17a
17b
4
D
xylosyltransferase is shown in Scheme 1.
4.48
1''-H (17a)
4.30
1'''-H (17a)
3. Conclusion
4.01
6b''-H (17a)
The enzymatic resolution of (RS)-1-phenylalkyl b-D-glucosides
has been achieved by stereoselective xylosylation with plant xylo-
3.78
5b'''-H (17a)
3.68
6a''-H (17a)
syltransferase from C. roseus. The enzyme catalyzed (R)-selective
xylosylation of (RS)-1-phenylalkyl b-
phenylalkyl b-primeverosides and unreacted (S)-1-phenylalkyl b-
-glucosides with high diastereoselectivity. The results of this
D-glucosides to give (R)-1-
D
experiment indicate that the enzyme can discriminate the absolute
configuration, at the stereogenic center apart from the xylosylation
δ (ppm)
Figure 4. Proton signals in the 1H NMR of the product 17 obtained by xylosylation
of 12 with xylosyltransferase from C. roseus.
part of the aglycones of the substrate (RS)-1-phenylalkyl b-
D-gluco-
sides. Analysis of the whole sequence of the xylosyltransferase

K. Shimoda, H. Katsuragi / Tetrahedron: Asymmetry 21 (2010) 2060–2065
2063
13
using molecular biological techniques is currently under investiga-
tion to understand the molecular basis of its properties.
H-6); C NMR (100 MHz, CD3OD): d 62.5 (C-60), 65.9 (C-7), 72.5
(C-40), 76.2 (C-20), 79.0 (C-30, C-50), 104.5 (C-10), 118.0 (C-3), 124.1
(C-5), 124.5 (C-1), 130.9 (C-4), 131.0 (C-6), 158.2 (C-2).
2-Hydroxymethylphenyl b- -glucoside 2: HRFABMS: m/z 309.1164
D
[M+Na]+ (calcd for C13H18O7Na: 309.1162); 1H NMR (400 MHz,
CD3OD): d 3.28–3.51 (4H, m, H-20, 30, 40, 50), 3.69 (1H, dd, J = 12.2,
5.1 Hz, H-6a0), 3.87 (1H, dd, J = 12.0, 1.6 Hz, H-6b0), 4.56 (1H, d,
J = 7.2 Hz, H-10), 4.77 (1H, d, J = 12.0 Hz, H-7a), 4.86 (1H, d,
J = 12.0 Hz, H-7b), 7.00 (1H, td, J = 7.2, 2.0 Hz, H-5), 7.20 (1H, dd,
J = 7.2, 2.0 Hz, H-3), 7.25 (1H, td, J = 7.2, 2.0 Hz, H-4), 7.32 (1H, dd,
4. Experimental section
4.1. General
Chemicals such as o-, m-, and p-hydroxybenzylalcohol, (R)- and
(S)-1-phenylethanol, and (R)- and (S)-1-phenylbutanol were pur-
chased from Sigma–Aldrich Co. The suspension cells of C. roseus
were cultured in 500 mL conical flasks containing 300 mL of Schenk
and Hildebrand (SH) medium supplemented with 3% sucrose and
10 mM 2,4-D under illumination (4000 lux).14 HPLC analyses were
carried out with Puresil C18 column (Waters) using MeOH/H2O
(1:3, v/v) as a solvent (detect, UV280; flow rate, 1 mL minꢀ1). GLC
analyses were performed with FID and a capillary column of Rt-
bDEX (0.25 mm ꢁ 30 m) using N2 as a carrier gas (injector, 180 °C;
detector, 180 °C; make up, 50 mL minꢀ1). Optical rotation data were
obtained on a Jasco DIP-360 using a 4 mL cuvette. 1H and 13C NMR,
H–H COSY, C–H COSY, and HMBC spectra were measured on a Varian
XL-400 spectrometer. HRFABMS spectra were taken on a JEOL MSta-
tion JMS-700 spectrometer.
13
J = 7.2, 2.0 Hz, H-6); C NMR (100 MHz, CD3OD): d 63.5 (C-60),
68.5 (C-7), 72.4 (C-40), 76.0 (C-20), 78.0, 78.7 (C-30, C-50), 104.1 (C-
10), 117.0 (C-3), 121.4 (C-5), 125.8 (C-1), 130.9 (C-4), 131.9 (C-6),
157.3 (C-2).
