An easy access to halide ion-catalytic alpha-glycosylation using carbon tetrabromide and triphenylphosphine as multifunctional reagents.

Article abstract of DOI:10.1039/b303984f

The reaction of a 2-O-benzyl-1-hydroxy sugar with CBr4 and Ph3P generates a glycosyl bromide in situ, which is coupled with an acceptor alcohol in the presence of N,N-tetramethylurea to afford an alpha-glycosyl product virtually quantitatively. In a proposed pathway, the reagent combination plays multiple roles such as the generation of a glycosyl donor, the activation of glycosylation, and the dehydration of the reaction system. These roles allow a simple alpha-glycosylation to be performed without special attention to dehydration. Various alpha-glycosyl (D-gluco-, D-galacto- and L-fuco-) products including glycosyl glycerols and cholesterols have been prepared with this method.

Full text of DOI:10.1039/b303984f

View Article Online / Journal Homepage / Table of Contents for this issue
An easy access to halide ion-catalytic ꢀ-glycosylation using carbon
tetrabromide and triphenylphosphine as multifunctional reagents
Yuko Shingu, Yoshihiro Nishida,* Hirofumi Dohi and Kazukiyo Kobayashi*
Department of Molecular Design and Engineering, Graduate School of Engineering,
Nagoya University, Furou-chou, Chikusa-ku, Nagoya, 464-8603, Japan.
E-mail: kobayash@mol.nagoya-u.ac.jp; Fax: ϩ81-52-789-2528
Received 14th April 2003, Accepted 3rd June 2003
First published as an Advance Article on the web 18th June 2003
The reaction of a 2-O-benzyl-1-hydroxy sugar with CBr4 and Ph3P generates a glycosyl bromide in situ, which is
coupled with an acceptor alcohol in the presence of N,N-tetramethylurea to afford an α-glycosyl product virtually
quantitatively. In a proposed pathway, the reagent combination plays multiple roles such as the generation of a
glycosyl donor, the activation of glycosylation, and the dehydration of the reaction system. These roles allow a simple
α-glycosylation to be performed without special attention to dehydration. Various α-glycosyl (-gluco-, -galacto-
and -fuco-) products including glycosyl glycerols and cholesterols have been prepared with this method.
reagent combination of CBr4 and Ph3P, called Appel agents9 in
this paper, is applied effectively.
Introduction
The development of practical 1,2-cis-glycosylation methods is a
challenging and meaningful objective in bio-organic chemistry.1
Results and discussion
This is because many human oligosaccharide antigens carry the
corresponding α-glycosyl linkage which constructs a key
determinant structure like α--fucopyranoside of sialyl LewisX
(sLeX)2 and α--galactopyranoside of globotriaosyl Pk anti-
gens.3 These glycosyl linkages are expected to possess a high
potential applicable to carbohydrate-based medicinal agents.
Among representative α-glycosylation methods now available,4
the halide ion-catalytic method developed by Lemieux et al.5 in
the 1970s provides one of the most definitive pathways; the
obtained α-selectivity depends scarcely on the reactivity of
acceptors, donors, and other reaction conditions. Moreover, the
Lemieux method possesses an advantage from an environ-
mental point of view since the glycosylation is performed under
mild conditions without heavy metal promoters and strong
Lewis acid catalysts.
Also in our stereospecific syntheses of 3-O-[6-O-phos-
phorylcholine-α--glucopyranosyl]-1,2-diacyl-sn-glycerides
(GGPL-I and GGPL-III, Scheme 1) as major cell surface
glycolipids of Mycoplasma fermentans,6 the halide ion-catalytic
method was applied effectively.7 In practice, however, a
synthetic problem, associated with the high instability of
2-O-benzylated glycosyl bromides used as α-glycosyl donors,
has suspended their large-scale syntheses for biochemical
studies. Therefore, we have investigated alternative access to
the α-glycosylation pathway.8 In this article, we propose
an advanced halide ion-catalytic α-glycosylation, in which the
In carbohydrate chemistry, the Appel and analogous phosphine
reagents have been applied to the halogeno-dehydroxylation
of anomeric10,11 and primary12 OH groups. The glycosyl
halides prepared in the former have been incorporated within
various glycosylation methods. Recently, Khatuntseva et al.11
reported that the reaction of the Appel agents and 2-O-
benzyl-1-hydroxy--fucose derivatives gave the corresponding
1-bromides (α- or β) quantitatively, which were applied without
isolation to α-glycosylation in the presence of Hg(CN)2 and
HgBr2. Previously,13 we reported that a 2-O-benzyl-1-hydroxy-
-glucose derivative 1a, treated with the Appel agents, gave
an α-glycosyl bromide 2a exclusively (Scheme 1). The glycosyl
donor 2a was used directly in a halide ion-catalytic α-glyco-
sylation with an acceptor alcohol 3a in the presence of tetra-
ethylammonium bromide, N,N-tetramethylurea (TMU),7 and
molecular sieves. The one-pot glycosylation quantitatively gave
a mixture of two α-glycosylated products 4a-1 and 4a-2 (ca. 7 :
3, the ratio was dependent on the reaction time) both of which
are useful for the synthesis of GGPLs and other α-glycosyl-sn-
glycerolipids. The one-pot reactions reported by Khatuntseva
et al. and us have proved that the presence of Appel agents and
possible side products does not affect the α-glycosylation
reaction substantially (Scheme 2 and Table 1, entries 1 and 2).
