Two approaches to the synthesis of 3-β-D-glucopyranosyl-D-glucitol

Article abstract of DOI:10.1016/j.carres.2004.07.019

Glycosidation of 1,2:5,6-di-O-isopropylidene-d-glucose with tetra-O-acetyl-glucosyl bromide in 1:1 benzene-MeNO₂ afforded approximately equal amounts of the 3-O-β-d-glycoside and the rearranged 6-O-β-d-glycoside, while in MeCN only the latter was formed. When tetra-O-acetyl-β-thiophenylglucoside was used as donor in CH ₂Cl₂ in the presence of NIS/TfOH as activator, the 6-O-β-d-glycoside and a 3-O-orthoester were formed in a 1:2 ratio at -20°C, while at 20°C only the former could be isolated. Glycosidation of 1-O-benzoyl-2,4-O-benzylidene-5,6-O-isopropylidene-d-glucitol with tetra-O-acetyl-glucosyl bromide in MeCN in the presence of Hg(CN)₂ afforded the corresponding 3-O-α- and 3-O-β-glycopyranoside in a 1:4 ratio in MeCN and 1:5 in 1:1 benzene-MeNO₂, respectively. When Hg(CN)₂/HgBr₂ was used as promoter, the corresponding orthoester was also formed. When tetra-O-acetyl-β-thiophenylglucoside was used as donor, the 3-O-β-anomer and the orthoester were obtained predominantly in a 3:2 ratio together with traces of the 3-O-α-glycoside. Both β-glycosides could be smoothly converted into 3-β-d- glucopyranosyl-d-glucitol.

Full text of DOI:10.1016/j.carres.2004.07.019

Carbohydrate
RESEARCH
Carbohydrate Research 339 (2004) 2407–2414
Note
Two approaches to the synthesis of 3-b-D-glucopyranosyl-D-glucitol
*
´ ´ ´
Janos Kuszmann, Gabor Medgyes and Sandor Boros
IVAX Drug Research Institute, H-1325 Budapest, PO Box 82, Hungary
Received 21 June 2004; accepted 28 July 2004
Available online 25 August 2004
Dedicated to the memory of Aleksander Zamojski who passed away February 23, 2004
Abstract—Glycosidation of 1,2:5,6-di-O-isopropylidene-D-glucose with tetra-O-acetyl-glucosyl bromide in 1:1 benzene–MeNO2
afforded approximately equal amounts of the 3-O-b-D-glycoside and the rearranged 6-O-b-D-glycoside, while in MeCN only the lat-
ter was formed. When tetra-O-acetyl-b-thiophenylglucoside was used as donor in CH2Cl2 in the presence of NIS/TfOH as activator,
the 6-O-b-D-glycoside and a 3-O-orthoester were formed in a 1:2 ratio at ꢀ20ꢀC, while at 20ꢀC only the former could be isolated.
Glycosidation of 1-O-benzoyl-2,4-O-benzylidene-5,6-O-isopropylidene-D-glucitol with tetra-O-acetyl-glucosyl bromide in MeCN in
the presence of Hg(CN)2 afforded the corresponding 3-O-a- and 3-O-b-glycopyranoside in a 1:4 ratio in MeCN and 1:5 in 1:1 benz-
ene–MeNO2, respectively. When Hg(CN)2/HgBr2 was used as promoter, the corresponding orthoester was also formed. When tetra-
O-acetyl-b-thiophenylglucoside was used as donor, the 3-O-b-anomer and the orthoester were obtained predominantly in a 3:2 ratio
together with traces of the 3-O-a-glycoside. Both b-glycosides could be smoothly converted into 3-b-D-glucopyranosyl-D-glucitol.
ꢁ 2004 Elsevier Ltd. All rights reserved.
Keywords: Glycosidation; Partially protected D-glucitol; Disaccharides
For our ongoing search on heparin degradation prod-
ucts and their analogs,1,2 a larger amount of 3-b-D-glu-
copyranosyl-D-glucitol (1) was needed as starting
material. This compound can be easily prepared by
borohydride reduction of the corresponding disaccha-
ride laminaribiose (3-b-D-glucopyranosyl-D-glucose 2),
but the latter is not commercially available and its prep-
aration by hydrolysis of natural polysaccharides is a
very tedious process.3 In the last decades, many syn-
thetic approaches were published4–16 for 2 via 6, using
1,2:5,6-di-O-isopropylidene-D-glucofuranose 3 as accep-
tor and tetra-O-acetyl-a-D-glucopyranosyl bromide 4 as
donor. All of these methods had the drawback that the
yields of the glycosidation reaction affording 6 were very
modest, due to the fact that under acidic conditions an
isopropylidene migration takes place resulting in
1,2:3,5-di-O-isopropylidene-D-glucofuranose 7, which
reacts with 4 yielding the corresponding 6-O-glycosyl-
ated derivative 8.15 In 1990, a high yielding (83%)
approach to 6 was published,17 which used the
thiophenylglycoside 5 as donor, N-iodosuccinimide as
promoter and TfOH as catalyst.