3-Hydroxybenzyl b- -glucoside 3: HRFABMS: m/z 309.1161
D
[M+Na]+ (calcd for C13H18O7Na: 309.1162); 1H NMR (400 MHz,
CD3OD): d 3.19–3.31 (4H, m, H-20, 30, 40, 50), 3.64 (1H, dd, J = 11.8,
5.1 Hz, H-6a0), 3.85 (1H, dd, J = 11.8, 2.0 Hz, H-6b0), 4.30 (1H, d,
J = 7.7 Hz, H-10), 4.55 (1H, d, J = 11.5 Hz, H-7a), 4.82 (1H, d,
J = 11.5 Hz, H-7b), 6.65 (1H, m, H-6), 6.81 (1H, m, H-2), 6.82 (1H,
dd, J = 8.2, 1.4 Hz, H-4), 7.08 (1H, t, J = 8.2 Hz, H-5); 13C NMR
(100 MHz, CD3OD): d 63.5 (C-60), 72.5 (C-40, C-7), 76.1 (C-20),
78.7, 78.9 (C-30, C-50), 103.9 (C-10), 116.7 (C-6), 117.0 (C-2), 121.0
(C-4), 131.2 (C-5), 141.5 (C-1), 159.5 (C-3).
4.2. Substrates
3-Hydroxymethylphenyl b- -glucoside 4: HRFABMS: m/z 309.1162
D
[M+Na]+ (calcd for C13H18O7Na: 309.1162); 1H NMR (400 MHz,
CD3OD): d 3.33–3.41 (4H, m, H-20, 30, 40, 50), 3.65 (1H, dd, J = 11.2,
5.1 Hz, H-6a0), 3.86 (1H, dd, J = 12.0, 2.0 Hz, H-6b0), 4.53 (2H, m, H-
7), 4.86 (1H, d, J = 7.6 Hz, H-10), 6.95–6.97 (2H, m, H-4, 6), 7.06 (1H,
t, J = 2.0 Hz, H-2), 7.20 (1H, t, J = 7.8 Hz, H-5); 13C NMR (100 MHz,
CD3OD): d 63.4 (C-60), 65.5 (C-7), 72.2 (C-40), 75.7 (C-20), 78.9 (C-30,
C-50), 103.2 (C-10), 117.0 (C-2), 117.6 (C-4), 122.7 (C-6), 131.2 (C-
5), 145.1 (C-1), 159.9 (C-3).
Substrates such as p-nitrophenyl b-
-galactoside 8, p-nitrophenyl b- -xyloside 9, and p-nitrophenyl
-arabinoside 10 were purchased from Sigma–Aldrich Co.
2-Hydroxybenzyl b- -glucoside 1, 2-hydroxymethylphenyl b-
glucoside 2, 3-hydroxybenzyl b- -glucoside 3, 3-hydroxymethyl-
phenyl b- -glucoside 4, 4-hydroxybenzyl b- -glucoside 5, and
4-hydroxymethylphenyl b- -glucoside 6 were prepared by almond
b- -glucosidase-catalyzed reverse-hydrolysis according to the pre-
viously reported procedures.2 A typical synthetic method using
almond b- -glucosidase is as follows. To the mixture (20 mL) of ace-
tonitrile and water (9:1 v/v) including 5 mmol of -glucose and
700 U of almond b- -glucosidase were added 1.5 mmol of the corre-
D-glucoside 7, p-nitrophenyl
b-D
D
a-L
D
D-
D
D
D
D
D
4-Hydroxybenzyl b- -glucoside 5: HRFABMS: m/z 309.1169
D
[M+Na]+ (calcd for C13H18O7Na: 309.1162); 1H NMR (400 MHz,
CD3OD): d 3.16–3.30 (4H, m, H-20, 30, 40, 50), 3.85 (1H, dd, J = 12.0,
2.0 Hz, H-6a0), 3.62 (1H, dd, J = 12.0, 5.5 Hz, H-6b0), 4.28 (1H, d,
J = 7.6 Hz, H-10), 4.51 (1H, d, J = 11.2 Hz, H-7a), 4.77 (1H, d,
J = 11.2 Hz, H-7b), 6.70 (2H, d, J = 8.6 Hz, H-2, 6), 7.19 (2H, d,
D
D
D
sponding aglycone alcohols. The reaction mixture containing an
aglycone alcohol was individually incubated for 48 h at 40 °C. Con-
centration of the reaction mixture by evaporation under reduced
pressure followed by purification using column chromatography
13
J = 8.6 Hz, H-3, 5); C NMR (100 MHz, CD3OD): d 63.5 (C-60), 72.5
(C-40, C-7), 75.9 (C-20), 78.8 (C-30, C-50), 103.7 (C-10), 116.8 (C-3,
C-5), 120.2 (C-1), 132.0 (C-2, C-6), 159.0 (C-4).
