Regioselective benzoylation of sugars mediated by excessive Bu2SnO: Observation of temperature promoted migration

Article abstract of DOI:10.1016/S0040-4020(02)00661-0

Regioselective benzoylation of carbohydrates using an excess of dibutyltin oxide (Bu₂SnO) at increased reaction temperatures has been developed for the synthesis of several glycoside benzoates with one or two free hydroxyl groups, including gal

Full text of DOI:10.1016/S0040-4020(02)00661-0

Tetrahedron 58 (2002) 6513–6519
Regioselective benzoylation of sugars mediated by excessive
Bu2SnO: observation of temperature promoted migration
*
Zhiyuan Zhang and Chi-Huey Wong
Department of Chemistry and the Skaggs Institute for Chemical Biology, The Scripps Research Institute, 10550 North Torrey Pines Road,
La Jolla, CA 92037, USA
Received 4 April 2002; accepted 10 June 2002
Abstract—Regioselective benzoylation of carbohydrates using an excess of dibutyltin oxide (Bu2SnO) at increased reaction temperatures
has been developed for the synthesis of several glycoside benzoates with one or two free hydroxyl groups, including galactosides, glucosides,
mannosides and lactosides in high yields. These compounds are useful as building blocks for the synthesis of complex saccharides and
derivatives. q 2002 Elsevier Science Ltd. All rights reserved.
1. Introduction
intermediate A and then give the 2,3,6-tri-O-benzoyl
product B. To our surprise, some unexpected results were
observed (Table 1). For example, p-methoxyphenyl b-D-
galactoside 114 was first treated with 3 equiv. of Bu2SnO in
toluene–benzene (1:1) followed by azeotropic removal of
water to give a clear solution of the stannylene complex.
Treatment of this complex with 3 equiv. of BzCl at room
temperature, however, gave only the diprotected 3,6-O-di-
benzoyl derivative 10 in 93% yield (Table 1). This result
appears to be consistent with an early study with 1.5 equiv.
of (Bu3Sn)2O and 3 equiv. of BzCl at room temperature.10
We thought that a higher temperature might facilitate
additional benzoylation. Therefore, in a second experiment
the tin complex was treated with 3 equiv. of BzCl at 1008C.
While additional benzoylation was observed, the 3,4,6-tri-
OBz derivative 11 was obtained in 86% yield instead of the
expected 2,3,6-derivative. It is known that, in general, the
2-OH position is more reactive than the 4-OH in a
b-galactoside series,15 even in the tin complex case.10
Oligosaccharides with immense molecular diversity can be
created from a few common monosaccharides through a
linear or branched connection between each multifunctional
sugar unit. However, creating such a high degree of
molecular diversity requires the design of building blocks
with orthogonal protecting groups for use in glycosylation
reactions. Development of new methods for the synthesis of
carbohydrate building blocks for this application is thus of
current interest.
Recently, the use of organotin reagents such as tributyltin
oxide, (Bu3Sn)2O, or dibutyltin oxide, Bu2SnO, has been
widely explored.1–4 Regioselective alkylation,5,6 silyl-
ation,7 acylation8–10 and even regioselective glycosyl-
ation11–13 have been reported with such reagents. It is
believed that the reactivity of a hydroxyl group can be
enhanced by forming a stannylene complex and, in most
cases, the stannylene complex of a sugar exists as cyclic
acetals, which can be selectively protected with an alkyl or
acyl group. Most of the studies, however, use a stoichio-
metric or slightly excessive amount of the tin reagent. Our
interest in this subject is to exploit the use of excess
organotin reagents to mediate the regioselective protection
of the hydroxyl groups of a sugar to give a product with one
free hydroxyl group, as these building blocks are useful for
selective glycosylation.
To confirm this unusual regioselectivity, galactosides 2 and
3 were also tested under this condition, and compounds 13
and 14 were obtained with the same selectivity in 89 and
91% yield, respectively (Table 1, entries 3 and 4). We were
excited by these results, since there was no known synthetic
method that would give this type of compound in few steps.
To further study these reactions, we carefully followed the
reaction at room temperature by TLC. We observed that the
3,6-di-OBz derivative 10 was first formed after only a few
minutes. When the tin complex was treated with 2.2 equiv.
of benzoyl chloride at room temperature followed by 1008C,
a new spot with low Rf appeared after approximately
10 min, then disappeared during the formation of the 3,4,6-
tri-Obz derivative 11. We finally isolated the first
intermediate and confirmed by NMR and high-resolution
mass spectrometry this to be the 4,6-O dibenzoate 12.
As suggested by previous studies,3 if the galactoside shown
in Fig. 1 is treated with 3 equiv. of Bu2SnO followed by
3 equiv. of BzCl, it is expected to generate the stannylene
Keywords: sugar; glycoside; benzoylation; regioselective.
