Fast galloylation of a sugar moiety: Preparation of three monogalloylsucroses as references for antioxidant activity. A method for the selective deprotection of tert-butyldiphenylsilyl ethers

Article abstract of DOI:10.1016/S0040-4020(00)00593-7

Three protected new gallotannins, namely the 6'-O-(tri-O-methylgalloyl)-2,3,4,6,1',3',4'-hepta-O-acetylsucrose, the 6'-O-(tri-O-methylgalloyl)-2,3,4,6,1',3',4'-hepta-O-benzoylsucrose and the 6,6'-di-O-tert-butyldiphenylsilyl-1'-O-(tri-O-methylgalloyl)-2,3, 4,3',4'-penta-O-acetylsucrose have been prepared in four short sequences from sucrose. Methods for rapid galloylation have been studied in order to avoid simultaneous acyl transfer reactions. A method for the deprotection of a tert-butyldiphenylsilyl ether has been developed which avoids the intramolecular migration of a benzoate group. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4020(00)00593-7

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 6511±6516
Fast Galloylation of a Sugar Moiety: Preparation of Three
Monogalloylsucroses as References for Antioxidant Activity.
A Method for the Selective Deprotection
of tert-Butyldiphenylsilyl Ethers
c
b
c
M. Teresa Barros,a,b, Christopher D. Maycock, Fernando SinÄeriz and Christine Thomassigny
*
a
Â
Instituto de Biologia Experimental e Tecnologica, Universidade Nova de Lisboa, Apartado 12, P-2780-156 Oeiras, Portugal
Ã
b
Departamento de Quimica, Faculdade de Ciencias e Tecnologia, Universidade Nova de Lisboa, P-2825-114 Caparica, Portugal
cInstituto de Tecnologia Quimica e Biologia, Universidade Nova de Lisboa, Rua da Quinta Grande 6,
Apartado 127, P-2780-156 Oeiras, Portugal
Received 22 May 2000; revised 29 June 2000; accepted 3 July 2000
AbstractÐThree protected new gallotannins, namely the 60-O-(tri-O-methylgalloyl)-2,3,4,6,10,30,40-hepta-O-acetylsucrose, the 60-O-(tri-
O-methylgalloyl)-2,3,4,6,10,30,40-hepta-O-benzoylsucrose and the 6,60-di-O-tert-butyldiphenylsilyl-10-O-(tri-O-methylgalloyl)-2,3,4,30,40-
penta-O-acetylsucrose have been prepared in four short sequences from sucrose. Methods for rapid galloylation have been studied in order to
avoid simultaneous acyl transfer reactions. A method for the deprotection of a tert-butyldiphenylsilyl ether has been developed which avoids
the intramolecular migration of a benzoate group. q 2000 Elsevier Science Ltd. All rights reserved.
Introduction
importance of such derivatives has been emphasised by
Dangles et al.,3 who prepared and tested some polygalloyl-
sucroses. We know of no other reference to galloylated
sucrose, which is surprising when we consider the good
antioxidant ability of these compounds. Many studies
have been done with the glucose core,5,6 but only few
natural galloylated furanoses have been isolated.6,13,14 As
an example, the 1-O-galloylfructose extracted from a
rhubarb13 or hamamelitannins6 whose oxidation tests have
shown that furanose rings can play an important role in
scavenging hydroxyl radicals. This lack of study has stimu-
lated us to concentrate our work in the fructose unit of the
sucrose. With the aim of comparing the effect of the position
of the galloyl group on its antioxidant activity, we have
functionalised selectively each one of the primary hydroxyl
groups of the fructose unit of sucrose.
Gallotannins are heterogeneous glycosidic esters of gallic
acid and its O-linked derivatives (meta-depside linkage),
which are secondary metabolites of higher plants.1,2 They
are ef®cient antioxidants by reason of their ability to trap
reactive oxygen species involved in cardiovascular diseases
and cancer.3 The strong contribution of the galloyl or similar
polyphenolic groups have been demonstrated several
times.4,5 The sugar moiety also displays an important role
which is demonstrated by the observation that most of the
gallotannins have a stronger effect than the smaller poly-
phenol molecules.5,6 As a general rule, the antioxidative
activity of each tannin is dependent on the type and number
of the phenolic groups in the molecule.5 Nevertheless, the
activity is dependent on other factors such as the position of
the galloyl groups,7 the active oxygen species (superoxide
anions, hydroxyl radicals, singlet oxygen),6 the interaction
with coexisting compounds8 or the alcohol core.5,6,8
We were stimulated to enlarge the collection of known
gallotannins, with the aim of studying the mechanistic
aspects of the oxidation reaction. Such tests need a com-
parison with a speci®c reference. The most employed refer-
ences are gallic acid and various ether gallates.5,6 As a ®rst
target we proposed to construct molecules more similar to
gallotannins and with all the supposed active groups
protected. A methylation of the phenolic groups9,10,11 induces
a loss of antioxidative activity.12 The introduction of only one
tri-O-methyl galloyl unit into a speci®c place of a protected
sucrose should avoid the intra- and/or intermolecular stack-
ing interactions suspected between the galloyl units.3
At the end of the 1980s, ®ve new gallotannins have been
isolated2 from commercial rhubarbs, each one displaying a
sucrose core acylated by one gallate unit (respectively the
60-O, 40-O, 6-O, 10-O and 2-O-monogalloylsucrose). The
Keywords: sucrose; galloylation; gallotannins; tert-butyldiphenylsilyl ether
(deprotection of).
