Stereoselective glycosylations using benzoylated glucosyl halides with inexpensive promoters

Article abstract of DOI:10.1016/j.carres.2008.03.019

Reactions of O-benzoylated glucopyranosyl halide (I, Br), isolated or generated in situ from per-benzoylated glucose (8a) and trimethylsilyl halide, with various alcohols were efficiently promoted by zinc halide (Cl, Br) or N-bromosuccinimide with a catal

Full text of DOI:10.1016/j.carres.2008.03.019

Available online at www.sciencedirect.com
Carbohydrate Research 343 (2008) 1297–1308
Stereoselective glycosylations using benzoylated
glucosyl halides with inexpensive promoters
Teiichi Murakami,* Yukari Sato and Motonari Shibakami
Institute for Biological Resources and Functions, National Institute of Advanced Industrial Science and Technology (AIST),
Tsukuba Central 5, Tsukuba, Ibaraki 305-8565, Japan
Received 30 January 2008; received in revised form 26 February 2008; accepted 11 March 2008
Available online 17 March 2008
Abstract—Reactions of O-benzoylated glucopyranosyl halide (I, Br), isolated or generated in situ from per-benzoylated glucose (8a)
and trimethylsilyl halide, with various alcohols were efficiently promoted by zinc halide (Cl, Br) or N-bromosuccinimide with a
catalytic ZnI2 to give the corresponding 1,2-trans-b-glucosides in good to high yields. When the anomeric halogenation of 8a
was carried out in the presence of reactive alcohols, 1,2-cis-a-glucosides were selectively formed.
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Glycosylation; Disarmed glycosyl halides; Lewis acids; NBS; Anomer
1. Introduction
In contrast, acyl-protected, ‘disarmed’ glycosyl
iodides have proved to be rather stable.11,12 These can
be purified by silica gel chromatography and be stored
for months.11h,12b,c Their reactions with alcohols have
usually been performed by using conventional Koe-
nigs–Knorr promoters such as AgOTf and Hg(CN)2.13
To avoid the use of those heavy metals, iodonium
reagents and/or Lewis acids have recently been
employed.14 However, the scope of these glycosylations
has not been well studied.
In the course of our synthetic studies of glycolipids, we
were involved in 1,2-trans-glycosylation reactions using
sugar peracetates by neighboring-group participation.15
Although the glycosylations promoted by ZnCl2 in
toluene have proved to be convenient, the yields are
moderate (40–65%) and the glycosyl acceptors are lim-
ited to simple alcohols. Our attention was then turned
to the reactivity of disarmed glycosyl iodides. In this pa-
per, we describe efficient glycosylation processes from
per-benzoylated glucose via the iodide, using inexpensive
and safe promoters such as ZnBr2 and N-bromosuccini-
mide (NBS). These promoters have also been found to
activate less reactive tetra-O-benzoyl-a-D-glucopyran-
osyl bromide to give the 1,2-trans-glucosides in good-
to-high yields. In addition, unexpected 1,2-cis-selective
Glycosylation of alcohols (O-glycosylation)1 is an essen-
tial process for the synthesis of oligosaccharides and
glycoconjugates such as glycolipids. In the last three
decades, a variety of glycosyl donors2 have been devel-
oped, including thioglycosides3 and 1-O-trichloroacet-
imidates.4 Nevertheless glycosyl halides, especially
bromides, are still commonly employed as donors since
the pioneering work of Koenigs and Knorr.5 Glycosyl
fluorides have now been widely employed for O-glycosyl-
ations6 due to the high thermal and chemical stability,
whereas glycosyl iodides have been considered impracti-
cal reagents in glycosylation reactions due to their insta-
bility.7 However, in recent years, the anomeric iodides
have attracted some interest8 and have been extensively
studied by Gervay-Hague’s group.9 Highly reactive
‘armed’ (typically O-benzyl protected) glycosyl iodides
are generated from the 1-chloride,8a 1-acetate,9 1-phos-
phate,8c or 1-silyl ether,10 and are reacted in situ with
nucleophiles by SN2 displacement in the presence or
absence of catalyst.
*
Corresponding author. Fax: +81 29 861 4549; e-mail: t-murakami@
aist.go.jp
0008-6215/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2008.03.019

1298
T. Murakami et al. / Carbohydrate Research 343 (2008) 1297–1308
glucosylation has been observed when the anomeric
halogenation is carried out in the presence of reactive
alcohols.
reddish-brown, and 2 was consumed with the appear-
ance of two major products on TLC. The polar product
was the expected b-glucoside 4 (30–40% yield based on
2) and the less polar product was orthoester 5
(30–40%), indicating that the NBS can activate the
glucosyl iodide 2. Since orthoesters would be rearranged
2. Results and discussion
to b-glycosides under Lewis acid conditions,18
a
catalytic amount of a Lewis acid such as trimethylsilyl
triflate (TMSOTf) was added to the mixture before or
during the reaction (e.g., entries 1 and 2 in Table).
However, 4 did not increase (<50% yields), while 2-O-
deacetylated glucosides 6 (a/b mixture) and O-acetyl-
ated alcohol 7 increased. Next, instead of NBS, several
Lewis acids were employed as promoter (e.g., entries 3
and 4). The glycosylation proceeded at rt, but the yields
of 4 were generally lower (20–40%). Therefore, the
glycosylation with iodide 2 has proved less efficient than
the direct glycosylation of 3 with the b-acetate 1
promoted by ZnCl2: 4 was obtained in 58% yield.15
2.1. Glycosylation with acetylated glucosyl iodide
Disarmed glycosyl iodides reported thus far are primarily
the acetyl-protected ones.11 These have been prepared
from the corresponding sugar peracetates with iodotri-
methylsilane (TMS-I)9b,11a,c,f or other HI equiva-
lents.11b,e,g Since TMS-I is unstable and expensive,
in situ generation from iodine and hexamethyldisilane
(HMDS)16 would be preferable for practical synthesis.11d,h
According to the literature,11h b-glucose pentaacetate
1 was treated with iodine (0.6 equiv) and HMDS
(0.6 equiv) in CH2Cl2 at rt to give tetra-O-acetyl-a-D-
glucopyranosyl iodide (2) in essentially pure form by
solvent evaporation or simple extractive workup
(Scheme 1). With iodide 2 in hand, we examined the
reactions with 12-bromo-1-dodecanol (3) using activa-
tors other than Ag and Hg salts. Iodonium reagents
such as N-iodosuccinimide (NIS) and I2 have been em-
ployed for the activation of glycosyl halides.17 However,
to our knowledge, bromonium reagents have not been
used with such donors. We first employed N-bromosuc-
cinimide (NBS) as a promoter because NBS is more sta-
ble and much cheaper than NIS. Thus, to a mixture of 2
and 3 (1.2 equiv) in CH2Cl2 with 4A MS was added
NBS (1.2 equiv) at 5 °C. The mixture soon turned to
2.2. Preparation of benzoylated glucosyl iodide and a
one-pot glycosylation process
There have been reported an increasing number of
unsuccessful examples of glycosylations17d,19 with an
acetate group at C-2 owing to the formation of 1,2-
orthoacetates, which would be decomposed to major
byproducts, 2-O-deacetylated glycosides (a/b mixture)
and O-acetylated alcohols as shown in Scheme 1. Such
byproducts can be reduced by using either a benzoyl19a,b
or a pivaloyl19c group at C-2. Therefore we set out
to prepare O-benzoyl-protected glycosyl iodides.
I2
OAc
Br-(CH2)12-OH
OAc
OAc
(Me3Si-)2
3
O
O
O
AcO
AcO
AcO
AcO
AcO
O-(CH2)12Br
AcO
OAc
promoter
MS4A
CH2Cl2
CH2Cl2
AcO
AcO
AcO
I
4 (β-glucoside)
2
1
+
(rearrangement)
OAc
O
AcO
AcO
OAc
O-(CH2)12Br
HO
O
AcO
AcO
6 (α/β mixture)
(degradation)
O
+
O
O-(CH2)12Br
Br-(CH2)12-OAc
5 (orthoester)
7
Yield (% based on 2)
Entry
Promoter (equiv)
Conditions
5 °C to rt, 4 h
4
6
7
NBS (1.2), TMSOTf (0.4), TMU (0.2)
38
47
24
28
20 42
19 31
16 46
22 32
1
2
3
4
NBS (1.2), Cu(OTf)2 (0.12)
ZnCl2 (1.4)
5 °C to rt, 28 h
rt, 12 h
CuOTf (1.2)
5 °C, 2 h
Scheme 1. Glycosylation with acetylated glucosyl iodide.

