Diastereoselective formation of disaccharides from (RS)-1-phenylethanol by cultured cells of Catharanthus roseus

Article abstract of DOI:10.1246/bcsj.74.539

The biotransformation of (RS)- 1-phenylethanol with cultured cells of Catharanthus roseus gave its vicianoside [α-L-arabinopyranosyl-(1-6)-β-D-glucopyranoside], primeveroside [β-D-xylopyranosyl-(1-6)-β-D-glucopyranoside] and gentiobioside [β-D-glucopyrano

Full text of DOI:10.1246/bcsj.74.539

© 2001 The Chemical Society of Japan
539
Bull. Chem. Soc. Jpn., 74, 539–542 (2001)
Diastereoselective Formation of Disaccharides from (RS)-1-
Phenylethanol by Cultured Cells of Catharanthus roseus
Toshifumi Hirata,* Kei Shimoda, Takeshi Fujino, Shin-ya Yamane, and Shinji Ohta†
Department of Mathematical and Life Sciences, Graduate School of Science, Hiroshima University,
1-3-1 Kagamiyama, Higashi-Hiroshima 739-8526
†Instrument Center for Chemical Analysis, Hiroshima University, 1-3-1 Kagamiyama, Higashi-Hiroshima 739-8526
(Received September 22, 2000)
The biotransformation of (RS)-1-phenylethanol with cultured cells of Catharanthus roseus gave its vicianoside [α-
L-arabinopyranosyl-(1-6)-β-D-glucopyranoside], primeveroside [β-D-xylopyranosyl-(1-6)-β-D-glucopyranoside] and
gentiobioside [β-D-glucopyranosyl-(1-6)-β-D-glucopyranoside] via the corresponding glucoside. It was found that pri-
meveroside and gentiobioside were enantioselectively composed of only (R)-1-phenylethanol as its aglycone moiety.
A glucosylation reaction of alcohols with biocatalysts is
useful for the preparation of pharmacologically important sub-
stances. Although there are many reports of the glycosylation
of alcohols with cultured plant cells,1–10 little attention has been
paid to the enantioselectivity in the glycosylation reaction.6 We
recently found that cultured cells of Catharanthus roseus are
able to transform an exogenous secondary alcohol to a mono-
glucoside having the (R)-configuration at its aglycone moi-
ety.11,12 During the course of a study on enantioselective glyco-
sylation by cultured plant cells, we took into consideration the
diastereoselective formation of a disaccharide, and report here
that the cultured cells of C. roseus were capable diastereoselec-
tively of transforming 1-phenylethanol into disaccharides, such
as vicianoside, primeveroside and gentiobioside, via the corre-
sponding mono-glucoside.
with the cultured cells of C. roseus occurred enantioselectively
to preferentially give the mono-glucoside of the (R)-alcohol.11
In order to clarify the enantioselectivity in the glycosylation
reaction, the time courses for the biotransformation of the
(RS)-alcohol 1 were examined; the results are given in Table 2.
From an early stage of the incubation, 1 was transformed to the
corresponding mono-glucoside 5 and disaccharides 2–4. Fur-
ther transformation products were not found during the time-
course period, despite careful analysis of the products by
HPLC. The yield of the mono-glucoside 5 was at its maximum
in the 5-days incubation, and then decreased. The diastereo-
meric excess (d.e.) of 5 was over 60% (R-enantioselective), but
gradually decreased with lapse of the incubation time. during
10 days of incubation, 5 was composed preferentially of the
(S)-alcohol. On the other hand, the yields of the disaccharides,
2–4, increased during the time course period, especially after
5–7 days of incubation. Interestingly, the configurations of the
aglycone moiety of the primeveroside 3 and gentiobioside 4
were R (100% e.e.). In contrast, a diastereomeric excess of the
vicianoside 2 decreased with decrease in the diastereomeric ex-
cess of the mono-glucoside 5. These observations indicate that
the (RS)-alcohol 1 was first transformed R-enantioselectively
to the mono-glucoside 5 (about 60% d.e.), and then the diaste-
reomer 5a was subsequently transformed to disaccharides,
such as vicianoside 2a, primeveroside 3 and gentiobioside 4,
whereas 5b was converted to only vicianoside 2b. Such a dif-
ference of the diastereoselectivity in the formation of disaccha-
rides may reflect decrease in the diastereomeric purity of the
remaining 5.