During additional attempts to optimise the one-pot reaction,
we have found two notable facts. The α-glycosylation of 3a
proceeds without Et4NϩBrϪ, though the quaternary ammonium
salt is essential when using the original method. Furthermore,
the reaction, causing little decomposition of the donor 2a, is
very clean even in the absence of molecular sieves (MS) and
drying gas. Thus, the observed one-pot α-glycosylation using
the Appel agents is apparently different from the authentic one
and is expected to provide a simpler α-glycosylation method.
To examine the generality of the observed phenomenon,
other primary and secondary acceptor alcohols 3b–3e were sub-
jected to the same reaction in the absence of Et4NϩBrϪ and MS
(Scheme 2 and Table 1, Entry 3–7). In every case, the corre-
sponding 1,2-cis-glycosides 4b–4e were derived in high yields
and with α-selectivity, although the reactions of secondary
hydroxyl groups in 3d and 3e required prolonged time for com-
pletion. Neither a decomposed product 1a nor a self-condensed
product could be observed in the reaction mixture even after the
Scheme 1
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 5 1 8 – 2 5 2 1
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
2518

View Article Online
Table 1 One-pot glycosylationa of various acceptor alcohols 3a–3e with a donor precursor 1a
Entry
Acceptors
Conditions
Time/hc
Products
Yield (%)d
Configuration α : β
1
2
3
4
5
6
7
3a
3d
3a
3b
3c
3d
3e
Et4NBr ϩ TMUb
Et4NBr ϩ TMUb
TMU
TMU
TMU
15
53
15
28
30
84
96
4a-1 ϩ 4a-2
98e
92
>99 : 1g
98 : 2
4d
4a-1 ϩ 4a-2
98f
96
>99 : 1g
>99 : 1g
95 : 5
4bh
4c
92
95
92
TMU
TMU
4d
4e
>99 : 1g
>99 : 1g
a All reactions were carried out at room temperature with 1a (200 mg, 0.41 mmol) in CH2Cl2 (3 mL), 3 mol equiv. of CBr4, Ph3P, and acceptor
alcohol, and N,N-tetramethylurea (TMU, 300 µL). b Et4NBr (1 mol equiv.) was added with TMU. c Reaction time/h until the glycosyl bromide donor
was completely consumed (TLC analysis) after the addition of acceptor alcohol and TMU. d Isolated yields (%) based on the amount of 1a used for
the reaction. e Total yields of 4a-1 (63%) and 4a-2 (35%). f 4a-1 (65%) and 4a-2 (33%). g No β-isomer was detected in 1H-NMR spectra of isolated
products (500 MHz, CDCl3). h The reaction gave a mixture of 1-O- and 3-O-α--glucosyl-sn-glycerols (6 : 5) due to epimerization (ref. 13).
Scheme 2
Table 2 One-pot glycosylationa of cholesterol 3d with various donor
precursor 1a–1e
These experimental data indicated that the present one-pot
α-glycosyation is applicable to a wide-range of glycosyl donors
and acceptors. Judging from the high α-selectivity, it is obvious
Donors
Time/h
Yield (%)b
Configuration (α : β)
that the α-glycosylation conforms to the pathway of a halide
ion-catalytic reaction (Scheme 4). That is, α-glycosyl bromide
1a
1b
1c
1d
1e
84
58
72
72
24
95
95
82
86
95
>99 : 1
90 : 10
>99 : 1
88 : 12
86 : 14
1
R1–Br(ꢀ), initially produced (as determined by H-NMR spec-
trum, cited in the Experimental section), is located in the
equilibrium with β-glycosyl bromide R1–Br(ꢁ) in the presence
of nucleophilic bromide anions. Then, the highly reactive
β-glycosyl bromide permits α-glycosylation in an S2 fashion.