*
Corresponding author. Tel.: þ1-619-784-2487; fax: þ1-619-784-2409;
e-mail: wong@scripps.edu
0040–4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(02)00661-0

6514
Z. Zhang, C.-H. Wong / Tetrahedron 58 (2002) 6513–6519
Table 1. Benzoylations using Bu2SnO
Starting material
Entry Bu2SnO (equiv.) BzCl (equiv.) Temperature (8C)
Reaction time
Products (yield)
1
2
3
3.0
3.0
3.0
3.3
3.3
2.2
22
100
22, then 100
5 h
30 min
2 h, then 30 min 12 (75%), 11 (20%)
10 (93%)
11 (93%)
4
5
3.0
3.0
2.2
2.2
100
100
30 min
30 min
13 (88%)
14 (85%)
6
7
2.0
2.0
3.3
2.2
22
80
5 h
3 h
15 (90%)
15 (23%), 16 (46%)
8
3.0
3.3
45
2.5 h
17 (42%), 18 (29%), 19 (12%)
20 (91%), 21 (6%)
9
3.0
3.0
3.2
3.3
100
90
5 min
10
20 min
20 min
22 (85%)
11
12
3.0
3.0
3.3
2.2
90
22, then 100
23 (89%), 24 (6%), 25 (2%)
2 h, then 30 min 24 (74%), 23 (20%)
13
14
6
7
10
10
120
120
2 h
7 h
26 (86%)
27 (85%)

Z. Zhang, C.-H. Wong / Tetrahedron 58 (2002) 6513–6519
6515
Figure 1.
As proposed in Fig. 2, it is possible that the galactoside
intermediate A was formed after reaction with 3 equiv. of
Bu2SnO. As stannylene complexes have different dimer and
polymer forms,3,16 we assume that the R0 group may be
another tin moiety or sugar residue. The 3-O and 6-O
positions of intermediate A are perhaps more reactive and
readily benzoylated to give the 3,6-di-O-benzoyl derivative
with a structure like compound C. This intermediate will
then give the diprotected product 10. Under higher
temperature, two pathways from intermediate C may be
possible. One is a slow benzoylation reaction at the 2-O-
position to give products B and H through intermediate G,
which is often isolated in a small amount in each reaction.
Another is a faster migration reaction, where the benzoyl
group at 3-O migrates to the 4-O-position, presumably with
an aid from the tin group at the 2-O position to give the
newly formed 2,3-O stannylene acetal D. Further benzoyl-
ation would occur at the 3-O position as the 3-O position is
more reactive than the 2-O position to give the hydrolyzed
product F through intermediate E.
refluxing benzene. After 10 h, only about 50% of the
migrated product was formed. Moreover, when a complex
of 2-azido-2-deoxy galactoside 4 and 2 equiv. of Bu2SnO
was treated with 3 equiv. of BzCl at room temperature, 3,6-
di-benzoate 15 was obtained in 90% yield and with 2 equiv.
of BzCl at 70–908C for 1.5–3 h the migrated product 16
was obtained in 45% yield along with ,20% of compound
15. All these examples show that the migration of Bz- from
position 3-O to 4-O is very slow without the anchimeric
effect from the tin moiety at the 2-O position.
The galactosides discussed above are b-linked compounds;
however, a result from Ogawa’s group indicated a different
reaction pattern with an a-galactopyranoside.8 Treatment of
methyl a-galactopyranoside 5 with 1.5 equiv. of (Bu3Sn)2O
and 3 equiv. of BzCl gave a mixture of different di-OBz and
tri-OBz compounds, and the 2,3,6-tri-OBz 17 was isolated
in 40% yield as the major product. We obtained a mixture
using 3 equiv. of Bu2SnO and 3.3 equiv. of BzCl under a
higher temperature (458C) condition. Three major products,
2,3,6-tri-OBz 17, 2,4-6-tri-OBz 18 and 3,4,6-tri-OBz 19
were isolated in 42, 29 and 12, respectively (entry 8). A
possible explanation for this is that a stannylene acetal at the
2-O position coordinates with the a-anomeric oxygen and
reduces the extent of the 3-O to 4-O benzoyl migration
(Scheme 1). In this case, the 2-O seems more reactive
towards benzoylation than its b-isomer, consequently
slowing down the 3-O to 4-O migration. In addition,
compound 17 was synthesized in 65% yield by a direct
benzoylation using 4.2 equiv. of BzCl in pyridine.15
In order to further support this migration mechanism, we
performed additional experiments. First, we treated com-
pound 1 (Table 1) with 3 equiv. of Bu2SnO, followed by
2.2 equiv. of BzCl at room temperature to form the 3,6-OBz
intermediate 10 (see Fig. 2, C). Raising the reaction
temperature to 1008C gave 4,6-di-benzoate 12 in 75%
along with 20% of 3,4,6-OBz 11. This example shows that
higher temperature is necessary for the migration. To
support the hypothesis involving the assistance from the tin
moiety at the 2-O position of intermediate C (Fig. 1),
2-(trimethylsilyl)ethyl 2,3,6-tri-O-benzoyl-b-D-galacto-
pyranoside17 was treated with 1 equiv. of Bu2SnO in
A glucoside is a 4-epimer of galactoside. Ogawa et al.
systematically studied the reactions of an a-glucoside with
Figure 2. Note: R0 is not identified (could be another tin moiety or another sugar).

