Aryl and Allyl C-Glycosidation Methods Using Unprotected Sugars

Article abstract of DOI:10.1021/jo972146v

Practical and highly stereoselective aryl and allyl C-glycosidation methods using unprotected sugars as glycosyl donors have been developed. Aryl C-glycosidations of several unprotected 2-deoxy sugars with phenol and naphthol derivatives by the combined u

Full text of DOI:10.1021/jo972146v

J . Org. Chem. 1998, 63, 2307-2313
2307
Ar yl a n d Allyl C-Glycosid a tion Meth od s Usin g Un p r otected Su ga r s
Kazunobu Toshima,* Goh Matsuo,1 Toru Ishizuka, Yasunobu Ushiki, Masaya Nakata, and
Shuichi Matsumura
Department of Applied Chemistry, Faculty of Science and Technology, Keio University,
3-14-1 Hiyoshi, Kohoku-ku, Yokohama 223-8522, J apan
Received November 24, 1997
Practical and highly stereoselective aryl and allyl C-glycosidation methods using unprotected sugars
as glycosyl donors have been developed. Aryl C-glycosidations of several unprotected 2-deoxy sugars
with phenol and naphthol derivatives by the combined use of trimethylsilyl trifluoromethane-
sulfonate (TMSOTf)-AgClO4 or TMSOTf exclusively gave the corresponding unprotected o-
hydroxyaryl â-C-glycosides which appear in many biologically attractive aryl C-glycoside antibiotics
as the key subunit. On the other hand, allyl C-glycosidations of several unprotected glycals with
allyltrimethylsilane by TMSOTf afforded the corresponding unprotected and 2,3-unsaturated allyl
R-C-glycosides in high yields which are versatile synthetic intermediates for the syntheses of
optically active natural products.
In tr od u ction
An efficient C-glycosidation2 with high regio- and
stereoselectivity is of particular interest as well as
O-glycosidation3 in the syntheses of biologically impor-
tant natural products. Many biologically attractive C-
glycosides such as aryl C-glycoside antibiotics have
already been found in nature, and several types of
C-glycosides such as alkyl and allyl C-glycosides are well
recognized to be useful chiral building blocks for the
synthesis of optically active natural products.2 Further-
more, carbon-linked glycosides, stable analogues of natu-
F igu r e 1. Molecular structures of the representative aryl
C-glycoside antibiotics.
rally occurring O- and N-glycosides, have become the
subject of considerable interest in bioorganic and medici-
nal chemistry. Although remarkable progress has been
made in the C-glycoside synthesis,2 the development of
simple and practical C-glycosidation methods is still one
of the central problems in synthetic organic chemistry.
In this context, the use of an unprotected sugar as a
glycosyl donor in the glycosidation reaction undoubtedly
has considerable advantages. However, practical C-
glycosidations using totally unprotected sugars have
never been reported. The main reasons why the glycosi-
dation of an unprotected sugar is difficult are the
undesirable generation of self-coupling products of the
glycosyl donor and the deactivation of a glycosidation
reagent by the free hydroxy groups of the glycosyl donor.
Therefore, we undertook development of novel C-glycosi-
dation methods employing unprotected sugars which
overcame such difficulties. For this purpose, we carried
out two approaches which were based on the differences
in the stability between the C-glycoside bond and O-
glycoside bond, and on their formation rate. Thus, if we
could find a reaction that cleaves any O-glycoside bond
and then forms a C-glycoside bond, or a reaction in which
the formation of the C-glycoside bond is much faster than
that of the O-glycoside bond, C-glycosidation using an
unprotected sugar as the glycosyl donor could be achieved.
In this paper, we report the aryl and allyl C-glycosida-
tions utilizing unprotected sugars based on these con-
cepts.4,5
Resu lts a n d Discu ssion
Ar yl C-Glycosid a t ion s of Un p r ot ect ed 2-Deoxy
Su ga r s. Over the past several years, aryl C-glycoside
antibiotics such as the angucyclin6 and pluramycin7
families have attracted considerable attention due to
their significant biological properties and architecturally
attractive structures (Figure 1). 2-Deoxy sugars are the
most common and important of the sugar residues.
Therefore, the effective and practical coupling of the
sugar part into the aglycon, the aromatic moiety, has
become an important task in contemporary organic
(4) For our preliminary communications on aryl C-glycosidation,
see: (a) Toshima, K.; Matsuo, G.; Tatsuta, K. Tetrahedron Lett. 1992,
33, 2175. (b) Toshima, K.; Matsuo, G.; Ishizuka, T.; Nakata, M.;
Kinoshita, M. J . Chem. Soc., Chem. Commun. 1992, 1641. (c) Toshima,
K.; Matsuo, G.; Nakata, M. J . Chem. Soc., Chem. Commun. 1994, 997.
(5) For our preliminary communication on allyl C-glycosidation,
see: Toshima, K.; Ishizuka, T.; Matsuo, G.; Nakata, M. Tetrahedron
Lett. 1994, 35, 5673.
(6) For a recent review of angucycline antibiotics, see: Rohr, J .;
Thiericke, R. Nat. Prod. Rep. 1992, 103.
(7) For a recent review of pluramycin antibiotics, see: Hansen, M.
R.; Hurley, L. H. Acc. Chem. Res. 1996, 29, 249.
* Tel: +81-45-563-1141, ext. 3429. Fax: +81-45-563-0446. E-Mail:
toshima@applc.keio.ac.jp.
(1) Taken in part from the Ph.D. Thesis of Goh Matsuo, Keio
University, 1997.
(2) For recent reviews of C-glycosidation method, see: (a) Postema,
M. H. D. Tetrahedron 1992, 40, 8545. (b) Levy, D. E.; Tang, C. The
Chemistry of C-Glycosides; Pergamon Press: Oxford, 1995.
(3) For a recent review of O-glycosidation method, see: Toshima,
K.; Tatsuta, K. Chem. Rev. 1993, 93, 1503.
S0022-3263(97)02146-4 CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/19/1998

2308 J . Org. Chem., Vol. 63, No. 7, 1998
Toshima et al.
F igu r e 2. Aryl C-glycosidations.
Ta ble 1. Ar yl C-Glycosid a tion s of 1a a n d 5 by
synthesis. Efficient aryl C-glycoside syntheses via OfC-
glycoside rearrangement were independently announced
by Kometani8 and Suzuki.9 In this context, we investi-
gated the aryl C-glycosidation of an unprotected sugar
based on the higher stability of the C-glycoside bond
compared to that of the O-glycoside bond (Figure 2).
