Enantioselective glucosylation of (±)-secondary alcohols with plant glucosyltransferases

Article abstract of DOI:10.1016/j.tetasy.2004.06.022

Two glucosyltransferases were isolated from plant cell cultures of Catharanthus roseus and Nicotiana tabacum. The enzyme from C. roseus enantioselectively glucosylated (±)-secondary alcohols to give the glucosides of (R)-alcohols, while the glucosylation with that from N. tabacum gave preferentially the glucosides of (S)-alcohols.

Full text of DOI:10.1016/j.tetasy.2004.06.022

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 15 (2004) 2319–2321
Enantioselective glucosylation of ( )-secondary alcohols with plant
glucosyltransferases
a
b
Kei Shimoda,a, Naoji Kubota and Hiroki Hamada
*
aDepartment of Pharmacology and Therapeutics, Faculty of Medicine, Oita University, 1-1 Hasama-machi, Oita 879-5593, Japan
bDepartment of Life Science, Faculty of Science, Okayama University of Science, 1-1 Ridai-cho, Okayama 700-0005, Japan
Received 18 May 2004; accepted 15 June 2004
Available online 28 July 2004
Abstract—Two glucosyltransferases were isolated from plant cell cultures of Catharanthus roseus and Nicotiana tabacum. The
enzyme from C. roseus enantioselectively glucosylated ( )-secondary alcohols to give the glucosides of (R)-alcohols, while the glu-
cosylation with that from N. tabacum gave preferentially the glucosides of (S)-alcohols.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
of column chromatographies.4–7 First, ( )-secondary
alcohols 1–5 (10mg each) were administered to 10mL
of 50mM HEPES buffer (pH7.0) containing ca. 50lg
of isolated GTF-I from C. roseus and 40mg of UDP-
glucose and incubated at 35ꢁC for 24 or 36h. The yields
of the product glucosides were determined by HPLC
analyses.8 Extraction from the reaction mixture with 1-
butanol followed by purification using column chroma-
tography on silica gel with CHCl3–MeOH (95:5, v/v)
Glycosylation using biocatalysts is very attractive in the
practical preparation of alkyl glycosides, because one-
stepenzymatic glycosylation is more advantageous than
chemical glycosylation, which requires tedious steps
such as protection and deprotection of sugar hydroxyl
groups. The enantioselectivities of enzymatic glycosyla-
tion have been studied in the transglycosylation and re-
verse hydrolysis with glycosidases.1,2 However, little
attention has been paid to the enantioselective glycosy-
lation by glycosyltransferases. Recently, enantioselective
glucosylation of a ( )-secondary alcohol with plant cell
cultures has been reported; the glucosylation with Ca-
tharanthus roseus occurred enantioselectively to give
the glucoside of the (R)-alcohol, while the glucosylation
with Nicotiana tabacum preferentially gave the glucoside
of the (S)-alcohol.3 Over the course of developing a new
enzymatic asymmetric synthesis method, we investigated
the enantioselective glucosylation with plant glucosyl-
transferases from C. roseus and N. tabacum.
1
gave the products. The J values in the H NMR signal
for the anomeric protons of the resulting glucosides
showed that the sugar moiety of the products was of
b-orientation. The absolute configurations of the agly-
cone moieties of the products were confirmed by direct
1
comparison of the H NMR spectra with the authentic
glucosides synthesized from enantiomerically pure
HO
H
HO
HO
HO
H3C
H
HO
R
CH3
1: R=Et
3
4
5
2: R=n-Hex
H
2. Results and discussion
GlcO
H3C
H
GlcO
R
CH3
Two glucosyltransferases named GTF-I and II were
isolated from the corresponding plant cell cultures of
C. roseus and N. tabacum, respectively, by three steps
6: R=Et (6a: R, 6b: S)
8 (8a: R, 8b: S)
7: R=n-Hex (7a: R, 7b: S)
6
GlcO
6
1
5
GlcO
HO
1
5
4
2
2
4
3
3
*
Corresponding author. Tel.: +81-97-586-5606; fax: +81-97-586-
5619; e-mail: shimoda@med.oita-u.ac.jp
9 (9a: 1R,2S,5R, 9b: 1S,2R,5S) 10 (10a: 1R,2R, 10b: 1S,2S)
0957-4166/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2004.06.022

2320
K. Shimoda et al. / Tetrahedron: Asymmetry 15 (2004) 2319–2321
Table 1. Enantioselective glucosylation of ( )-secondary alcohols with GTF-I from C. roseus
Substrates
Products
Reaction time (h)
Conversion (%)
De (%)a
Eb
Configurationc
1
2
3
4
5
6a
7a
8a
9a
10a
24
24
36
36
24
25
22
28
17
39
78
84
11
15
20
R
R
R
R
R
87
98
>100
>100
>99
a % De(s) were determined on the basis of the intensities of the 1-methyl proton signals in the 1H NMR of glucosides 6 and 7 and the anomeric proton
signals in the 1H NMR of glucosides 8–10.