6516
Z. Zhang, C.-H. Wong / Tetrahedron 58 (2002) 6513–6519
Scheme 1.
different ratios of (Bu3Sn)2O to BzCl.10 They found that
1.5 equiv. of (Bu3Sn)2O and 2–3 equiv. of BzCl gave 2,6-
di-OBz 21 in 73–82%. A larger excess of (Bu3Sn)2O and
BzCl gave the 2,3,6-O-tri-benzoate 20 in 63%. A higher
yield (90%) was obtained with a modified procedure.18 We
hypothesized that the stannylene complex of an a-glucoside
with 3 equiv. of Bu2SnO may be similar to the a-galactoside
case. With our high temperature procedure, the methyl
a-glucoside 6 was treated with 3 equiv. of Bu2SnO,
followed by 3.2 equiv. of BzCl at 708C to give 2,3,6-O-
tri-OBz 20 in 91% yield. TLC analysis of the reaction
showed that the intermediate 2,6-di-OBz 21 was formed
first. Based on these results, the reactivities of these hydroxy
groups as tin derivatives in the a-glucoside seem to be
6-O.2-O.3-O.4-O, a trend similar to those found under
imidazole benzoate treatment.19 This high regioselectivity
makes this method attractive for preparation of tri-
benzoates. We have also applied this procedure to the
modification of 2-deoxy globotriosyl neo-glycosyl lipid.20
We treated the 2-deoxy glucosyl lipid 7 (entry 10) with
3 equiv. of Bu2SnO followed by 2.4 equiv of BzCl at 1008C
for 5 min, giving the desired 3,6-dibenzoate in 85% yield.
different partially protected building blocks by varying the
amount of BzCl and temperature.
We have found an interesting trend under these conditions.
In a pyranoside structure, when an equatorial hydoxy group
is cis to an adjacent axial hydroxy group, this equatorial
hydoxy group is much more reactive than another equatorial
hydoxy group which is trans to an adjacent equatorial
hydroxyl group. For instance, the 3-O group is more reactive
than the 2-O group in the b-galactoside series, the 2-O
group is more reactive than the 3-O and 4-O groups in the
a-glucoside, and the 3-O group is more reactive than the
4-O group in the a-mannoside series.
Encouraged by these results, the disaccharide lactoside 9
was chosen as the substrate for multiple stannylation. We
treated compound 9 with 6 equiv. of Bu2SnO, followed by
9 equiv. of BzCl at 1208C for 2 h, and we isolated a major
product (86% yield). The 1H NMR spectrum of this product
indicated the structure 2,6,30,40,60-O-penta-benzoyl lactosyl-
pyranoside 26. In another experiment, lactoside 9 was
treated with 7 equiv. of Bu2SnO, followed by 10 equiv. of
BzCl at 1208C for a longer reaction time (7 h). Surprisingly,
a new product, 2,6,20,30,40,60-hexa-benzoate 27 was isolated
in 85% yield. The mechanisms in these two cases are
unclear. Perhaps, with a longer reaction time and lager
excess of Bu2SnO at high temperature, an 8-membered
stannylene acetal from the 3-O and 20-O position of the 26 is
formed (Scheme 2), but further studies are needed to
understand the mechanism. The speculated existence of
3,20-O-stannylene complex is based on the known structure
of 3,20-O-benzylidene lactoside.21 In the newly formed
stannylene complex, the 20-O position may be less hindered
towards benzoylation to give the hexabenzoate 27. This
result is different from Ogawa’s early result, where they
treated a free lactose with slightly more than 3 equiv. of
(Bu3Sn)2O and 6.6 equiv. of BzCl at 458C for 2 days and
2,6,3060-tetra-OBz lactose was isolated in 72% yield. In their
procedure, no migration reaction is involved.10 Therefore,
Another model structure studied was a methyl a-mannoside,
2-epimer of a glucoside. Based on the stannylene complex,
we expected that the migration reaction would be involved
in the benzolation process. Indeed, treatment of a-methyl
mannoside 8 with 3 equiv. of Bu2SnO followed by 3 equiv.
of 3.3 BzCl at 908C for 15 min gave the product 2,3,6-tri-
OBz mannoside 23 in 89% yield. The 2,6-di-OBz 24 and
3,6-di-OBz 25 were also isolated in 6 and 3%, respectively.
Using 2.2 equiv. of BzCl gave 3,6-O-di-OBz 24 in 61%
yield which is comparable to a reported result (66%).10
Taking advantage of the observation, we treated the
stannylene complex (3 equiv. of Bu2SnO) with BzCl
(2.2 equiv.) at room temperature for 2 h, then a higher
temperature (1008C) for 20 min. We were able to isolate the
2,6-diester 24 in 74% yield together with 20% of the 2,3,6-
tri-OBz 23. This procedure thus enables us to prepare three
Scheme 2.

Z. Zhang, C.-H. Wong / Tetrahedron 58 (2002) 6513–6519
6517
the high temperature induced migration procedure is crucial
for the syntheses of compounds 26 and 27.
118.43, 114.43, 102.14, 77.50, 76.99, 72.35, 71.69, 69.91,
62.65, 55.53.
In summary, we have observed a different migration pattern
in organotin-mediated benzoylations at high temperature
using excess amounts of (Bu3Sn)2O and BzCl. We have
used this method in preparation of building blocks that
normally require many steps with traditional synthesis. High
regioselectivities have been found in different sugar
structures including galactosides, glucosides, mannosides,
and even a disaccharide. The conditions used in this study
represent a new method for the synthesis of partially
benzoylated glycosides. The mechanisms of these trans-
formations described above are speculative and require
further investigation.