* Corresponding author. Fax: 1351-212948550;
e-mail: mtbarros@dq.fct.unl.pt
0040±4020/00/$ - see front matter q 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)00593-7

6512
M. T. Barros et al. / Tetrahedron 56 (2000) 6511±6516
Scheme 1. (i) TBDPSCl, pyr.; (ii) Ac2O, pyr.; (iii) nBu4NF, THF; (iv)
(OMe)3galloyl chloride, TMEDA, CH2Cl2.
Scheme 3. (i) Br2 in MeOH, re¯ux; (ii) (OMe)3galloyl chloride, TMEDA,
CH2Cl2.
We have prepared three references, composed of a sucrose
core acylated with one tri-O-methylgalloyl in the position 10
or 60 of the fructose moiety, and with the hydroxyl groups
fully protected (benzoates, acetates and/or silyloxy groups).
One of the routes required a speci®c desilylation without
migration of an adjacent benzoyl group. We present here a
procedure, using bromine in methanol, which has allowed
us to obtain our target compound.
followed by an acetylation by acetic anhydride in pyridine,
as described in the literature,15 leading to 2. Selective depro-
tection of the silyloxy function using tetrabutylammonium
¯uoride in THF gave the expected alcohol 3 with a moderate
yield (50%) by reason of the formation of products resulting
from acyl migration and ¯uorolysis of acetate groups. First
attempts at the esteri®cation of partially protected sucrose
by tri-O-methyl galloylchloride in a mixture of dimethyl-
formamide±triethylamine in the presence of 4-dimethyl-
aminopyridine3 led to the expected compounds but with
again only moderate yields (32±64% depending on the
initial). An adaptation of the recent method described by
Oriyama et al.,16 consisting in a fast acylation in the
presence of TMEDA, afforded the expected galloylated
sucroses with medium to good yields. Following this pro-
cedure, reaction of 3 with the trimethylgalloyl chloride led
to the 60-tri-O-methylgalloyl 4 with the medium yield of 52%.
Results and Discussion
The ®rst way to our targets (Scheme 1) consisted of a selec-
tive silylation of the 60 hydroxyl group of the sucrose (1)
Compound 4 was also obtained from sucrose (Scheme 2), in
just one extra step than the previous way but with better
overall yields. An attempt to remove the silyloxy groups
of polyacetylated 6,60-di-O-silyloxy sucrose 515 using tetra-
butylammonium ¯uoride in THF occurred (6) with the
1
migration of one acetate. Indeed, a study of the H NMR
spectrum of the obtained derivative showed a triplet at
3.46 ppm, that we have attributed to the proton H-4 after
some irradiation experiments. We have con®rmed this
migration by the observation of the two doublets of doub-
lets, respectively at 4.26 and 4.48 ppm, attributed to the
protons H-6 or H-60. These signals were too deshielded to
belong to an hydroxymethylene, but are typical of an acetyl-
ated one. Unfortunately, it was impossible to determine at
that time if the migration had occurred from the position
O-4 to the position O-6 or O-60. An acylation of 6 using
2 equiv. of galloyl chloride, and following the same pro-
cedure described for 3, led to a major product (84%)
whose NMR spectral data agree with the monogalloyl-
sucrose 7. In particular, the signal at 7.31 ppm (two
aromatic protons) and the singlets at around 3.9 ppm (nine
Scheme 2. (i) nBu4NF, THF; (ii) (OMe)3galloyl chloride, TMEDA,
CH2Cl2; (iii) Ac2O, pyr.

M. T. Barros et al. / Tetrahedron 56 (2000) 6511±6516
6513
As a continuation of studies already carried out in our
laboratory,19 and with the aim of functionalising selectively
the 10 position of the sucrose, we have used (Scheme 4) the
yet-obtained 6,60-di-O-silyloxy sucrose, for a selective
protection of the 10-hydroxyl free function of the sucrose.
The action of TBDMSCl in pyridine in the presence of 4-
dimethylaminopyridine led to the pentol 11, which has been
characterized as its pentabenzoate 12. An acetylation of 11
gave an inseparable mixture of two compounds, containing
as a major one the pentaacetate 13. The reaction with a
solution of iodine in methanol gave the expected 14
whose physical data are in agreement with the compound
described in the literature.20 The galloylation of 14 by the
procedure previously described for 3 led to 15 (78%).
Conclusion
We have selectively prepared the precursors of two new
gallotannins, containing one phenolic unit speci®cally at
the 10 or the 60 position of the fructose unit of the sucrose.
These compounds have been obtained in four short-step
syntheses, using in particular a new method for the fast
galloylation of sugar moieties.