Our reaction, however, employs the Appel agents with multiple
functionalities. First, the reagent converts a 2-O-benzyl-1-
hydroxy sugar R1–OH to an α-glycosyl bromide R1–Br(ꢀ).
The reaction proceeds in the same manner as the dehydroxy-
bromination of primary alcohols. Second, a nucleophilic
bromide (BrϪ) in the form of triphenylphosphonium salt
(Ph3PϩCBr3BrϪ) and HBr–TMU may attain the equilibration
of the α- and β-bromides required for the halide ion-catalytic
reaction. The existence of such species is suggested from the
reaction of 2a which permitted opening at the epoxide group by
a bromide anion. Then, the unstable β-isomer reacts with an
acceptor R2–OH to construct an α-glycoside linkage R1–O–R2
more readily than for the α-isomer. Third, the Appel agents
work also as dehydrating agents to ensure an anhydrous system.
a All reactions were carried out at room temperature with 1a–1b
(200 mg, 0.41 mmol) in CH2Cl2 (3 mL) and 3 mol equiv. of CBr4, Ph3P,
and cholesterol 3d, and N,N-tetramethylurea (TMU, 300 µL). b Isolated
yields (%) based on the amount of each donor precursor used for the
reaction.
Scheme 3
prolonged reactions of 3d and 3e. Though 1,2-O-isopropyl-
idene-sn-glycerol 3b underwent a partial epimerization, the
α-glycosylation per se was nearly perfect. An azido compound
3c was also useful in spite of the use of a reductive Ph3P in
excess amounts for the donor.
Reactions of other donors 1b–1e with -gluco-, -galacto-
and -fucose configurations were also examined with choles-
terol 3d (Scheme 3 and Table 2). All of them showed
α-glycosylation in reasonable yields (>80%). There, the donors
1a and 1c with 6-O-Ac showed higher α-selectivity than 1b and
1d with 6-O-Bn and 1e with 6-Me. A similar phenomenon has
been observed for 4-O-,11,14 and 6-O-acyl15 donors in variable
α-glycosylation reactions and may be explained in terms of a
long-range participation effect. On the other hand, the reactiv-
ity increased in the reverse order of 6-O-Ac < 6-O-Bn < 6-Me.
Thus, an -fucosyl donor 1e with 6-Me showed the highest
activity with a slight decrease in the α-selectivity.
Scheme 4 Overview of one-pot α-glycosylation with Appel agents.
When water molecules contaminate the reaction, the labile
donor R1–Br(ꢁ) may be decomposed to R1–OH in competition
with the α-glycosylation. In this case, the efficiency of the glyco-
sylation is lowered depending on the extent of water contamin-
ation and donor decomposition. In our pathway, the
decomposed R1–OH seems reusable as the donor since the
Appel agents are added in excess amounts. In a preceding
study,13 we found that 3 mol equiv. of CBr4 and Ph3P for a
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 5 1 8 – 2 5 2 1
2519

View Article Online
donor precursor are required to complete the 1-bromination
reaction at room temperature, though smaller amounts were
reported11 for similar reactions at elevated temperature. The
amount of 3 molar equivalents was also sufficient for the com-
pletion of α-glycosylation without MS and special attention to
the contamination of water molecules. It is highly probable that
the Appel agents can capture water molecules more effectively
4.69 and 4.57 (dd, 2H × 3, –CH2C6H5), 4.29 (dd, 1H, J = 3.5 and
12.0 Hz, H-6S), 4.28 (dd, 1H, J = 2.5 and 12.0 Hz, H-6R), 4.12
(m, 1H, H-5), 4.06 (dd, 1H, J = 9.0 and 9.5 Hz, H-3), 3.58 (dd,
1H, J = 9.0 and 10.0 Hz, H-4), 3.52 (dd, 1H, J = 4.0 and 9.5 Hz,
H-2), 2.01 (s, 3H, –Ac); FAB-MS: m/z 577 [M ϩ Naϩ].
(S )-3-O-(6-O-Acetyl-2,3,4-tri-O-benzyl-ꢀ-D-glucopyrano-
than what we expect for the conversion of Ph P to Ph P᎐O.
syl)glycidol (4a-1). IR (KBr film) ν/cmϪ1; 3029, 2873, 1741,
᎐
3
3
1
CBr4 may also be associated with the dehydration to finally
afford carbon dioxide. Although a precise mechanism for
the dehydration as well as the glycosylation pathways is still
being investigated, the present reaction using the Appel agents
provides a very clean and simple α-glycosylation protocol.