In our initial attempts at searching for such a reaction,
we examined the aryl C-glycosidation of a glycosyl donor
possessing a methyl glycoside because the methyl glyco-
side bond is one of the most stable O-glycoside bonds. If
the methyl glycoside is converted into the C-glycoside,
any O-glycoside bond could be cleaved and then converted
into the C-glycoside bond. Therefore, we first tested the
C-glycosidations of the methyl glycoside 1a with 2-naph-
thol (5) using several Lewis acids such as trimethylsilyl
trifluoromethanesulfonate (TMSOTf), trifluoromethane-
sulfonic anhydride (Tf2O), TMSOTf-AgClO4, Tf2O-Ag-
ClO4, TMSOTf-silver trifluoromethanesulfonate (AgO-
Tf), or Tf2O-AgOTf. Among them, it was found that only
the combined use of TMSOTf-AgClO4 (1:1) showed a
sharp contrast to the other activators and worked ef-
ficiently in this case. As seen in the results shown in
Table 1, the methyl glycoside 1a was smoothly glycosi-
dated to the ortho-position of 5 by using TMSOTf-
AgClO4 (1:1) in CH2Cl2 at 25 °C for 1 h to give the aryl
â-C-glycoside of 11 (Figure 3) in 91% yield together with
1% of its R-anomer (entry 1 in Table 1).10 Furthermore,
even the use of 20 mol % of the present activator was
found to be good enough to perform the reaction with a
quite satisfactory chemical yield and stereoselectivity
(entry 3 in Table 1), while the use of 10 mol % of the
activator gave a significant amount of an O-glycosidated
product (entry 4 in Table 1). We further confirmed that
a
TMSOTf-AgClO4
mol % of
activator
yield
(%)b
entry
R/âc
1
2
3
4
100
50
20
92
98
99
72
1:94
1:>99
1:>99
1:>99
10
a
All reactions were carried out by use of 2.0 equiv of 5 to 1a .
b
Isolated yields after purification by column chromatography.
c R:â Ratios were determined by 1H-NMR (270 MHz) spectroscopy
and/or isolation of pure isomers.
the O-glycosidated product was smoothly converted into
the corresponding C-glycoside under the conditions for
the entry 3 in Table 1. This reaction process involved
the OfC-glycoside rearrangement which was similar to
the Lewis acid-catalyzed C-glycosidation mechanism.8,9,11
To enhance the synthetic utility of this reaction, the
C-glycosidations of several other methyl glycosides 2a -
4a , which occurred as subunits in a variety of C-glycoside
antibiotics,6,7 with 5 were next examined and showed an
additional feature. Not only the neutral sugars 1a -3a
but also the amino sugar 4a were smoothly glycosidated
in a similar manner to give the corresponding aryl
C-glycoside 15a with high â-stereoselectivity in high yield
as shown in Table 2.
From these observations, it was made clear that the
novel TMSOTf-AgClO4 catalyst system cleanly cleaved
the alkyl O-glycosidic bond and then smoothly formed
the aryl C-glycosidic bond. Therefore, we next expected
that if the TMSOTf-AgClO4 combined activator was not
(8) Kometani, T.; Kondo, H.; Fujimori, Y. Synthesis 1988, 1005.
(9) (a) Matsumoto, T.; Katsuki, M.; Suzuki, K. Tetrahedron Lett.
1988, 29, 6935. (b) Matsumoto, T.; Katsuki, T.; J ona, H.; Suzuki, K. J .
Am. Chem. Soc. 1991, 113, 6982.
(10) The structures of all o-hydroxyaryl â-C-glycosides were assigned
from their 1H NMR data which exhibited typical chemical shifts and
coupling constants for both R- and â-anomers; see ref 9. The regiose-
lectivity of the glycosidic bond could be derived from the 1H NOE
experiments of the aromatic moiety.
(11) For other representative aryl C-glycosidations via OfC-glyco-
side migration, see: (a) Ramesh, N. G.; Balasubramanian, K. K.
Tetrahedron Lett. 1992, 33, 3061. (b) Gasiraghi, G.; Cornia, M.; Rassu,
G.; Zetta, L.; Fava, G. G.; Belicchi, M. F. Tetrahedron Lett. 1988, 29,
3323. (c) Mahling, J .-A.; Schmidt, R. R. Synthesis 1993, 325.

Aryl and Allyl C-Glycosidation Methods
J . Org. Chem., Vol. 63, No. 7, 1998 2309
F igu r e 3. Aryl C-glycosides.
Ta ble 2. Ar yl C-Glycosid a tion s of 2a , 3a , a n d 4a w ith 5
only the 1-OMe group but also the 1-OH group of the
sugars. Our next attempts were the aryl C-glycosidations
of the unprotected methyl glycosides 1c-4c with 5 to
examine the activating capacity of the TMSOTf-AgClO4
system in the presence of other hydroxyl groups of the
glycosyl donor. In the case of the glycosyl donors 1c-
3c, CH3CN was used as an appropriate solvent instead
of CH2Cl2, considering their solubility. However, the use
of CH2Cl2 as a solvent was crucial for the effective
glycosidation of the amino sugar 4c. The results sum-
marized as entries 5-8 in Table 3 showed an additional
feature of the present method. Even the trihydroxy sugar
1c and monohydroxy amino sugar 4c were smoothly
glycosidated with 5 by use of 50 mol % of the present
activator to give the corresponding unprotected aryl â-C-
glycosides of 12b and 15b, respectively, in high yields
(entries 5 and 8 in Table 3). These results suggested that
the ability of the TMSOTf-AgClO4 system as a catalytic
activator was not significantly influenced by any hydroxyl
group of the glycosyl donor.
a
by TMSOTf-AgClO4
temp
(°C)
time
(h)
yield
(%)b
entry
sugar
product
R/âc
1
2
3
2a
3a
4a
25
25
40
1
1
2
13a
14a
15a
99
98
99
1:>99
1:87
1:>99
a
All reactions were carried out by use of 2.0 equiv of 5 to the
b
glycosyl donor. Isolated yields after purification by column
chromatography. c R:â Ratios were determined by 1H-NMR (270
MHz) spectroscopy and/or isolation of pure isomers.
a
Ta ble 3. Ar yl C-Glycosid a tion s by TMSOTf-AgClO4
mol % of
temp time
yield
entry sugar activator solvent (°C) (h) product (%)b R/âc
Having these favorable results, we tried the aryl
C-glycosidations of totally unprotected sugars by using
the present catalyst system. Although the totally un-
protected sugars 1d and 4d were not able to be applied
to the glycosidation reaction due to their low solubility
in both CH3CN and CH2Cl2, both aryl C-glycosidations
of 2d and 3d with 5 in CH3CN were effectively achieved
under similar conditions to afford the unprotected aryl
â-C-glycosides of 13b and 14b, respectively, with high
chemical yield and stereoselectivity (entries 9 and 10 in
Table 3). At this stage, it was unfortunately found that
other polar solvents, MeOH, i-PrOH, t-BuOH, DMF, and
THF, were found to be not suitable for the present
glycosidation reaction.