Scheme 4. (i) Ac2O, pyr. or BzCl, pyr.; (ii) I2 in MeOH, re¯ux; (iii)
(OMe)3galloyl chloride, TMEDA, CH2Cl2.
protons in total) proved the presence of only one galloyl
function. Here also, it was impossible to certify which one
was the newly acylated atom (O-6 or O-60). Some irradia-
tion experiments have con®rmed the free hydroxyl function
at C-4. A classical acetylation of 7 led to a compound whose
Experimental
General
Reagents and solvents were puri®ed before use. Flash chro-
matography was performed on silica gel from Macherey±
Nagel (Kieselgel 60 M). Preparative TLC was performed on
glass plates coated with 1 mm of silica gel (Macherey±
Nagel, Kieselgel DGF254). Analytical TLC: Aluminium-
backed silica gel Merck 60 F254. Optical rotations were
measured at 208C on a Perkin±Elmer 241 polarimeter
(1 dm cell). NMR spectra were recorded on a Bruker
AMX-400 apparatus (1H at 400 MHz and 13C at
100 MHz) in CDCl3 with chemical shift values (d) in ppm
down®eld from tetramethylsilane. Microanalyses were
performed by the IST analytical services in Lisbon using a
combustion apparatus. Mass spectroscopies were performed
1
physical data (optical rotation, H and 13C NMR spectra)
were identical to those of the previously obtained 60-galloyl
sucrose 4. This result has led us to suppose that the inter-
mediate 6 was the 60-OH free sucrose, and that 7 was tri-O-
methyl galloylated in 60. During the formation of 6, the
acetate had migrated from O-4 to O-6 under the in¯uence
of the ¯uoride.
We have also prepared the benzoylated 60-galloyl sucrose
10 in order to carry out further experiments (Scheme 3) with
the aim of con®rming that the protective groups of the
sucrose unit would not in¯uence the proposed oxidation
tests of our references.
Â
by the `Unidade de Espectroscopõa de Masas' de Santiago
de Compostela, Spain.
We have developed in the laboratory a new method of
deprotection of tert-butyldiphenylsilyloxy (TBDPS) groups,
occurring without migration of the esters groups. This
method is derived from the recently published procedure
of Vaino and Szarek17 who have demonstrated its ef®ciency
for the deprotection of tert-butyldimethylsilyl (TBDMS)
ethers using iodine in methanol. This procedure was unsuc-
cessful with the compound 815 as the OTBDPS is usually
more stable than an OTBDMS.18 By using a smaller halogen
(e.g. bromine), we have been able to deprotect the TBDPS
function of 8, with good selectivity (69%) and without
observation of any migration. Unfortunately, as the mecha-
nism of the reaction was suspected to occur with the forma-
tion of acid-like species and traces of HBr, degradation of
the O-glycosidic link of the sucrose was expected and this
fact can explain the somewhat disappointing yield. Galloyl-
ation of the so obtained alcohol 9 in the presence of
TMEDA led to the expected 10 with a good yield (85%).
60-O-tert-Butyldiphenylsilyl-2,3,4,6,10,30,40-hepta-O-acetyl-
sucrose (2). This compound was prepared as described in
the literature.15 [a]Dˆ151.6 (c 1, CHCl3), Lit.15 150 (c 1.2,
CHCl3); 1H NMR (CDCl3) d 7.66 (m, 4H, Ar), 7.41 (m, 6H,
Ar), 5.64 (d, Jˆ3.7 Hz, 1H, H-1), 5.52 (t, Jˆ5.7 Hz, 1H,
H-40), 5.40 (m, 1H, H-3), 5.40 (d, Jˆ5.4, 1H, H-30), 5.03 (t,
Jˆ9.9 Hz, 1H, H-4), 4.83 (dd, Jˆ10.4, 3.7 Hz, 1H, H-2),
4.19 (d, Jˆ12.3 Hz, 1H, H-10a), 4.15 (m, 3H, H-5, H-6a and
H-50), 4.14 (d, 1H, H-10b), 3.95 (dd, Jˆ7.8, 1.4 Hz, 1H,
H-6b), 3.87 (dd, Jˆ11.0, 5.4 Hz, 1H, H-60a), 3.83 (dd,
Jˆ11.0, 5.8 Hz, 1H, H-60b), 2.14, 2.10, 2.09, 2.05, 2.01,
t
1.98, 1.97 (7s, 3H each, OCOMe), 1.06 (s, 9H, Bu). 13C
NMR (CDCl3) d 19,10 (CMe3), 20.45, 20.47, 20.53, 20.55,
20.68 (Me), 26.64 (CMe3), 61.50, 62.80, 63.80 (C-6, C-10
and C-60), 68.00, 68.20, 69.70, 70.10 (C-2, C-3, C-4 and
C-5), 75.00, 76.10 (C-30 and C-40), 81.30 (C-50), 89.60
(C-1), 103.60 (C-20), 127.70, 129.80, 132.80, 132.90,

6514
M. T. Barros et al. / Tetrahedron 56 (2000) 6511±6516
135.50 (Ar), 169.40, 169.70, 169.80, 169.90, 170.00, 170.50
(CO).
described in the literature.15 [a]Dˆ154.4 (c 1, CHCl3),
1
Lit.15 154.0 (c 1.2, CHCl3); H NMR (CDCl3) d 7.66±
7.60 (m, 8H, Ar), 7.43±7.27 (m, 12H, Ar), 5.65 (d,
Jˆ3.5 Hz, 1H, H-1), 5.38 (m, 4H, H-3, H-4, H-30 and
H-40), 4.82 (dd, Jˆ9.5, 3.5 Hz, 1H, H-2), 4.31 (d, Jˆ
12.0 Hz, 1H, H-10a), 4.23 (d, 1H, H-10b), 4.00 (d, Jˆ
9.0 Hz, 1H, H-50), 3.81 (m, 2H, H-60a and H-60b), 3.63 (d,
Jˆ11.5 Hz, 1H, H-6a), 3.52 (d, 1H, H-6b), 2.11, 2.09, 2.07,
2.04, 2.02, 1.84 (6s, 3H each, OCOMe), 1.02 (s, 18H, tBu).