1454, 1363, 1238, 1072, 1031, 750, 700. H-NMR (500 MHz,
CDCl3); δH 7.40–7.23 (m, 5H × 3, –CH2C6H5), 4.55–5.02 (dd,
2H × 3, CH2C6H5), 4.87 (d, 1H, J = 4.0 Hz, H-1), 4.26 (dd, 1H,
J = 4.0 and 12.0 Hz, H-6proR), 4.22 (dd, 1H, J = 2.5 and 12.0 Hz,
H-6proS), 4.02 (dd, 1H, J = 9.0 and 9.5 Hz, H-3), 3.88 (m, 1H,
H-5), 3.76 (dd, 1H, J = 3.5 and 12.0 Hz, glycidol H-3proR), 3.48
(dd, 1H, J = 6.0 and 12.0 Hz, glycidol H-3proS), 3.53 (dd, 1H
J = 3.5 and 9.5 Hz, H-2), 3.48 (dd, 1H, J = 9.0 and 9.5 Hz, H-4),
3.20 (m, 1H, glycidol H-2), 2.57 and 2.78 (dd, 1H × 2, J = 4.0
and J = 5.0, J = 3.0 and 5.0 Hz, glycidol H-1proR or H-1proS), 1.99
(s, 3H, –Ac). HRMS (FAB); m/z calcd for C32H37O8BrNa
[M ϩ Naϩ] 571.2308; found 571.2285.
Conclusion
We have demonstrated one-pot halide ion-catalytic α-glyco-
sylation by using the reagent combination of CBr4 and Ph3P.
The reagent is responsible for the generation, equilibration, and
regeneration of glycosyl donors as well as the dehydration of
the reaction system. The multiple roles enable us to conduct a
simple α-glycosylation conductible at room temperature
without paying special attention to the decomposition of the
donors by water molecules. Moreover, the glycosylation, con-
forming to the halide ion-catalytic pathway, shows a near per-
fect α-selectivity. Mechanistic studies and further applications
of the methodology are in progress and will appear elsewhere.
3-O-(6-O-Acetyl-2,3,4-tri-O-benzyl-ꢀ-D-glucopyranosyl)-1-
bromo-1-deoxy-sn-glycerol (4a-2). IR (KBr film) ν/cmϪ1; 3488,
3062, 3029, 2915, 1741, 1496, 1454, 1363, 1330, 1238, 1159,
1
1072, 1031, 912, 852, 740, 700. H-NMR (500 MHz, CDCl3);
δH 7.40–7.23 (m, 5H × 3, –CH2C6H5), 4.55–5.00 (dd, 2H × 3,
–CH2C6H5), 4.75 (d, 1H, J = 3.5 Hz, H-1), 4.25 (m, 2H, H-6),
4.02 (m, 1H, glycerol H-2), 3.97 (t, 1H, J = 9.5 and 9.5 Hz, H-3),
3.86 (m, 1H, H-5), 3.80 (dd, 1H, J = 4.0 and 10.5 Hz, glycerol
H-3proR), 3.55 (dd, 1H J = 3.5 and 9.5 Hz, H-2), 3.52 (dd, 1H,
J = 5.5 and 10.5 Hz, glycerol H-3proS), 3.52 and 2.91 (m and d,
1H × 2, J = 5.5 Hz (d), glycerol H-1proR or H-1proS), 3.47 (dd,
1H, J = 9.5 and 10.0 Hz, H-4), 2.17 (s, 3H, –Ac). 13C-NMR
(500MHz, CDCl3); δC 170.4, 165.4, 138.3. HRMS (FAB): m/z
calcd for C32H37O8BrNa [M ϩ Naϩ] 653.1555; found 653.1545.
Experimental
General methods
Infrared (IR) spectra were recorded on a JASCO FT/IR-230
Fourier transform infrared spectrometer in the form of KBr
1
disks. All H NMR (500 MHz) spectra were recorded using
Varian INOVA-500 or Varian Gemini 200 instruments at ambi-
1
ent temperature. H chemical shifts are expressed in parts per
3-O-(6-O-Acetyl-2,3,4-tri-O-benzyl-ꢀ-D-glucopyranosyl)-1,2-
O-isopropylidene-sn-glycerol (4b) and sn-2,3 isomer. IR (KBr
film) ν/cmϪ1 3031, 2985, 2929, 2877, 1741, 1496, 1454, 1369,
1330, 1238, 1157, 1072, 1031, 921, 840, 738, 700; 1H-NMR (500
MHz, CDCl3): δH 7.40–7.23 (m, 5H × 3, –CH2C6H5), 4.56–
4.99 (dd, 2H × 3, –CH2C6H5), 4.83 (d, 1H, J = 3.5 Hz, H-1),
4.35 (t, 1H, J = 5.5 and 6.5 Hz, glycerol H-2), 4.20–4.28 (dd, 1H
× 2 H-6), 3.98 (dd, 1H, J = 9.0 and 9.5 Hz, H-3), 3.85 (m, 1H,
H-5), 4.07 and 3.74 (dd, 1H × 2, J = 8.5 and 6.5, J = 6.0 and 8.0
Hz, glycerol H-3proR or H-3proS), 3.60 and 3.55 (dd, 2H, J = 6.0
and 10.5, J = 6.5 and 10.5 Hz, glycerol H-1proR or H-1proS), 3.54
(dd, 1H, J = 3.5 and 9.5 Hz, H-2), 3.47 (dd, 1H, J = 9.0 and 10.0
Hz, H-4), 2.02 (s, 3H, –Ac), 1.42 and 1.36 (s, 3H × 2, isopropyl).