on silica gel afforded b-D-glucoside. The yields of 1–6 were 9%, 6%,
5%, 3%, 5%, and 3%, respectively.
4-Hydroxymethylphenyl b- -glucoside 6: HRFABMS: m/z 309.1160
D
[M+Na]+ (calcd for C13H18O7Na: 309.1162); 1H NMR (400 MHz,
CD3OD): d 3.35–3.41 (4H, m, H-20, 30, 40, 50), 3.65 (1H, dd, J = 12.0,
4.2 Hz, H-6a0), 3.85 (1H, dd, J = 12.0, 1.8 Hz, H-6b0), 4.51 (2H, m, H-
7), 4.70 (1H, d, J = 7.3 Hz, H-10), 7.03 (2H, d, J = 8.8 Hz, H-2, 6), 7.23
(2H, d, J = 8.8 Hz, H-3, 5); 13C NMR (100 MHz, CD3OD): d 63.3 (C-
60), 65.5 (C-7), 72.2 (C-40), 75.5 (C-20), 78.8, 78.9 (C-30, C-50), 103.2
(C-10), 118.2 (C-3, C-5), 120.2 (C-1), 130.9 (C-2, C-6), 159.2 (C-4).
In the case of preparation of substrates composed of a 1:1 mix-
ture of (RS)-diastereomers, that is, (RS)-1-phenylethyl b-
side 11 or (RS)-1-phenylbutyl b- -glucoside 12, each 0.05 mmol
of (R)- and (S)-1-phenylalkyl b- -glucosides, which had been indi-
vidually synthesized from enantiomerically pure alcohols by al-
mond b- -glucosidase-catalyzed reverse-hydrolysis under similar
D-gluco-
D
D
D
conditions to the standard reaction system, except that the scale
was fivefold enlarged (yields of 11a, 11b, 12a, and 12b were 3%,
5%, 2%, and 4%), and were combined to be a substrate. The R/S ratio
(RS)-1-Phenylethyl b- -glucoside 11: HRFABMS: m/z 307.1158
D
[M+Na]+ (calcd for C14H20O6Na: 307.1158); 1H NMR (400 MHz,
CD3OD): d 1.47 (3H, d, J = 6.4 Hz, H-2), 3.41–3.76 (6H, m, H-200,
300, 400, 500, 600), 4.11 (0.5H, d, J = 8.0 Hz, H-100), 4.41 (0.5H, d,
J = 7.9 Hz, H-100), 4.82 (0.5H, q, J = 6.4 Hz, H-1), 4.95 (0.5H, q,
J = 6.4 Hz, H-1), 7.20 (1H, m, H-40), 7.29–7.41 (4H, m, H-20, 30, 50,
of these substrates, that is, (RS)-1-phenylalkyl b-D-glucosides 11
and 12, was confirmed to be 1:1 as judged by 1H NMR analysis
of H-100 anomeric proton signals [Fig. 3(A) for 11 and Fig. 5(A)
for 12].
13
60); C NMR (100 MHz, CD3OD): d 22.9, 24.3 (C-2), 69.9 (C-600),
70.1 (C-400), 75.2, 75.3 (C-200), 76.0, 78.0 (C-1), 78.5, 78.9 (C-300, C-
500), 102.0, 102.5 (C-100), 128.0, 128.5 (C-20, C-60), 128.9, 129.0 (C-
40), 130.0, 130.2 (C-30, C-50), 143.9, 145.8 (C-10).