1
2
3
4
5
6
7
8
9
1b
2b
3b
4a
1c
2c
3c
4c
2d
3d
20
20
20
50
50
20
20
50
20
20
CH2Cl2 25
CH2Cl2 25
CH2Cl2 25
CH2Cl2 40
CH3CN 25
CH3CN 25
CH3CN 25
CH2Cl2 40
CH3CN 25
CH3CN 25
1
12a
13a
14a
15a
12b
13b
14b
15b
13b
14b
85 1:>99
99 1:70
90 1:15
83 1:>99
86 1:>99
91 1:>99
92 1:32
72 1:>99
92 1:>99
84 1:97
0.5
0.5
2
1
1
1
2
1
1
10
a
All reactions were carried out by use of 2.0 equiv of 5 to the
b
glycosyl donor. Isolated yields after purification by column
chromatography. c R:â Ratios were determined by 1H-NMR (270
MHz) spectroscopy and/or isolation of pure isomers.
deactivated by any hydroxyl group and could effectively
activate the 1-OH group of the glycosyl donor, aryl
C-glycosidation of a totally unprotected sugar would be
achieved. We examined the aryl C-glycosidations of the
benzoylated glycoses 1b-4b with 2-naphthol (5) to assess
the ability of the TMSOTf-AgClO4 catalyst to activate
the 1-OH group of the sugars. The results summarized
as entries 1-4 in Table 3 showed these glycosidations
proceeded smoothly under mild conditions to afford the
corresponding benzoylated aryl â-C-glycosides of 12a -
15a 9b with high stereoselectivity in very good to excellent
yields. These results clearly indicated that the TM-
SOTf-AgClO4 catalyst was effective in activating not
In our extended studies of this project, we further
investigated a more practical method without AgClO4,
which would not be employed for large scale experiments
and industrial processes due to its hazardous and explo-
sive properties. Since sugars having acyl protecting
groups such as acetyl and benzoyl groups could become
useful glycosyl donors in a wide variety of glycosidation
reactions, we tried the aryl C-glycosidations of the
acylated methyl glycosides 1a and 2a with 2-naphthol
(5) using only TMSOTf as the catalyst. The results
summarized in Table 4 as entries 1 and 2 showed that
the chemical yields of these reactions were much lower

2310 J . Org. Chem., Vol. 63, No. 7, 1998
Toshima et al.
Ta ble 4. P r otectin g Gr ou p Effect in Ar yl
Ta ble 6. Ar yl C-Glycosid a tion s of 2c a n d 6-10 by
a
a
C-Glycosid a tion s by TMSOTf
TMSOTf
glycosyl
acceptor
mol % of
activator
yield
(%)b
yield
entry
product
R/âc
entry
sugar
product
(%)b
R/âc
1
2
3
4
5
6
7
8
9
10
20
20
20
20
50
16
17
18
19
20
76
79
98
91
64
1:>99
1:>99
1:>99
1:>99
1:>99
1
2
3
4
1a
2a
1e
2e
11
19
57
99
89
1:>99
1:>99
1:>99
1:>99
13a
12c
13c
a
All reactions were carried out by use of 2.0 equiv of 5 to the
a
All reactions were carried out by use of 2.0 equiv of the glycosyl
b
glycosyl donor. Isolated yields after purification by column
b
acceptor to 2c. Isolated yields after purification by column
chromatography. c R:â Ratios were determined by 1H-NMR (270
chromatography. c R:â Ratios were determined by 1H-NMR (270
MHz) spectroscopy and/or isolation of pure isomers.
MHz) spectroscopy and/or isolation of pure isomers.
a
Ta ble 5. Ar yl C-Glycosid a tion s by TMSOTf
Ta ble 7. Ar yl C-Glycosid a tion s of 2d a n d 6-10 by
a
TMSOTf or TMSOTf-AgClO4
mol % of
temp time
yield
entry sugar activator solvent (°C) (h) product (%)b R/âc
1
2
3
4
5
6
1c
2c
3c
4c
2d
3d
50
20
20
120
20
20
CH3CN 40
CH2Cl2 25
CH2Cl2 25
CH2Cl2 40
CH2Cl2 25
CH2Cl2 25
1
1
1
8
1
1
12b
13b
14b
15b
13b
14b
89 1:>99
98 1:>99
91 1:>99
93 1:>99
97 1:>99
72 1:>99
glycosyl
activator
(mol %)
time
yield
entry acceptor
(h) product (%)b R/âc
1
2
3
4
5
6
7
8
9
6
6
7
7
8
8
9
9
10
10
TMSOTf (20)
TMSOTf-AgClO4 (20) 0.5
TMSOTf (20)
TMSOTf-AgClO4 (20) 0.5
TMSOTf (20)
TMSOTf-AgClO4 (20) 0.5
TMSOTf (20)
TMSOTf-AgClO4 (20) 0.5
TMSOTf (20)
TMSOTf-AgClO4 (20) 0.5
1
16
16
17
17
18
18
19
19
20
20
77 1:>99
87 1:>99
63 1:>99
91 1:>99
89 1:>99
98 1:>99
71 1:>99
90 1:>99
42 1:>99
75 1:>99
1
a
All reactions were carried out by use of 2.0 equiv of 5 to the
1
b
glycosyl donor. Isolated yields after purification by column
chromatography. c R:â Ratios were determined by 1H-NMR (270
MHz) spectroscopy and/or isolation of pure isomers.
1
1
than those with TMSOTf-AgClO4 as expected from our
earlier observations. After many attempts to optimize
the new activator which has no hazardous or explosive
properties, our attention turned to the effect of the
protecting groups of the glycosyl donors. Therefore, we
next examined the glycosidations of the methylated
methyl glycosides 1e and 2e with 5 by TMSOTf. It was
found that the results including both chemical yields and
stereoselectivity of these glycosidations were quite sat-
isfactory as shown in Table 4 as entries 3 and 4.