1. Reinvestigation of the synthesis of 6
In a first attempt, we repeated the glycosylation reaction
of 3 with 4 at 20ꢀC, using the conditions described by
15
Liptak et al. (1:1 benzene–MeNO2, Hg(CN)2) and
´
could isolate the two isomeric glycosides 6 and 8 after
repeated column chromatography in 25% and 28%
yields, respectively. When acetonitrile was used as sol-
vent, only the 6-O-glycoside 8 could be isolated in a very
poor yield (15%). Thereafter, we applied the thioglyco-
side 5 as donor and used the conditions according to
lit.17 but, contrary to the published data, the following
results were obtained. When the reaction was carried
out at ꢀ40ꢀC, two compounds, the rearranged 6-O-gly-
coside 8 as well as the orthoester 9 were formed in a 1:2
ratio. The structure of the latter is evident from its
*
Corresponding author. Tel.: +36 1 399 3300; fax: +36 1 399 3356;
e-mail: janos.kuszmann@idri.hu
1
NMR spectra as in the H NMR spectrum the methyl
0008-6215/$ - see front matter ꢁ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2004.07.019

2408
J. Kuszmann et al. / Carbohydrate Research 339 (2004) 2407–2414
HO
HO
OH
O
HO
OH
HO
HO
HO
OH
HO
HO
O
O
O
OH
O
OH
HO
OH
HO
OH
1
2
O
R1
R2
O
O
Me2C
O
O
Me2C
O
O
O
AcO
O
CMe2
+
O
CMe2
AcO
OAc
O
O
O
AcO
HO
OAc
4 R1 = H, R2 = Br
5 R1 = SPh, R2 = H
3
AcO
OAc
OAc
O
6
AcO
Me2C
O
O
HO
O
O
O
AcO
O
CMe2
O
O
O
O
O
O
O
AcO
AcO
O
CMe2
OAc
OAc
O
Me2C
O
O
CMe2
Me
O
Me2C
O
O
7
9
OAc
8
Scheme 1.
signal of the orthoester appears at d 1.76ppm, well sep-
arated from the three acetyl signals (d 2.06, 2.06 and
2.07ppm), and in the 13C NMR spectrum the signal of
the quaternary carbon atom can be detected at d
121.4ppm, while the three carbonyl atoms appear at d
168.9, 169.5 and 170.0ppm, respectively. The NOE
effect observed between the methyl group of the ortho-
ester, H-30 and H-50 proved the exo orientation of the
aglycon that is the S-configuration of the orthoester car-
bon. When the temperature of the glycosidation reaction
was raised to 20ꢀC for 5min, only the 6-O-glycoside 8
could be isolated from the multicomponent mixture in
a modest yield (24%) (Scheme 1).
approach was taken into consideration, that is the
glycosidation of an appropriately protected D-glucitol
derivative.
3. Synthesis of a partially protected D-glucitol
derivative 15
D-Glucitol can be easily converted into its 2,4-O-benzyl-
idene derivative 13, which, on treatment with acetone in
the presence of CuSO4, afforded18 the corresponding
5,6-O-isopropylidene derivative 14. The relative low
yield of this latter reaction (ꢁ25%) can be ascribed to
the low solubility of 13 in acetone and the fact that
the formed diacetal 14 reacts further with acetone
affording triacetal 16 (Scheme 3). To overcome this
problem, 2,2-dimethoxypropane was used as reagent in
the presence of a catalytic amount of p-TsOH and differ-
ent solvents, for example, DMF, Me2SO, dioxane and
methanol, but only DMF proved to be superior to ace-
tone. The optimal reaction time as well as the optimal
excess of the reagent was established by following the
reaction by GLC (after acetylation of the withdrawn
samples).