13C NMR (CDCl3) d 19,11 (CMe3), 20.17, 20.47, 20.57,
20.74 (Me), 26.65, 26.67 (CMe3), 61.30, 62.20, 63.90
(C-6, C-10 and C-60), 68.00, 70.20, 70.50, 70.60 (C-2,
C-3, C-4 and C-5), 75.10, 76.20 (C-30 and C-40), 81.90
(C-50), 90.50 (C-1), 104.50 (C-20), 127.60, 127.62, 127.70,
129.50, 129.60, 129.70, 132.90, 135.40, 135.60, 135.63
(Ar), 169.10, 169.62, 169.64, 170.10, 170.20, 170.60
(CO); Anal. Calcd for C56H70O17Si2 (1070.42): C, 62.78;
H, 6.59. Found C, 62.71; H, 6.86.
2,3,4,6,10,30,40-Hepta-O-acetylsucrose (3). To a solution of
2 (250.0 mg, 0.29 mmol) in dry THF (7 mL) was added,
under argon, tetrabutylammonium ¯uoride (1.8 equiv.,
135.0 mg). When the reaction mixture showed no more
starting material (TLC in AcOEt/hexane, 3/2), the THF
was evaporated. Water was then added, and the organic
compounds were extracted with dichloromethane. After
drying and concentration, a puri®cation by ¯ash chroma-
tography gave the expected but unstable 60-free sucrose 3
(91.0 mg, 50%). 1H NMR (CDCl3) d 5.68 (d, Jˆ3.5 Hz, 1H,
H-1), 5.47 (m, 3H, H-3, H-30 and H-40), 5.09 (t, Jˆ9.4 Hz,
1H, H-4), 4.88 (dd, Jˆ12.0, 3.5 Hz, 1H, H-2), 4.18 (d,
Jˆ11.7 Hz, 1H, H-10a), 4.16 (m, 4H, H-5, H-6a, H-6b and
H-50), 4.10 (d, 1H, H-10b), 3.84 (dd, Jˆ12.8, 2.8 Hz, 1H,
H-60a), 3.68 (dd, Jˆ12.8, 4.4 Hz, 1H, H-60b), 2.70 (wide
signal, 1H, OH), 2.19, 2.13, 2.11, 2.10, 2.08, 2.05, 2.02 (7s,
3H each, OCOMe). 13C NMR (CDCl3) d 20.41, 20.45,
20.47, 20.50, 20.55, 20.61, 20.87 (Me), 61.10, 61.40,
63.80 (C-6, C-10 and C-60), 67.90, 68.60, 69.30, 70.10
(C-2, C-3, C-4 and C-5), 73.70, 76.00 (C-30 and C-40),
81.60 (C-50), 89.90 (C-1), 103.20 (C-20), 169.40, 169.80,
169.90, 170.00, 170.40, 170.50, 170.60 (CO).
2,3,6,10,30,40-Hexa-O-acetylsucrose (6). To a solution of 5
(0.5 g, 0.5 mmol) in anhydrous THF (10 mL) under argon
was added 0.3 mg (2.5 equiv.) of tetrabutylammonium
¯uoride. After 2.5 h of stirring, TLC (AcOEt/hexane, 2/1)
showed no trace of the starting material. Evaporation of the
solvent followed by puri®cation on silica gel (AcOEt/
hexane, 2/1, then AcOEt) gave the title compound 6
(219.0 mg, 79%). [a]Dˆ136.9 (c 1, CHCl3); 1H NMR
(CDCl3) d 5.55 (d, Jˆ3.6 Hz, 1H, H-1), 5.40 (m, 2H, H-30
and H-40), 5.23 (t, Jˆ4.8±5.2 Hz, 1H, H-3), 4.76 (dd,
Jˆ10.4, 3.6 Hz, 1H, H-2), 4.48 (dd, Jˆ12.8, 3.6 Hz, 1H,
H-6a), 4.26 (dd, Jˆ12.8, 2.4 Hz, 1H, H-6b), 4.10 (d,
Jˆ11.6±12.0 Hz, 1H, H-10a), 4.04 (m, 2H, H-5 and H-50),
4.00 (d, 1H, H-10b), 3.77 (dd, Jˆ12.8, 2.8 Hz, 1H, H-60a),
3.60 (dd, Jˆ12.8, 4.0 Hz, 1H, H-60b), 3.46 (t, Jˆ9.6 Hz, 1H,
H-4), 2.11, 2.07, 2.05, 2.04, 2.03, 2.03 (6s, 3H each,
OCOMe). 13C NMR (CDCl3**) d 20.28, 20.35, 20.37,
20.44, 20.47, 20.56 (Me), 61.26, 62.43, 63.53 (C-6, C-10
and C-60), 68.47, 70.17, 71.11, 71.62 (C-2, C-3, C-4 and
C-5), 73.89, 76.03 (C-30 and C-40), 81.69 (C-50), 90.26
(C-1), 103.28 (C-20), 170.12, 170.24, 170.29, 170.81,
171.13, 171.80 (CO); Anal. Calcd for C24H34O17 (594.18):
C, 48.49; H, 5.76. Found C, 48.17; H, 5.65.