Furthermore, unexpected favorable results were obtained
for the glycosidations of the corresponding unprotected
methyl glycosides 1c-4c with 5 by TMSOTf. The results
illustrated in Table 5 as entries 1 and 2 showed that
these glycosidations proceeded much more effectively
than those of the corresponding acylated methyl glyco-
sides 1a and 2a to afford only the unprotected o-
hydroxyaryl â-C-glycosides of 12b and 13b in high yields,
respectively. These results clearly indicated that unpro-
tected sugars could become very versatile glycosyl donors
in the aryl C-glycosidation reaction using TMSOTf. Our
attention next turned to the scope and limitations of the
present method. The results summarized in Table 5 as
entries 3 and 4 showed that other unprotected methyl
glycosides 3c and 4c also reacted with 5 to give high
yields of the unprotected aryl â-C-glycosides of 14b and
15b, respectively. Furthermore, it was found that both
glycosidations of totally unprotected sugars 2d and 3d
with 5 using a catalytic amount of TMSOTf were ef-
fectively achieved under similar conditions to afford the
aryl â-C-glycosides of 13b and 14b with quite satisfactory
chemical yield and stereoselectivity (entries 5 and 6 in
Table 5).
10
a
All reactions were carried out by use of 2.0 equiv of the glycosyl
b
acceptor to 2d . Isolated yields after purification by column
chromatography. c R:â Ratios were determined by 1H-NMR (270
MHz) spectroscopy and/or isolation of pure isomers.
the unprotected glycose, olivose 2d , with some other
phenol and naphthol derivatives 6-10 by TMSOTf-
AgClO4 or TMSOTf because olivose is a very representa-
tive sugar which exists as a glycosidic component in many
aryl C-glycoside antibiotics.6,7 These results are sum-
marized in Tables 6 and 7. It was found that all
glycosidations of 2c proceeded only using a catalytic
amount of TMSOTf under mild conditions to give the
corresponding unprotected aryl â-C-glycosides of 16-20
in very good to excellent yields (Table 6). On the other
hand, in the case of 2d , the combined use of TMSOTf-
AgClO4 gave significantly better results than the use of
TMSOTf especially when 7 and 10 were employed as the
glycosyl acceptors (Table 7). In the case of the combined
use of TMSOTf-AgClO4, the true activating species for
12
the glycosyl donor is presumably TMSClO4 and/or
HClO4 which is generated due to the presence of the free
hydroxy groups of the glycosyl acceptor and donor, while
TMSOTf and/or TfOH is the activating spices when
TMSOTf is used.13 Indeed, it was confirmed that when
TfOH was used instead of TMSOTf as the activating
reagent for the glycosyl donors, a 20-30% decrease in
chemical yields was observed.
Allyl C-Glycosid a tion s of Un p r otected Glyca ls.
The C-glycosidations of unprotected sugars based on the
(12) Barton, T. J .; Tully, C. R. J . Org. Chem. 1978, 43, 3649.
(13) Kondo, H.; Aoki, S.; Ichikawa, Y.; Halcomb, R. L.; Ritzen, H.;
Wong, C.-H. J . Org. Chem. 1994, 59, 864.
Finally, we tried the aryl C-glycosidations of the
unprotected methyl glycoside, methyl olivoside 2c, and

Aryl and Allyl C-Glycosidation Methods
J . Org. Chem., Vol. 63, No. 7, 1998 2311
F igu r e 5. 2,3-Unsaturated allyl C-glycosides.
Ta ble 9. Allyl C-Glycosid a tion s of 21a -24a a n d 25 by
TMSOTfa
F igu r e 4. Allyl C-glycosidations.
solvent
time
yield
entry sugar
(M for sugar)
(h) product (%)b R/âc
Ta ble 8. Allyl C-Glycosid a tion s of 21a a n d 25a
1
2
3
4
5
6
7
8
9
21a CH2Cl2 (0.2)
21a CH2Cl2 (0.1)
22a CH2Cl2 (0.2)
0.5
0.5
0.5
26a
26a
27a
27a
27a
28a
28a
28a
29a
29a
91 >99:1
94 >99:1
14 >99:1
85 >99:1
91 >99:1
22a CH2Cl2[2]-CH3CN[1] (0.2) 0.5
22a CH2Cl2[2]-CH3CN[1] (0.05) 0.5
23a CH2Cl2 (0.2)
0.5
1
1
0.5
2
2
>99:1
yield
23a CH2Cl2[2]-CH3CN[1] (0.05)
23a CH2Cl2[2]-CH3CN[1] (0.02)
24a CH2Cl2 (0.2)
59 >99:1
entry
activator
(%)b
R/âc
90 >99:1
9
>99:1
1
2
3
4
5
6
TMSOTf
TBSOTf
BF3‚Et2O
Tf2O
TfOH
CSA
94
46
trace
0
73
0
>99:1
>99:1
-
-
35:1
-
10
24a CH2Cl2[2]-CH3CN[1] (0.02)
66 >99:1
a
All reactions were carried out by use of 2.0 equiv of 25 to the
b
glycosyl donor. Isolated yields after purification by column
chromatography. c R:â Ratios were determined by 1H-NMR (270
MHz) spectroscopy and/or isolation of pure isomers.
a
All reactions were carried out by use of 2.0 equiv of 25 to 21a .
b Isolated yields after purification by column chromatography.c R:â
Ratios were determined by 1H-NMR (270 MHz) spectroscopy and/
or isolation of pure isomers.
only the unprotected and 2,3-unsaturated allyl R-C-
glycoside of 26a (Figure 5) in 94% yield. Since self-
coupling products came from the O-glycosidation of 21a
were not detected during the reaction at any stage, the
present method depended on the faster trapping of 25
than any hydroxyl group of the glycosyl donor 21a .
Furthermore, it was confirmed that use of low temper-
ature (-78 °C) was important for the selective formation
of the C-glycoside bond.
To enhance the synthetic utility of this reaction, the
allyl C-glycosidations of several other unprotected glycals,
D-glucal (22a ), D-galactal (23a ), and D-fucal (24a ), with
25 were next examined. The results summarized in
Table 9 as entries 5, 8, and 10 showed that although the
yield of the glycosidation of 24a was not very high, the
glycosidations of 22a and 23a proceeded under similar
conditions to give the allyl R-C-glycosides of 27a and 28a ,
respectively, in high yields. Remarkably, the stereose-
lectivity of these glycosidations was quite R-selective in
all cases.17 In the glycosidations of 22a -24a , the use of
MeCN as a cosolvent and the low concentration of the
glycals in the solvent were necessary to get high yields
of the desired allyl C-glycosides owing to the low solubil-
ity of 22a -24a in CH2Cl2 at low temperature (-78 °C)
(entries 3-10 in Table 9).
faster formation of the C-glycoside bond compared to that
of the O-glycoside bond were investigated and realized
as the allyl C-glycosidations of unprotected glycals and
allyltrimethysilane at low temperature. These glycosi-
dations exclusively gave the corresponding unprotected
and 2,3-unsaturated allyl R-C-glycosides, which are very
versatile synthetic intermediates for natural products
syntheses,14 in a fashion similar to the carbon-Ferrier
rearrangement15,16 (Figure 4).