2.1.3. p-Methylphenyl 3,4,6-tri-O-benzoyl-1-thio-b-D-
galactopyranoside (13).22 p-Methylphenyl 1-thio-b-D-
galactopyranoside23 (100 mg, 0.350 mmol) was treated
with Bu2SnO (266 mg, 1.04 mmol), then BzCl (130 mL,
1.12 mmol) at 1008C as described in the synthesis of
compound 11. Chromatography of the residue (hexane–
EtOAc, 4:1 to 2:1 to 1:1) furnished 13 (192 mg, 91.7%) as a
1
syrup: [a ]2D2¼þ1508 (c 0.5, CHCl3); H NMR (250 MHz,
CDCl3), d 5.91 (d, 1H, J¼3.1 Hz, H-4), 5.43 (dd, 1H,
J¼3.3, 9.7 Hz, H-3), 4.720 (d, 1H, J¼9.6 Hz, H-1), 4.29
(bt, 1H, J¼6.5 Hz, H-5), 4.62 (dd, 1H, J¼6.6 Hz, H-6) 4.37
(dd, 1H each, J¼11.5 Hz, H-60), 4.03 (t, 1H, J¼9.6 Hz,
H-2), 2.38 (s, 3H, Me); 13C NMR (62 MHz, CDCl3) d
165.98, 165.90, 165.29, 88.26, 74.97, 74.47, 68.59, 67.34,
62.45, 21.28.
2. Experimental
2.1.4. 2-(Trimethylsilyl)ethyl 3,4,6-tri-O-benzoyl-b-D-
galactopyranoside (14). 2-(Trimethylsilyl)ethyl b-D-galac-
topyranoside17 (150 mg, 0.535 mmol) was treated with
Bu2SnO (400 mg, 1.6 mmol), followed by BzCl (205 mL,
1.77 mmol) at 808C as described in the preparation of 11.
Work up and chromatography (AcOEt–hexanes, 4:1 to 2:1)
gave compound 14 (271 mg, 86%): [a ]2D5¼þ1.48 (c 2.0,
2.1. Data for compounds
2.1.1. p-Methoxylphenyl 3,4,6-tri-O-benzoyl-b-D-galac-
topyranoside (11). A suspended mixture of p-methoxyl-
phenyl b-D-galactopyranoside14 (100 mg, 0.350 mmol) and
Bu2SnO (266 mg, 1.04 mmol) in toluene–benzene (1:1,
40 mL) was refluxed for 20 min. Approximately 30 mL
solvent was azeotropicly removed using a Dean–stark
apparatus under Ar, then the hot solution (,1008C) was
treated with benzoyl chloride (130 mL, 1.12 mmol). After
being stirred for 1 h at this temperature, the reaction mixture
was diluted with EtOAc (50 mL), cooled to 08C and filtered
through Celite. The solid cake was washed with cold (08C)
EtOAc, and concentrated. Chromatography of the residue
(hexane–EtOAc, 4:1 to 2:1 to 1:1) gave 11 (242 mg, 93%)
1
CHCl3); H NMR (300 MHz, CDCl3) partial data, d 5.91
(dd, 1H, J¼1.0, 3.4 Hz, H-4), 5.41 (dd, 1H, J¼3.5,
10.2 Hz, H-3), 4.66 (dd, 1H, J¼68, 11.2 Hz, H-6a), 4.57
(d, 1H, J¼7.8 Hz, H-1), 4.38 (dd, 1H, J¼6.6, 11.2 Hz,
H-6b), 0.05 (s, 9H, SiMe3); HRMS (MþNaþ) calcd for
C32H36O9SiNa: 615.2026, found 615.2010.
2.1.5. 2-(Trimethylsilyl)ethyl 2-azido-2-deoxy-3,6-di-O-
benzoyl-b-D-galactopyranoside (15) and 2-(trimethyl-
silyl)ethyl 2-azido-2-deoxy-4,6-di-O-benzoyl-b-D-galac-
topyranoside (16). Method a. 2-(Trimethylsilyl)ethyl
1
as a foam: H NMR (250 MHz, CDCl3), d 8.20–7.26 (m,
15H), 7.07 and 6.73 (d, each 2H, J¼9.0 Hz), 5.93 (d, 1H,
J¼3.4 Hz, H-4), 5.47 (dd, 1H, J¼3.4, 10.1 Hz, H-3), 5.02
(d, 1H, J¼7.8 Hz, H-1), 4.62 (dd, 1H, J¼7.7, 11.3 Hz, H-
6), 4.49–4.30 (m, 3H), 3.75 (s, 3H, OCH3). Selected 13C
NMR (62.5 MHz, CDCl3) d 166.4, 166.3, 166.0, 155.6,
150.9, 118.6, 114.5, 102.6, 73.2, 71.6, 69.7, 68.1, 62.3, 55.6.
2-azido-2-deoxy-b-D-galactopyranoside24
(50 mg,
0.175 mmol) was treated with Bu2SnO (107 mg,
0.438 mmol), followed by BzCl (60 mL, 0.524 mmol) at
room temperature as described in the preparation of 11.
Work up and chromatography (AcOEt–hexanes, 4:1 to 2:1)
gave compound 15 (77 mg, 89%) and compound 16 (6 mg,
7%).