An sn-2,3-glycerol isomer: δH 7.40–7.23 (m, 15H, –CH2C6H5),
4.56–4.99 (dd, 2H × 3, –CH2C6H5), 4.74 (d, 1H, J = 3.5 Hz,
H-1), 4.32 (t, 1H, J = 5.5 and 6.5 Hz, glycerol H-2), 4.20–4.28
(dd, 1H × 2, H-6), 4.00 (dd, 1H, J = 9.0 and 9.5 Hz, H-3), 3.88
(m, 1H, H-5), 4.07 and 3.78 (dd, 1H × 2, J = 6.5 and 8.5, J = 5.5
and J = 8.0 Hz, glycerol H-3proR or H-3proS), 3.69 and 3.42 (dd,
1H × 2, J = 6.0 and 10.5, J = 6.5 and 10.5 Hz, glycerol H-1proR or
H-1proS), 3.54 (dd, 1H, J = 3.5 and 9.5 Hz, H-2), 3.48 (dd, 1H,
J = 9.0 and 10.0 Hz, H-4), 2.02 (s, 3H, –Ac), 1.41 and 1.35 (s, 3H
× 2, isopropyl); HRMS (FAB): m/z calcd for C35H42O9Na
[M ϩ Naϩ] 629.2727; found 629.2704.
million (ppm) downfield of tetramethylsilane (TMS). Mass
spectra were recorded by a JEOL JMS 700 spectrometer for fast
atom bombardment (FAB) spectra. For thin layer chromato-
graphy (TLC) analysis, Merck pre-coated TLC plates (silica gel
60 F254, layer thickness 0.25 mm) and Merck TLC aluminium
roles (silica gel 60 F254, layer thickness 0.2 mm) were used.
Silica gel column chromatography was performed on silica gel
60 (Merck 0.063–0.200 mm and 0.040–0.063 mm). All other
chemicals for the synthesis of target compounds were pur-
chased from Tokyo Kasei Kogyo Co., Ltd., Kishida Chemical
Co., Ltd., and Sigma-Aldrich Chemical Company, and were
used without further purification. 2-O-benzyl-1-hydoxy sugars
1a–1b were prepared from the corresponding reducing sugars
according to methods in the literature.11,13,16
A general protocol for one-pot ꢀ-glycosylation. Reactions were
carried out in a glass vessel closed with a septum cap. Neither
molecular sieves nor drying gas were used. A 2-O-benzyl-1-
hydroxy sugar (1a–1b, 200 mg, 0.41 mmol) in 3 mL of CH2Cl2
was treated with Ph3P (3 mol equiv.) and CBr4 (3 mol equiv.)
and stirred for 3 h at room temperature. Then, N,N-tetra-
methylurea (TMU, 300 µL) and acceptor alcohol 3a–3e (3 mole
equiv.) were added and stirred at room temperature. In all cases,
the reaction was continued until the bromide donor was con-
sumed completely as evidenced by TLC analysis. The reaction
mixture diluted with CHCl3 was washed with saturated aq.
NaHCO3 and aq. NaCl solution and dried (Na2SO4) and con-
centrated. The product was purified by silica gel column
chromatography.