The spectral data of b-
as follows.
2-Hydroxybenzyl b-
D
-glucoside substrates 1–6, 11, and 12 are
D
-glucoside 1: HRFABMS: m/z 309.1166
[M+Na]+ (calcd for C13H18O7Na: 309.1162); 1H NMR (400 MHz,
CD3OD): d 3.23–3.37 (4H, m, H-20, 30, 40, 50), 3.67 (1H, dd, J = 12.2,
5.1 Hz, H-6a0), 3.87 (1H, dd, J = 12.0, 1.6 Hz, H-6b0), 4.38 (1H, d,
J = 7.6 Hz, H-10), 4.70 (1H, d, J = 12.0 Hz, H-7a), 4.91 (1H, d,
J = 12.0 Hz, H-7b), 6.76 (1H, d, J = 7.8 Hz, H-3), 6.79 (1H, t,
J = 7.6 Hz, H-5), 7.08 (1H, t, J = 7.8 Hz, H-4), 7.30 (1H, d, J = 7.2 Hz,
(RS)-1-Phenylbutyl b- -glucoside 12: HRFABMS: m/z 335.1471
D
[M+Na]+ (calcd for C16H24O6Na: 335.1469); 1H NMR (400 MHz,
CD3OD): d 0.90 (3H, m, H-4), 1.28–1.41 (2H, m, H-3), 1.64–1.70
(2H, m, H-2), 3.09–3.51 (6H, m, H-200, 300, 400, 500, 600), 4.15 (0.5H, d,
J = 8.0 Hz, H-100), 4.49 (0.5H, d, J = 8.0 Hz, H-100), 4.85 (0.5H, t,
J = 7.6 Hz, H-1), 4.90 (0.5H, t, J = 7.6 Hz, H-1), 7.22 (1H, m, H-40),

2064
K. Shimoda, H. Katsuragi / Tetrahedron: Asymmetry 21 (2010) 2060–2065
7.27–7.30 (4H, m, H-20, 30, 50, 60); 13C NMR (100 MHz, CD3OD): d
14.1 (C-4), 20.0, 20.1 (C-3), 42.4 (C-2), 70.4 (C-600), 71.9 (C-400),
75.9, 76.0 (C-200), 76.4, 77.7 (C-1), 77.9, 78.8 (C-300, C-500), 102.2,
102.5 (C-100), 126.8 (C-20, C-60), 128.0, 128.2 (C-40), 128.9, 129.0
(C-30, C-50), 145.1, 146.3 (C-10).
by evaporation gave the aglycone alcohol sample for chiral GLC
analysis. The absolute configuration, diastereomeric excess, and
enantiomeric excess of the aglycone moiety of 17 were determined
by the same method as the case of 16.
The retention times of enantiomerically pure alcohols in the chi-
ral GLC analyses were as follows: (R)-1-phenylethanol, 15.1 min;
(S)-1-phenylethanol, 16.0 min; (R)-1-phenylbutanol, 17.4 min;
and (S)-1-phenylbutanol, 17.9 min.
4.3. Purification of the enzyme
The plant xylosyltransferase was purified from the cultured C.
roseus cells. A typical purification method is as follows: cultured
cells of C. roseus (ca. 5 kg) were homogenated in 50 mM phosphate
buffer (pH 6.8). The homogenate (ca. 7.5 L) was centrifuged at
10,000g for 30 min and the supernatant was further centrifuged
at 100,000g for 2 h to give a microsomal enzyme fraction. Micro-
somal enzymes were solubilized from the microsomal fraction
with 0.1% Triton X-100. The enzyme, which has xylosylation activ-
The spectroscopic data of b-primeveroside products are as
follows.