T. Murakami et al. / Carbohydrate Research 343 (2008) 1297–1308
1299
Stachulski’s group12c and Field’s group14b reported that
perpivaloylated methyl b-glucuronate and perbenzoylated
glucose are converted to the a-anomeric iodides by
treatment with I2 (0.6 equiv) and HMDS (0.6 equiv) in
CH2Cl2 at room temperature for 5–6 h, respectively. In
our hands, however, the iodination of a-D-glucopyr-
anose pentabenzoate 8a20 under identical reaction condi-
tions was incomplete at rt for 12 h (ca. 80% conversion).
As shown in Scheme 2, we found that the addition of
ZnI2 (0.1–0.3 equiv) as a cheap Lewis acid iodide source
accelerated the reaction to afford iodide 9a21 in nearly
quantitative yield. Molecular sieves were not included
in the reaction mixture since these retarded the rates
of the reaction considerably. Trimethylsilyl benzoate
10, a byproduct necessarily generated from 8a and
TMSI, can be removed by extractive workup with aq
NaHCO3–Na2S2O3, and iodide 9a is isolated in a suffi-
ciently pure form. Addition of ZnBr2 instead of ZnI2
gave a mixture of 9a and the a-glucosyl bromide 9b.
We initially examined a one-pot glycosylation process
from 8a without the isolation of iodide 9a since 9a is
essentially pure and ZnI2 in the reaction mixture might
serve as an acid catalyst, which is often necessary for
Fig. 1) was obtained in 62% yield. Although 8a
disappeared at the iodination stage, b-glucose penta-
benzoate 8b was obtained in 18% yield. Byproduct 8b
would be formed by the reaction of 9a with trimethyl-
silyl benzoate (10). Other byproducts were orthoester
12 (7%) and trimethysilylated alcohol. Neither 2-O-
debenzoylated glucoside nor O-benzoylated alcohol
was obtained.
This NBS-promoted one-pot glucosylation was ap-
plied to other alcohols (in Fig. 1), and the results are
summarized in Table 1. Addition of 2-phenylethanol
(13), cyclohexanol (15), and 1,2:3,4-di-O-isopropyl-
idene-a-D-galactopyranose (Ip-Gal-OH, 17) afforded
the corresponding b-glucosides in good yields (entries
2–4). In contrast, N-Cbz-serine methyl ester (Z-Ser-
OMe, 19) gave glucoside 20 in very low yield (10%)
under identical conditions, along with several byproducts:
orthoester 12S (40%), b-benzoate 8b (30%), iodide 9a
(15%), and trimethysilylated alcohol (36%) (entry 5).
Formation of larger amounts of 8b and orthoester 12S
would result from the lower reactivity of 19 compared
to the other alcohols. Polt and co-workers reported that
N-acylated b-amino-alcohols such as N-Cbz-serine
derivatives have decreased nucleophilicity due to an
unfavorable hydrogen-bonding pattern.23
the
N-halosuccinimide-promoted
glycosylations.22
Thus, to this reaction mixture at 5 °C were added 12-
bromododecanol (3, 1.5 equiv) and NBS (1.2 equiv)
with 4A molecular sieves (Scheme 3) Color of the
reaction mixture turned reddish-purple and 9a soon
decreased. To our delight, b-glucopyranoside 11 (in
Under Lewis acid conditions, a 1,2-orthoester can be
rearranged to the b-glycoside, and a trimethylsilylated
alcohol can be glycosylated.24 Thus we examined the
effect of Lewis acids in place of NBS. To the reaction
I2 (0.6 equiv)
OBz
O
OBz
O
(Me3Si-)2 (0.6 equiv)
BzO
BzO
BzO
BzO
+
Me3Si-OBz
CH2Cl2
rt, 6 h
BzO
BzO
10
OBz
I
9a
8a
no additive
(80%)
with ZnI2 (0.2 equiv)
(>95%)
Scheme 2. Preparation of benzoylated glucosyl iodide.
R-OH
(1.3-1.7 equiv)
I2
OBz
O
OBz
OBz
promoter
(Me3Si-)2
(1.0-1.5 equiv)
O
O
BzO
BzO
BzO
BzO
BzO
BzO
8a
O-R
OBz
ZnI2 (0.3 equiv)
CH2Cl2
rt, 6 h
MS4A
CH2Cl2
BzO
BzO
BzO
I
1-β-benzoate 8b
β-glucoside
9a
OBz
O
BzO
BzO
Me3Si-OR
O
O
Ph
silylated alcohol
O-R
orthoester 12
Scheme 3. One-pot glucosylation via the anomeric iodide 9a.

1300
T. Murakami et al. / Carbohydrate Research 343 (2008) 1297–1308
OBz
R'OCH2CH2Ph
R'O
R'O(CH2)12Br
R' = H
11 R' = Bz-β-Glc-
O
Bz-β-Glc- =
BzO
BzO
13 R' = H
3
15 R' = H
14 R' = Bz-β-Glc-
BzO
16 R' = Bz-β-Glc-
OR'
O
NHCO2Bn
O
O
R'O
R'O
CO2Me
O
O
21 R' = H
22 R' = Bz-β-Glc-
19 R = H (Z-Ser-OMe)
20 R = Bz-β-Glc-
17 R' = H (Ip-Gal-OH)
18 R' = Bz-β-Glc-
OBn
O
R'O
BnO
BnO
O
R'O
BnO
BnO
BnO
OMe
OMe
25 R' = H (Glc-6-OH)
26 R' = Bz-β-Glc-
27 R' = H (Glc-4-OH)
28 R' = Bz-β-Glc-
R'O
23 R' = H
24 R' = Bz-β-Glc-
OBz
O
OBz
BzO
BzO
OBz
O
O
BzO
BzO
BzO
BzO
O
NHCO2Bn
CO2Me
O
Ph
OH
BzO
29
BzO
X
O
12S
9b X = Br
9c X = Cl
Figure 1. Acceptor alcohols and products in Tables 1 and 2.
Table 1. One-pot glycosylation of alcohols with benzoylated glucosyl iodide 9a prepared in situ from the pentabenzoate 8a
Entry
(For glycosylation)
Conditionsb
Yieldc (%)
R–OHa
Promoter (equiv)
b-Glucoside
8b
12
1
2
3
4
5
6
7
8
9
Br(CH2)12–OH (3)
Ph(CH2)2–OH (13)
Cyclohexanol (15)
Ip-Gal-OH (17)
Z-Ser-OMe (19)
3
NBS (1.2)
NBS (1.1)
NBS (1.5)
NBS (1.2)
NBS (1.5)
ZnI2 (1.0)
ZnCl2 (1.6)
ZnCl2 (1.1)
ZnCl2 (1.2)
ZnBr2 (1.4)
ZnBr2 (1.0)
NBS (1.5)
5–10 °C, 6 h
5–10 °C, 6 h
5 °C, 3 h
5–10 °C, 6 h
5 °C, 4 h
40 °C, 5 h
rt, 5 h
rt, 20 h
rt, 6 h
rt, 30 h
rt, 4 h
11
14
16
18
20
11
11
22
16
18
11
62
67
86
67
10d
82e
79
84
84
70
78
18
22
0
7
30
7
16
14
6
15
11
7
0
0
20
40
3
4-Penten-1-ol (21)
15
17
3
10
11f
12f
3
rt, 4 h
No reaction
a Alcohol used was 1.3–1.7 equiv to 8a.
b Reaction was carried out in CH2Cl2 until 9a was consumed. However, in several entries, a small amount (3–6%) of the donor remained as a mixture
of 9a and 9b or 9a and 9c.
c Isolated yield after chromatography, based on 8a.
d 9a was recovered (15%).
e a-Anomer was obtained in 4% yield.
f TMSBr was employed instead of I2–HMDS.
mixture of the glucosyl iodide containing ZnI2
(0.3 equiv) in CH2Cl2 were added 12-bromododecanol
(3, 1.5 equiv) and ZnI2 (1.0 equiv) with 4A molecular
sieves. The reaction was slow at rt, but proceeded at
40 °C to give glucoside 11 in 82% yield (entry 6). A small
amount of the a-glucoside 11a was obtained (4%). Addi-
tion of ZnCl2 instead of ZnI2 was found to promote the
reaction at rt, giving 11 in 79% yield along with 8b (entry
7). Under Lewis acid conditions, the orthoester was not
obtained. This ZnCl2-promoted one-pot glycosylation