60-O-(Tri-O-methylgalloyl)-2,3,4,6,10,30,40-hepta-O-acetyl-
sucrose (4). First way: general procedure. To a solution of
TMEDA (2.0 equiv., 94.2 mL) in dry dichloromethane
Ê
(1 mL), was added molecular sieves 3 A (165 mg), and
then a solution of 200.0 mg of 3 dissolved in dry dichloro-
methane (6 mL) under argon. The mixture was cooled to
2788C, and then tri-O-methylgalloyl chloride (2.5 equiv.,
179 mg) in dry dichloromethane (1 mL) was added. The
reaction mixture was allowed to warm until room tempera-
ture, then stirred overnight. After neutralization by a
solution of phosphate buffer (pH 7) and extraction with
dichloromethane, the combined organic phases were
washed with brine, dried and concentrated. A puri®cation
of the crude by ¯ash chromatography (AcOEt/hexane, 1/1)
led to the expected 4 (135.0 mg, 52%). [a]Dˆ136.9 (c 0.4,
1
CHCl3); H NMR (CDCl3) d 7.33 (m, 2H, Ar), 5.76 (d,
Jˆ3.6 Hz, 1H, H-1), 5.47 (m, 3H, H-3, H-30 and H-40),
5.07 (t, Jˆ9.6 Hz, 1H, H-4), 4.87 (dd, Jˆ11.8, 3.4 Hz,
1H, H-2), 4.63 (dd, Jˆ11.6, 6.8 Hz, 1H, H-60a), 4.51 (dd,
Jˆ11.6, 5.6 Hz, 1H, H-60b), 4.40±4.13 (m, 6H, H-5, H-6a,
H-6b, H-50, H-10a and H-10b), 3.93 (s, 6H, ArOMe), 3.91 (s,
3H, ArOMe), 2.18, 2.12, 2.11, 2.10, 2.08, 2.03, 2.02 (7s, 3H
each, OCOMe). 13C NMR (CDCl3) d 20.38, 20.46 (Me),
56.15 (ArOMe), 60.75 (ArOMe), 61.64, 62.78, 64.31 (C-6,
C-10 and C-60), 68.00, 68.49, 69.39, 70.16 (C-2, C-3, C-4
and C-5), 75.53, 75.80 (C-30 and C-40), 79.07 (C-50), 90.03
(C-1), 104.15 (C-20), 107.08, 124.40, 142.66, 153.07 (Ar),
165.76, 169.59, 169.77, 169.90, 170.08, 170.21, 170.71
(CO); HRMS Calcd for C36H46O22: 830.2480. Found:
830.2484.
60-O-(Tri-O-methylgalloyl)-2,3,6,10,30,40-hexa-O-acetyl-
sucrose (7). The reaction of galloylation of 6 (133.0 mg,
0.2 mmol) was carried out according to the general pro-
cedure given for compound 3 (149 mg, 84%). [a]Dˆ
1
128.7 (c 1.2, CHCl3); H NMR (CDCl3) d 7.31 (m, 2H,
Ar), 5.64 (d, Jˆ3.6 Hz, 1H, H-1), 5.46 (m, 2H, H-30 and
H-40), 5.40 (t, Jˆ9.6 Hz, 1H, H-3), 4.79 (dd, Jˆ10.4, 3.2±
3.6 Hz, 1H, H-2), 4.47 (dd, Jˆ14.0, 4.8 Hz, 1H, H-6 or
H-60), 4.40 (m, 2H, H-6 and H-60, or H-5 and H-50), 4.31
(dd, Jˆ12.4, 1.6 Hz, 1H, H-6 or H-60), 4.22 (d, Jˆ12.4 Hz,
1H; H-10a), 4.18 (d, 1H, H-10b), 4.12 (m, 1H, H-5, H-50, H-6
or H-60), 4.10 (m, 1H, H-5, H-50, H-6 or H-60), 3.92 (s, 6H,
OMe), 3.91 (s, 3H, OMe), 3.53 (t, 1H, H-4), 3.39 (s, 1H,
OH), 2.15, 2.12, 2.11, 2.10, 2.08, 2.04 (6s, 3H each,
OCOMe). 13C NMR (CDCl3) d 20.23, 20.36, 20.41,
20.48, 20.62 (Me), 56.17, 56.19 (ArOMe), 60.77
(ArOMe), 62.55, 62.73, 64.68 (C-6, C-10 and C-60), 68.84,
70.30, 70.96, 71.82 (C-2, C-3, C-4 and C-5), 75.65, 75.83
(C-30 and C-40), 79.39 (C-50), 90.44 (C-1), 104.31 (C-20),
107.15, 124.40, 142.71, 153.07 (Ar), 166.03, 169.82,
Second way. Acetylation of 7 in the presence of 4-DMAP,
followed by a puri®cation on silica gel (AcOEt/hexane, 2/1)
gave the previously reported 4 (90%).
6,60-Di-O-tert-butyldiphenylsilyl-2,3,4,10,30,40-hexa-O-
acetylsucrose (5). This compound was prepared as

M. T. Barros et al. / Tetrahedron 56 (2000) 6511±6516
6515
170.00, 170.23, 170.48, 171.15, 171.61 (CO); Anal. Calcd
for C34H44O21 (788.24): C, 51.78; H, 5.62. Found C, 51.36;
H, 5.62.