2. Conversion of 6 into 1
The disaccharide derivative 6 was first deacetylated
´
according to Zemplen and thereafter the 5,6-O-isoprop-
ylidene group of the resulting glycoside 10 could be
removed selectively with aqueous sulfuric acid at rt
affording 11. The remaining 1,2-O-isopropylidene group
of the latter could be hydrolyzed with aqueous sulfuric
acid at 60ꢀC and the resulting free disaccharide 12 af-
forded, on treatment with sodium borohydride, the re-
quired glucitol derivative 1 (Scheme 2). Because of the
low yield and tedious preparation of 6, an independent
As can be seen from the Tables 1 and 2 the reaction
rate increases with the ratio of the reagent, but simulta-
neously the amount of triacetal 16 increases too. As,

J. Kuszmann et al. / Carbohydrate Research 339 (2004) 2407–2414
2409
6
1
O
O
O
HO
HO
O
HO
Me2C
O
O
O
O
CMe2
OH
OH
O
HO
O
CMe2
O
O
O
O
O
HO
HO
HO
HO
OH
HO
OH
HO
OH
OH
OH
12
OH
11
10
Scheme 2.
O
RO
HO
O
HO
HO
HO
Me2C
Me2C
O
O
O
O
Ph
O
O
Ph
Ph
O
O
+
OH
Me2C
O
O
13
16
14 R = H
15 R = Bz
Scheme 3.
crude product from benzene. On the other hand, forma-
tion of the triacetal 16 seems to be a slower process than
the dissolution of the starting material, therefore the
amount of solvent could be diminished without signifi-
cantly lowering the yield. The diacetal 14 was treated
in pyridine at low temperature (ꢀ20ꢀC) with 1.1equiv
of benzoyl chloride to afford the 1-O-benzoate 15, which
was contaminated with some 1,4-di-O-benzoate. The lat-
ter could be removed by column chromatography, but
for the large scale glycosidation reaction the crude mix-
ture could be used, as the diester remained unchanged
during this reaction and the required glycoside had to
be separated by column chromatography anyway.
Table 1. Reaction of 13 with 1.6 equiv of 2,2-dimethoxypropane in
DMF according to GLC data
Time (h)
13 (%)
14 (%)
16 (%)
Yielda (%)
0.5
1
35.0
28.2
24.9
24.0
19.2
40.3
45.4
47.1
49.3
48.1
8.4
11.5
13.2
13.8
17.4
15
24
30
34
36
2
4
20
a Isolated yield of 14.
Table 2. Reaction of 13 with 3 equiv of 2,2-dimethoxypropane in
DMF according to GLC data
Time (h)
13 (%)
14 (%)
16 (%)
Yielda (%)
44
0.5
1
15.7
7.6
3.5
1.7
1.1
47.0
50.1
47.2
44.4
39.1
17.6
27.6
37.1
45.2
50.1
2
4. Glycosidation of 15
3
5
38
When bromide 4 was used as donor, acetonitrile as sol-
vent and Hg(CN)2 as promoter for the glycosydation of
15, the crude reaction product (formed at 20ꢀC; 20h)
contained besides several decomposition products two
anomeric glycosides 17 and 18 in a 1:4 ratio. From this
mixture, the needed crystalline b-glycoside 18 could be
separated in a relatively modest yield (ꢁ20%) after re-
peated column chromatography only because of the
similar Rf value of these isomers. This purification step
could not be avoided, as the purity of the final product
a Isolated yield of 14.
during the work up process, it is difficult to remove the
unchanged starting material 13, the yield of the isolated
product 14 is not proportional to its amount present in
the reaction mixture as detected by GLC. For this rea-
son, it is worthwhile to drive the conversion as far as
possible, as triacetal 16, which is formed as by-product,
can be easily removed by a simple recrystallization of the

2410
J. Kuszmann et al. / Carbohydrate Research 339 (2004) 2407–2414
Table 3. Influence of the reaction conditions on the glycosidation reaction
Donor
Promoter
Solventa
Temp/time
17:18:19
Yieldb (%)
4
4
4
4
4
5
4
4
4
Hg(CN)2
Hg(CN)2
Hg(CN)2
Hg(CN)2/HgBr2
AgOTf
A
A
A
A
A
B
20ꢀC; 20h
20ꢀC; 5h
20ꢀC; 48h
20ꢀC; 20h
1:4:Tr
1:3:Tr
1:4:Tr
1:4:2
20
10
20
5
20ꢀC; 20h
––
––
30
30
35
25
NIS/AgOTf
Hg(CN)2
Hg(CN)2
Hg(CN)2
ꢀ30ꢀC; 1.5h
20ꢀC; 15h + 45ꢀC; 2h
20ꢀC; 15h + 45ꢀC; 3h
50ꢀC; 5h
Tr:3:2
1:5:Tr
1:5:Tr
1:3:Tr
C
C
C
a A: MeCN; B: CH2Cl2; C: 1:1 benzene:MeNO2.
b Yield of isolated 18.