2.1.2. p-Methoxylphenyl 4,6-di-O-benzoyl-b-D-galacto-
pyranoside (12). A suspended mixture of p-methoxyl-
phenyl b-D-galactopyranoside (50 mg, 0.175 mmol) and
Bu2SnO (133 mg, 0.52 mmol) in toluene–benzene (1:1,
40 mL) was refluxed for 20 min. Approximately 30 mL
solvent was azeotropically removed using a Dean–stark
apparatus under Ar, then the solution (about 508C) was
treated with benzoyl chloride (45 mL, 0.38 mmol). After
being stirred for 1 h at this temperature, the reaction mixture
was heated to 1008C for 30 min, then diluted with EtOAc
(50 mL), cooled to 08C and filtered through Celite. The solid
cake was washed with cold (08C) EtOAc, and concentrated.
Chromatography of the residue (toluene–acetone, 20:1 then
5:1) gave 11 (33 mg, 32%) and compound 12 (57 mg, 66%).
Compound 15: [a ]2D5¼þ27.68 (c 1.0, CHCl3); H NMR
1
(300 MHz, CDCl3) partial data, d 4.98 (dd, 1H, J¼10.7,
3.2 Hz, H-3), 4.66 (dd, 1H, J¼6.8, 11.5 Hz, H-6), 4.53 (dd,
1H, J¼6.4, 11.5 Hz, H-6), 4.46 (d, 1H, J¼8.1 Hz, H-1),
4.20 (m, 1H), 2.51 (d, 1H, J¼5.3 Hz, OH), 0.03 (s, 9H,
SiMe3); HRMS (MþNaþ) calcd for C25H31O7SiNa:
536.1829, found 536.1820.
Compound 16 had: [a ]D25¼220.08 (c 2.0, CHCl3); 1H NMR
(300 MHz, CDCl3) partial data, d 5.65 (dd, 1H, J¼1.0,
3.4 Hz, H-4), 4.58 (dd, 1H, J¼6.9, 11.4 Hz, H-6), 4.43 (d,
1H, J¼7.9 Hz, H-1), 4.37 (dd, 1H, J¼6.4, 11.4 Hz, H-6),
3.78 (dd, 1H, J¼3.4, 10.3 Hz, H-3), 0.04 (s, 9H, SiMe3);
HRMS (MþNaþ) calcd for C25H31O7SiNa: 536.1829,
found 536.1825.
1
Compound 12 had: H NMR (250 MHz, CDCl3), d 8.12–
8.00 (m, 4H), 7.60–7.38 (m, 6H), 7.01 and 6.68 (bd, each
2H, J¼9.1 Hz), 5.68 (d, 1H, J¼2.8 Hz, H-4), 4.80 (d, 1H,
J¼7.4 Hz, H-1), 4.55–4.40 (m, 2H, H-6), 4.14–3.90 (m,
3H, H-2, H-5, H-3), 3.72 (s, 3H, OCH3); 13C NMR
(62 MHz, CDCl3), d 166.39, 166.03, 155.49, 150.98,
133.55, 133.25, 1330.05, 129.73, 129.08, 128.49, 128.40,
Method b. 2-(Trimethylsilyl)ethyl 2-azido-2-deoxy-b-D-
galactopyranoside (50 mg, 0.175 mmol) was treated with

6518
Z. Zhang, C.-H. Wong / Tetrahedron 58 (2002) 6513–6519
1
Bu2SnO (96 mg, 0.386 mmol), followed by BzCl (45 mL,
0.385 mmol) at 908C as described in the preparation of 11.
Work up and chromatography (AcOEt–hexanes, 4:1 to 2:1)
gave compound 15 (20 mg, 23%) and compound 16 (41 mg,
47%).
[a ]2D2¼22.78 (c 1.0, CHCl3); H NMR data (CDCl3): d
5.15 (m, 1H), 4.77 (dd, 1H, J¼12.2, 4.2 Hz), 4.72 (dd, 1H,
J¼9.6, 1.9 Hz), 4.63 (dd, 1H, J¼12.1, 2.0 Hz), 4.02, 3.96
(dd, 1H each, J¼10.0, 5.0, 4.8 Hz), 2.45 (ddd, 1H, J¼12.3,
5.1, 1.7 Hz), 0.88 (t, 6H, J¼6.6 Hz); HRMS calcd for
C56H92O11S2Na (MþNa): 1027.5979; found: 1027.5958.
2.1.6. Methyl 2,3,6-tri-O-benzoyl-a-D-galactopyranoside
(17),10 methyl 2,4,6-tri-O-benzoyl-a-D-galactopyrano-
side (18)19 and methyl 3,4,6-tri-O-benzoyl-a-D-galacto-
pyranoside (19).25 Methyl a-D-galactopyranoside (50 mg,
0.257 mmol) was treated with Bu2SnO (200 mg,
0.80 mmol), followed by BzCl (120 mL, 1.03 mmol) at
458C for 2.5 h as described in the preparation of 11.
Chromatography (AcOEt–hexanes, 3:1 to 1:1) gave com-
pound 17 (38 mg, 42%), 18 (55 mg, 29%) and 19 (18 mg,
14%).