b Calculated from eesubstrate and eeproduct using standard equation.15
c Preferred configuration at the aglycone moieties of the products.
1
anomeric proton signals in the H NMR spectra.11–13
The enantioselectivity was greatly improved upon when
using the isolated GTF-I, whereas glucosylation with the
microsomal crude enzyme resulted in lower de(s) of 57–
65%, suggesting that the enantioselectivity of the glu-
cosylation with the microsomal enzyme might be af-
fected by impurities such as glucosidases or other
glucosyltransferases. It was found that 1–4 could be glu-
cosylated to give the corresponding glucosides of (R)-al-
cohols by GTF-I (Table 1). Although the glucosylation
of 1 showed relatively low enantioselectivity (78% de), it
was improved to 84% de when 2, which has a long alkyl
chain, was used as the substrate. In the case of 4, the glu-
cosylation resulted in a high diastereomeric excess
of 98%. A considerably challenging diol-substrate for
enantioselective glucosylation, 5, was also glucosylated
to the mono-glucoside with an (R)-configuration in its
aglycone part by GTF-I, allowing us to achieve the high-
est de of >99% (Fig. 1a).14 These results demonstrate
that glucosylation with the glucosyltransferase from C.
roseus occurred enantioselectively to give the glucosides
of (R)-alcohols and that the substrates with hydroxyl
group(s) attached to the cyclohexane ring could be glu-
cosylated with excellent enantioselectivity.
Substrates 1–5 were next subjected to glucosylation with
GTF-II from N. tabacum and then glucosylated to the
corresponding glucosides having an (S)-configuration
at their aglycone moieties (Table 2). It is noteworthy
that GTF-II glucosylated 3–5 to (S)-alkyl glucosides
8b–10b with very high enantioselectivity [de(s) of
>99% and 100%] (Fig. 1b), suggesting that enantioselec-
tive glucosylation with GTF-II is useful, as a new enzy-
matic enantiomer discriminating reaction, for the
practical preparation of alkyl glucosides in diastereo-
merically pure form. The results obtained herein reveal
that the glucosylation with the glucosyltransferase from
Figure 1. Pairs of anomeric proton signals in 1H NMR of the products
10 obtained by the glucosylation of 5 with (a) GTF-I from C. roseus
and (b) GTF-II from N. tabacum.
(R)- and (S)-alcohols by chemical glucosylation.9,10 The
diastereomeric compositions of the products were deter-
mined based on the intensities of the 1-methyl or
Table 2. Enantioselective glucosylation of ( )-secondary alcohols with GTF-II from N. tabacum
Substrates
Products
Reaction time (h)
Conversion (%)
De (%)a
Eb
Configurationc
1
2
3
4
5
6b
7b
24
24
36
36
24
24
20
26
21
33
90
93
25
35
S
S
S
S
S
8b
9b
>99
100
>99
>100
>100
>100
10b
a % De(s) were determined on the basis of the intensities of the 1-methyl proton signals in the 1H NMR of glucosides 6 and 7 and the anomeric proton
signals in the 1H NMR of glucosides 8–10.
b Calculated from eesubstrate and eeproduct using standard equation.15
c Preferred configuration at the aglycone moieties of the products.

K. Shimoda et al. / Tetrahedron: Asymmetry 15 (2004) 2319–2321
2321
6. Schenk, R. U.; Hildebrand, A. C. Can. J. Botany 1972, 50,
199.