In our initial attempts to find a suitable activator, we
examined several acid activators such as TMSOTf, tert-
butyldimethylsilyl trifluoromethanesulfonate (TBSOTf),
BF3‚Et2O, Tf2O, TfOH, and dl-10-camphorsulfonic acid
(CSA) for the allyl C-glycosidation of the unprotected
L-rhamnal (21a ) and allyltrimethylsilane (25). To avoid
the self O-glycosidation of the unprotected glycal 21a ,
these reactions were carried out at low temperature (-78
°C). From the results shown in Table 8, it was found
that TMSOTf, in sharp contrast to the other activators,
worked effectively. Thus, the unprotected glycal 21a was
smoothly coupled with 25 by using 100 mol % of TMSOTf
in CH2Cl2 (0.1 M for 21a ) at -78 °C for 0.5 h to afford
We then compared the unprotected glycals 21a -24a
and the corresponding acetylated glycals 21b-24b from
the viewpoint of chemical yield and stereoselectivity in
their allyl C-glycosidations with 25 by TMSOTf. Al-
though acetylated glycals have been frequently used as
suitable glycosyl donors in the carbon-Ferrier rearrange-
ment,14,16 the allyl C-glycosidations of the acetylated
glycals 21b-24b using TMSOTf have never been re-
(14) For representative application to natural products syntheses,
see: (a) Nicolaou, K. C.; Duggan, M. E.; Hwang, C.-K.; Somers, P. K.
J . Chem. Soc., Chem. Commun. 1985, 1359. (b) Wincott, F. E.;
Danishefsky, S. J .; Schulte, G. Tetrahedron Lett. 1987, 28, 4951. (c)
Danishefsky, S. J .; DeNinno, S.; Lartey, P. J . Am. Chem. Soc. 1987,
109, 2082. (d) Ichikawa, Y.; Isobe, M.; Goto, T. Tetrahedron 1987, 43,
4749.
(15) (a) Ferrier, R. J . Chem. Soc. 1964, 5443. (b) Ferrier, R.; Prasad,
N. J . Chem. Soc. C 1969, 570.
(16) (a) Danishefsky, S.; Kerwin, J . F. J r. J . Org. Chem. 1982, 47,
3803. (b) Ichikawa, Y.; Isobe, M.; Konobe, M.; Goto, T. Carbohydr. Res.
1987, 171, 193. (c) Toshima, K.; Ishizuka, T.; Matsuo, G.; Nakata, M.
Chem. Lett. 1993, 2013. (d) Toshima, K.; Miyamoto, N.; Matsuo, G.;
Nakata, M.; Matsumura, S. Chem. Commun. 1996, 1379.
(17) The R-configurations of the anomeric positions were confirmed
by the comparison between the corresponding acetylated glycosides,
which were obtained by standard acetylation, and the authentic
samples reported in ref 16.

2312 J . Org. Chem., Vol. 63, No. 7, 1998
Toshima et al.
Ta ble 10. Allyl C-Glycosid a tion s of 21b-24b a n d 25 by
of glycosyl donor (0.1 mmol), glycosyl acceptor (0.2 mmol), and
silver perchlorate (0.02 or 0.05 mmol) in dry CH2Cl2 or CH3CN
(1 mL) was added trimethylsilyl trifluoromethanesulfonate
(0.02 or 0.05 mmol) dropwise at 0 °C under argon. The
reaction mixture was stirred at 25 or 40 °C for 0.5, 1, or 2 h
as described in Tables 3 and 7. In the case of neutral sugars,
the reaction mixture was quenched with triethylamine under
ice-cooling and then concentrated in vacuo. On the other hand,
in the case of amino sugar, the reaction mixture was quenched
with ice-cold and saturated aqueous NaHCO3, the resultant
mixture was extracted with ethyl acetate several times, and
the extracts were washed with saturated aqueous NaCl, dried
over anhydrous Na2SO4, and concentrated in vacuo. Purifica-
tion of the residue by flash column chromatography gave the
corresponding unprotected aryl C-glycoside.
a
TMSOTf
solvent
(M for sugar)
time
yield
entry sugar
(h) product (%)b
R/âc
1
2
3
4
21b CH2Cl2 (0.1)
22b CH2Cl2[2]-CH3CN[1] (0.05) 0.5
23b CH2Cl2[2]-CH3CN[1] (0.02)
24b CH2Cl2[2]-CH3CN[1] (0.02)
0.5
26b
27b
28b
29b
95
63
9.9:1
37:1
1
2
19 >99:1
20 >99:1
a
All reactions were carried out by use of 2.0 equiv of 25 to the
b
glycosyl donor. Isolated yields after purification by column
chromatography. c R:â Ratios were determined by 1H-NMR (270
MHz) spectroscopy and/or isolation of pure isomers.
Ar yl C-Glycosid a tion s of Un p r otected 2-Deoxy Su ga r s
by TMSOTf. Gen er a l P r oced u r e. To a mixture of glycosyl
donor (0.1 mmol) and glycosyl acceptor (0.2 mmol) in dry
CH2Cl2 or CH3CN (1 mL) was added trimethylsilyl trifluo-
romethanesulfonate (0.02, 0.05, or 0.12 mmol) dropwise at 0
°C under argon. The reaction mixture was stirred at 25 or 40
°C for 1, 2, or 8 h as described in Tables 5, 6, and 7. The
reaction mixture was worked up and purified as described
above to give the corresponding unprotected aryl C-glycoside.
1-(2′-Deoxy-â-D-a r a bin o-h exop yr a n osyl)-2-n a p h t h ol
ported. The results of the C-glycosidations of 21b-24b
under similar conditions as those for 21a -24a , respec-
tively, are summarized in Table 10. Notably, it was
found that the allyl C-glycosidations of the totally
unprotected glycals 22a -24a with 25 proceeded much
more effectively than those of the acetylated glycals 22b-
24b to give high yields of the allyl R-C-glycosides of 27a -
29a , respectively. Furthermore, the stereoselectivity of
the glycosidations of 21a and 22a was higher than that
of the corresponding acetylated glycals 21b and 22b.