6,60-Di-O-tert-butyldiphenylsilyl-10-O-tert-butyldimethyl-
silyl-2,3,4,30,40-penta-O-benzoylsucrose (12). tert-Butyl-
dimethylchlorosilane (0.7 g, 2.0 equiv.) was added to a
solution of 1.9 g (2.3 mmol) of 6,60-di-O-tert-butyldi-
phenylsilyl sucrose15 in pyridine (35 mL) in the presence
of a catalytic amount of 4-DMAP. After stirring overnight
at room temperature under argon, the pyridine was evapo-
rated. Chromatography (AcOEt) gave the 6,60-di-O-tert-
butyldiphenylsilyl-10-O-tert-butyldimethylsilylsucrose (11;
1.5 g, 71%). Benzoylation in the presence of 4-DMAP,
followed by puri®cation on silica-gel (AcOEt/hexane, 1/4)
gave 12 (1.7 g, 70%). [a]Dˆ24.8 (c 0.9, CHCl3); 1H NMR
(CDCl3) d 8.01±7.12 (m, 45H, Ar), 6.18±6.02 (m, 5H, H-1,
H-3, H-4, H-30 and H-40), 5.46 (dd, Jˆ9.8±10.2, 3.4±
3.8 Hz, 1H, H-2), 4.32 (m, 2H, H-5 and H-50), 4.04 (d,
Jˆ10.5 Hz, 1H, H-10a), 3.97 (m, 2H, H-6ab or H-60ab),
3.63 (d, 1H, H-10b), 3.54 (d, Jˆ11.1 Hz, 1H, H-6a or
H-60a), 3.45 (d, Jˆ10.7 Hz, 1H, H-6b or H-60b), 1.09,
SiMe2); 13C NMR (CDCl3) d 25.98, 25.86 (SiMe2), 17.97,
18.86, 18.95 (CMe3), 25.63, 25.75, 26.54, 26.57 (CMe3),
60.90, 63.83, 65.18 (C-6, C-10 and C-60), 68.04, 71.12,
71.49 (C-2, C-3, C-4 and C-5), 75.04, 76.61 (C-30 and
C-40), 80.45 (C-50), 89.81 (C-1), 104.85 (C-20), 127.48±
135.85 (Ar), 164.85, 165.27, 165.58, 165.95, 166.10 (CO);
Anal. Calcd for C85H92O16Si3 (1452.57): C, 70.22; H, 6.38.
Found C, 70.31; H, 6.30.
60-O-tert-Butyldiphenylsilyl-2,3,4,6,10,30,40-hepta-O-ben-
zoylsucrose (8). This compound was prepared as described
in the literature.15 [a]Dˆ136.1 (c 1.2, CHCl3), Lit.15 139 (c
1
1, CHCl3); H NMR (CDCl3) d 8.01±7.27 (m, 45H, Ar),
6.17 (t, Jˆ10.0 Hz, 1H, H-3), 6.06 (t, Jˆ6.2 Hz, 1H, H-40),
6.02 (d, Jˆ3.6 Hz, 1H, H-1), 5.93 (t, 1H, H-30), 5.72 (t,
Jˆ10.0 Hz, 1H, H-4), 5.41 (dd, Jˆ10.4, 3.6 Hz, 1H, H-2),
4.65 (d, Jˆ11.9 Hz, 1H, H-10a), 4.59 (m, 1H, H-5), 4.54 (d,
1H, H-10b), 4.33 (m, 2H, H-6a and H-50), 4.21 (dd, Jˆ12.5,
3.1 Hz, 1H, H-6b), 3.99 (d, Jˆ5.3 Hz, 2H, H-60a and H-60b),
1.00 (s, 9H, tBu). 13C NMR (CDCl3) d 18.87, 21.21 (CMe3),
26.51, 26.55 (CMe3), 62.13, 63.72, 65.33 (C-6, C-10 and
C-60), 68.88, 70.24, 71.02 (C-2, C-3, C-4 and C-5), 75.33,
76.68 (C-30 and C-40), 81.17 (C-50), 90.15 (C-1), 103.94
(C-20), 125.31±135.66 (Ar), 165.13, 165.34, 165.60,
165.66, 165.67, 165.75, 166.02 (CO).
t
1.08, 0.93 (3s, 9H each, Bu), 20.01, 20.03 (2s, 3H each,
2,3,4,6,10,30,40-Hepta-O-benzoylsucrose (9). 850.0 mg
(0.65 mmol) of 8 were re¯uxed in a solution of bromine
1% in methanol (4 mL) during 3 h. When TLC (AcOEt/
hexane, 1/2) showed no more initial, the mixture was cooled
and the excess of bromine was destroyed by a solution of
5% of Na2S2O3. The aqueous layer was extracted by
dichloromethane, and the organic layer was dried, concen-
trated, and puri®ed on silica gel (AcOEt/hexane, 1/2) to give
the heptabenzoate 9 (480 mg, 69%); [a]Dˆ165.4 (c 1.0,
CHCl3); 1H NMR (CDCl3) d 8.05±7.16 (m, 35H, Ar),
6.22 (m, 2H, H-1 and H-3), 6.03 (t, Jˆ5.9 Hz, 1H, H-40),
5.97 (t, 1H, H-30), 5.78 (t, Jˆ10.0 Hz, 1H, H-4), 5.34 (dd,
Jˆ10.4, 3.5 Hz, 1H, H-2), 4.72 (d, Jˆ11.8 Hz, 1H, H-10a),
4.61 (m, 1H, H-5), 4.56 (d, 1H, H-10b), 4.22 (m, 3H, H-6a,
H-6b and H-50), 3.81 (m, 2H, H-60a and H-60b), 3.19 (br t,
1H, OH). 13C NMR (CDCl3) d 60.80, 61.74, 65.54 (C-6,
C-10 and C-60), 68.60, 69.24, 69.79, 71.94 (C-2, C-3, C-4
and C-5), 74.88, 77.99 (C-30 and C-40), 81.54 (C-50), 90.80
(C-1), 104.43 (C-20), 128.33±133.75 (Ar), 165.15, 165.63,
165.67, 165.74, 166.14, 167.03 (CO); Anal. Calcd for
C61H50O18 (1070.30): C, 68.41; H, 4.71. Found C, 68.14;
H, 4.68.