N. tabacum affords the glucosides of the (S)-alcohols
with excellent enantioselectivity.
7. Murashige, T.; Skoog, F. Physiol. Plant 1962, 15, 473.
8. Conditions for HPLC analysis: column, Puresil C18
column (Waters); detector, differential refractometer
(Waters); solvent, MeOH–H2O (1:3, v/v); flow rate,
3. Conclusion
1mLminꢀ1
.
The enantioselective glucosylation with two glucosyl-
transferases from C. roseus and N. tabacum has been
accomplished with high enantioselectivity. It should be
emphasized that the enantioselectivities in the glucosyla-
tion of the ( )-secondary alcohols were opposite be-
tween these enzymes and that each diastereomer of
alkyl glucosides can be synthesized by selective use of
these glucosyltransferases.
9. Authentic glucosides 6–10 were prepared according to the
previously reported procedure.10 Enantiomerically pure
alcohols (12mg each) were added to a mixture of 2,3,4,6-
tetra-O-acetyl a-D-glucopyranosyl fluoride (53mg),
1,1,3,3-tetramethylguanidine (35mg) and BF3–OEt2
(0.05mL) in acetonitrile (2mL) and stirred for 2h at room
temperature. Extraction of organic materials with ethyl
acetate followed by purification using column chromato-
graphy on silica gel with hexane–ethyl acetate (9:1, v/v)
afforded alkyl 2,3,4,6-tetra-O-acetyl D-glucopyranosides
as a ca. 1:9 a/b-mixture, which were hydrolyzed with
saturated K2CO3 to give 5–10mg of authentic glucosides
as a mixture of a- and b-anomers (ca. 1:9).
References and notes
1. (a) Mitsuo, N.; Takeichi, H.; Satoh, T. Chem. Pharm. Bull.
1984, 32, 1183; (b) Ooi, Y.; Mitsuo, N.; Satoh, T. Chem.
Pharm. Bull. 1985, 33, 5547; (c) Gais, H. J.; Zeissler, A.;
Maidonis, P. Tetrahedron Lett. 1988, 29, 5743; (d) Toone,
E. J.; Simon, E. S.; Bednarski, M. D.; Whitesides, G. M.
Tetrahedron 1989, 45, 5365; (e) Tricone, A.; Nicolaus, B.;
Lama, L.; Morzillo, P.; De Rosa, M.; Gambacorta, A.
Biotechnol. Lett. 1991, 13, 235; (f) Tricone, A.; Nicolaus,
B.; Lama, L.; Gambacorta, A. J. Chem. Soc., Perkin
Trans. 1, 1991, 2841; (g) Vic, G.; Biton, J.; Beller, D. L.;
Michel, J. M.; Thomas, D. Biotechnol. Bioeng. 1995, 46,
109; (h) Matsumura, S.; Yamazaki, H.; Toshima, K.
Biotechnol. Lett. 1997, 19, 583.
10. Yamaguchi, T.; Horiguchi, A.; Fukuda, A.; Minori, T. J.
Chem. Soc., Perkin Trans. 1, 1990, 1079.
11. The intensities of the pair of 1-methyl proton signals in the
1H NMR (CD3OD) spectra were used for the determina-
tion of the diastereomeric excesses of 6 and 7. 1-Methyl
proton signals of the products were as follows: d 1.15 (d,
J=6.4Hz, for 6a) and 1.23 (d, J=6.1Hz, for 6b); d 1.15 (d,
J=6.4Hz, for 7a) and 1.21 (d, J=6.4Hz, for 7b). The
intensities of the pair of anomeric proton signals in the 1H
NMR (CD3OD) spectra were used for the determination
of the diastereomeric excesses of 8–10. Anomeric proton
signals of the products were as follows: d 4.40 (d,
J=7.3Hz, for 8a) and 4.10 (d, J=8.0Hz, for 8b); d 4.36
(d, J=7.6Hz, for 9a) and 4.32 (d, J=7.6Hz, for 9b); d 4.35
(d, J=7.6Hz, for 10a) and 4.46 (d, J=7.6Hz, for 10b). The
enantiomeric compositions of the aglycone moieties of
products 6–10 were confirmed by chiral GLC analysis of
the hydrolyzed alcohols, which had been prepared by
2. Itano, K.; Yamasaki, K.; Kihara, C.; Tanaka, O. Carbo-
hydr. Res. 1980, 87, 27.