1 depends on the purity of 18 because the further inter-
mediates could not be separated from the accompanying
by-products. For increasing the yield of 18 different
reaction conditions (solvents, promoters, reaction tem-
perature) were applied (Table 3), the formed glycosides
were isolated and their structure was established by
NMR spectroscopy (Tables 4–6). Depending on the
reaction conditions, three main products were formed
in different ratios the two anomers 17 and 18 as well
as the orthoester 19 (Scheme 4). The chemical shifts of
the anomeric protons of these products differed signifi-
cantly (17: 5.39ppm, J1,2 3.8Hz; 18: 4.78ppm, J1,2
8Hz; and 19: 5.9ppm, J1,2 5Hz), consequently their ra-
tio could be determined in the crude reaction mixtures
by recording the NMR spectra. When the reaction time
was shortened to 5h, the reaction was incomplete, con-
sequently the isolated yield of 18 dropped. A prolonged
reaction time of 48h had no influence on the yield.
When Hg(CN)2/HgBr2 was used as promoter, a signifi-
cant amount of the orthoester 19 was formed, diminish-
ing the yield. In the presence of AgOTf no reaction took
place. When the thioglycoside 5 was used as donor in the
presence of NIS/AgOTf, the a-anomer 17 was only
formed in traces, but the amount of the orthoester in-
creased substantially. The best results were obtained
using 4 as donor, Hg(CN)2 as promoter and 1:1 ben-
zene–nitromethane as solvent, but in this system the
reaction was slower, therefore after 20h at 20ꢀC the
temperature had to be increased to 45ꢀC to complete
the reaction. In this way the yield of 18 could be in-
creased to 35%.
6. Experimental
6.1. General methods
Organic solns were dried over MgSO4 and concentrated
under diminished pressure at or below 40ꢀC. TLC: E.
Merck precoated Silica Gel 60 F254 plates, with
EtOAc–hexane mixtures (A, 1:1; B, 1:2; C, 2:1),
EtOAc–MeOH mixtures (D, 3:1; E, 1:1); detection by
spraying the plates with a 0.02Msoln of I
0.3Msoln of KI in 10% aq H SO4 soln followed by
and a
2
2
heating at ca. 200ꢀC. For column chromatography, Kie-
selgel 60 was used. Melting points are uncorrected. Opti-
cal rotations were determined on 1.0% solns in CHCl3 at
20ꢀC unless stated otherwise. GLC were performed on a
Chrompack CP9000 equipment using a RH-5ms column
(30m · 0.25mm), N2 as carrier gas at 25kPa, tempera-
ture, 2ꢀCminꢀ1 from 185 to 325ꢀC and a CP-Maitre
data system. The NMR spectra were recorded on a
Bruker Avance 500 spectrometer at 500 (1H) and 125
(13C)MHz, respectively, at ambient temperature. The
chemical shifts were referenced to dTMS = 0ppm. The
1
solvent is indicated at the H NMR spectral data. For
structure determination 1H, 1H-COSY, TOCSY,
HMQC or HSQC, HMBC as well as selective 1D
TOCSY and NOESY spectra were recorded. For data
see Tables 4–6.
6.2. 3-O-(b-D-Glucopyranosyl)-D-glucitol (1)
(a) To a stirred soln of 12 (0.9g, 2.6mmol) in water
(10mL), NaBH4 (0.2g) was added at rt. After 2h, so-
dium ions were removed with an ion exchange resin
and the filtered soln was concentrated. The residue
was reevaporated with MeOH (2 · 10mL), the obtained
syrup was dissolved in water and freeze–dried to give 1
(0.85g, 94%) as an amorphous hygroscopic powder,
[a]D ꢀ17 (c 1, water). Anal. Calcd for C12H24O11: C,
41.86; H, 7.03. Found: C, 41.64; H, 7.15.
5. Conversion of 18 into 1
´
The ester groups of 18 were removed by the Zemplen
method and the benzylidene group of the resulting pen-
tahydroxy derivative 20 was removed by catalytic
hydrogenation over Pd/C. Finally, the terminal 5,6-O-
isopropylidene group of the formed 21 was hydrolyzed
with aqueous sulfuric acid to yield, after freeze–drying,
the target disaccharide 1 in excellent yield.