From these results, the allyl C-glycosidations of the
unprotected glycals with allyltrimethylsilane using TM-
SOTf exhibits two significant advantages. The first one
is the higher reactivity of such glycosyl donors compared
to the acylated derivatives. The second advantage is the
extremely high R-stereoselectivity obtained in the gly-
cosidation reactions. Since the epimerization at the
anomeric position of the resulting C-glycoside was not
observed during the C-glycosidation, the high R-stereo-
selectivity must arise from a kinetic anomeric effect.18
(12b-â). Rf 0.38 (5:1 chloroform-methanol); [R]26 +92.6° (c
D
0.81, MeOH); 1H NMR (CD3OD) δ 2.02 (1H, ddd, J ) 13.6,
11.9, and 11.9 Hz), 2.17 (1H, ddd, J ) 13.6, 4.6, and 2.4 Hz),
3.45-3.55 (2H, m), 3.8-4.0 (3H, m), 5.60 (1H, dd, J ) 11.9
and 2.4 Hz), 7.0-8.15 (6H). Anal. Calcd for C16H18O5: C,
66,20; H, 6.25. Found: C, 66.14; H, 6.43.
1-(2′,6′-Did eoxy-â-D-a r a bin o-h exop yr a n osyl)-2-n a p h -
th ol (13b-â). Rf 0.63 (4:1 chloroform-methanol); [R]28D +155.5°
(c 1.24, CHCl3); mp 152.0-153.0 °C (ethyl acetate-hexane);
1H NMR δ 1.46 (3H, d, J ) 6.0 Hz), 1.98 (1H, ddd, J ) 13.6,
11.9, and 11.9 Hz), 2.37 (1H, ddd, J ) 13.6, 4.4, and 2.1 Hz),
2.75 (1H, br s), 3.03 (1H, br s), 3.30 (1H, br dd, J ) 9.2 and
9.2 Hz), 3.60 (1H, dq, J ) 9.2 and 6.0 Hz), 3.8-3.95 (1H, m),
5.50 (1H, dd, J ) 11.9 and 2.1 Hz), 7.05-7.8 (6H), 8.89 (1H,
s). Anal. Calcd for C16H18O4: C, 70.06; H, 6.61. Found: C,
69.99; H, 6.69.
1-(2′,6′-Did eoxy-r-D-r ibo-h exop yr a n osyl)-2-n a p h t h ol
(14b-r). Rf 0.37 (1:1 hexane-acetone); [R]24 -85.1° (c 0.83,
D
Con clu sion s
1
CHCl3); H NMR δ 1.45 (3H, d, J ) 6.4 Hz), 2.04 (1H, ddd, J
The present novel aryl and allyl C-glycosidations using
unprotected sugars as simple glycosyl donors offered
promising entries to the practical and effective syntheses
of totally unprotected aryl and allyl C-glycosides. Ap-
plication of these methods to natural product synthesis
will be reported elsewhere in detail.19
) 13.2, 5.2, and 3.0 Hz), 2.19 (1H, br s), 2.26 (1H, ddd, J )
13.2, 11.8, and 11.8 Hz), 2.66 (1H, br s), 3.86 (1H, br s), 4.15-
4.35 (1H, m), 4.53 (1H, dq, J ) 6.4 and 1.0 Hz), 5.71 (1H, dd,
J ) 11.8 and 3.0 Hz), 7.1-7.8 (6H), 9.10 (1H, s). HRMS (EI)
m/z 274.1212 (274.1205 calcd for C16H18O4, M+).
1-(2′,6′-Did eoxy-â-D-r ibo-h exop yr a n osyl)-2-n a p h t h ol
(14b-â). Rf 0.65 (5:1 chloroform-methanol); [R]31D +159.1° (c
1
0.90, CHCl3); mp 207.5-208.5 °C (ethyl acetate-hexane); H
NMR δ 1.45 (3H, d, J ) 6.0 Hz), 2.12 (1H, ddd, J ) 14.2, 11.8,
and 1.9 Hz), 2.19 (1H, d, J ) 6.4 Hz), 2.29 (1H, ddd, J ) 14.2,
3.6, and 2.4 Hz), 2.43 (1H, br s), 3.51 (1H, ddd, J ) 9.6, 6.4,
and 4.4 Hz), 4.03 (1H, dq, J ) 9.6 and 6.0 Hz), 4.2-4.3 (1H,
m), 5.88 (1H, dd, J ) 11.8 and 2.4 Hz), 7.05-7.8 (6H), 9.00
(1H, s). Anal. Calcd for C16H18O4: C, 70.06; H, 6.61. Found:
C, 69.81; H, 6.66.
1-(3′-(Dim eth yla m in o)-2′,3′,6′-tr ideoxy-â-D-r ibo-h exop y-
r a n osyl)-2-n a p h t h ol (15b -â). Rf 0.62 (2:1 chloroform-
methanol); [R]31D +159.7° (c 1.15, CHCl3); 1H NMR δ 1.50 (3H,
d, J ) 6.0 Hz), 1.87 (1H, ddd, J ) 13.6, 11.8, and 11.8 Hz),
2.13 (1H, ddd, J ) 13.6, 3.7, and 2.0 Hz), 2.3-2.8 (1H, br),
2.39 (6H, s), 2.92 (1H, ddd, J ) 11.8, 9.2, and 3.7 Hz), 3.33
(1H, dd, J ) 9.2 and 9.2 Hz), 3.67 (1H, dq, J ) 9.2 and 6.0
Hz), 5.88 (1H, dd, J ) 11.8 and 2.0 Hz), 7.05-7.8 (6H), 8.9-
9.1 (1H, br s). Anal. Calcd for C18H23NO3: C, 71.73; H, 7.69;
N, 4.65. Found: C, 71.59; H, 7.86; N, 4.64.
Exp er im en ta l Section
Gen er a l Meth od s. Melting points are uncorrected. 1H
NMR spectra were measured in CDCl3 using TMS as internal
standard unless otherwise noted. High-resolution mass spec-
tra (HRMS) were recorded under electron impact (EI) condi-
tions. Silica gel TLC and column chromatography were
performed on Merck TLC 60F-254 (0.25 mm) and Fuji-Davison
BW-820MH or BW-300, respectively. Air- and/or moisture-
sensitive reactions were carried out under an atmosphere of
argon with oven-dried glassware. In general, organic solvents
were purified and dried by the appropriate procedure, and
evaporation and concentration were carried out under reduced
pressure below 30 °C, unless otherwise noted.