2.1.9. Methyl 2,3,6-tri-O-benzoyl-a-D-mannopyranoside
(23).10
Methyl
a-D-mannopyranoside
(100 mg,
0.524 mmol) was treated with Bu2SnO (400 mg,
1.60 mmol), followed by BzCl (200 mL, 1.72 mmol) at
1008C as described in the preparation of 11. The residue was
chromatographed (AcOEt–hexanes, 3:1 to 1:1) to give
1
compound 23 (232 mg) in 89% yield: H NMR (500 MHz,
CDCl3), d 8.12, 7.98 and 7.9 (bd, 2H each, J¼7.6 Hz),
7.65–7.29 (m, 9H), 5.64–5.55 (m, 2H, H-2 and H-3), 4.92
(d, 1H, J¼1.6 Hz, H-1), 4.86 (dd, 1H, J¼4.0, 12.0 Hz,
H-6a), 4.66 (dd, 1H, J¼4.0, 12.0 Hz, H-6b), 4.31 (ddd, 1H,
J¼5.1, 9.5, 9.5 Hz, H-4), 3.48 (w, 3H, OCH3), 3.07 (d, 1H,
J¼5.1 Hz, OH); 13C NMR (100 MHz, CDCl3), d 166.82,
166.53, 165.29, 133.32, 133.22, 133.02, 129.76, 129.72,
129.66, 128.42, 128.25, 98.59, 72.51, 71.11, 70.40, 66.13,
63.36, 55.25.
1
Compound 17 had: H NMR (250 MHz, CDCl3), d 8.10–
7.96 (m, 6H), 7.62–7.3 (m, 9H), 5.76 (dd, 1H, J¼2.9,
10.7 Hz, H-3), 5.67 (dd, 1H, J¼3.3, 10.7 Hz, H-2), 5.14 (d,
1H, J¼3.3 Hz, H-1), 4.68 (dd, 1H, J¼6.0, 11.4 Hz, H-6),
4.62 (dd, 1H, J¼6.7, 11.4 Hz, H-6), 4.39 (m, 2H), 3.45 (s,
3H, OCH3), 2.55 (d, 1H, J¼4.2 Hz, OH).
1
Compound 18 had: H NMR (250 MHz, CDCl3), d 8.16–
2.1.10. Methyl 2,6-di-O-benzoyl-a-D-mannopyranoside
(24).10 Methyl a-D-mannopyranoside (50 mg, 0.257 mmol)
was treated with Bu2SnO (200 mg, 0.80 mol) as described in
the preparation of 11. To the mixture at room temperature
was added BzCl (66 mL, 0.56 mmol) and stirred for 3 h. The
mixture was brought to 1008C and stirred for 1 h. Work up
and chromatography (AcOEt–hexanes, 3:1 to 1:1) gave
compound 24 (78 mg, 76%):1H NMR (500 MHz, CDCl3), d
8.11 and 7.91 (bd, 2H each, J¼7.9 Hz), 7.61 and 7.50 (bt,
1H each, J¼7.5 Hz), 7.46 and 7.24 (t, 2H each, J¼7.6 Hz),
5.37 (dd, 1H, J¼1.5, 3.0 Hz, H-2), 4.89 (dd, 1H, J¼2.6,
12.1 Hz, H-6), 4.86 (d, 1H, J¼1.5 Hz, H-1), 4.53 (dd, 1H,
J¼1.4, 12.1 Hz, H-6b), 4.16 (bs, 1H, H-3), 3.91 (m, 2H,
H-4, H-5), 3.44 (s, 3H, OCH3).
8.01 (m, 6H), 7.64–7.3 (m, 9H), 5.83 (d, 1H, J¼3.5 Hz,
H-4), 5.40 (dd, 1H, J¼3.7, 10.3 Hz, H-2), 5.20 (d, 1H,
J¼3.7 Hz, H-1), 4.6–4.35 (m, 4H), 3.45 (s, 3H, OCH3).
1
Compound 19 had: H NMR (250 MHz, CDCl3), d 8.10–
7.20 (m, 15H), 5.91 (d, 1H, J¼3.3 Hz, H-4), 5.57 (dd, 1H,
J¼3.3, 10.3 Hz, H-3), 5.05 (d, 1H, J¼3.8 Hz, H-1), 4.58
(dd, 1H, J¼6.8, 10.3 Hz, H-6), 4.63–4.22 (m, 4H), 3.54 (s,
3H, OCH3).