T. Murakami et al. / Carbohydrate Research 343 (2008) 1297–1308
1301
was applied to other alcohols: 4-pentenol 21, which can-
not be used with NBS, and cyclohexanol 15 afforded the
corresponding b-glucopyranosides in high yields with no
a-anomer (entries 8 and 9). Thus ZnCl2 would be a more
effective promoter than ZnI2. However, when the
reaction proceeded slowly with ZnCl2, anomeric
halogen exchange occurred to give less reactive glucosyl
chloride 9c.
The halogen exchange from I (9a) to Br (9b) occurred
in the NBS-activation as well as the ZnBr2-activation
(entry 10), and a small amount (3–6%) of the mixture
(9a and 9b) was recovered in several entries. These
results led us to examine the reaction of glucopyranosyl
bromide 9b with 12-bromododecanol 3 in the presence
of NBS or ZnBr2. Treatment of 8a with TMSBr
(1.6 equiv) and ZnBr2 (0.3 equiv) gave bromide 9b in
nearly quantitative yield,11a,c,25 though the bromination
of 8a required a longer reaction time (at rt for 18 h) than
the iodination (rt, 5 h). To this mixture were added 3
(1.3 equiv) and promoter (1.0 equiv) with molecular
sieves 4 A. Addition of ZnBr2 afforded the glucoside
11 in 78% yield (entry 11), whereas the addition of
NBS did not promote the reaction (entry 12). These
results suggest that the NBS-activation of glycosyl
bromides would require ‘I’ source to form more active
species such as NIS or I–Br.17b,d
2.3. Comparison of reactivities
Due to the presence of non-volatile trimethylsilyl benzo-
ate 10, this convenient one-pot glycosylation should give
significant amounts of b-benzoate 8b as shown in Table
1. Gervay-Hague reported that magnesium oxide effec-
tively deactivates TMSOAc to suppress the formation
of a protected glucose b-acetate.26 In our case, however,
MgO (1 equiv) suppressed the glycosylation. Addition
of CaCO3 in the place of MgO showed little effect for
the deactivation of TMSOBz.
As mentioned above, essentially pure glucosyl iodide
9a was obtained simply by treatment with aqueous
NaHCO3–Na2S2O3 to remove TMS-OBz 10 and excess
TMS-I. Thus we employed the isolated iodide 9a, and
compared the donor reactivity to tetra-O-benzoyl-a-D-
glucopyranosyl bromide 9b. Bromide 9b is more stable
than iodide 9a, and it is easily available from 8a with
HBr–AcOH20a or from a commercial source.
We examined the glucosylation using either NBS with
a catalytic amount of acid or Lewis acid alone (ZnX2) as
Table 2. Glycosylation of various alcohols with glucosyl bromide 9b or iodide 9a
OBz
OBz
O
promoter
O
BzO
R-OH
+
BzO
BzO
O-R
BzO
MS4A
CH2Cl2
BzO
BzO
X
9a X = I
9b X = Br
β-glucoside
Entry
R–OHa
Promoter (equiv)
Conditions
Yield (%) of b-glucoside
From 9bb (X = Br) From 9ac (X = I)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
3
13
NBS (1.3) + ZnI2 (0.2)
NBS (1.5) + ZnI2 (0.3)
NBS (1.5) + Zn(OTf)2 (0.3)
NBS (1.5) + TMSOTf (0.3)
NBS (1.5) + ZnI2 (0.3)
NBS (1.5) + ZnI2 (0.6)
NBS (1.5) + ZnI2 (0.3) + TMSOTf (0.5)
NBS (1.1) + TMSOTf (0.5)
ZnCl2 (1.5)
ZnCl2 (2.0)
ZnBr2 (2.5)
ZnBr2 (1.5)
ZnBr2 (1.8)
5 °C to rt, 24 h
5 °C to rt, 14 h
5 °C to rt, 20 h
5 °C to rt, 3 h
5 °C to rt, 5 h
5 °C to rt, 5 h
5 °C, 3 h to rt, 6 h
5 °C, 3 h
rt, 4 h
rt, 48 h
40 °C, 8 h
rt, 3 h
40 °C, 6 h
rt, 10 h
rt, 24 h
11
14
14
14
18
20
85
90
83
90
13
13
7d
0d
90
83
64e
87
17
19
31f
nah
nah
85
84i
60j
90
nah
48g
56e
91
19
Glc-6-OH 25
3
17
20
26
11
18
20
22
24
26
28
75i
55j
85
90
74
48j
19
21
Cholesterol 23
25
93
75
53j
ZnBr2 (2.0)
ZnBr2 (2.0)
Glc-4-OH 27
a Alcohol used was 1.3–1.7 equiv to 9, except for 19 and 27 (2 equiv).
b Isolated yield based on 9b.
c Overall yield from benzoate 8a, not based on 9a.
d Donor 9b was recovered (ca. 90%).
e A mixture of 9a/9b was recovered (ca. 20%).
f,g A mixture of mono-hydroxy-glucoses was obtained (f: 66%, g: 50%).
h Not attempted.
i A small amount (<10%) of glucosyl chloride 9c was obtained.
j 1-Hydroxy-glucose 29 was obtained (25–40%).

1302
T. Murakami et al. / Carbohydrate Research 343 (2008) 1297–1308
promoter. The results are summarized in Table 2. As
expected, the yields of b-glucosides from the isolated
iodide 9a were higher than those by the one-pot process.
We were pleased to find that bromide 9b also reacted
with a variety of alcohols by both promoter systems to
give the b-glucosides with high stereoselectivity in com-
parable good yields except entries 3 and 4, in which little
or no glucoside 14 was obtained from 9b by the NBS-
activation. These entries and entry 12 in Table 1 have
revealed that a catalytic ‘I’ source such as ZnI2 is neces-
sary for the activation of glycosyl bromides with NBS.
In the NBS–ZnI2 activation (entries 1, 2, 5–7), iodide
9a was consumed faster (at 5 °C for 1 h) than bromide
9b (at rt for several hours) to afford the b-glucoside
and the orthoester, which was generally less polar than
the former on TLC, in a variable ratio. The orthoester
gradually decreased with the increase of b-glucoside by
further stirring at rt for 5–24 h in most cases. In the case
of Z-Ser-OMe 19 (entry 7), orthoester 12S (ca. 80%) and
b-glucoside 20 (ca. 10%) were formed at 5 °C within 1 h.
To smoothly promote the rearrangement, TMSOTf
(0.5 equiv) was added to the mixture, and the mixture
was stirred at rt until 12S disappeared. However, the
yield of 20 was modest, and a mixture of 1-hydroxy-
(29) and 2-hydroxy-benzoylated glucoses was obtained
in 50% yield.
Since the orthoesters have not been obtained with
ZnX2 alone, the ZnX2 activation seems more general
and operationally simpler than the NBS activation. In
the ZnX2 activation, the reaction rates of 9b with alco-
hols were similar to those of 9a. Noteworthy is that yield
from bromide 9b was slightly higher than that from io-
dide 9a in many entries. This result might be explained
by two reasons: one is that the latter is not based on
9a, but an overall yield from benzoate 8a; the other rea-
son is the hydrolytic instability of 9a. Although ZnCl2
and ZnBr2 showed similar activities, ZnBr2 would be
preferred toward less reactive alcohols since ZnCl2
would cause the halogen exchange of the donor to form
less reactive glucosyl chloride 9c (e.g., entry 10). By
using ZnBr2, cholesterol 23 reacted with donor 9a or
9b at 40 °C to afford the b-glucoside 24 in high yield (en-
try 13). For less reactive alcohols such as Z-Ser-OMe
(19) and methyl 2,3,6-tri-O-benzyl-a-D-glucopyranoside
(Glc-4-OH, 27) the yields of b-glucosides were still mod-
est, and substantial amounts of 1-hydroxy-glucose 29
were obtained (entries 11 and 15).
promoters. They also reported the glucuronidation
using Lewis acid alone [ZnCl2, CeCl3, or Sc(OTf)3],
but the substrate used was only 2-phenylethanol.14a
The results described herein indicate that zinc halides
and NBS would be promising substitutes for the
Koenigs–Knorr promoters such as AgOTf. The present
methods appear to be more dependent on the nucleo-
philicity of alcohols than the conventional Ag+-pro-
moted dehalogenative glycosylations, and thus the
yields with reactive alcohols are comparable. Indeed,
the yields of the b-glucosides 18 (84%), 22 (90%), and
24 (93%) by the ZnX2-activation shown in Table 2 are
higher than those reported for the AgOTf-promoted
glucosylations (18: 75%,19a 22: 68%,29 24: 60%19c).
2.4. 1,2-cis-Selective glucosylation
We next attempted a tandem bromination and glucosyl-
ation process: the reaction of a-benzoate 8a with
TMSBr–ZnBr2 in the presence of the alcohol 3. The
reaction proceeded smoothly at rt, and 8a disappeared
within 2 h. Two major products obtained were a-gluco-
pyranosyl bromide 9b (49%) and bromododecyl a-
glucopyranoside 11a (45%), and only a small amount
of b-glucoside 11 (2%) was obtained. To examine this
unexpected 1,2-cis-a-selectivity despite the presence of
participating benzoyl group at C-2, other alcohols as
well as the iodination reagents, TMSI–ZnI2, were
employed. The results are summarized in Table 3. The
standard reaction was carried out with TMSBr
(1.6 equiv) and ZnBr2 (1.2 equiv) until 8a was con-
sumed. Under such conditions (entries 1, 5, 7, 9 and
10), reactive primary alcohols (3, 13) afforded corre-
sponding a-glucosides (11a, 14a) and the a-bromide 9b
with a small amount of the b-glucosides. Cyclohexanol
15 gave glucoside 16a in lower yield than the primary
alcohols with a larger amount of 9b (entry 7). In con-
trast, cholesterol 23 and Z-Ser-OMe (19) gave a little
and no a-glucoside, respectively. Hence, the yield of
a-glucosides appears to be dependent on the reactivity
of the parent alcohols. In the presence of 23, bromide
9b was obtained in 83% yield (entry 9), whereas the yield
of 9b was only 20% in the presence of 19, and 76% of 8a
was recovered (entry 10). These results suggest that
acceptor 19 would deactivate the promoter ZnBr2 due
to its Lewis base nature, which can also be responsible
for the low reactivity of 19 as shown in Tables 1 and
2. Longer reaction times at rt to promote further
glucosylation and/or anomerization did not show a clear
tendency. In all such cases (entries 2–4, 6 and 8) the
a-halides 9a/9b apparently decreased, and the a-gluco-
sides increased in entries 3 (3 with a catalytic amount
of ZnBr2) and 8 (15 with TMSI–ZnI2). However, the
other entries did not have a positive effect, and signifi-
cant amounts of the hydrolyzed product 29 were formed
instead. TLC monitoring of the reactions indicated that
Several Lewis acids,27 including zinc halides,12c,14a,28
have been reported to promote the reaction of disarmed
glycosyl halides (mainly O-acetylated bromides, except
glucosamine-type donors) with alcohols. However, these
have not been widely employed for the O-glycoside syn-
thesis, and the scope has not been sufficiently explored.
For example, Stachulski and co-workers recently
reported the glycosylation of various alcohols with
pivaloylated glucuronate iodide using FeCl3 with I2 as