(b) To a soln of 18 (4.46g, 6mmol) in dry MeOH
(20mL), 2MNaOMe (0.2mL) was added. After 1h,
when according to TLC the reaction was complete, the

Table 4. Characteristic 1H NMR chemical shifts
No.
Solvent
Glucose/glucitol unit
Glucose unit
Others
Ha-1
Hb-1
3.54
H-2
H-3
H-4
H-5
Ha-6
Hb-6
H-10
H-20
H-30
H-40
H-50
Ha-60
Hb-60
1
Me2SO
CDCl3
CDCl3
Me2SO
Me2SO
Me2SO
CDCl3
CDCl3
CDCl3
CDCl3
3.41
5.94
5.82
5.86
5.81
3.74
4.53
4.46
4.60
4.60
3.86
4.31
4.26
4.27
4.23
3.83
4.15
4.29
4.24
4.20
3.57
3.89
3.92
3.94
3.82
3.41
4.18
4.05
4.22
3.93
3.80
3.83
3.74
3.84
3.71
3.64
3.72
4.17
4.34
3.71
4.23
4.40
4.47
4.51
4.44
3.41
3.64
3.92
3.85
3.36
3.91
4.07
3.58
3.97
4.08
3.91
3.57
3.98
4.14
4.30
4.59
5.76
4.27
4.34
3.07
4.97
4.37
2.90
2.99
3.16
5.15
5.14
3.13
3.16
2.97
5.04
4.85
3.01
3.03
3.14
3.67
3.88
3.13
3.19
3.34
4.11
4.16
3.41
3.40
3.72
4.22
8
9
1.76 (orthoacetate); 2 · 2.06, 2.07 (OAc)
10
11
14
15
17
18
19
3.68
3.70
3.50–3.63
4.60
4.67
4.47
4.46
4.66
4.04–4.11
4.06–4.15
4.03 4.12
5.39
5.13
4.62
5.11
5.21
5.19
5.57
5.15
4.91
5.12
3.65
4.00
4.34
4.18
4.15
4.25
4.24
4.24
4.33
4.59
4.72
4.79
5.88
1.88 (orthoacetate); 2.04, 2.07, 2.09 (OAc)
Table 5. Characteristic 1H, 1H coupling constants
No.
Glucose/glucitol unit
Glucose unit
2J (1a,1b) 3J (1a,2) 3J (1b,2) 3J (2,3) 3J (3,4) 3J (4,5) 3J (5,6a) 3J (5,6b) 2J (6a,6b) 3J (10,20) 3J (20,30) 3J (30,40) 3J (40,50) 3Jð50; 6a0 Þ 3Jð50; 6b0 Þ 2Jð60a; 60bÞ
1
11.3
4.2
6.3
ꢁ2.5
ꢁ3.0
3.2
8.5
7.3
7.7
4.3
9.0
7.2
8.2
9.2
6.6
8.7
ꢁ6.0
7.3
5.8
6.2
5.9
5.5
4.6
2.9
1.8
7.0
6.4
3.5
6.4
6.2
11.1
11.1
8.6
7.9
7.9
5.1
7.8
7.9
8.6
9.7
2.8
8.0
8.5
9.1
9.5
1.9
8.5
9.0
9.1
9.7
9.5
8.5
9.1
8.0
2.3
2.0
4.8
11.1
12.2
8
3.6
3.5
3.5
3.7
6.1
6.8
ꢁ0
9
ꢁ0
ꢁ0
ꢁ0
ꢁ1
ꢁ1
ꢁ1
ꢁ1
ꢁ1
(4.2)
10
11
14
15
17
18
19
ꢁ3.0
2.9
8.7
5.3
7.0
5.1
10.0
10.8
11.2
8.4
45
11.3
11.7
6.1
5.4
ꢁ1
ꢁ1
8.6
(5.8)
ꢁ1
(3.8)
(6.1)
3.8
7.8
5.3
9.9
9.5
4.5
9.9
9.6
4.5
9.5
9.6
9.2
11.3
11.8
6.0
7.3
6.8
4.8
ꢁ1
2.3
2.8
4.0
5.6
12.1
12.5
ꢁ1
5.3
6.4
8.6

2412
J. Kuszmann et al. / Carbohydrate Research 339 (2004) 2407–2414
sodium ions were removed by an ion exchange resin and
the filtered soln was concentrated. The residue, which
contained besides 20 methyl benzoate, was dissolved in
MeOH (50mL), water (3mL), AcOH (1mL) and 5%
Pd/C catalyst (2g) were added and the mixture was
hydrogenated at rt for 15h. The filtered soln was con-
centrated, the residue partitioned between CHCl3 and
water, the aq soln was concentrated to a volume of
15mL, 1.5mL Msulfuric acid was added and the mix-
ture was heated at 60ꢀC for 1h. After neutralization
with an ion exchange resin, the filtered soln was concen-
trated to a vol of 15mL and was freeze–dried to give 1
(2g, 97%) identical with that described above.