2.1.7. Methyl 2,3,6-tri-O-benzoyl-a-D-glucopyranoside
(20)10 and methyl 2,6-di-O-benzoyl-a-D-glucopyranoside
(21).10 Methyl a-D-glucopyranoside (100 mg, 0.524 mmol)
was treated with Bu2SnO (400 mg, 1.60 mmol), followed by
BzCl (200 mL, 1.72 mmol) at 1008C for 30 min as described
in the preparation of 11. Chromatography of the residue
(AcOEt–hexanes, 3:1 to 1:1) gave compound 20 (233 mg)
in 91% yield: 1H NMR (500 MHz, CDCl3), d 8.10 (m, 2H),
7.97 (m, 4H), 7.65 (m, 1H), 7.51–7.33 (m, 8H), 5.78 (dd,
1H, J¼9.5, 10.0 Hz, H-3), 5.26 (dd, 1H, J¼4.0, 10.0 Hz,
H-2), 5.14 (d, 1H, J¼4.0 Hz, H-1), 4.78 (dd, 1H, J¼5.0,
12.5 Hz, H-6), 4.62 (dd, 1H, J¼2.5, 12.5), 4.11 (ddd, 1H,
J¼2.5, 4.5, 10.0 Hz, H-5), 3.87 (t, 1H, J¼10.0 Hz, H-4);
13C NMR (125 MHz, CDCl3), d 167.3, 166.9, 165.9, 133.4,
133.3, 129.8, 129.79, 128.4, 128.39, 128.38, 73.8, 71.3,
70.0, 69.7, 63.4, 55.4. Further elution gave compound 21
2.1.11. Methyl 3,6-di-O-benzoyl-a-D-mannopyranoside
(25).10 Methyl a-D-mannopyranoside (50 mg, 0.257 mmol)
was treated with Bu2SnO (200 mg, 0.80 mol) as described in
the preparation of 11. To the mixture at room temperature
was added BzCl (91 mL, 0.77 mmol) and stirred for 3 h. The
residue was worked up and chromatographed (AcOEt–
hexanes, 3:1 to 1:1) to give compound 25 (92 mg, 88%):1H
NMR (400 MHz, CDCl3), d 8.10 (d, 4H, J¼7.9 Hz), 7.59
(m, 2H), 7.46 (m, 4H), 5.38 (dd, 1H, J¼3.2, 9.7 Hz, H-3),
4.82 (d, 1H, J¼1.8 Hz, H-1), 4.79 (dd, 1H, J¼4.7, 12.0 Hz,
H-6a), 4.61 (dd, 1H, J¼2.3, 12.0 Hz, H-6b), 4.20 (m, 1H),
4.11 (ddd, 1H, J¼5.0, 10.0, 10.0 Hz, H-4), 4.01 (dddd, 1H,
J¼2.0, 4.4, 10.0 Hz, H-5), 3.46 (s, 3H, OCH3), 3.07 and
2.07 (d, 1H each, J¼5.3, 4.7 Hz, OH).
1
(12.7 mg, 6.8%): H NMR (250 MHz, CDCl3), d 8.08 (m,
4H), 7.59–7.25 (m, 6H), 5.05 (d, 1H, J¼3.7 Hz, H-1), 4.95
(dd, 1H, J¼3.7, 9.9 Hz, H-2), 4.79 (dd, 1H, J¼5.0,
12.5 Hz, H-6a), 4.53 (dd, 1H, J¼2.5, 12.5 Hz, H-6b), 4.20
(t, 1H, J¼9.9 Hz), 3.95 (m, 1H, H-5), 3.59 (t, 1H,
J¼10.0 Hz, H-4), 3.40 (s, 3H); 13C NMR (62 MHz,
CDCl3), d 167.3, 133.4, 129.9, 129.8, 128.4, 79.3, 73.7,
71.5, 70.6, 69.6, 63.6, 55.4.
2.1.12. 2-(Trimethylsilyl)ethyl 2,6-di-O-benzoyl-4-O-
(3,4,6-tri-O-benzoyl-b-D-galactopyranosyl)-b-D-gluco-
pyranoside (26). 2-(Trimethylsilyl)ethyl 4-O-b-D-galacto-
pyranosyl-b-D-glucopyranoside17 (50 mg, 0.113 mmol)
was treated with Bu2SnO (169 mg, 0.678 mmol), followed
by BzCl (118 mL, 1.01 mmol) at 1208C for 2 h as described
in the preparation of 11. Work up and chromatography
(toluene–ether, 30:1) gave compound 26 (93.5 mg, 86%):
[a ]2D5¼þ38.28 (c 0.9, CHCl3); 1H NMR (300 MHz, CDCl3)
2.1.8. 3-(Hexadecylsulfonyl)-2-[(hexadecylsulfonyl)-
methyl]propyl 3,6-di-O-benzoyl-2-deoxy-a-D-arabino-
hexopyranoside (22). Yield 84%, see Ref. 20.

Z. Zhang, C.-H. Wong / Tetrahedron 58 (2002) 6513–6519
6519
partial data, d 5.87 (d, 1H, J¼3.4 Hz, H-40), 5.40 (dd, 1H,
J¼3.4, 10.2 Hz, H-30), 5.26 (dd, 1H, J¼8.1, 9.6 Hz, H-2),
4.71 (d, 1H, J¼7.8 Hz, H-10), 4.65 (d, 1H, J¼8.0 Hz, H-1),
20.08 (s, 9H, SiMe3); 13C NMR (75 MHz, CDCl3) partial
data, d 167.0, 166.2, 165.9, 165.4, 165.3, 104.7, 100.3, 82.7,
73.9, 73.6, 73.1, 73.0, 69.7, 68.2, 67.4, 17.9, 21.5; HRMS
(MþNaþ) calcd for C52H54O16SiNa: 985.3079, found
985.3057.
5. David, S.; Thieffry, A.; Veyrieres, A. J. Chem. Soc., Perkin
Trans. 1 1981, 1796–1801.
6. Fernandex, P.; Jimenez-Barbero, J.; Martin-Lomas, M.
Carbohydr. Res. 1994, 254, 61–79.