6-Azidohexyl
6-O-Acetyl-2,3,4-tri-O-benzyl-ꢀ-D-gluco-
pyranoside (4c). IR (KBr film) ν/cmϪ1 3031, 2935, 2865, 2094,
1743, 1496, 1454, 1363, 1236, 1157, 1074, 1031, 910, 734, 698;
1H-NMR (500 MHz, CDCl3) δH 7.40–7.23 (m, 5H × 3,
–CH2C6H5), 4.55–5.02 (dd, 2H × 3, –CH2C6H5), 4.72 (d, 1H,
J = 3.5 Hz, H-1), 4.27 (dd, 1H, J = 4.0 and 12.0 Hz, H-6S), 4.23
(dd, 1H, J = 2.5 and 12.0 Hz, H-6R), 4.01 (dd, 1H, J = 9.0 and
9.5 Hz, H-3), 3.83 (m, 1H, H-5), 3.61 and 3.40 (dt, 1H × 2,
–OCH2CH2(CH2)2CH2CH2N3), 3.53 (dd, 1H, J = 3.5 and 9.5
6-O-Acetyl-2,3,4-tri-O-benzyl-ꢀ-D-glucopyranosyl
(2a). H-NMR (500 MHz. CDCl3) δH 7.40–7.23 (m, 5H × 3,
–CH2C6H5), 6.37 (d, 1H, J = 4.0 Hz, H-1) 5.01, 4.89, 4.83, 4.72,
bromide
1
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 5 1 8 – 2 5 2 1
2520

View Article Online
Hz, H-2), 3.48 (dd, 1H, J = 9.0 and 10.0 Hz, H-4), 3.24 (t, 2H,
J = 7.0 and 7.0 Hz, –OCH2CH2(CH2)2CH2CH2N3), 2.02 (s, 3H,
–Ac), 1.59 and 1.66 (br, 2H × 2, –OCH2CH2(CH2)2CH2CH2N3),
1.39 (br, 2H × 2, –OCH2CH2(CH2)2CH2CH2N3), 4.40 (d, 1H,
J = 8.0 Hz, H-1β); HRMS (FAB): m/z calcd for C35H43O7N3Na
[M ϩ Naϩ] 640.2999; found 640.2988.
O-[2,3,4-Tri-O-benzyl-ꢀ-L-fucopyranosyl]cholesterol (5e). IR
(KBr film) ν/cmϪ1 3062, 3031, 2937, 2933, 1724, 1457, 1454,
1369, 1101, 1037, 738, 700; 1H-NMR (500 MHz, CDCl3)
δH 7.40–7.23 (m, 5H × 3, –CH2C6H5), 5.34 (m, 1H, choles-
terol H-6), 4.64–4.99 (dd, 2H × 3, –CH2C6H5), 4.95 (d, 1H,
J = 4.0 Hz, H-1), 4.02 (dd, 1H, J = 3.5 and 10.0 Hz, H-2), 3.95
(m, 1H, H-5), 3.96 (m, 1H, H-5), 3.94 (dd, 1H, J = 3.0 and 10.0
Hz, H-3), 3.66 (br d, 1H, J = 2.0 Hz, H-4), 3.44 (m, 1H, choles-
terol H-3), 1.08 (d, 2H, J = 6.5 Hz, H-6). HRMS (FAB); m/z
calcd for C54H74O5Na [M ϩ Naϩ] 825.5434; found 825.5402.
O-(6-O-Acetyl-2,3,4-tri-O-benzyl-ꢀ-D-glucopyranosyl)-
cholesterol (4d). IR (KBr film) ν/cmϪ1; 3031, 2938, 2867,
1
1741, 1457, 1369, 1238, 1155, 1072, 1031, 742, 700. H-NMR
(500 MHz, CDCl3); δH 7.40–7.23 (m, 5H × 3, –CH2C6H5), 5.32
(m, 1H, cholesterol H-6), 4.54–5.03 (dd, 2H × 3, –CH2C6H5),
4.89 (d, 1H, J = 4.0 Hz, H-1), 4.25 (dd × 2, 2H, H-6S and R), 4.03
(dd, 1H, J = 9.0 and 9.5 Hz, H-3), 3.96 (m, 1H, H-5), 3.53 (dd,
1H, J = 4.0 and 9.5 Hz, H-2), 3.47 (dd, 1H, J = 9.0 and 10.0 Hz,
H-4), 3.43 (m, 1H, cholesterol H-3). HRMS (FAB): m/z calcd
for C56H76O7Na [MϩNaϩ] 883.5489; found 883.5425.
Acknowledgements
This study was supported in part by a grant of the 21st Century
COE program “Nature-Guided Materials Processing” of the
Ministry of Education, Culture, Sport, Science and Technology
of the Japanese Government. And the study was also funded by
the Sasagawa Scientific Research Grant from The Japan Science
Society.