T. Murakami et al. / Carbohydrate Research 343 (2008) 1297–1308
Table 3. cis-Selective glucosylation of reactive alcohols
1303
Me3Si-X
ZnX2
OBz
O
OBz
O
OBz
O
BzO
BzO
BzO
BzO
BzO
BzO
R-OH
+
+
CH2Cl2
5 °C to rt
BzO
BzO
BzO
OBz
O-R
X
α-glucoside
α-X
9a X = I
9b X = Br
8a
11α R = Br(CH2)12
-
14α R = Ph(CH2)2-
16α R = cyclohexyl
Entry
R–OHa
Xb,c
Conditions
Yieldd (%)
a-Glc
b-Glc
a-X
1
2
3
4
5
6
7
8
9
Br(CH2)12–OH (3)
3
Br
Br
Brc
I
Br
I
Br
I
Br
Br
5 °C to rt, 2 h
5 °C to rt, 24 h
5 °C to rt, 24 h
5 °C to rt, 48 h
5 °C to rt, 2 h
5 °C to rt, 40 h
5 °C to rt, 3 h
5 °C to rt, 70 h
5 °C, 3 h
45
48
63
47
68
73
32
58
(2)e
0
2
3
4
5
2
5
0
0
0
0
49
32
27
25
16
9
54
20
83f
20g
3
3
Ph(CH2)2–OH (13)
13
Cyclohexanol (15)
15
Cholesterol (23)
Z-Ser-OMe (19)
10
5 °C to rt, 18 h
a Alcohol used was 1.3–1.7 equiv to 8a, except for 15 (2 equiv).
b TMS-X used was 1.5–2.0 equiv to 8a, and TMSI was prepared from I2 and HMDS.
c ZnX2 was 1.1–1.5 equiv to 8a, except entry 3 (0.3 equiv).
d Isolated yield after chromatography.
e The structure has not been confirmed.
f Cholesterol 23 was not recovered in this reaction since 23 reacted with the brominating agents to give a much less polar product.
g 8a was recovered (76%).
the a-bromide 9b was formed faster than the a-gluco-
sides at lower temperatures (0–10 °C).
a-glucoside selectively (a:b = 7.5:1), whereas the ben-
zoylated counterpart 9a gave only the b-glucoside.14b
These precedents suggest that particular reactants and/
or conditions would be necessary for the 1,2-cis-selectiv-
ity. To our knowledge, cis-selective glycosylations with
benzoylated sugars have rarely been reported.31
Preferential formation of the a-glucoside might be
explained as shown in Scheme 4. Halogenation of 8a
would give b-halide 9-b as a kinetically favored anomer,
which quickly equilibrates to the more stable a-halide
9-a under the acidic and halide-rich conditions.9b In
the presence of a reactive alcohol, the b-halide 9-b, being
much more reactive than the a form 9-a, would react
1,2-cis-Glycosylations in the presence of a participat-
ing group at C-2 have occasionally been observed with
acetylated sugars.27c,30 For example, Gutman and
co-workers reported that, in the reaction of acetylated
glucuronate bromide with 3-O-acetyl-morphine, cis-a-
glycoside was selectively formed (a:b = 8:1) by using
two and more equivalents of ZnBr2 as promoter,
whereas the b-glycoside was preferentially formed
(a:b = 1:7) with 0.8 equiv of ZnBr2.28b Field and
co-workers reported that the iodine-promoted reaction
of acetylated glucosyl iodide 2 with methanol gave the
OBz
OBz
R-OH
fast
O
O
BzO
BzO
BzO
BzO
X
Me3Si-X
ZnX2
(
)
BzO
BzO
9-β
O-R
α-glucoside
R-OH
8a
(
fast)
(
slow
)
OBz
O
OBz
R-OH
slow
O
BzO
BzO
BzO
BzO
O-R
(
)
BzO
BzO
X
β-glucoside
9-α
Scheme 4. Plausible reaction pathway.

1304
T. Murakami et al. / Carbohydrate Research 343 (2008) 1297–1308
with the alcohol in an SN2 fashion to give a-glucoside.
SN1-type reaction of 9-b by the assistance of a C-2
benzoyloxy group or the reaction of 9-a with the alcohol
described in the above sections, both leading to
b-glucoside, seems to be much slower. Anomerization
of the b-glucoside to the a anomer would also be
slow.
CH2Cl2 (3 mL) was added a solution of hexamethyldisi-
lane (38 mg, 0.25 mmol) in CH2Cl2 (0.5 mL) followed by
iodine (65 mg, 0.25 mmol). The deep-purple mixture was
stirred at rt until 8a was consumed and the color of the
solution became pale pink. The resulting mixture was
diluted with CH2Cl2 (20 mL) and an aq solution
(20 mL) containing NaHCO3 (120 mg) and Na2S2O3
(80 mg), and was stirred for 10 min. The mixture was
transferred into a separatory funnel, and the layers were
separated. The organic layer was washed with aq NaCl
(10 mL). The combined aq phase was extracted with
CH2Cl2 (2 ꢁ 10 mL), and the combined organic layers
were dried over Na2SO4.
3. Conclusions
We have developed comparatively mild, inexpensive,
less-toxic, and efficient ways to prepare 1,2-trans-O-
glucopyranosides from glucose pentabenzoate via the
anomeric halides. Not only the glucosyl iodide but also
the bromide can be activated by zinc halide or NBS with
catalytic ZnI2 to react with a variety of alcohols, indicat-
ing that these promoters would be promising substitutes
for silver and mercury salts employed in the Koenigs–
Knorr glycosylation.
Removal of the solvent gave the crude glucosyl iodide
9a (285 mg, quantitative) as a colorless foam, which was
used in the glycosylation step without further purifica-
24
tion. Compound 9a: ½aꢂD +140 (c 2.4, CHCl3), lit.14b
1
+138 (c 1, CHCl3), lit.20a +139.5 (CHCl3); H NMR
(CDCl3): d 4.52 (m, 2H, H-5, 6a), 4.67 (m, 1H, H-6b),
4.74 (dd, 1H, J1,2 4.4, J2,3 9.8 Hz, H-2), 5.86 (t, 1H,
J3,4 = J4,5 9.8 Hz, H-4), 6.19 (t, 1H, J2,3 = J3,4 9.8 Hz,
H-3), 7.25 (d, 1H, J1,2 4.4 Hz, H-1), 7.27–7.61 (m,
12H), 7.87 (m, 2H) 7.96 (m, 2H), 8.01 (m, 2H), 8.07
(m, 2H); 13C NMR (CDCl3): d 61.8 (C-6), 67.6, 71.0,
72.2, 73.0, 75.4 (C-1), 128.31, 128.33, 128.43, 128.46,
128.54, 128.73, 129.4, 129.7, 129.8, 129.9, 130.1, 133.2,
4. Experimental
4.1. General methods
1
Melting points were determined with a Yanaco melting
point apparatus MP-500D. Optical rotations were mea-
sured with a JASCO DIP-1000 polarimeter at 24 2 °C,
133.3, 133.6, 133.8, 165.0, 165.1, 165.5, 166.0. The H
and 13C NMR data are in accord with those
reported.12b,14b
1
and [a]D values are given in 10ꢀ1 deg cm2 gꢀ1. H and
13C NMR spectra were recorded on a JEOL JNM-
4.2.1. General one-pot glycosylation procedure (A) via
iodide 9a. To a mixture of a-D-glucose pentabenzoate
8a (142 mg, 0.20 mmol) and ZnI2 (16 mg, 0.05 mmol)
in CH2Cl2 (2 mL) was added a solution of hexamethyl-
disilane (18 mg, 0.12 mmol) in CH2Cl2 (0.2 mL)
followed by iodine (31 mg, 0.12 mmol). The deep-purple
mixture was stirred at rt until most of 8a was consumed
and the color of the solution became pale pink. To this
mixture containing 9a were added powdered 4A mole-
cular sieves (200 mg), alcohol (ca. 0.3 mmol) in CH2Cl2
(1 mL), and promoter (0.2–0.6 mmol) in that order
(when NBS was used as a promoter, the mixture was
cooled in an ice-water bath before the addition). The
resulting suspension was stirred until 9a (or the orthoes-
ter if formed) was almost consumed (monitored by
TLC). The reaction mixture was quenched by dilution
with EtOAc (10 mL) and an aq solution (10 mL) con-
taining NaHCO3 (80 mg) and Na2S2O3 (120 mg), and
the mixture was stirred for 10 min, filtered through
Celite, and washed thoroughly with EtOAc (10 mL).
The combined filtrate and washings were transferred
into a separatory funnel, and the layers were separated.
The organic layer was washed with aq NaCl (10 mL).
The combined aq layers were extracted with EtOAc
(2 ꢁ 10 mL), and the combined organic layers were
dried over Na2SO4. Removal of the solvent gave a
1
GSX-270 (270 MHz for H and 67.8 MHz for 13C) or
1
a Varian INOVA 400 (400 MHz for H and 100 MHz
for 13C) spectrometer in CDCl3. Chemical shifts (d)
are given in parts per million (ppm) relative to tetra-
1
methylsilane as the internal standard (d = 0.0) for H
NMR, and the central line of CDCl3 (d = 77.0) for
13C NMR spectra. Coupling constants (J) are given in
Hertz. Elemental analyses were performed in the analyt-
ical section in this Institute (AIST). High-resolution
mass (HRMS) and fast-atom-bombardment mass spec-
tra (FABMS) were obtained on a Hitachi M80B and a
JEOL MS600H mass spectrometer, respectively. Rou-
tine monitoring of reactions was carried out using E.
Merck Silica Gel 60F254 TLC plates, and compounds
were detected by UV absorbance (254 nm) and/or dip-
ping the plates in 10% aq H2SO4 followed by heating.
Column chromatography was performed with indicated
solvents on silica gel (Kanto Chemicals, neutral, 100–
210 lm, or Wakogel C-300, 45–75 lm).
4.2. 2,3,4,6-tetra-O-benzoyl-a-D-glucopyranosyl iodide
(9a)
To a mixture of a-D-glucopyranose pentabenzoate 8a
(284 mg, 0.40 mmol) and ZnI2 (32 mg, 0.10 mmol) in