Ar yl C-Glycosid a tion s of Un p r otected 2-Deoxy Su ga r s
by TMSOTf-AgClO4. Gen er a l P r oced u r e. To a mixture
(18) (a) Lewis, M. D.; Cha, J . K.; Kishi, Y. J . Am. Chem. Soc. 1982,
104, 4976. (b) Babirad, S. A.; Wang, Y.; Kishi, Y. J . Org. Chem. 1987,
52, 1370.
(19) For our preliminary communications of those works, see: (a)
Matsuo, G.; Miki, Y.; Nakata, M.; Matsumura, S.; Toshima, K. Chem.
Commun. 1996, 225. (b) Matsuo, G.; Matsumura, S.; Toshima, K.
Chem. Commun. 1996, 2173.
2-(2′,6′-Did eoxy-â-D-r ibo-h exop yr a n osyl)-5-m eth oxy-1-
n a p h th ol (16-â). Rf 0.48 (6:1 chloroform-methanol); [R]28
D
+32.8° (c 0.58, CH3OH); mp 138.5-139.5 °C (ethyl ether-
hexane); 1H NMR δ 1.48 (3H, d, J ) 6.2 Hz), 2.01 (1H, ddd, J
) 13.9, 11.8, and 11.8 Hz), 2.14 (1H, br d, J ) 2.8 Hz), 2.34
(1H, ddd, J ) 13.9, 4.8, and 2.2 Hz), 2.37 (1H, br), 3.28 (1H,

Aryl and Allyl C-Glycosidation Methods
J . Org. Chem., Vol. 63, No. 7, 1998 2313
ddd, J ) 9.2, 9.2, and 2.8 Hz), 3.59 (1H, dq, J ) 9.2 and 6.2
Hz), 3.81 (1H, m), 3.98 (3H, s), 4.88 (1H, dd, J ) 11.8 and 2.2
Hz), 6.82 (1H, d, J ) 8.2 Hz), 7.02 (1H, d, J ) 8.4 Hz), 7.37
(1H, dd, J ) 8.4 and 8.4 Hz), 7.73 (1H, d, J ) 8.4 Hz), 7.82
(1H, d, J ) 8.2 Hz), 8.67 (1H, s). Anal. Calcd for C17H20O5:
C, 67.09; H, 6.62. Found: C, 66.90; H, 6.61.
saturated aqueous NaHCO3. The resultant mixture was then
extracted with ethyl acetate several times, and the extracts
were washed with saturated aqueous NaCl, dried over anhy-
drous Na2SO4, and concentrated in vacuo. Purification of the
residue by flash column chromatography gave the correspond-
ing unprotected and 2,3-unsaturated allyl R-C-glycoside.
3-(2′,3′,6′-Tr id eoxy-r-L-er yth r o-h ex-2′-en op yr a n osyl)-1-
2-(2′,6′-Did eoxy-â-D-r ibo-h exop yr a n osyl)-4-m eth oxy-1-
n a p h th ol (17-â). Rf 0.48 (6:1 chloroform-methanol); [R]23
p r op en e (21a -r): Rf 0.60 (3:2 hexane-acetone); [R]27 -2.3°
D
D
1
+38.2° (c 1.20, CH3OH); H NMR δ 1.47 (3H, d, J ) 6.0 Hz),
(c 0.77, CHCl3); 1H NMR δ 1.27 (3H, d, J ) 6.2 Hz), 1.73 (1H,
d, J ) 8.4 Hz), 2.31 (1H, br ddd, J ) 13.8, 6.8, and 6.8 Hz),
2.42 (1H, br ddd, J ) 13.8, 6.8, and 6.8 Hz), 3.65-3.75 (1H,
m), 3.77 (1H, dq, J ) 6.2 and 4.6 Hz), 4.20 (1H, m), 5.05-5.2
(2H, m), 5.75-5.95 (3H, m); HRMS (EI) m/z 155.1085 (155.1072
calcd for C9H15O2, M + H+). Anal. Calcd for C9H14O2: C,
70.10; H, 9.15. Found: C, 69.98; H, 9.33.
2.02 (1H, ddd, J ) 13.2, 11.5, and 11.5 Hz), 2.23 (1H, br d, J
) 2.6 Hz), 2.37 (1H, ddd, J ) 13.2, 5.0, and 2.2 Hz), 2.42 (1H,
br d, J ) 2.6 Hz), 3.29 (1H, ddd, J ) 9.2, 9.2, and 2.6 Hz), 3.57
(1H, dq, J ) 9.2 and 6.0 Hz), 3.80 (1H, m), 3.92 (3H, s), 4.81
(1H, dd, J ) 11.5 and 2.2 Hz), 6.35 (1H, s), 7.41-7.57 (2H),
8.1-8.24 (2H), 8.24 (1H, s). Anal. Calcd for C17H20O5: C,
67.09; H, 6.62. Found: C, 66.98; H, 6.85.
3-(2′,3′-Dideoxy-r-D-er yth r o-h ex-2′-en opyr an osyl)-1-pr o-
2-(2′,6′-Did eoxy-â-D-r ibo-h exop yr a n osyl)-3,5-d im eth ox-
p en e (22a -r): Rf 0.50 (1:1 benzene-acetone); [R]27 -33.4°
D
yp h en ol (18-â). Rf 0.54 (5:1 chloroform-methanol); [R]28
(c 0.94, CHCl3); mp 27.5-28.5 °C (hexane, needle); 1H NMR δ
2.06 (2H, br s), 2.31 (1H, br ddd, J ) 13.8, 6.8, and 6.8 Hz),
2.47 (1H, br ddd, J ) 13.8, 6.8, and 6.8 Hz), 3.55 (1H, ddd, J
) 7.9, 5.9, and 4.3 Hz), 3.77 (1H, dd, J ) 11.7 and 5.9 Hz),
3.83 (1H, dd, J ) 11.7 and 4.3 Hz), 4.10 (1H, br d, J ) 7.9
Hz), 4.24 (1H, m), 5.05-5.2 (2H, m), 5.75-5.95 (3H, m); HRMS
(EI) m/z 171.1003 (171.1021 calcd for C9H15O3, M + H+). Anal.
Calcd for C9H14O3: C, 63.51; H, 8.29. Found: C, 63.44; H, 8.46.