2-Hydroxybenzyl b-primeveroside 13: HRFABMS: m/z 441.1376
[M+Na]+ (calcd for C18H26O11Na: 441.1373); 1H NMR (400 MHz,
CD3OD): d 3.15–3.70 (8H, m, H-20, 200, 30, 300, 40, 400, 50, 5a00), 3.72
(1H, dd, J = 11.0, 5.5 Hz, H-6a0), 3.82 (1H, dd, J = 11.0, 5.5 Hz, H-
5b00), 4.08 (1H, dd, J = 11.0, 1.8 Hz, H-6b0), 4.33 (1H, d, J = 7.6 Hz,
H-100), 4.36 (1H, d, J = 7.8 Hz, H-10), 4.69 (1H, d, J = 12.0 Hz, H-7a),
4.85 (1H, d, J = 12.0 Hz, H-7b), 6.75 (1H, d, J = 7.4 Hz, H-3), 6.77
(1H, t, J = 7.4 Hz, H-5), 7.08 (1H, t, J = 7.4 Hz, H-4), 7.30 (1H, d,
ity against the substrate 3-hydroxybenzyl b-D-glucoside 3, was
purified by chromatography on a DEAE-Toyopearl column, Sepha-
dex G-150 column, and REACTIVE GREEN 19 agarose-gel column to
give homogeneous xylosyltransferase (ca. 22 mL).
13
J = 7.4 Hz, H-6); C NMR (100 MHz, CD3OD): d 67.7 (C-500), 68.8
(C-7), 70.5 (C-60), 72.0 (C-400), 72.2 (C-40), 75.7 (C-200), 76.0 (C-20),
78.0, 78.8 (C-30, C-300, C-50), 103.5 (C-10), 106.0 (C-100), 117.2 (C-3),
121.4 (C-5), 125.9 (C-1), 130.9 (C-4), 132.0 (C-6), 157.5 (C-2).
3-Hydroxybenzyl b-primeveroside 14: HRFABMS: m/z 441.1370
[M+Na]+ (calcd for C18H26O11Na: 441.1373); 1H NMR (CD3OD): d
3.17–3.51 (8H, m, H-20, 200, 30, 300, 40, 400, 50, 5a00), 3.70 (1H, dd,
J = 12.0, 6.0 Hz, H-6a0), 3.82 (1H, dd, J = 11.5, 5.5 Hz, H-5b00), 4.07
(1H, dd, J = 12.0, 2.0 Hz, H-6b0), 4.30 (1H, d, J = 7.6 Hz, H-100), 4.32
(1H, d, J = 7.6 Hz, H-10), 4.54 (1H, d, J = 11.5 Hz, H-7a), 4.80 (1H,
d, J = 11.5 Hz, H-7b), 6.65 (1H, dt, J = 7.8, 2.5 Hz, H-6), 6.81–6.85
(2H, m, H-2, 4), 7.09 (1H, t, J = 7.8 Hz, H-5); 13C NMR (CD3OD): d
67.7 (C-500), 70.5 (C-60), 71.9 (C-400), 72.2 (C-40), 72.5 (C-7), 75.5
(C-200), 75.7 (C-20), 77.7, 78.8 (C-30, C-300, C-50), 103.9 (C-10), 106.0
(C-100), 116.8 (C-6), 119.0 (C-2), 120.9 (C-4), 130.9 (C-5), 141.0 (C-
1), 159.0 (C-3).
4.4. Xylosylation conditions
In the enzyme purification procedures, the enzyme assay was
performed using a vial containing 0.5 mL of enzyme fraction,
300 lg of 3-hydroxybenzyl b-D-glucoside 3, and 1 mg of UDP-xy-
lose. The reaction mixture was incubated at 37 °C for 30 h and
was then extracted with n-butanol (1 mL ꢁ 3). The n-butanol frac-
tion was analyzed by HPLC on a Puresil C18 column.
For experiments on substrate specificity and kinetic resolution,
the substrate (10 mg) was administered to 15 mL of 50 mM HEPES
buffer (pH 7.0) containing ca. 15 lg of isolated xylosyltransferase
and 40 mg of UDP-xylose and was incubated at 37 °C. The reaction
was monitored by HPLC analyses using a Puresil C18 column. The
incubation times of each substrate were 36 h for 1–10, 52 h for 11,
and 61 h for 12. The conversion yield of the glycoside products was
determined by HPLC analyses and the yield was calculated using
calibration curves. Extraction from the reaction mixture with
n-butanol (20 mL ꢁ 3) followed by preparative HPLC gave the
product. In order to obtain products adequate for specific rotation
analysis, the reaction was performed under similar conditions to
the standard reaction system except that the scale was 10-fold
enlarged.