T. Murakami et al. / Carbohydrate Research 343 (2008) 1297–1308
1305
residue, which was purified by chromatography using an
appropriate solvent system as indicated. When the sepa-
ration was incomplete, the product ratio in fractions of
6.3 Hz), 4.14 (m, 2H), 4.50 (dd, 1H, J 5.2, 12.1 Hz,
H-6a), 4.64 (dd, 1H, J 3.3, 12.1 Hz, H-6b), 4.83 (d,
1H, J 7.8 Hz, H-1), 5.55 (dd, 1H, J 8.1, 9.8 Hz, H-2),
5.67 (t, 1H, J 9.8 Hz, H-4), 5.88 (t, 1H, J 9.5 Hz, H-3),
7.04 (m, 5H), 7.28 (m, 2H), 7.38 (m, 6H), 7.52 (m,
4H), 7.83 (m, 2H), 7.91 (m, 4H), 8.02 (m, 2H); 13C
NMR (CDCl3): d 35.9, 63.1, 69.8, 70.8, 71.8, 72.2,
72.9, 101.1, 126.1, 128.17, 128.27, 128.30, 128.36,
128.38, 128.75, 128.77, 129.3, 129.6, 129.71, 129.73,
129.81, 129.82, 133.11, 133.15, 133.21, 133.4, 138.2,
165.0, 165.2, 165.8, 166.1. Anal. Calcd for C42H36O10:
C, 71.99; H, 5.18. Found: C, 72.13; H, 5.12.
1
the mixture was determined by H NMR.
4.2.2. General glycosylation procedure (B) using ZnX2 as
promoter. Zinc halide and powdered molecular sieves
4A were dried in vacuo at 110 °C for 1 h prior to use.
To a mixture of sugar 9a or 9b (0.20 mmol), alcohol
(0.3–0.4 mmol), ZnX2 (0.3–0.6 mmol), and 4A molecu-
lar sieves (200 mg) was added CH2Cl2 (3 mL), and the
suspension was stirred until 9a/9b was almost con-
sumed. The reaction mixture was worked up as
described in procedure (A).
4.5. Cyclohexyl 2,3,4,6-tetra-O-benzoyl-b-D-glucopyran-
oside (16)
4.2.3. General glycosylation procedure (C) using NBS as
promoter. To a mixture of sugar 9a or 9b (0.20 mmol),
alcohol (0.3–0.4 mmol), ZnI2 (0.04–0.06 mmol), and 4A
molecular sieves (200 mg) was added CH2Cl2 (3 mL),
and the mixture was cooled to 5 °C. NBS (0.22–
0.30 mmol) was added in two portions, and the suspen-
sion was stirred at 5 °C to rt until most of 9a/9b (or the
orthoester if formed) was consumed. The reaction mix-
ture was worked up as described in procedure (A).
Chromatographic purification with CH2Cl2 as eluent
gave 16 as a colorless solid: mp 155–158 °C, lit.32 mp
97.5–100 °C; [a]D +12.3 (c 2.0, CHCl3); Rf 0.29 (4:1 hex-
1
ane–EtOAc); H NMR (CDCl3): d 1.05–1.32 (m, 4H),
1.43 (m, 2H), 1.52–1.76 (m, 3H), 1.88 (m, 1H), 3.66
(tt, 1H, J 3.9, 9.1 Hz), 4.15 (ddd, 1H, J5,6b 3.4, J5,6a
5.6, J4,5 9.8 Hz, H-5), 4.51 (dd, 1H, J5,6a 5.6, J6a,6b
12.0 Hz, H-6a), 4.62 (dd, 1H, J5,6b 3.4, J6a,6b 12.0 Hz,
H-6b), 4.94 (d, 1H, J1,2 7.8 Hz, H-1), 5.51 (dd, 1H, J1,2
7.8, J2,3 9.8 Hz, H-2), 5.64 (t, 1H, J 9.6 Hz, H-4), 5.90
(t, 1H, J 9.6 Hz, H-3), 7.25–7.45 (m, 8H), 7.51 (m,
4H), 7.84 (m, 2H), 7.91 (m, 2H), 7.96 (m, 2H), 8.01
(m, 2H); 13C NMR (CDCl3): d 23.5, 23.7, 25.3, 31.6,
33.2, 63.4, 70.0, 71.98, 72.03, 73.0, 78.5, 99.8, 128.24,
128.29, 128.36, 128.77, 128.80, 129.4, 129.58, 129.64,
129.66, 129.72, 129.78, 133.1, 133.2, 133.4, 165.0,
165.2, 165.8, 166.1. Anal. Calcd for C40H38O10: C,
70.78; H, 5.64. Found: C, 70.84; H, 5.48.
4.3. 12-Bromododecyl 2,3,4,6-tetra-O-benzoyl-b-D-gluco-
pyranoside (11)
Chromatographic purification on silica gel with 7:1:1
hexane–EtOAc–CH2Cl2 as eluent gave 11 as a colorless
oil: [a]D +14.8 (c 1.3, CHCl3); Rf 0.32 (4:1 hexane–
EtOAc); 1H NMR (CDCl3): d 1.00–1.30 (m, 14H),
1.41 (m, 2H), 1.52 (m, 2H), 1.84 (quint, 2H, J 7.1 Hz),
3.40 (t, 2H, J 6.8 Hz, –CH2Br), 3.54 (dt, 1H, J 9.6,
6.7 Hz), 3.92 (dt, 1H, J 9.6, 6.1 Hz), 4.17 (ddd, 1H, J
3.3, 5.0, 9.8 Hz, H-5), 4.51 (dd, 1H, J 5.1, 12.1 Hz,
H-6a), 4.65 (dd, 1H, J 3.3, 12.1 Hz, H-6b), 4.85 (d,
1H, J 7.8 Hz, H-1), 5.54 (dd, 1H, J 7.8, 9.5 Hz, H-2),
5.69 (t, 1H, J 9.6 Hz, H-4), 5.92 (t, 1H, J 9.6 Hz, H-3),
7.30 (m, 2H), 7.35 (m, 2H), 7.40 (m, 4H), 7.52 (m,
4H), 7.84 (m, 2H), 7.90 (m, 2H), 7.96 (m, 2H), 8.02
(m, 2H); 13C NMR (CDCl3): d 25.7, 28.1, 28.7, 29.2,
29.36, 29.41, 29.43, 32.8, 34.1, 63.2, 69.8, 70.4, 71.9,
72.1, 72.9, 101.3 (C-1), 128.26, 128.32, 128.37, 128.8,
129.3, 129.6, 129.7, 129.8, 133.0, 133.1, 133.2, 133.4,
165.0, 165.2, 165.8, 166.1; FABMS (matrix NBA) m/z
(%) 867 ([M(81Br)+Na]+, 25), 865 ([M(79Br)+Na]+,
21), 579 (100).
4.6. 6-O-(2,3,4,6-tetra-O-benzoyl-b-D-glucopyranosyl)-
1,2:3,4-di-O-isopropylidene-a-D-galactopyranose (18)
Chromatographic purification with 30:1 CH2Cl2–EtOAc
gave 18 as a colorless foam: [a]D ꢀ15.3 (c 1.45, CHCl3),
lit.19a ꢀ18 (CHCl3), lit.22e ꢀ15.7 (c 1.0, CHCl3); Rf 0.38
1
(2:1 hexane–EtOAc); H NMR (CDCl3): d 1.19 (s, 3H),
1.20 (s, 3H), 1.24 (s, 3H), 1.37 (s, 3H), 3.86 (dd, 1H,
J5,6a 6.8, J6a,6b 10.5 Hz, H-6a), 3.90 (m, 1H, H-5), 4.02
(dd, 1H, J5,6b 3.5, J6a,6b 10.5 Hz, H-6b), 4.10 (dd, 1H,
0
0
J4,5 1.8, J3,4 8.0 Hz, H-4), 4.18 (ddd, 1H, J5 ;6 b 3.3,
J5 ;6 a 5.3, J4 ;5 10.0 Hz, H-50), 4.22 (dd, 1H, J2,3 2.5,
J1,2 5.1 Hz, H-2), 4.43 (dd, 1H, J2,3 2.3, J3,4 8.0 Hz,
0
0
0
0
H-3), 4.49 (dd, 1H, J5 ;6 a 5.3, J6 a;6 b 12.1 Hz, H-60a),
0
0
0
0
4.4. 2-Phenylethyl 2,3,4,6-tetra-O-benzoyl-b-D-gluco-
pyranoside (14)
4.65 (dd, 1H, J5 ;6 b 3.1, J6 a;6 b 12.1 Hz, H-60b), 5.05 (d,
0
0
0
0
1H, J1 ;2 7.8 Hz, H-10), 5.42 (d, 1H, J1,2 5.1 Hz, H-1),
0
0
5.54 (dd, 1H, J1 ;2 7.9, J2 ;3 9.6 Hz, H-20), 5.68 (t, 1H, J
9.8 Hz, H-40), 5.91 (t, 1H, J 9.6 Hz, H-30), 7.27 (m,
2H), 7.31–7.44 (m, 6H), 7.50 (m, 4H), 7.83 (m, 2H),
7.90 (m, 2H), 7.98 (m, 2H), 8.03 (m, 2H); 13C NMR data
0
0
0
0
Chromatographic purification with 30:1 CH2Cl2–EtOAc
gave 14 as a colorless solid: mp 136–138 °C; [a]D +27.4
1
(c 1.3, CHCl3); Rf 0.33 (3:1 hexane–EtOAc); H NMR
(CDCl3): d 2.86 (m, 2H), 3.74 (dt, 1H, J = 9.8,