3-(2′,3′-Did eoxy-r-D-th r eo-h ex-2′-en op yr a n osyl)-1-p r o-
D
+82.8° (c 2.07, CH3OH); mp 185.0-186.0 °C (acetone-hexane);
1H NMR δ 1.41 (3H, d, J ) 6.0 Hz), 1.81 (1H, ddd, J ) 13.4,
11.6, and 11.6 Hz), 2.16 (1H, d, J ) 4.4 Hz), 2.21 (1H, ddd, J
) 13.4, 4.8, and 2.2 Hz), 2.47 (1H, d, J ) 3.5 Hz), 3.22 (1H,
ddd, J ) 9.2, 9.2, and 4.4 Hz), 3.49 (1H, dq, J ) 9.2 and 6.0
Hz), 3.7-3.82 (1H, m), 3.75 (3H, s), 3.76 (3H, s), 5.06 (1H, dd,
J ) 11.6 and 2.2 Hz), 6.00 (1H, d, J ) 2.0 Hz), 6.07 (1H, d, J
) 2.0 Hz), 8.45 (1H, s). Anal. Calcd for C14H20O6: C, 59.14;
H, 7.09. Found: C, 59.05; H, 7.31.
p en e (23a -r): Rf 0.44 (1:1 benzene-acetone); [R]27 -285.5°
D
2-(2′,6′-Dideoxy-â-D-r ibo-h exopyr an osyl)-3,4,5-tr im eth ox-
(c 0.77, CHCl3); mp 50.0-51.0 °C (hexane, needle); 1H NMR δ
2.06 (2H, br s), 2.28 (1H, br ddd, J ) 13.8, 6.8, and 6.8 Hz),
2.46 (1H, br ddd, J ) 13.8, 6.8, and 6.8 Hz), 3.75-3.95 (4H,
m), 4.32 (1H, m), 5.12 (1H, br d, J ) 10.2 Hz), 5.19 (1H, br d,
J ) 17.2 Hz), 5.84 (1H, ddt, J ) 17.2, 10.2, and 6.8 Hz), 5.94
(1H, dd, J ) 10.2 and 3.2 Hz), 6.05 (1H, ddd, J ) 10.2, 5.4,
and 2.0 Hz); HRMS (EI) m/z 170.0957 (170.0942 calcd for
C9H14O3, M+). Anal. Calcd for C9H14O3: C, 63.51; H, 8.29.
Found: C, 63.37; H, 8.48.
yp h en ol (19-â). Rf ) 0.39 (6:1 chloroform-methanol); [R]27
D
+56.7° (c 0.67, CH3OH); mp 106.5-107.5 °C (ethyl acetate-
hexane); 1H NMR δ 1.47 (3H, d, J ) 6.0 Hz), 1.88 (1H, ddd, J
) 13.4, 11.6, and 11.6 Hz), 2.17 (1H, ddd, J ) 13.4, 4.8, and
2.3 Hz), 2.40 (1H, br s), 2.66 (1H, br s), 3.21 (1H, br dd, J )
9.2 and 9.2 Hz), 3.50 (1H, dq, J ) 9.2 and 6.0 Hz), 3.7-3.85
(1H, m), 3.76 (3H, s), 3.80 (3H, s), 3.89 (3H, s), 4.95 (1H, dd,
J ) 11.6 and 2.3 Hz), 6.77 (1H, s), 8.11 (1H, s). Anal. Calcd
for C15H22O7: C, 57.32; H, 7.05. Found: C, 57.29; H, 7.35.
2-(2′,6′-Did eoxy-â-D-r ibo-h exop yr a n osyl)-9,10-d im eth -
oxy-1-h yd r oxya n th r a cen e (20-â). Rf 0.42 (8:1 chloroform-
methanol); [R]23D +72.1° (c 1.17, CH3OH); 1H NMR δ 1.47 (3H,
d, J ) 6.0 Hz), 1.6-2.5 (2H, br), 1.75 (1H, ddd, J ) 13.8, 11.6,
and 11.6 Hz), 2.38 (1H, ddd, J ) 13.8, 5.0, and 2.1 Hz), 3.28
(1H, dd, J ) 9.2 and 9.2 Hz), 3.58 (1H, dq, J ) 9.2 and 6.0
Hz), 3.89 (1H, ddd, J ) 11.6, 9.2, and 5.0 Hz), 4.07 (3H, s),
4.12 (3H, s), 5.19 (1H, dd, J ) 11.6 and 2.1 Hz), 7.45-7.55
(2H), 7.60 (1H, d, J ) 8.6 Hz), 7.82 (1H, d, J ) 8.6 Hz), 8.13-
8.21 (1H), 8.26-8.30 (1H), 11.5 (1H, s). Anal. Calcd for
3-(2′,3′,6′-Tr id eoxy-r-D-th r eo-h ex-2′-en op yr a n osyl)-1-
p r op en e (24a -r): Rf 0.62 (3:2 hexanes-ethyl acetate); [R]28
D
1
-292.2° (c 0.98, CHCl3); H NMR δ 1.26 (3H, d, J ) 6.2 Hz),
1.55 (1H, br s), 2.27 (1H, br ddd, J ) 13.8, 6.8, and 6.8 Hz),
2.44 (1H, br ddd, J ) 13.8, 6.8, and 6.8 Hz), 3.68 (1H, br dd,
J ) 5.4 and 2.1 Hz), 3.92 (1H, dq, J ) 6.2 and 2.1 Hz), 4.23
(1H, m), 5.10 (1H, br d, J ) 10.1 Hz), 5.12 (1H, br d, J ) 17.3
Hz), 5.86 (1H, ddt, J ) 17.3, 10.1, and 6.8 Hz), 5.89 (1H, dd,
J ) 10.2 and 3.2 Hz), 6.04 (1H, ddd, J ) 10.2, 5.4, and 2.0
Hz). HRMS (EI) m/z 154.1008 (154.0994 calcd for C9H14O2,
M+). Anal. Calcd for C9H14O2: C, 70.10; H, 9.15. Found: C,
69.91; H, 9.42.
C
22H24O6: C, 68.74; H, 6.29. Found: C, 68.50; H, 6.49.
Allyl C-Glycosid a t ion s of Un p r ot ect ed Glyca ls b y
TMSOTf. Gen er a l P r oced u r e. To a mixture of glycosyl
donor (0.1 mmol) and allyltrimethysilane (0.2 mmol) in dry
solvent described in Table 9 was added trimethylsilyl trifluo-
romethanesulfonate (0.1 mmol) dropwise at -78 °C under
argon. The reaction mixture was stirred at the same temper-
ature for 0.5, 1, or 2 h and quenched with ice-cold and
Ack n ow led gm en t. We are indebted to Mr. K.
Hokazono of Keio University for elemental analyses.
Financial support by The Nissan Science Foundation
is gratefully acknowledged.
J O972146V