3. (a) Lee, Y. S.; Ito, D. I.; Koya, K.; Izumi, S.; Hirata, T.
Chem. Lett., 1996, 719; (b) Hirata, T.; Koya, K.; Sarfo, K.
J.; Shimoda, K.; Ito, D. I.; Izumi, S.; Ohta, S.; Lee, Y. S.
J. Mol. Catal. B: Enzym. 1999, 6, 67; (c) Hirata, T.;
Shimoda, K.; Fujino, T.; Yamane, S.; Ohta, S. Bull. Chem.
Soc. Jpn. 2001, 74, 539.
hydrolysis
of
the
products
with
almond-b-
glucosidase.12,13
4. Homogenates of cultured cells of C. roseus5 in 100mM
HEPES buffer (pH7.0) were centrifuged at 10,000g for
30min to give a cell free extract, which itself was
centrifuged at 100,000g for 2h to give a microsomal
enzyme fraction as the precipitation. After solubilization
of the enzymes from the microsomal fraction with 0.1%
Triton X-100, purification of the solubilized enzymes
by chromatographies on a REACTIVE GREEN 19
agarose gel column, a Sephadex G-200 column and then
a diethylaminoethyl-Toyopearl column gave homogene-
ous glucosyltransferase as judged by SDS-PAGE: GTF-I,
monomeric form with molecular mass of ca. 72kDa. The
glucosyltransferase was isolated from cultured cells of N.
tabacum5 by the same procedure: GTF-II, monomeric
form with molecular mass of ca. 51kDa.
5. The suspension cells of C. roseus were cultured in 500mL
conical flasks containing 300mL of SH medium supple-
mented with 3% sucrose and 10mM 2,4-dichlorophenoxy-
acetic acid (2,4-D) under illumination (4000lux).6 The
suspension cells of N. tabacum were cultured in 500mL
conical flasks containing 300mL of Murashige and SkoogÕs
(MS) medium supplemented with 1% sucrose and 5mM
2,4-D under illumination (4000lux).7 Each suspension cells
were incubated on a rotary shaker (75rpm) at 25ꢁC for
3weeks prior to use for enzyme preparation.
12. Product glucosides were incubated with almond-b-glu-
cosidase (90U) and 1mL of phosphate buffer (0.1M,
pH6.0) at 37ꢁC for 24h. After 24h incubation, the
glucosides were completely hydrolyzed to the alcohols as
judged by HPLC and TLC analyses of the reaction
mixture. The reaction mixture was extracted with diethyl
ether to give a crude alcohol fraction, which was purified
by column chromatography on silica gel with pentane-
ethyl acetate (9:1, v/v) to yield the hydrolyzed alcohols.
13. Conditions for capillary GLC analysis: column, Rt-bDEX
(Restek, 0.25mm·30m); injector, 180ꢁC; detector, 180ꢁC;
oven, 100ꢁC; carrier gas, N2 (50mLminꢀ1). Retention
times for alcohols 1–5 in the GLC were as follows: (S)-
and (R)-1, 9.7 and 10.2min; (S)- and (R)-2, 12.6 and
13.2min; (R)- and (S)-3, 15.1 and 16.5min; (1S,2R,5S)-
and (1R,2S,5R)-4, 20.0 and 20.3min; (1S,2S)- and
(1R,2R)-5, 27.7 and 28.9min.
25
D
14. Product 10a converted from 5 by GTF-I: ½aŠ ¼ ꢀ15
20
(c 0.20, MeOH) {lit.2 ½aŠ ¼ ꢀ17}; 10b glucosylated by
D
25
D
20
D
GTF-II: ½aŠ ¼ ꢀ23 (c 0.18, MeOH) {lit.2 ½aŠ ¼ ꢀ22}.
The optical rotation data of the products 6–9 could not be
obtained due to the low transformation rate and lack of
products.
15. Chen, C. S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. Am.
Chem. Soc. 1982, 104, 7294.