1306
T. Murakami et al. / Carbohydrate Research 343 (2008) 1297–1308
were essentially identical with those reported.22e FABMS
9.8, 6.7 Hz), 3.93 (dt, 1H, J 9.8, 6.2 Hz), 4.17 (ddd,
1H, J5,6b 3.3, J5,6a 5.1, J4,5 9.8 Hz, H-5), 4.51 (dd, 1H,
J5,6a 5.2, J6a,6b 12.1 Hz, H-6a), 4.65 (dd, 1H, J5,6b 3.3,
J6a,6b 12.1 Hz, H-6b), 4.80 (m, 1H), 4.83 (m, 1H), 4.84
(d, 1H, J1,2 7.8 Hz, H-1), 5.54 (dd, 1H, J1,2 7.8, J2,3
9.8 Hz, H-2), 5.64 (tdd, 1H, J 6.6, 9.1, 18.2 Hz,
–CH@CH2), 5.69 (t, 1H, J 9.7 Hz, H-4), 5.92 (t, 1H, J
9.7 Hz, H-3), 7.27 (m, 2H), 7.38 (m, 6H), 7.50 (m,
4H), 7.84 (m, 2H), 7.90 (m, 2H), 7.97 (m, 2H), 8.02
(m, 2H); 13C NMR (CDCl3): d 28.5, 29.7, 63.2, 69.4,
69.8, 71.9, 72.1, 72.9, 101.2, 114.9, 128.25, 128.32,
128.36, 128.77, 128.78, 129.3, 129.6, 129.69, 129.71,
129.74, 129.78, 133.08, 133.16, 133.19, 133.4, 137.7,
165.0, 165.2, 165.8, 166.1. Anal. Calcd for C39H36O10:
C, 70.47; H, 5.46. Found: C, 70.34; H, 5.37.
(matrix NBA) m/z (%) 861 ([M+Na]+, 43), 579 (100).
4.7. N-Benzyloxycarbonyl-O-(2,3,4,6-tetra-O-benzoyl-b-
D-glucopyranosyl)-L-serine methyl ester (20)
Chromatographic purification with 2:1 hexane–EtOAc
gave 20 as a colorless foam: [a]D +31.4 (c 1.45, CHCl3);
1
Rf 0.25 (2:1 hexane–EtOAc); H NMR (CDCl3): d 3.66
(s, 3H), 3.91 (dd, 1H, J 3.3, 10.3 Hz), 4.11 (ddd, 1H, J
3.2, 5.1, 10.3 Hz, H-5), 4.33 (dd, 1H, J 2.9, 10.3 Hz),
4.47 (dd, 1H, J 5.1, 12.2 Hz, H-6a), 4.48 (m, 1H,
–CH–NH), 4.64 (dd, 1H, J 3.2, 12.2 Hz, H-6b), 4.83
(d, 1H, J 7.8 Hz, H-1), 4.96 and 5.04 (2 ꢁ d, 2H, Jgem
12.5 Hz, CH2Ph), 5.48 (dd, 1H, J 7.8, 9.5 Hz, H-2),
5.56 (d, 1H, J 8.1 Hz, NH), 5.66 (t, 1H, J 9.8 Hz,
H-4), 5.88 (t, 1H, J 9.6 Hz, H-3), 7.24–7.58 (m, 17H),
7.82 (m, 2H), 7.89 (m, 2H), 7.94 (m, 2H), 8.02 (m,
2H); 13C NMR (CDCl3): d 52.6, 54.2, 62.9, 66.9, 69.3,
69.4, 71.7, 72.3, 72.6, 101.3, 128.07, 128.11, 128.28,
128.38, 128.39, 128.47, 128.71, 128.72, 129.0, 129.5,
129.71, 129.76, 129.79, 133.1, 133.24, 133.29, 133.4,
136.2, 155.8, 165.04, 165.11, 165.7, 166.1, 169.8; FAB-
MS (matrix NBA) m/z (%) 854.5 ([M+Na]+, 20), 832.5
([M+H]+, 4), 579.3 (34).
4.9. Cholest-5-en-3b-yl 2,3,4,6-tetra-O-benzoyl-b-D-
glucopyranoside (24)
Chromatographic purification with CH2Cl2 gave 24 as a
colorless solid: mp 205–208 °C, lit.33a 214–216 °C, lit.19c
204–206°C; [a]D +17.3 (c 1.6, CHCl3), lit.33a +15 (c 0.9,
CHCl3), lit.33b +14.1(c 0.71, CHCl3), lit.33c +13.5 (c 1.0,
CHCl3); Rf 0.40 (4:1 hexane–EtOAc); 1H and 13C NMR
data were essentially identical with those reported.33c
Anal. Calcd for C61H72O10: C, 75.91; H, 7.52. Found:
C, 75.82; H, 7.39.
4.7.1. 3,4,6-tri-O-benzoyl-a-D-glucopyranose orthoester
(12S). Chromatographic purification with 40:1
CH2Cl2–EtOAc gave 12S as a colorless foam: [a]D
1
4.10. Methyl 2,3,4,6-tetra-O-benzoyl-b-D-glucopyranosyl-
(1?6)-2,3,4-tri-O-benzyl-a-D-glucopyranoside (26)
+1.3 (c 1.65, CHCl3); Rf 0.32 (2:1 hexane–EtOAc); H
NMR (CDCl3): d 3.63 (dd, 1H, J 3.3, 10.0 Hz), 3.70
(s, 3H), 3.72 (dd, 1H, J 3.1, 10.0 Hz), 4.08 (ddd, 1H, J
3.0, 5.0, 8.6 Hz, H-5), 4.36 (dd, 1H, J 5.1, 12.1 Hz,
H-6a), 4.44 (m, 1H, –CH–NH), 4.51 (dd, 1H, J 2.8,
12.2 Hz, H-6b), 4.72 (m, 1H, H-2), 5.07 and 5.11
(2 ꢁ d, 2H, Jgem 12.2 Hz, CH2Ph), 5.47 (d-like, 1H,
J4,5 8.8 Hz, H-4), 5.58 (d, 1H, J 8.6 Hz, NH), 5.73 (dd,
1H, J3,4 1.6, J2,3 2.9 Hz, H-3), 6.01 (d, 1H, J1,2 5.1 Hz,
H-1), 7.24–7.35 (m, 7H), 7.37–7.51 (m, 8H), 7.59 (m,
2H), 7.69 (m, 2H), 7.91 (m, 2H), 7.93 (m, 2H), 8.07
(m, 2H); 13C NMR (CDCl3): d 52.5, 53.8, 63.8, 64.0,
67.1, 67.6, 68.3, 69.0, 72.3, 97.6 (C-1), 121.1 (Ph–CO3–),
126.2, 128.0, 128.1, 128.2, 128.39, 128.46, 128.53,
128.8, 129.0, 129.5, 129.6, 129.8, 129.9, 130.0, 133.0,
133.5, 133.7, 134.4, 136.1, 155.8, 164.5, 165.1, 165.9,
170.3; FABMS (matrix NBA) m/z (%) 854 ([M+Na]+,
39), 832 ([M+H]+, 2), 579 (100).
Chromatographic purification with 40:1 CH2Cl2–EtOAc
gave 26 as a colorless foam: [a]D +21.5 (c 2.2, CHCl3),
lit.2e +18 (c 1.4, CHCl3), lit.33b +22.1 (c 2.7, CHCl3),
lit.34 +21.7 (c 1.0, CHCl3); Rf 0.40 (2:1 hexane–EtOAc);
1H and 13C NMR data were essentially identical with
those reported.2e,34 FABMS (matrix NBA) m/z (%)
1065 ([M+Na]+, 31), 579 (100).
4.11. Methyl 2,3,4,6-tetra-O-benzoyl-b-D-glucopyranosyl-
(1?4)-2,3,6-tri-O-benzyl-a-D-glucopyranoside (28)
Chromatographic purification with 40:1 CH2Cl2–EtOAc
gave 28 as a colorless foam: [a]D ꢀ1.0 (c 2.6, CHCl3),
lit.2e +1 (c 1.5, CHCl3), lit.33b ꢀ3.0 (c 1.7, CHCl3),
lit.34 +1.0 (c 1.0, CHCl3); Rf 0.36 (2:1 hexane–EtOAc);
1H and 13C NMR data were essentially identical with
those reported.2e,g,34 Anal. Calcd for C62H58O15ꢃ0.5H2O:
C, 70.77; H, 5.65. Found: C, 70.78; H, 5.42.
4.8. 4-Pentenyl 2,3,4,6-tetra-O-benzoyl-b-D-glucopyran-
oside (22)
Chromatographic purification with 3:1 hexane–EtOAc
gave 22 as a colorless solid: mp 112–114 °C, lit.29 mp
113–114 °C; [a]D +19.3 (c 2.1, CHCl3), lit.29 +18.7 (c
1, CHCl3); Rf 0.35 (3:1 hexane–EtOAc); 1H NMR
(CDCl3): d 1.64 (m, 2H), 1.98 (m, 2H), 3.56 (dt, 1H, J
4.12. General procedure for 1,2-cis selective glycosylation
To a cooled mixture of a-D-glucose pentabenzoate 8a
(0.20 mmol), alcohol (0.3–0.4 mmol), and ZnBr2
(55 mg, 0.24 mmol) in CH2Cl2 (2 mL) was added a