Results and Discussion
(RS)-, (R)-, and (S)-1-phenylethanol (1, 1a, and 1b) were
separately administered to a cultured cell suspension of C. ro-
seus; the cultures were then incubated at 25 °C for 5–10 days.
The products were isolated by preparative TLC and HPLC to
give disaccharides, such as vicianoside 2, primeveroside 3 and
gentiobioside 4 (Chart 1, Table 1), as well as 1-phenylethyl β-
D-glucopyranoside (5), which was reported in our previous pa-
per.11 The structures of the products were determined by means
of FABMS and 1H- and 13C NMR measurements. The viciano-
side 2 obtained from incubating (RS)-1-phenylethanol (1) was
found to be a mixture of diastereomers, 2a composed of the R-
alcohol and 2b composed of the S-alcohol. The primeveroside
3 and gentiobioside 4 were composed of only the aglycone
with the R-configuration (each 100% e.e.). On the other hand,
1-phenylethyl β-D-glucopyranoside (5) obtained from the
(RS)-alcohol was preferentially composed of the (S)-alcohol
(12% e.e.), although we previously found that glucosylation
It was thus demonstrated that the cultured cells of C. roseus
were able to glycosylate 1-phenylethanol stepwise to give the
corresponding vicianoside, primeveroside and gentiobioside.
The xylosylation and glucosylation of 1-phenylethyl glucoside
(5) occurred diastereospecifically to give the disaccharides

540 Bull. Chem. Soc. Jpn., 74, No. 3 (2001)
Formation of Disaccharides by Cultured Cells
Table 1. Biotransformation of (RS )-, (R)-, and (S )-1-
Phenylethanol (1, 1a, and 1b) by the Cultured Cells of
C. roseus
Substrate Incubation time Product Yield d.e. Configurationa)
/d
/% /%
1
10
2
3
4
5
2
3
4
5
2
5
7.8
5.3 100
2.6 100
0
RS
R
R
S
R
R
R
R
S
4.1
14
1a
1b
5
5
1.5 100
2.1 100
0.9 100
9.5 100
1.0 100
3.7 100
S
a) Preferred configuration of the aglycone.
Table 2. Time Courses in the Biotransformation of (RS )-
1-Phenylethanol (1) by the cultured Cells of C. roseus
Incubation time Product Yield/% d.e./% Configurationa)
/d
1
2
3
4
5
2
3
4
5
2
3
4
5
2
3
4
5
2
3
4
5
0.8
tr.b)
tr.
51
—
—
63
48
100
100
62
R
—
—
R
R
R
R
R
R
R
R
R
R
R
R
R
RS
R
R
1.8
2.0
0.7
0.6
5.2
2.8
1.1
1.0
15.0
4.5
2.5
2.1
13.2
8.0
6.1
3.8
3.5
3
5
29
100
100
49
7
25
100
100
35
10
0
100
100
12
S
Chart 1.
a) Preferred configuration of the aglycone. b) Trace amount.
with only (R)-1-phenylethanol, but the arabinosylation was not
high in diastereoselectivity. To our knowledge, there are only
a few enzymatic studies on enantioselective glycosylation.