6.3. Glycosidation of 3 using phenyl 1-thioglucoside
5 as donor
A soln of 3 (2.6g, 10mmol) and 519 (4.4g, 10mmol) in
CH2Cl2 (50mL) was stirred in the presence of molecular
˚
sieves (4A) (6g) for 1h. Thereafter the mixture was
cooled to ꢀ40ꢀC and NIS (3.4g, 15mmol) as well as
TfOH (0.2mL) were added. Stirring was continued at
ꢀ40ꢀC for 15min and then the reaction was quenched
with Et3N (3mL). The filtered soln was diluted with
CH2Cl2 (50mL) washed with 10% aq thiosulfate soln,
4% NaHCO3 soln and water. The residue of the dried
and concentrated soln was submitted to column chro-
matography (A). The fractions having Rf 0.65 afforded
3,4,6-tri-O-acetyl-1,2-(S)-O-[10-O-(1,2:5,6-di-O-isopropyl-
idene-D-glucofuranos-3-yl)-ethylidene]-a-D-glucopyra-
nose 9 (1.0g, 17%) as a syrup. Anal. Calcd for
C26H38O15: C, 52.88; H, 6.49. Found: C, 52.65; H, 6.72.
Concentration of the fractions having Rf 0.60 afforded
8 (0.5g, 8.5%), mp 103–104ꢀC (ether–hexane); lit.15 mp
104–105ꢀC.
When the temperature of the reaction was raised for
5min to 20ꢀC before quenching with Et3N, only 8
(1.4g, 24%) could be separated by column
chromatography.
6.4. 1,2:5,6-Di-O-isopropylidene-3-O-(b-D-glucopyranos-
yl)-D-glucofuranose (10)
To a soln of 615 (3.5g, 5.9mmol) in dry MeOH (60mL)
2MNaOMe (0.2mL) was added. After 1h, when accord-
ing to TLC the reaction was complete, the sodium ions
were removed by an ion exchange resin and the filtered
soln was concentrated to give 10 (2.5g, ꢁ100%) as a syr-
up, [a]D ꢀ33 (c 1, water). Anal. Calcd for C18H30O11: C,
51.18; H, 7.16. Found: C, 51.02; H, 7.26.
6.5. 1,2-O-Isopropylidene-3-O-(b-D-glucopyranosyl)-D-
glucofuranose (11)
A soln of 10 (2.4g, 5.7mmol) in 0.1N H2SO4 (20mL)
was kept at rt for 24h. The soln was neutralized with

J. Kuszmann et al. / Carbohydrate Research 339 (2004) 2407–2414
2413
15 + 4 or 5
AcO
AcO
AcO
BzO
AcO
BzO
O
O
O
BzO
O
AcO
O
O
O
O
O
O
O
O
AcO
O
Ph
O
Ph
Ph
AcO
AcO
AcO
OAc
OAc
Me2C
18
Me
O
O
O
Me2C
Me2C
O
O
O
19
17
HO
HO
HO
HO
O
HO
O
O
O
O
O
HO
OH
1
Ph
OH
HO
HO
OH
OH
O
O
Me2C
Me2C
O
O
21
20
Scheme 4.
an ion exchange resin, filtered and concentrated to give
11 (2.0g, 92%). [a]D ꢀ23 (c 1, water); Rf 0.4 (D). Anal.
Calcd for C15H26O11: C, 47.12; H, 6.85. Found: C,
46.88; H, 7.12.
lized from benzene (70mL) to yield 14 (14.29g, 46%),
mp 175–179ꢀC, Rf 0.4 (C); lit.18 mp 179ꢀC.
For checking the course of the reaction by GLC, sam-
ples (0.1mL) were withdrawn at different intervals and
acetylated with pyridine (0.5mL)–Ac2O (0.3mL) before
GLC investigation. The retention time of the three
components was the following: 13 (14.77min); 14
(13.77min); 16 (12.32min).