T. Murakami et al. / Carbohydrate Research 343 (2008) 1297–1308
1307
solution of bromotrimethylsilane (50 mg, 0.32 mmol) in
CH2Cl2 (0.3 mL). The mixture was stirred at ca. 5 °C for
1 h, and then at rt until 8a was consumed. The reaction
was quenched by dilution with EtOAc (10 mL) and aq
NaHCO3 (10 mL), and the mixture was worked up as
described in procedure (A).
colorless solid; mp 168–169 °C, lit.32 mp 91–92 °C; [a]D
1
+88.0 (c 1.3, CHCl3); Rf 0.37 (4:1 hexane–EtOAc); H
NMR (CDCl3): d 1.08–1.33 (m, 4H), 1.46 (m, 1H),
1.53 (m, 1H), 1.62 (m, 1H), 1.71 (m, 2H), 1.96 (m,
1H), 3.62 (tt, 1H, J 3.9, 9.4 Hz), 4.47 (m, 1H),
4.59 (m, 2H), 5.28 (dd, 1H, J1,2 3.7, J2,3 10.1 Hz, H-2),
5.51 (d, 1H, J1,2 3.7 Hz, H-1), 5.65 (t, 1H, J 9.9 Hz,
H-4), 6.21 (t, 1H, J 9.9 Hz, H-3), 7.28 (m, 2H), 7.32–
7.45 (m, 6H), 7.52 (m, 4H), 7.88 (m, 2H), 7.96 (m,
2H), 7.99 (m, 2H), 8.05 (m, 2H); 13C NMR (CDCl3): d
23.6, 23.9, 25.4, 31.6, 33.4, 63.2, 67.8, 69.7, 70.6,
72.1, 77.5, 94.5 (C-1), 128.22, 128.31, 128.35 128.9,
129.1, 129.2, 129.63, 129.67, 129.77, 129.83, 133.0,
133.26, 133.34, 165.3, 165.8 (2C), 166.1. Anal. Calcd
for C40H38O10: C, 70.78; H, 5.64. Found: C, 70.78; H,
5.43.
4.12.1. 12-Bromododecyl 2,3,4,6-tetra-O-benzoyl-a-D-
glucopyranoside (11a). The crude product mixture
from the reaction of 8a (144 mg, 0.20 mmol) and 3
(80 mg, 0.30 mmol) was purified by silica gel chromato-
graphy eluting with 5:1?4:1 hexane–EtOAc to give 9b
(65 mg, 49%) and 11a (76 mg, 45%): colorless oil; [a]D
1
+67.0 (c 2.3, CHCl3); Rf 0.39 (4:1 hexane–EtOAc); H
NMR (CDCl3): d 1.10–1.35 (m, 14H), 1.41 (m, 2H),
1.61 (m, 2H), 1.84 (quint, 2H, J 7.1 Hz), 3.40 (t, 2H, J
6.8 Hz), 3.49 (dt, 1H, J 10.0, 6.6 Hz), 3.80 (dt, 1H, J
9.8, 6.4 Hz), 4.48 (m, 2H, H-5 and 6a), 4.61 (m, 1H,
H-6b), 5.30 (dd, 1H, J1,2 3.7, J2,3 10.0 Hz, H-2), 5.35
(d, 1H, J1,2 3.7 Hz, H-1), 5.68 (t, 1H, J 9.6 Hz, H-4),
6.20 (t, 1H, J 9.8 Hz, H-3), 7.25–7.45 (m, 8H), 7.50
(m, 4H), 7.87 (m, 2H), 7.94 (m, 2H), 7.99 (m, 2H),
8.05 (m, 2H); 13C NMR (CDCl3): d 26.0, 28.1, 28.7,
29.25, 29.29, 29.35, 29.45, 32.8, 34.0, 63.1, 67.7, 68.8,
69.7, 70.6, 72.1, 95.9 (C-1), 128.23, 128.33, 128.35,
128.9, 129.1, 129.2, 129.64, 129.69, 129.71, 129.83,
129.86, 133.0, 133.28, 133.34, 165.3, 165.76, 165.78,
166.1; FABMS (matrix NBA) m/z (%) 866 ([M+Na]+,
12), 843 ([M]+, 4), 579 (100).
References
1. For reviews of glycosylations, see: (a) Toshima, K.;
Tatsuta, K. Chem. Rev. 1993, 93, 1503–1531; (b) Boons,
G.-J. Tetrahedron 1996, 52, 1095–1121; (c) Davis, B. G. J.
Chem. Soc., Perkin Trans. 1 2000, 2137–2160; (d) Dem-
chenko, A. V. Synlett 2003, 1225–1240.
2. For representative examples, see: O-pentenyl glycosides:
(a) Fraser-Reid, B.; Konradsson, P.; Mootoo, D. R.;
Udodong, U. J. Chem. Soc., Chem. Commun. 1988, 823–
825. Sulfoxides: (b) Kahne, D.; Walker, S.; Cheng, Y.;
Van Engen, D. J. Am. Chem. Soc. 1989, 111, 6881–6882.
(c) Hashimoto, S.; Umeo, K.; Sano, A.; Watanabe, N.;
Nakajima, M.; Ikegami, S. Tetrahedron Lett. 1995, 36,
2251–2254. Glycals: (d) Danishefsky, S. J.; Bilodeau, M.
T. Angew. Chem., Int. Ed. 1996, 35, 1380–1419. 1-OH
sugars: (e) Garcia, B. A.; Gin, D. Y. J. Am. Chem. Soc.
2000, 122, 4269–4279. O-Phosphates: (f) Plante, O. J.;
Palmacci, E. R.; Andrade, R. B.; Seeberger, P. H. J. Am.
Chem. Soc. 2001, 123, 9545–9554. 1-Thiazolyl-thio sugars:
(g) Pornsuriyasak, P.; Demchenko, A. V. Chem. Eur. J.
2006, 12, 6630–6646.
4.12.2. 2-Phenylethyl 2,3,4,6-tetra-O-benzoyl-a-D-gluco-
pyranoside (14a). The crude product mixture from the
reaction of 8a (144 mg, 0.20 mmol) and 13 (40 mg,
0.34 mmol) was purified by chromatography eluting
with CH2Cl2 to give 9b (21 mg, 16%) and 14a (95 mg,
68%): colorless foam; [a]D +111 (c 1.6, CHCl3); Rf
1
0.40 (3:1 hexane–EtOAc); H NMR (CDCl3): d 2.95 (t,
3. For a review: (a) Garegg, P. J. Adv. Carbohydr. Chem.
Biochem. 1997, 52, 179–266; (b) Crich, D.; Smith, M. J.
2H, J 6.5 Hz), 3.75 (dt, 1H, J 9.4, 6.1 Hz), 3.96 (m,
1H), 3.98 (dt, 1H, J 9.4, 6.8 Hz), 4.32 (dd, 1H, J 5.1,
12.1 Hz), 4.46 (dd, 1H, J 2.8, 12.1 Hz), 5.28 (dd, 1H,
J1,2 3.7, J2,3 10.1 Hz, H-2), 5.33 (d, 1H, J1,2 3.7 Hz,
H-1), 5.62 (t, 1H, J 9.9 Hz, H-4), 6.16 (t, 1H, J 9.9 Hz,
H-3), 7.04 (m, 5H), 7.28 (m, 2H), 7.38 (m, 6H), 7.52
(m, 4H), 7.83 (m, 2H), 7.91 (m, 4H), 8.02 (m, 2H); 13C
NMR (CDCl3): d 35.9, 62.8, 67.6, 68.8, 69.3, 70.4,
71.9, 95.5 (C-1), 126.3, 128.25, 128.33, 128.34, 128.38,
128.86, 129.02, 129.14, 129.62, 129.67, 129.82, 129.91,
133.06, 133.32, 133.37, 138.7, 165.16, 165.74, 165.79,
166.1; FABMS (matrix NBA) m/z (%) 723 ([M+Na]+,
64), 700 ([M]+, 12), 579 (100).
´
Am. Chem. Soc. 2001, 123, 9015–9020; (c) Codee, J. D. C.;
Litjens, R. E. J. N.; den Heeten, R.; Overkleeft, H. S.; van
Boom, J. H.; van der Marel, G. A. Org. Lett. 2003, 5,
1519–1522.
4. For a review: Schmidt, R. R.; Kinzy, W. Adv. Carbohydr.
Chem. Biochem. 1994, 50, 21–123.
5. Koenigs, W.; Knorr, E. Ber. Dtsch. Chem. Ges. 1901, 34,
957–981.
6. For a review: Toshima, K. Carbohydr. Res. 2000, 327, 15–
26.
7. Jacobsson, M.; Malmberg, J.; Ellervik, U. Carbohydr. Res.
2006, 341, 1266–1281 (see Fig. 4 in p 1269, concerning
aromatic O-glycosylation using glycosyl halides).
8. (a) Kronzer, F. J.; Schuerch, C. Carbohydr. Res. 1974,
34, 71–78; (b) Hashimoto, S.-i.; Honda, T.; Ikegami, S.
Tetrahedron Lett. 1990, 31, 4769–4772; (c) Schmid, U.;
Waldmann, H. Tetrahedron Lett. 1996, 37, 3837–
3840.
4.12.3. Cyclohexyl 2,3,4,6-tetra-O-benzoyl-a-D-gluco-
pyranoside (16a). The crude product mixture from
the reaction of 8a (144 mg, 0.20 mmol) and 15 (40 mg,
0.4 mmol) was purified by chromatography eluting with
CH2Cl2 to give 9b (71 mg, 54%) and 16a (44 mg, 32%):
9. For a review, see: (a) Gervay, J. In Comprehensive Organic
Synthesis: Theory and Applications; JAI Press: Oxford,
1998; Vol. 4, pp 121–153; (b) Gervay, J.; Nguyen, T. N.;