7. Glen, A.; Leigh, D. A.; Martin, R. P.; Smart, J. P.; Truscello,
A. M. Carbohydr. Res. 1993, 248, 365–369.
8. Ogawa, T.; Nozaki, M.; Matsui, M. Carbohydr. Res. 1978, 60,
C7–C10. Ogawa, T.; Matsui, M. Carbohydr. Res. 1977, 56,
C1–C6. Ogawa, T.; Matsui, M. Tetrahedron 1981, 37,
2363–2369.
2.1.13. 2-(Trimethylsilyl)ethyl 2,6-di-O-benzoyl-4-O-
(2,3,4,6-tri-O-benzoyl-b-D-galactopyranosyl)-b-D-gluco-
pyranoside (27). 2-(Trimethylsilyl)ethyl 4-O-b-D-galacto-
pyranosyl-b-D-glucopyranoside (50 mg, 0.113 mmol) was
treated with Bu2SnO (183 mg, 0.734 mmol), followed by
BzCl (160 mL, 1.13 mmol) at 1208C for 7 h as described in
the preparation of 11. Work up and chromatography
(toluene–ether, 30:1) gave compound 27 (102 mg, 85%):
[a]2D5¼þ79.98 (c 0.9, CHCl3); 1H NMR (300 MHz, CDCl3)
partial data, d 5.97 (0d, 1H, J¼3.2 Hz, H-40), 5.91 (dd, 1H,
J¼8.1, 10.5 Hz, H-2 ), 5.60 (dd, 1H, J¼3.4, 10.5 Hz, H-30),
5.25 (dd, 1H, J¼8.1, 9.5 Hz, H-2), 5.04 (d, 1H, J¼7.8 Hz,
H-10), 4.61 (d, 1H, J¼8.1 Hz, H-1), 20.14 (s, 9H, SiMe3);
HRMS (MþNaþ) calcd for C59H58O17SiNa: 1089.334,
found 1089.3340.
9. Roelens, S. J. Org. Chem. 1996, 61, 5257–5263.
10. Bredenkamp, M. W.; Spies, H. S. C. Tetrahedron Lett. 2000,
41, 543–546. Bredenkamp, M. W.; Spies, H. S. C.; van der
Merwe, M. J. Tetrahedron Lett. 2000, 41, 547–550.
11. Hodosi, G.; Kovac, P. J. Am. Chem. Soc. 1997, 119,
2335–2336.
12. Hodosi, G.; Kovac, P. Carbohydr. Res. 1997, 303, 239–243.
13. Nicolaou, K. C.; van Delft, F. L.; Conley, S. R.; Mitchell, H. J.;
Jin, Z.; Rodriguez, R. M. J. Am. Chem. Soc. 1997, 119,
9057–9058.
14. De Bruyne, C. K.; Wouters-Leysen, J. Carbohydr. Res. 1971,
18, 124–126.
15. Williams, J. M.; Richardson, A. C. Tetrahedron 1967, 23,
1369–1378.
16. Qin, H.; Grindley, T. B. J. Carbohydr. Chem. 1994, 13,
475–490.
17. Jansson, K.; Ahlfors, S.; Frejd, T.; Kihlberg, J.; Magnusson,
G.; Dahmen, J.; Noori, G.; Stenvall, K. J. Org. Chem. 1988,
53, 5629–5647.
Acknowledgments
Support of the research by the Skaggs Funds is
acknowledged.
18. Li, K. C.; Helm, R. F. Carbohydr. Res. 1995, 273, 249–253.
19. Varela, O.; Cicero, D.; De Lederkremer, R. M. J. Org. Chem.
1989, 54, 1884–1890.
20. Zhang, Z.; Magnusson, G. J. Org. Chem. 1996, 61,
2383–2393.
References
21. Sakairi, N.; Kuzuhara, H. Carbohydr. Res. 1993, 246, 61–73.
22. Zhang, Z.; Ollmann, I. R.; Ye, X.-S.; Wischnat, R.; Baasov, T.;
Wong, C.-H. J. Am. Chem. Soc. 1999, 121, 734–753.
23. Vargas-Berenguel, A.; Meldal, M.; Paulsen, H.; Jensen, K. J.;
Bock, K. J. Chem. Soc., Perkin Trans. 1 1994, 3287–3294.
24. Wilstermann, M.; Kononov, L. O.; Nilsson, U.; Ray, A. K.;
Magnusson, G. J. Am. Chem. Soc. 1995, 117, 4742–4754.
25. Mulard, L. A.; Glaudemans, C. P. J. Carbohydr. Res. 1995,
274, 209–222. Zhang, Z. Thesis, LUTKDH/(TKOK-1027)1-
82/(1996), Lund University, Sweden.
1. David, S. Prep. Carbohydr. Chem. 1997, 69–86.
2. David, S.; Hanessian, S. Tetrahedron 1985, 41, 643–663.
3. Grindley, T. B. Applications of Stannyl Ethers and Stannylene
Acetals to Oligosaccharide Synthesis; American Chemical
Society: Washington, DC, 1994; Vol. 560. Grindley, T. B.
Adv. Carbohydr. Chem. Biochem. 1998, 53, 17–142.
4. David, S.; Thieffry, A. Tetrahedron Lett. 1981, 22,
2647–2650. Grindley, T. B.; Thangarasa, R. Can. J. Chem.
1990, 68, 1007–1019.