3-O-(6-O-Acetyl-2,3,4-tri-O-benzyl-ꢀ-D-glucopyranosyl)-
1,2;5,6-di-O-isopropylidene-ꢀ-D-glucofuranose (4e). IR (KBr
film) ν/cmϪ1; 3031, 2987, 2933, 1741, 1496, 1454, 1373, 1309,
1236, 1162, 1074, 1029, 912, 848, 781, 736, 698. 1H-NMR (500
MHz, CDCl3); δH 7.40–7.23 (m, 5H × 3, –CH2C6H5), 5.83 (d,
1H, J = 3.5 Hz, H-1Ј), 5.21 (d, 1H, J = 3.5 Hz, H-1), 4.58–5.00
(dd, 2H × 3, –CH2C6H5), 4.75 (t, 1H, J = 4.0 and 4.5 Hz, H-2Ј),
4.23–4.26 (m, 2H × 2, H-6S and R and H-6ЈS and R), 4.14 (m, 1H,
H-5), 4.08 (dd, 1H, J = 8.5 and 9.0 Hz, H-3Ј), 4.08 (m, 1H,
H-5Ј), 4.00 (dd, 1H, J = 6.0 and 8.5 Hz, H-4Ј), 3.98 (t, 1H,
J = 9.0 and 9.0 Hz, H-3), 3.58 (dd, 1H J = 4.0 and 9.5 Hz, H-2),
3.50 (dd, 1H, J = 9.0 and 10.0 Hz, H-4), 2.00 (s, 3H, –Ac), 1.32,
1.36, 1.45 and 1.56 (s, 3H × 4, isopropyl). HRMS (FAB); m/z
calcd for C41H50O12Na [M ϩ Naϩ] 757.3200; found 757.3192.
References
1 H. Paulsen, Angew. Chem., Int.Ed. Engl., 1982, 21, 155–173;
H. Kunz, Angew. Chem.,Int.Ed. Engl., 1987, 26, 294–308;
K. Toshima and K. Tatsuta, Chem. Rev., 1993, 93, 1503–1531.
2 E. E. Simanek, G. J. McGarvey, J. A. Jablonwski and C.-H. Wong,
Chem. Rev., 1998, 98, 833–862; F. M. Unger, Adv. Carbohydr. Chem.
Biochem., 2001, 57, 210–435.
3 Y. Nishida, H. Dohi, H. Uzawa and K. Kobayashi, Tetrahedron
Lett., 1998, 39, 8681–8684; K. Matsuoka, M. Terabatake, Y. Esumi,
D. Terumuma and H. Kuzuhara, Tetrahedron Lett., 1999, 40, 7839–
7843; P. I. Kitov, J. M. Sadowska, G. Mulvey, G. D. Armstrong,
H. Ling, N. S. Pauun, R. J. Read and D. R. Bundle, Nature, 2000,
403, 669–671.
4 R. R. Schmidt, Adv. Carbohydr. Chem. Biochem., 1994, 50, 21–123;
P. J. Garegg, Adv. Carbohydr. Chem. Biochem., 1997, 52, 179–205.
5 R. U. Lemieux, K. B. Hendriks, R. V. Stick and K. James, J. Am.
Chem. Soc., 1975, 97, 4056–4062.
6 K. Matsuda, T. Kasama, I. Ishizuka, S. Handa, N. Yamamoto and
T. Taki, J. Biol. Chem., 1994, 269, 33123–33128; K. Matsuda,
I. Ishizuka, T. Kasama, S. Handa and N. Yamamoto, Biochim.
Biophys. Acta, 1997, 1349, 1–12.
7 Y. Nishida, H. Ohrui, H. Meguro, M. Ishizawa, K. Matsuda,
T. Taki, S. Handa and N. Yamamoto, Tetrahedron Lett., 1994, 35,
5465–5468; Y. Nishida, Y. Takamori, H. Ohrui, I. Ishizuka,
K. Matsuda and K. Kobayashi, Tetrahedron Lett., 1999, 40, 2371–
2374.
O-[2,3,4,6-Tetra-O-benzyl-ꢀ-D-glucopyranosyl]cholesterol
(5b). IR (KBr film) ν/cmϪ1 3029, 2935, 2933, 1718, 1454, 1361,
1101, 1060, 740, 696; 1H-NMR (500 MHz, CDCl3) δH 7.40–7.23
(m, 5H × 4, –CH2C6H5), 5.28 (m, 1H, cholesterol H-6), 4.44–
5.01 (dd, 2H × 4, –CH2C6H5), 4.93 (d, 1H, J = 3.5 Hz, H-1),
4.00 (dd, 1H, J = 9.0 and 9.5 Hz, H-3), 3.87 (m, 1H, H-5), 3.74
and 3.64 (dd × 2, 2H, H-6S and R), 3.64 (dd, 1H, J = 9.0 and 10.0
Hz, H-4), 3.35 (dd, 1H, J = 3.5 and 9.5 Hz, H-2), 3.46 (m, 1H,
cholesterol H-3); HRMS (FAB): m/z calcd for C61H80O6Na [M
ϩ Naϩ] 931.5853; found 931.5820.