6.6. 3-O-(b-D-Glucopyranosyl)-D-glucose (12)
A soln of the O-isopropylidene derivative 11 (1.2g,
3.1mmol) in 0.5N sulfuric acid (10mL) was heated at
60ꢀC for 90min. The cooled soln was neutralized with
an ion exchange resin, filtered and freeze–dried to give
amorphous 12 (1g, 93%): [a]D +25 ! +20 (24h, c 1,
water); Rf 0.4 (E); lit.3 [a]D +19 (c 0.6, water).
6.8. 1-O-Benzoyl-2,4-O-benzylidene-5,6-O-isopropylid-
ene-D-glucitol (15)
To a stirred soln of 14 (3,1g, 10mmol) in pyridine
(20mL), benzoyl chloride (1.3mL, 11mmol) was added
dropwise at ꢀ20ꢀC. The mixture was stirred at this tem-
perature for 15min and subsequently at rt for 30min to
give after usual processing a syrup (4.2g, ꢁ100%), which
after purification by column chromatography (B) af-
forded 15 (2.9g, 70%); Rf 0.6, [a]D ꢀ24 (c 1, CHCl3).
Anal. Calcd for C23H26O7: C, 66.65; H, 6.32. Found:
C, 66.78; H, 6.55.
6.7. 2,4-O-Benzylidene-5,6-O-isopropylidene-D-glucitol
(14)
To a stirred slurry of 2,4-O-benzylidene-D-glucitol (13,
27g, 0.1mol) in DMF (75mL), 2,2-dimethoxypropane
(22.5mL, 0.3mol) and p-TsOH (200mg) were added at
20ꢀC. Stirring was continued until a clear solution was
obtained (ꢁ50min), and after further 30min the reac-
tion was quenched with Et3N (1.5mL). The residue of
the concentrated mixture was boiled with CHCl3
(200mL) for 30min. The undissolved 13 (ꢁ2g) was fil-
tered off without cooling and was washed with CHCl3
(20mL). The combined filtrate was washed with water,
dried and concentrated. The solid residue was recrystal-
6.9. Glycosidation reactions of 15
(a) To a stirred soln of 15 (2.5g, 6mmol) in dry MeCN
˚
(20mL) molecular sieves (4A) (5g) were added. After
30min, acetobromoglucose 4 (2.5g, 6mmol) and
Hg(CN)2 (1.6g, 6.5mmol) were added and the stirring

2414
J. Kuszmann et al. / Carbohydrate Research 339 (2004) 2407–2414
was continued at rt for 20h. The filtered mixture was
diluted with threefold CHCl3, washed with 5% NaHCO3
soln and an 10% aq soln of KBr, dried and concen-
trated. The residue was submitted to column chroma-
tography (A). The fractions having Rf 0.5–0.7 were
concentrated and the residue separated by a repeated
column chromatography. Concentration of the fractions
having Rf 0.65 gave 1-O-benzoyl-2,4-O-benzylidene-5,6-
O-isopropylidene-3-O-(2,3,4,6-tetra-O-acetyl-a-D-gluco-
pyranosyl)-D-glucitol (17) (125mg, 7%): mp 177–179ꢀC;
[a]D +37 (c 1, CHCl3). Anal. Calcd for C37H44O16: C,
59.67; H, 5.96. Found: C, 59.55; H, 6.09.
dried and concentrated. The residue was submitted to
column chromatography (A). The residue obtained on
concentration of the fractions having Rf 0.55 was recrys-
tallized from fivefold EtOH to give 18 (19.5g, 35%) iden-
tical to that, described above.
Acknowledgements
The authors are very much indebted to Dr. Ferenc Sze-
´
derkenyi and Mr. Attila Gyepes for the GC results.
Concentration of the fractions having Rf 0.55 gave
1-O-benzoyl-2,4-O-benzylidene-5,6-O-isopropylidene-3-
O-(2,3,4,6-tetra-O-acetyl-b-D-glucopyranosyl)-D-glucitol
(18) (900mg, 20%); mp 170–171ꢀC (after two recrystal-
lizations from EtOH), [a]D ꢀ4 (c 1, CHCl3). Anal. Calcd
for C37H44O16: C, 59.67; H, 5.96. Found: C, 59.63; H,
6.00.
References
1. Kuszmann, J.; Medgyes, G.; Boros, S. Carbohydr. Res.
2004, 339, 1569–1579.