4-Hydroxybenzyl b-primeveroside 15: HRFABMS: m/z 441.1377
[M+Na]+ (calcd for C18H26O11Na: 441.1373); 1H NMR (CD3OD): d
3.15–3.51 (8H, m, H-20, 200, 30, 300, 40, 400, 50, 5a00), 3.70 (1H, dd,
J = 12.0, 6.0 Hz, H-6a0), 3.82 (1H, dd, J = 11.5, 5.5 Hz, H-5b00), 4.08
(1H, dd, J = 12.0, 2.0 Hz, H-6b0), 4.30 (1H, d, J = 7.6 Hz, H-100), 4.35
(1H, d, J = 7.8 Hz, H-10), 4.51 (1H, d, J = 11.2 Hz, H-7a), 4.77 (1H,
d, J = 11.2 Hz, H-7b), 6.70 (2H, d, J = 8.6 Hz, H-2, 6), 7.19 (2H, d,
13
J = 8.6 Hz, H-3, 5); C NMR (CD3OD): d 67.6 (C-500), 70.4 (C-60),
71.9 (C-400), 72.2 (C-40), 72.5 (C-7), 75.5 (C-200), 75.7 (C-20), 77.8,
78.8 (C-30, C-300, C-50), 103.7 (C-10), 106.0 (C-100), 116.8 (C-3, C-5),
120.0 (C-1), 132.0 (C-2, C-6), 159.0 (C-4).
4.5. Product identification
(R)-1-Phenylethyl b-primeveroside 16a obtained by xylosylation of
The structures of products 13–15 were determined by spectro-
scopic methods including HRFABMS, 1H, and 13C NMR, H–H COSY,
C–H COSY, and HMBC spectra as 2-hydroxybenzyl b-primeveroside
13, 3-hydroxybenzyl b-primeveroside 14, and 4-hydroxybenzyl
b-primeveroside 15. Product 16 was identified as (R)-1-phenyl-
ethyl b-primeveroside by direct comparison of its 1H NMR
spectrum with those of authentic (R)- and (S)-1-phenylethyl
b-primeverosides, which had been prepared by chemosynthetic
procedures according to the previously reported method12 to give
(RS)-1-phenylethyl b-D-glucoside 11 with C. roseus xylosyltransferase:
HRFABMS: m/z 439.1580 [M+Na]+ (calcd for C19H28O10Na:
439.1580); 1H NMR (CD3OD): d 1.45 (3H, d, J = 6.4 Hz, H-2), 3.09–
3.45 (8H, m, H-200, 2000, 300, 3000, 400, 4000, 500, 5a000), 3.68 (1H, dd,
J = 11.6, 5.8 Hz, H-6a00), 3.79 (1H, dd, J = 11.6, 4.1 Hz, H-5b000), 3.97
(1H, dd, J = 11.6, 1.9 Hz, H-6b00), 4.27 (1H, d, J = 7.6 Hz, H-1000),
4.46 (1H, d, J = 8.2 Hz, H-100), 4.96 (1H, q, J = 6.4 Hz, H-1), 7.20
(1H, t, J = 7.3 Hz, H-40), 7.30 (2H, t, J = 7.3 Hz, H-30, 50), 7.41 (2H,
d, J = 7.3 Hz, H-20, 60); 13C NMR (CD3OD): d 23.0 (C-2), 67.7 (C-
5000), 70.4 (C-600), 72.0 (C-4000), 72.2 (C-400), 75.7 (C-2000), 76.2 (C-200),
78.0, 79.0 (C-300, C-3000, C-500), 78.3 (C-1), 103.3 (C-100), 106.0
(C-1000), 128.2 (C-20, C-60), 129.0 (C-40), 130.0 (C-30, C-50), 145.9
(C-10).
a mixture of a- and b-anomers (ca. 1:49). The diastereomeric com-
position of product 16 was determined based on the intensities of
H-1000 anomeric proton signals of (R)- and (S)-1-phenylethyl
b-primeverosides in the 1H NMR spectrum of product 16. In addi-
tion, the enantiomeric excess of the aglycone moiety of product 16
was confirmed based on the peak areas of each enantiomer in the
chiral GLC analysis of the aglycone alcohol which had been pre-
pared by hydrolysis of 16 with 4.0 M HCl. Product 16 was added
to a vial containing 100 lL of 4.0 M HCl, heated to 80 °C for 2 h,
then cooled to room temperature, and diluted with 1 mL of water.