6,60-Di-O-tert-butyldiphenylsilyl-2,3,4,30,40-penta-O-acetyl-
sucrose (14). Acetylation of 11 followed by a puri®cation
on silica-gel (AcOEt/hexane, 1/4) gave 1.7 g of a mixture of
13 contaminated by another product. This impure 13 was
re¯uxed in a solution of iodine 1% in methanol (50 mL)
during 1 h. After cooling, the reaction mixture was treated
with Na2S2O5 1% and extracted with dichloromethane. The
organic layer was washed by Na2S2O3 5%, dried and
concentrated. After chromatography (AcOEt/hexane, 1/2),
the product 14 (0.7 g, 41% in two steps) was obtained.
1
[a]Dˆ146.8 (c 0.2, CHCl3), Lit.20 150 (c 1, CHCl3); H
NMR (CDCl3) d 7.63±7.27 (m, 20H, Ar), 5.61 (d, Jˆ
3.6 Hz, 1H, H-1), 5.55 (t, Jˆ5.1 Hz, 1H, H-40), 5.38 (t,
Jˆ9.8±10.0 Hz, 1H, H-3), 5.32 (d, Jˆ5.2 Hz, 1H, H-30),
5.77 (t, Jˆ10.0 Hz, 1H, H-4), 4.87 (dd, Jˆ10.4, 3.7 Hz,
1H, H-2), 4.11 (m, 1H, H-5 or H-50), 3.98 (m, 1H, H-5 or
H-50), 3.80 (m, 2H, H-6 or H-60), 3.72 (d, Jˆ12.6 Hz, 1H,
H-10a), 3.65 (d, Jˆ12.7 Hz, 1H, H-10b), 3.53 (dd, Jˆ11.6,
1.6 Hz, 1H, H-6 or H-60), 3.44 (dd, Jˆ11.6, 3.8 Hz, 1H, H-6
or H-60), 2.13, 2.10, 2.02, 2.01, 1.86 (5s, 3H each, OCOMe),
60-O-(Tri-O-methylgalloyl)-2,3,4,6,10,30,40-hepta-O-ben-
zoylsucrose (10). The galloylation of 9 (500.0 mg,
0.47 mmol) as described previously, followed by a puri®ca-
tion by ¯ash chromatography (AcOEt/hexane, 1/2) afforded
1
t
ester 10 (413.0 mg, 70%); [a]Dˆ136.4 (c 0.3, CHCl3); H
1.02, 1.01 (2s, 9H each, Bu). 13C NMR (CDCl3) d 18.94,
NMR (CDCl3) d 8.09±7.16 (m, 37H, Ar), 6.20 (m, 2H, H-1
and H-3), 5.96 (m, 2H, H-30 and H-40), 5.77 (t, Jˆ10.0 Hz,
1H, H-4), 5.41 (dd, Jˆ10.4, 3.0 Hz, 1H, H-2), 4.79±4.59
(m, 6H, H-5, H-50, H-10a, H-10b, H-60a and H-60b), 4.50 (dd,
Jˆ12.5, 2.2 Hz, 1H, H-6a), 4.38 (dd, Jˆ12.4, 2.3 Hz, 1H,
H-6b), 3.91 (s, 3H, OMe), 3.85 (m, 6H, OMe). 13C NMR
(CDCl3) d 56.02 (ArOMe), 60.75 (ArOMe), 62.13, 64.01,
64.76 (C-6, C-10 and C-60), 68.87, 69.19, 69.98, 71.36 (C-2,
C-3, C-4 and C-5), 76.74, 77.54 (C-30 and C-40), 79.06
(C-50), 90.91 (C-1), 104.95 (C-20), 107.04, 124.58,
128.40±133.77, 142.37, 152.99 (Ar), 165.23, 165.50,
165.69, 165.88, 165.99, 166.16 (CO); Anal. Calcd for
C71H60O22 (1264.36): C, 67.40; H, 4.78. Found C, 67.13;
H, 4.69.
19.03 (CMe3), 20.35, 20.39, 20.48, 20.59, 20.66 (Me),
26.52, 26.55, 26.58 (CMe3), 61.55, 63.16, 64.21 (C-6,
C-10 and C-60), 68.17, 70.07, 70.31, 70.46 (C-2, C-3, C-4
and C-5), 76.40, 77.89 (C-30 and C-40), 81.28 (C-50), 90.59
(C-1), 105.86 (C-20), 127.71, 127.77, 129.72, 129.81,
129.86, 132.85, 133.10, 133.13, 135.56, 135.59, 135.75,
135.79 (Ar), 169.37, 169.87, 170.28, 170.46, 170.79 (CO).