1308
T. Murakami et al. / Carbohydrate Research 343 (2008) 1297–1308
Hadd, M. J. Carbohydr. Res. 1997, 300, 119–125; (c)
Acta Chem. Scand. B 1985, 39, 567–577; (c) Harreus, A.;
Kunz, H. Liebigs Ann. Chem. 1986, 717–730.
Gervay, J.; Hadd, M. J. J. Org. Chem. 1997, 62, 6961–
6967; (d) Hadd, M. J.; Gervay, J. Carbohydr. Res. 1999,
320, 61–69; (e) Gervay-Hague, J.; Lam, S. N. Org. Lett.
2002, 4, 2039–2042.
20. (a) Ness, R. K.; Fletcher, H. G., Jr.; Hudson, C. S. J. Am.
Chem. Soc. 1950, 72, 2200–2205; (b) Egusa, K.; Kusum-
oto, S.; Fukase, K. Eur. J. Org. Chem. 2003, 3435–3445.
21. Compound 9a was first prepared from 8a with HI in
AcOH; see Ref. 20a.
22. (a) Veeneman, G. H.; van Leeuwen, S. H.; van Boom, J.
H. Tetrahedron Lett. 1990, 31, 1331–1334; (b) Fukase, K.;
Hasuoka, A.; Kusumoto, S. Tetrahedron Lett. 1993, 34,
2187–2190; (c) Yamago, S.; Kokubo, K.; Murakami, H.;
Mino, Y.; Hara, O.; Yoshida, J. Tetrahedron Lett. 1998,
39, 7905–7908; (d) Qin, Z.-H.; Li, H.; Cai, M.-S.; Li, Z.-J.
Carbohydr. Res. 2002, 337, 31–36; (e) Valerio, S.; Iadonisi,
10. (a) Uchiyama, T.; Hindsgaul, O. Synlett 1996, 499–501;
(b) Bhat, A. S.; Gervay-Hague, J. Org. Lett. 2001, 3, 2081–
2084.
11. (a) Thiem, J.; Meyer, B. Chem. Ber. 1980, 113, 3075–
3085; (b) Klemer, A.; Bieber, M. Liebigs Ann. Chem. 1984,
1052–1055; (c) Montero, J.-L.; Winum, J.-Y.; Leydet, A.;
Kamal, M.; Pavia, A. A.; Roque, J.-P. Carbohydr. Res.
1997, 297, 175–180; (d) Kartha, K. P. R.; Field, R. A.
Carbohydr. Lett. 1998, 3, 179–182; (e) Chervin, S. M.;
Abada, P.; Koreeda, M. Org. Lett. 2000, 2, 369–372;
(f) Ying, L.; Gervay-Hague, J. Carbohydr. Res. 2003,
`
A.; Adinolfi, M.; Ravida, A. J. Org. Chem. 2007, 72, 6097–
6106.
`
´
338, 835–841; (g) Adinolfi, M.; Iadonisi, A.; Ravida, A.;
23. Polt, R.; Szabo, L.; Treiberg, J.; Li, Y.; Hruby, V. J. J.
Schiattarella, M. Tetrahedron Lett. 2003, 44, 7863–
7866; (h) Mukhopadhyay, B.; Kartha, K. P. R.; Russell,
D. A.; Field, R. A. J. Org. Chem. 2004, 69, 7758–
7760.
Am. Chem. Soc. 1992, 114, 10249–10258.
24. (a) Hashimoto, S.; Hayashi, M.; Noyori, R. Tetrahedron
Lett. 1984, 25, 1379–1382; (b) Mukaiyama, T.; Katsurada,
M.; Iida, Y. Chem. Lett. 1992, 2105–2108.
12. (a) Ernst, B.; Winkler, T. Tetrahedron Lett. 1989, 30,
3081–3084; (b) Caputo, R.; Kunz, H.; Mastroianni, D.;
Palumbo, G.; Pedatella, S.; Solla, F. Eur. J. Org. Chem.
1999, 3147–3150; (c) Bickley, J.; Cottrell, J. A.; Ferguson,
J. R.; Field, R. A.; Harding, J. R.; Hughes, D. L.; Kartha,
K. P. R.; Law, J. L.; Scheinmann, F.; Stachulski, A. V.
Chem. Commun. 2003, 1266–1267.
13. (a) Paulsen, H.; Rutz, V.; Brockhausen, I. Liebigs Ann.
Chem. 1992, 747–758; (b) Chervin, S. M.; Lowe, J. B.;
Koreeda, M. J. Org. Chem. 2002, 67, 5654–5662.
14. (a) Perrie, J. A.; Harding, J. R.; King, C.; Sinnott, D.;
Stachulski, A. V. Org. Lett. 2003, 4545–4548; (b) van Well,
R. M.; Kartha, K. P. R.; Field, R. A. J. Carbohydr. Chem.
2005, 24, 463–474.
15. Murakami, T.; Hirono, R.; Sato, Y.; Furusawa, K.
Carbohydr. Res. 2007, 342, 1009–1020.
16. (a) Olah, G. A.; Narang, S. C.; Gupta, B. G. B.; Malhotra,
R. Angew. Chem., Int. Ed. Engl. 1979, 18, 612–614; (b)
Sakurai, H.; Shirahata, A.; Sasaki, K.; Hosomi, A.
Synthesis 1979, 740–741.
25. (a) Gillard, J. W.; Israel, M. Tetrahedron Lett. 1981, 22,
513–516; (b) Gravier-Pelletier, C.; Ginisty, M.; Le Merrer,
Y. Tetrahedron: Asymmetry 2004, 15, 189–193.
26. Dabideen, D. R.; Gervay-Hague, J. Org. Lett. 2004, 6,
973–975.
27. Sn(OTf)2: (a) Lubineau, A.; Malleron, A. Tetrahedron
Lett. 1985, 26, 1713–1716. InCl3: (b) Mukherjee, D.; Ray,
P. K.; Chowdhury, U. S. Tetrahedron 2001, 57, 7701–7704.
FeCl3: (c) Koto, S.; Hirooka, M.; Tashiro, T.; Sakashita,
M.; Hatachi, M.; Kono, T.; Shimizu, M.; Yoshida, N.;
Kurasawa, S.; Sakuma, N.; Sawazaki, S.; Takeuchi, A.;
Shoya, N.; Nakamura, E. Carbohydr. Res. 2004, 339,
2415–2424.
28. ZnF2: (a) DeNinno, M. P.; McCarthy, P. A.; Duplantier,
K. C.; Eller, C.; Etienne, J. B.; Zawistoski, M. P.; Wen
Bangerter, F.; Chandler, C. E.; Morehouse, L. A.;
Sugarman, E. D.; Wilkins, R. W.; Woody, H. A.; Zaccaro,
L. M. J. Med. Chem. 1997, 40, 2547–2554. ZnBr2: (b)
Rukhman, I.; Yudovich, L.; Nisnevich, G.; Gutman, A. L.
Tetrahedron 2001, 57, 1083–1092.
17. (a) Kartha, K. P. R.; Aloui, M.; Field, R. A. Tetrahedron
Lett. 1996, 37, 8807–8810; (b) Kartha, K. P. R.; Field, R.
A. Tetrahedron Lett. 1997, 38, 8233–8236; (c) Kartha, K.
P. R.; Ballell, L.; Bilke, J.; McNeil, M.; Field, R. A. J.
Chem. Soc., Perkin Trans. 1 2001, 770–772; (d) Stachulski,
A. V. Tetrahedron Lett. 2001, 42, 6611–6613; (e) Carpen-
ter, C.; Nepogodiev, S. A. Eur. J. Org. Chem. 2005, 3286–
3296.
29. Madsen, R.; Fraser-Reid, B. J. Org. Chem. 1995, 60, 772–
779.
30. Banoub, J.; Bundle, D. R. Can. J. Chem. 1979, 57, 2085–
2090.
31. Zeng, Y.; Ning, J.; Kong, F. Tetrahedron Lett. 2002, 43,
3729–3733.
32. Petersen, L.; Jensen, K. J. J. Chem. Soc., Perkin Trans. 1
2001, 2175–2182.
18. (a) Ogawa, T.; Beppu, K.; Nakabayashi, S. Carbohydr.
Res. 1981, 93, C6–C9. For a recent review of sugar
orthoesters: (b) Kong, F. Carbohydr. Res. 2007, 342, 345–
373.
33. (a) Uvarora, N. I.; Atopkina, L. N.; Elyakov, G. B.
Carbohydr. Res. 1980, 83, 26–33; (b) Hashimoto, S.;
Nakajima, T.; Ikegami, S. J. Chem. Soc., Chem. Commun.
´
1989, 685–687; (c) Trujillo, M.; Morales, E. Q.; Vazquez,
19. (a) Garegg, P. J.; Norberg, B. Acta Chem. Scand. B 1979,
33, 116–118; (b) Garegg, P. J.; Konradsson, P.; Kvarn-
stro¨m, I.; Norberg, T.; Svensson, S. C. T.; Wigilius, B.
J. T. J. Org. Chem. 1994, 59, 6637–6642.
34. Kim, W.-S.; Hosono, S.; Sasai, H.; Shibasaki, M. Hetero-
cycles 1996, 42, 795–809.