2. Kuszmann, J.; Medgyes, G.; Boros, S. Pharmazie 2004, 59,
344–348.
3. Whelan, W. J. In Methods in Carbohydrate Chemistry I;
Whistler, R. L., Wolfrom, M. L., Eds.; Academic: New
York, 1962; Vol. I, pp 330–333.
(b) To a stirred soln of 15 (4.3g, 10.4mmol) in dry
˚
CH2Cl2 (65mL), molecular sieves 4A (8g) and phenyl
4. Ba¨chli, P.; Percival, E. G. V. J. Chem. Soc. 1952,
1243–1246.
5. Freudenberg, K.; Ortzen, K. Liebigs Ann. Chem. 1952,
574, 37–53.
6. Matsuda, K. Chem. Ind. (London), 1958, 1627.
7. Kochetkov, N. K.; Khorlin, A. J.; Bochkov, A. F.
Tetrahedron 1967, 23, 693–707.
8. Kochetkov, N. K.; Bochkov, A. F.; Sokolovskaya, T. A.;
Snyatkova, V. J. Carbohydr. Res. 1971, 16, 17–21.
9. Kochetkov, N. K.; Bochkov, A. F. In Methods in
Carbohydrate Chemistry VI; Whistler, R. L., BeMiller, J.
N., Eds.; Academic: New York, 1972; Vol. VI, pp
480–486.
thioglycoside 5 (4.4g, 10mmol) were added. After
30min, the mixture was cooled to ꢀ30ꢀC, NIS (3.4g,
15mmol) and AgOTf (0.4g, 1.5mmol) were added,
and stirring was continued ꢀ30ꢀC for 90min. Thereafter
the reaction was quenched with Et3N (2mL), the tem-
perature was raised to 20ꢀC, the mixture was filtered,
diluted with CH2Cl2 (60mL), washed with 10% aq
Na2S2O3, 5% NaHCO3 and water, dried and concen-
trated. The residue was separated by column chroma-
tography (A). Concentration of the fractions having Rf
0.7 gave 3,4,6-tri-O-acetyl-1,2-(S)-O-[10-O-(1-O-ben-
zoyl-2,4-O-benzylidene-5,6-O-isopropylidene-D-glucitol-
3-yl)-ethylidene]-a-D-glucopyranose (19) (1.4g, 19%),
[a]D +14.5 (c 1, CHCl3). Anal. Calcd for C37H44O16:
C, 59.67; H, 5.96. Found: C, 59.42; H, 6.17.
10. Koeppen, B. H. Carbohydr. Res. 1972, 24, 154–158.
11. Khorlin, A. J.; Nesmeyanov, V. A.; Zurabyan, S. E.
Carbohydr. Res. 1975, 43, 69–77.
12. Excoffier, G.; Gagnaire, D. Y.; Vignon, M. R. Carbohydr.
Res. 1976, 51, 280–286.
13. Thiem, J.; Karl, H. Chem. Ber. 1979, 112, 1046–1056.
14. Wozney, Y. V.; Backinowsky, L. V.; Kochetkov, N. K.
Carbohydr. Res. 1979, 73, 282–286.
Concentration of the fractions having 0.55 gave 18
(2.2g, 29%) identical with that described above.
(c) A soln of 15 (31g, 75mmol) in benzene (125mL)
and MeNO2 (125mL) was concentrated to a vol of
´
´ ´
´
15. Liptak, A.; Nanasi, P.; Neszmelyi, A.; Wagner, H.
Carbohydr. Res. 1980, 86, 133–136.
˚
200mL. Molecular sieves (4A) (24g) were added, the
16. Garegg, P. J.; Johansson, R.; Samuelsson, B. Acta Chem.
Scand. B 1982, 36, 249–250.
mixture was stirred for 30min at rt and thereafter 4
(34g, 83mmol) and Hg(CN)2 (21g) were added. Stirring
was continued at 20ꢀC for 15h, and at 45ꢀC for further
3h. The filtered soln was concentrated, the residue dis-
solved in CHCl3, washed with 10% KBr soln and water,
17. Kondrasson, P.; Mootoo, D. R.; McDevitt, R. E.; Fraser-
Reid, B. J. Chem. Soc., Chem. Commun. 1990, 270–272.
18. Vargha, L. Chem. Ber. 1935, 68, 18–24, and 1377–1384.
´
19. Tsvetkov, Y. E.; Byramova, N. E.; Backinowsky, L. V.
Carbohydr. Res. 1983, 115, 254–258.