Matsumura et al. recently reported that β-galactosidase from
E. coli are able to catalyze R-enantioselective galactosylation
of 1-phenylethanol by its reverse hydrolytic reaction.13 It is
worth noting that the enantioselective glycosylation with bio-
catalysts may be useful for preparing of diastereomeric pure
glycosides and olygosaccharides. For the practical use of those
enzymes, further investigations at the enzymatic level are nec-
essary, and are now in progress in our laboratory.
gel 60; GF254). NMR spectra were measured in MeOH-d4 on a
JEOL GSX500 [500 MHz (1H) and 125.8 MHz (13C)] NMR spec-
trometer. FABMS were taken on a JEOL SX102A mass spectrom-
eter. GLC was carried out with FID and an Alltech Chirasil-Val
capillary column (0.25 mm × 25 m) using He as carrier gas (1
cm3 min–1) at column temp: 50–100 °C (1 °C min–1) and HPLC
with Puresil C18 (Waters) column using MeOH:H2O (1:3 v/v) as
eluent. (RS)-1-Phenylethanol (1), (R)-1-phenylethanol (1a), and
(S)-1-phenylethanol (1b) were purchased from Wako Pure Chemi-
cal Industries. Ltd.
Incubation of (RS)-1-Phenylethanol (1) with the Cultured
Cells of C. roseus. Suspension cells of Catharanthus roseus14
were cultivated at 25 °C for 14 days in 1000 cm3 conical flasks con-
taining 600 cm3 of SH medium15 supplemented with 3% of sucrose
and 10 mM (1 M = 1 mol dm–3) of 2,4-dichlorophenoxyacetic acid
under illumination (4000 lux) on a rotary shaker (75 rpm).
Experimental
General. Analytical and prep. TLC was carried out on glass
sheets (0.25 mm and 0.5 mm) coated with silica gel (Merck silica

T. Hirata et al.
Bull. Chem. Soc. Jpn., 74, No. 3 (2001) 541
1′), 3.23 (1H, dd, J = 9.3, 7.8 Hz, H-2′), 3.31 (1H, m, H-3′), 3.37
(1H, t, J = 7.8 Hz, H-4′), 3.41 (1H, ddd, J = 7.8, 5.9, 2.0 Hz, H-5′),
3.73 (1H, dd, J = 11.7, 5.9 Hz, H-6a′), 4.03 (1H, dd, J = 11.7, 2.0
Hz, H-6b′), 4.31 (1H, d, J = 7.8 Hz, H-1′′), 3.16 (1H, dd, J = 8.3,
7.8 Hz, H-2′′), 3.30 (1H, m, H-3′′), 3.26 (1H, t, J = 8.8 Hz, H-4′′),
3.15 (1H, m, H-5′′), 3.63 (1H, dd, J = 11.7, 5.9 Hz, H-6a′′), 3.82
(1H, dd, J = 11.7, 2.4 Hz, H-6b′′); 13C NMR (CD3OD) see Table 3.
Incubation of (R)-1-Phenylethanol (1a) and (S)-1-Phenyl-
Table 3. 13C Chemical Shifts of the Products, 2a, 2b, 3
and 4, in the Biotransformation of 1-Phenylethanol (1)
by the Cultured Cells of C. roseus
Position
2a
2b
3
4
C-1
C-2
78.3
23.1
77.0
25.5
78.3
23.0
78.2
23.0
C-3
C-4
C-5
C-6
C-7
C-8
146.0
128.2
129.9
129.0
129.9
128.2
103.4
76.0
144.9
128.8
130.2
129.4
130.2
128.8
102.0
76.0
145.9
128.3
129.9
129.0
129.9
128.3
103.3
76.0
145.9
128.3
129.9
129.0
129.9
128.3
103.2
76.0
ethanol (1b) with the Cultured Cells of C. roseus.
According
to the same procedure as described above, 1a and 1b were trans-
formed into the corresponding glucosides. Samples of 1a and 1b
(60 mg) in MeOH (0.6 cm3) were administered to flask containing
the suspended cells (150 g), and the cultures were incubated at 25
°C for 5 days on a rotary shaker (75 rpm). The products were ex-
tracted and purified in the same way as for the biotransformation of
1. Purification of the products with preparative TLC and HPLC
gave the glycosides, 2a (3 mg), 3 (4 mg), 4 (2 mg) and 5a (14 mg)
from 1a, while the glycosides, 2b (2 mg) and 5b (5 mg) from 1b.