6,60-Di-O-tert-butyldiphenylsilyl-10-O-(tri-O-methylgal-
loyl)-2,3,4,30,40-penta-O-acetylsucrose (15). The galloyl-
ation of 14 (0.7 g, 0.7 mmol) as described previously,
followed by a puri®cation by ¯ash chromatography
(AcOEt/hexane, 1/2) led to the title compound 15 (0.6 g,
1
78%). [a]Dˆ148.5 (c 1, CHCl3); H NMR (CDCl3) d

6516
M. T. Barros et al. / Tetrahedron 56 (2000) 6511±6516
5. Yoshida, T.; Mori, K.; Hatano, T.; Okumura, T.; Uehara, I.;
Komagoe, K.; Fujita, Y.; Okuda, T. Chem. Pharm. Bull. 1989,
37, 1919±1921.
7.61±7.27 (m, 22H, Ar), 5.65±5.31 (m, 5H, H-1, H-3, H-4,
H-30 and H-40), 4.83 (dd, Jˆ9.9, 3.6 Hz, 1H, H-2), 4.56 (d,
Jˆ12.4 Hz, 1H, H-10a), 4.39 (d, 1H, H-10b), 3.95±3.90 (m,
11H, OMe, H-5 and H-50), 3.62 (d, Jˆ11.7 Hz, 1H, H-6 or
H-60), 3.51 (d, Jˆ11.1 Hz, 1H, H-6 or H-60), 3.40 (m, 2H,
H-6 or H-60), 2.05, 2.02, 2.00, 1.99, 1.96 (5s, 3H each,
OCOMe), 1.01, 1.00 (2s, 9H each, tBu). 13C NMR
(CDCl3) d 18.91, 18.93 (CMe3), 20.23, 20.27, 20.30,
20.51, 20.54 (Me), 25.53, 26.48 (CMe3), 56.11, 56.23,
59.03 (ArOMe), 60.74, 62.81, 63.65 (C-6, C-10 and C-60),
69.96, 70.17, 70.47, 70.57 (C-2, C-3, C-4 and C-5), 75.02,
76.35 (C-30 and C-40), 82.32 (C-50), 90.68 (C-1), 104.89
(C-20), 106.96, 107.39, 124.40±135.74, 153.11 (Ar),
165.57, 169.22, 169.50, 169.67, 170.25, 170.30 (CO);
HRMS Calcd for C64H78O20Si2 (C64H78O20Si2±Na):
1245.4522. Found: 1245.4490.
6. Masaki, H.; Atsumi, T.; Sakurai, H. Phytochemistry 1994, 37,
337±343.
7. Hatano, T.; Yasuhara, T.; Yoshihara, R.; Agata, I.; Noro, T.;
Okuda, T. Chem. Pharm. Bull. 1990, 38, 1224±1229.
8. Tanaka, T.; Zhang, H.; Jiang, Z.-H.; Kouno, I. Chem. Pharm.
Bull. 1997, 45, 1891±1897.
9. Saijo, R.; Nonaka, G..-I.; Nishioka, I. Phytochemistry 1989, 28,
2443±2446.
È
10. Khanbabaee, K.; Lotzerich, K. Tetrahedron 1997, 53, 10725±
10732.
11. (a) Lipshutz, B. H.; Liu, Z.-P.; Kayser, F. Tetrahedron Lett.
1994, 35, 5567±5570. (b) Itoh, T.; Chika, J.-I.; Shirakami, S.; Ito,
H.; Yoshida, T.; Kubo, Y.; Uenishi, J.-I. J. Org. Chem. 1996, 61,
3700±3705.
12. Valgimigli, L.; Banks, J. T.; Ingold, K. U.; Lusztyk, J. J. Am.
Chem. Soc. 1995, 117, 9966±9971.
Acknowledgements
13. Kashiwada, Y.; Nonaka, G..-I.; Nishioka, I. Chem. Pharm.
Bull. 1984, 32, 3461±3470.
This work has been supported by NATO SFS (PO-
Ã
SUGARS) and the FundacËaÄo para a Ciencia e Tecnologia
(grant PRAXIS XXI/BPD/16348/98 for C. T. and PRAXIS
XXI/BD/17007/98 for F. S.).
14. Chevalley, I.; Marston, A.; Hostettmann, K. Phytochemistry
1999, 50, 151±154.
15. Karl, H.; Lee, C. K.; Khan, R. Carbohydr. Res. 1982, 101, 31±
38.
References
16. Sano, T.; Ohashi, K.; Oriyama, T. Synthesis 1999, 7, 1141±
1144.
1. Salisbury, F. B.; Ross, C. W. Plant Physiology, Wadsworth:
Belmont, California, 1992 (p 321).
17. Vaino, A. R.; Szarek, W. A. Chem. Commun. 1996, 2351±
2352.
2. Kashiwada, Y.; Nonaka, G.-I.; Nishioka, I. Phytochemistry
1988, 27, 1469±1472 (and references cited).
18. Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic
Synthesis, 2nd ed.; Wiley: New York, 1991.
19. Barros, M. T. B.; Maycock, C. D. M.; Thomassigny, C.
Carbohydr. Res. 2000 (in press).
3. Potier, P.; Maccario, V.; Giudicelli, M.-B.; Queneau, Y.;
Dangles, O. Tetrahedron Lett. 1999, 40, 3387±3390 (and references
cited).
20. Khan, R.; Patel, G. Carbohydr. Res. 1987, 162, 209±215.
È
4. Khanbabaee, K.; Lotzerich, K. J. Org. Chem. 1998, 63, 8723±
8728 (and references cited).