Extraction from the mixture with ethylacetate (2 mL ꢁ 3) followed
(R)-1-Phenylbutyl b-primeveroside 17a obtained by xylosylation of
(RS)-1-phenylbutyl b- -glucoside 12 with C. roseus xylosyltransferase:
D
HRFABMS: m/z 467.1899 [M+Na]+ (calcd for C21H32O10Na:
467.1893); 1H NMR (CD3OD): d 0.92 (3H, t, J = 7.6 Hz, H-4), 1.28–
1.39 (2H, m, H-3), 1.64–1.73 (2H, m, H-2), 3.08–3.51 (8H, m, H-
200, 2000, 300, 3000, 400, 4000, 500, 5a000), 3.68 (1H, dd, J = 11.6, 5.9 Hz, H-
6a00), 3.78 (1H, dd, J = 11.5, 3.9 Hz, H-5b000), 4.01 (1H, dd, J = 11.6,

K. Shimoda, H. Katsuragi / Tetrahedron: Asymmetry 21 (2010) 2060–2065
2065
2.0 Hz, H-6b00), 4.30 (1H, d, J = 7.8 Hz, H-1000), 4.48 (1H, d, J = 7.9 Hz,
H-100), 4.90 (1H, t, J = 7.6 Hz, H-1), 7.22 (1H, m, H-40), 7.28–7.32 (4H,
m, H-20, 30, 50, 60); 13C NMR (CD3OD): d 14.3 (C-4), 20.0 (C-3), 42.4
(C-2), 67.7 (C-5000), 70.4 (C-600), 71.9 (C-4000), 72.0 (C-400), 75.7 (C-2000),
76.0 (C-200), 77.8 (C-1), 77.9, 78.9 (C-300, C-3000, C-500), 102.9 (C-100),
105.5 (C-1000), 126.8 (C-20, C-60), 127.9 (C-40), 129.0 (C-30, C-50),
146.3 (C-10).
3. Minami, A.; Uchida, R.; Eguchi, T.; Kakinuma, K. J. Am. Chem. Soc. 2005, 127,
6148.
4. Minami, A.; Kakinuma, K.; Eguchi, T. Tetrahedron Lett. 2005, 46, 6187.
5. Wang, L.-X. Carbohydr. Res. 2008, 343, 1509.
6. Marton, X.; Tran, V.; Tellier, C.; Dion, M.; Drone, J.; Rabiller, C. Carbohydr. Res.
2008, 343, 2939.
7. Gais, H. J.; Zeissler, A.; Maidonis, P. Tetrahedron Lett. 1988, 29, 5743.
8. Shimoda, K.; Kubota, N.; Hamada, H. Tetrahedron: Asymmetry 2004, 15, 2319.
9. Lim, E.-K.; Doucet, C. J.; Hou, B.; Jackson, R. G.; Abrams, S. R.; Bowles, D. J.
Tetrahedron: Asymmetry 2005, 16, 143.
10. Hirata, T.; Shimoda, K.; Fujino, T.; Yamane, S.; Ohta, S. Bull. Chem. Soc. Jpn. 2001,
74, 539.
Acknowledgment
11. Ushiyama, M.; Furuya, T. Phytochemistry 1989, 28, 2333.
12. Ma, S. J.; Mizutani, M.; Hiratake, J.; Hayashi, K.; Yagi, K.; Watanabe, N.; Sakata,
K. Biosci., Biotechnol., Biochem. 2001, 65, 2719.
13. Chen, C. S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. Am. Chem. Soc. 1982, 104, 7294.
14. Schenk, R. U.; Hildebrand, A. C. Can. J. Bot. 1972, 50, 199.
This work was supported by Grant-in-Aid for Young Scientists
(B) (No. 19790018) from The Ministry of Education, Culture, Sports,
Science and Technology of Japan.
References
1. Vic, G.; Crout, D. H. G. Tetrahedron: Asymmetry 1994, 5, 2513.
2. Vic, G.; Crout, D. H. G. Carbohydr. Res. 1995, 279, 315.