C-1ꢀ
C-2ꢀ
C-3ꢀ
C-4ꢀ
C-5ꢀꢀ
C-6ꢀꢀ
C-1ꢀꢀ
C-2ꢀꢀ
C-3
78.8a)
72.5
78.6b)
72.5
78.9c)
72.3
78.7d)
72.4
78.0a)
70.0
105.8
73.2
74.9
70.3
77.6b)
70.3
106.0
73.2
75.1
70.3
78.0c)
70.4
78.1d)
70.5
1-O-[α-
L
-arabinopyranosyl-(1-6)-β-D-glucopyranosyl]-(R)-1-
phenylethanol (2a): FABMS m/z 439 [M+Na]+; 1H NMR
(CD3OD) δ 4.96 (1H, q, J = 6.4 Hz, H-1), 1.48 (3H, d, J = 6.4 Hz,
H-2), 7.41 (2H, d, J = 7.3 Hz, H-4,8), 7.30 (2H, t, J = 7.3 Hz, H-
5,7), 7.21 (1H, t, J = 7.3 Hz, H-6), 4.46 (1H, d, J = 7.3 Hz, H-1′),
3.23 (1H, dd, J = 7.3, 8.7 Hz, H-2′), 3.32 (1H, m, H-3′), 3.27 (1H,
m, H-4′), 3.36 (1H, m, H-5′), 3.67 (1H, dd, J = 11.6, 6.7 Hz, H-6a′),
3.97 (1H, dd, J = 11.6, 2.1 Hz, H-6b′), 4.24 (1H, d, J = 7.0 Hz, H-
1′′), 3.53 (1H, dd, J = 8.8, 6.7 Hz, H-2′′), 3.41 (1H, dd, J = 8.8, 3.4
Hz, H-3′′), 3.73 (1H, m, H-4′′), 3.30 (1H, m, H-5a′′), 3.77 (1H, dd,
J = 12.5, 3.4 Hz, H-5b′′); 13C NMR (CD3OD) see Table 3.
106.2
75.7
105.6
76.0
78.3c)
72.0
78.9d)
72.4
C-4ꢀꢀ
C-5ꢀꢀ
C-6ꢀꢀ
67.4
—
67.5
—
67.6
—
78.7d)
63.5
a)–d) Assignments may be interchanged.
(RS)-1-phenylethanol (1) (20 mg) in MeOH (0.35 cm3) and glu-
cose (1.3 g) were administered to the cultured suspension cells (50
g), and the cultures were incubated at 25 °C for 10 days on a rotary
shaker (75 rpm). After incubation, the cells and medium were sep-
arated by filtration with suction. The cells were extracted with
MeOH and the extract was concentrated by evaporation in vacuo.
After the methanolic fraction was partitioned between H2O and
EtOAc, the aqueous layer was further extracted with BuOH. The
filtered medium was extracted with BuOH. The BuOH extracts
from the cells and the culture medium were combined and purified
by preparative TLC with MeOH:EtOAc (1:3) and HPLC with a
Puresil C18 column using MeOH:H2O (1:3 v/v) to give viciano-
side 2, primeveroside 3, gentiobioside 4, and 1-phenylethyl β-D-
glucopyranoside 5.12 The structures of the products were deter-
mined by means of FABMS, 1H and 13C NMR, H–H COSY, C–H
COSY, and HMBC measurements. The diastereomeric excess of 2
was determined by the intensities of the anomeric proton signals of
the arabinose moiety in its 1H NMR spectra.