8 H. Dohi, Y. Nishida, H. Tanaka and K. Kobayashi, Synlett, 2001,
1446–1448.
O-[6-O-Acetyl-2,3,4-tri-O-benzyl-ꢀ-D-galactopyranosyl]-
cholesterol (5c). IR (KBr film) ν/cmϪ1; 3029, 2938, 1743, 1457,
9 R. Appel, Angew. Chem., Int. Ed. Engl., 1975, 14, 801–811; J. P. H.
Verheyden and J. G. Moffatt, J. Org. Chem., 1972, 37, 2289–2299.
10 M. Petitou, P. Duchaussoy, I. Lederman, J. Choay, P. Sinay,
J.-C. Jacquinet and G. Torri, Carbohydr. Res., 1986, 147, 221–236;
H. Paulsen and K. Adermann, Carbohydr. Res., 1988, 175, 163–172;
Y. Itoh and T. Ogawa, Tetrahedron, 1990, 46, 89–102.
11 E. A. Khatuntseva, N. E. Ustuzhanina, G. V. Zatonskii, A. S.
Shashkov, A. I. Usov and N. E. Nifant’ev, J. Carbohydr. Chem.,
2000, 19, 1151–1173.
1
1369, 1236, 1132, 1099, 1035, 736, 698. H-NMR (500 MHz,
CDCl3); δH 7.40–7.23 (m, 5H × 3, –CH2C6H5), 5.32 (m, 1H,
cholesterol H-6), 4.60–4.98 (dd, 2H × 3, –CH2C6H5), 4.98 (d,
1H, J = 4.0 Hz, H-1), 4.10 and 4.04 (dd × 2, 2H, H-6S and R), 4.03
(dd, 1H, J = 3.5 and 10.0 Hz, H-2), 4.02 (m, 1H, H-5), 3.95 (dd,
1H, J = 2.5 and 10.0 Hz, H-3), 3.88 (br d, 1H, J = 2.5, H-4), 3.43
(m, 1H, cholesterol H-3). HRMS (FAB); m/z calcd for
C56H76O7Na [M ϩ Naϩ] 883.5489; found 883.5427.
12 M. A. Alim, L. Hough and A. C. Richardson, Carbohydr. Res., 1992,
216, 271–287.
13 Y. Shingu, Y. Nishida, H. Dohi, K. Matsuda and K. Kobayashi,
J. Carbohydr. Chem., 2002, 21, 601–607.
14 E. Dejeter-Juszynski and H. M. Flowers, Carbohydr. Res., 1971, 18,
219–226; R. Windmueller and R. R. Schmidt, Tetrahedron Lett.,
1994, 35, 7927–7930; S. J. Danishefski, J. Gervay, J. M. Peterson,
F. M. Mcdonald, K. Koseki, D. A. Griffith, T. Oriyama and
S. P. Marsden, J. Am. Chem. Soc., 1995, 117, 1940.
O-[2,3,4,6-Tetra-O-benzyl-ꢀ-D-galactopyranosyl]cholesterol
(5d). IR (KBr film) ν/cmϪ1; 3029, 2937, 2933, 1720, 1457, 1454,
1367, 1101, 1033, 736, 698. 1H-NMR (500 MHz, CDCl3);
δH 7.40–7.23 (m, 5H × 4, –CH2C6H5), 5.25 (m, 1H, choles-
terol H-6), 4.39–4.96 (dd, 2H × 4, –CH2C6H5), 4.98 (d, 1H,
J = 3.5 Hz, H-1), 4.05 (br t, 1H, J = 6.5 and 6.5 Hz, H-5), 4.01
(dd, 1H, J = 4.0 and 10.0 Hz, H-2), 3.98 (m, 1H, H-4), 3.95
(dd, 1H, J = 3.0 and 10.0 Hz, H-3), 3.53 (dd × 2, 2H, H-6S and R),
3.46 (m, 1H, cholesterol H-3). HRMS (FAB); m/z calcd for
C61H80O6Na [M ϩ Naϩ] 931.5853; found 931.5798.
15 K. Fukase, T. Yoshimura, S. Kotani and S. Kusumoto, Bull. Chem.
Soc. Jpn., 1994, 67, 473–482; Y. Nishida, Y. Takamori, K. Matsuda,
H. Ohrui, T. Yamada and K. Kobayashi, J. Carbohydr. Chem., 1999,
18, 65–72.
16 K. Toshima, Carbohydr. Res., 2000, 327, 15–26.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 5 1 8 – 2 5 2 1
2521