1-O-[α-
L
-arabinopyranosyl-(1-6)-β-D-glucopyranosyl]-(S)-1-
phenylethanol (2b): FABMS m/z 439 [M+Na]+; 1H NMR
(CD3OD) δ 5.01 (1H, q, J = 7.3 Hz, H-1), 1.45 (3H, d, J = 7.3 Hz,
H-2), 7.42 (2H, d, J = 7.3 Hz, H-4,8), 7.31 (2H, t, J = 7.3 Hz, H-
5,7), 7.24 (1H, t, J = 7.3 Hz, H-6), 4.06 (1H, d, J = 7.3 Hz, H-1′),
3.23 (1H, dd, J = 7.3,8.5 Hz, H-2′), 3.20 (1H, t, J = 8.5 Hz, H-3′),
3.32 (1H, dd, J = 8.5, 10.1 Hz, H-4′), 3.28 (1H, m, H-5′), 3.73 (1H,
dd, J = 11.0, 6.1 Hz, H-6a′), 4.06 (1H, dd, J = 11.0, 1.8 Hz, H-6b′),
4.34 (1H, d, J = 6.4 Hz, H-1′′), 3.62 (1H, dd, J = 8.2, 6.4 Hz, H-2′′),
3.55 (1H, dd, J = 8.2, 1.8 Hz, H-3′′), 3.81 (1H, m, H-4′′), 3.55 (1H,
dd, J = 12.2, 1.8 Hz, H-5a′′), 3.87 (1H, dd, J = 12.2, 2.4 Hz, H-
5b′′); 13C NMR (CD3OD) see Table 3.
Determination ofAbsolute Configuration of Sugar Moieties.
Each of the disaccharides, 2a, 2b, 3, and 4, was added to the vial in
4.0 M HCl (100 mm3) and heated to 80 °C for 2 h and then cooled.
After removing of the solvent in a stream of N2, each hydrolysate
was converted into a pentafluoropropionate with pentafluoropropi-
onic anhydride (400 mm3) in CH2Cl2 (400 µL) in a sealed tube at
120 °C for 2 h. Excess reagents were removed under a stream of N2
and the derivatives were analyzed by chiral GC on Chirasil-Val.
The peaks of the derivatives from vicianosides 2 and 2b were as-
signed to those of L-arabinose and D-glucose, peaks of the deriva-
tives from primeveroside 3 to those of D-xylose and D-glucose, and
peaks of the derivatives from gentiobioside 4 to that of D-glucose.
Time Course Experiments in the Glycosylation of (RS)-1-
Phenylethanol (1). Each 50 g of the suspension cells of C. roseus
was portioned to 5 flasks containing 100 ml of the SH medium.
(RS)-1-Phenylethanol (1) (21 mg) and glucose (1.3 g) were admin-
istered to the flasks and the mixtures were incubated on a rotary
shaker (75 rpm) at 25 °C. At a regular time interval, the cells and
medium were separated by filtration. The cells were extracted with
MeOH and the extract was concentrated by evaporation in vacuo.
The methanolic fraction was combined with the medium and the
1-O-[β-
D
-xylopyranosyl-(1-6)-β-D-glucopyranosyl]-(R)-1-
phenylethanol (3) (primeveroside): FABMS m/z 439 [M+Na]+;
1H NMR (CD3OD) δ 4.96 (1H, q, J = 6.4 Hz, H-1), 1.45 (3H, d, J =
6.4 Hz, H-2), 7.41 (2H, d, J = 7.3 Hz, H-4,8), 7.30 (2H, t, J = 7.3
Hz, H-5,7), 7.21 (1H, t, J = 7.3 Hz, H-6), 4.46 (1H, d, J = 8.3 Hz, H-
1′), 3.25 (1H, m, H-2′), 3.38 (1H, m, H-3′), 3.32 (1H, m, H-4′),
3.40 (1H, m, H-5′), 3.67 (1H, dd, J = 11.6, 5.8 Hz, H-6a′), 3.97 (1H,
dd, J = 11.6, 1.8 Hz, H-6b′), 4.26 (1H, d, J = 7.3 Hz, H-1′′), 3.16
(1H, dd, J = 8.8, 7.6 Hz, H-2′′), 3.28 (1H, m, H-3′′), 3.44 (1H, m,
H-4′′), 3.08 (1H, dd, J = 11.6, 10.1 Hz, H-5a′′), 3.79 (1H, dd, J =
11.6, 3.4 Hz, H-5b′′); 13C NMR (CD3OD) see Table 3.
1-O-[β-
D
-glucopyranosyl-(1-6)-β-D-glucopyranosyl]-(R)-1-
phenylethano
l
(4) (gentiobioside): FABMS m/z 447 [M+H]+;
1H NMR (CD3OD) δ 4.97 (1H, q, J = 6.4 Hz, H-1), 1.47 (3H, d, J =
6.4 Hz, H-2), 7.42 (2H, d, J = 7.3 Hz, H-4,8), 7.30 (2H, t, J = 7.3
Hz, H-5,7), 7.22 (1H, t, J = 7.3 Hz, H-6), 4.47 (1H, d, J = 7.8 Hz, H-

542 Bull. Chem. Soc. Jpn., 74, No. 3 (2001)
Formation of Disaccharides by Cultured Cells
mixture was extracted with EtOAc. The aqueous phase was further
extracted with n-BuOH. The butanolic fraction was subjected to
preparative TLC on silica gel with MeOH:EtOAc (1:3 v/v) to give
the glucoside and disaccharide fractions. The glucoside fraction
was further subjected to preparative HPLC with Puresil C18 col-
umn using MeOH:H2O (1:3 v/v) to give glucoside 5, of which the
diasteromeric excess was determined by the peak intensities of the
anomeric proton signals for 5a and 5b in its 1H NMR spectrum. On
the other hand, the disaccharide fraction was subjected to reverse
phase TLC with MeOH:H2O (1:1 v/v) and then prep. HPLC with
a Puresil C18 column using MeOH:H2O (1:3 v/v) to give disac-
charides 2–4. The diasteromeric excesses of the disaccharides 2–4
were determined by the ratio of the peak intensities of their corre-
sponding anomeric proton signals in their 1H NMR spectra.
3
M. Tabata, Y. Umetani, M. Ooya, and S. Tanaka, Phy-
tochemistry, 27, 809 (1988).
4
M. Ushiyama, Y. Asada, T.Yoshikawa, and T. Furuya, Phy-
tochemistry, 28, 1859 (1989).
5
T. Furuya, Y. Orihara, and H. Miyatake, J. Chem. Soc., Per-
kin Trans. 1, 1989, 1711.
6
M. Ushiyama and T. Furuya, Phytochemistry, 28, 2333 and
3009 (1989).
7
Y. Orihara, K. Saiki, and T. Furuya, Phytochemistry, 30,
3989 (1991).
8
Y. Orihara, T. Furuya, N. Hashimoto, Y. Deguchi, K.
Tokoro, and T. Kanisawa, Phytochemistry, 31, 827 (1992).
9
T. Furuya, Y. Asada, Y. Matsuura, S. Mizobata, and H.
Hamada, Phytochemistry, 46, 1355 (1997).
10 K. Kawaguchi, S. Koike, M. Hirotani, M. Fujihara, T.
Furuya, R. Iwata, and K. Morimoto, Phytochemistry, 47, 1261
(1998).
11 Y. S. Lee, D. I. Ito, K. Koya, S. Izumi, and T. Hirata, Chem.
Lett., 1996, 719.
12 T. Hirata, K. Koya, K. J. Sarfo, K. Shimoda, D. I. Ito, S.
Izumi, S. Ohta, and Y. S. Lee, J. Mol. Cat. B: Enzym, 6, 67 (1999).
13 S. Matsumura, H. Yamazaki, and K. Toshima, Biotech.
Lett., 19, 583 (1997).
The authors thank Dr. L. J. Goad of the University of Liver-
pool, UK for critical reading of the manuscript. The authors
also thank the Instrument Center for Chemical Analysis of
Hiroshima University for the measurements of 1H and
13C NMR and GC-MS.
References
1
H. Mizukami, T. Terao, A. Amano, and H. Ohashi, Plant
Cell Physiol., 27, 645 (1986).
T. Furuya, M. Ushiyama, Y. Asada, T. Yoshikawa, and Y.
Orihara, Phytochemistry, 27, 803 (1988).
14 H. Hamada, Y. Fuchikami, Y. Ikematsu, T. Hirata, H.
Williams, and A. I. Scott, Phytochemistry, 37, 1037 (1994).
15 R. U. Shenk and A. C. Hildebrand, Can. J. Botany, 50, 199
(1972).
2