Hydroxymethyl rotamer populations in disaccharides

Article abstract of DOI:10.1021/jo026913o

Sixteen methyl glucopyranosyl glucopyranoside disaccharides (methyl β-D-Glcp(p-Br-Bz)-(1→x)-β/ α-D-Glcp) containing β-glycosidic linkages (1→2, 1→3, 1→4, and 1→6) were synthesized and analyzed by means of CD and NMR spectroscopy in three different solvent

Full text of DOI:10.1021/jo026913o

Hyd r oxym eth yl Rota m er P op u la tion s in Disa cch a r id es
Alfredo Roe¨n, J uan I. Padro´n, and J esu´s T. Va´zquez*
Instituto Universitario de Bio-Orga´nica “Antonio Gonza´lez”, Universidad de La Laguna,
Avda. Astrofı´sico Fco. Sa´nchez 2, 38206 La Laguna, Tenerife, Spain
jtruvaz@ull.es
Received December 27, 2002
Sixteen methyl glucopyranosyl glucopyranoside disaccharides (methyl â-D-Glcp(p-Br-Bz)-(1fx)-â/
R-D-Glcp) containing â-glycosidic linkages (1f2, 1f3, 1f4, and 1f6) were synthesized and analyzed
by means of CD and NMR spectroscopy in three different solvents. For each of these four types of
disaccharides, a correlation was observed between the hydroxymethyl rotational populations around
the C5-C6 bond of the glucopyranosyl residue II with the substituents and the anomeric
configuration of the methoxyl group in residue I, as well as with the solvent. Nonbonded interactions,
the stereoelectronic exo-anomeric effect, and hydrogen bonding were found to be responsible for
the observed rotameric differences. Whereas the rotational populations of the (1f6)-linked
disaccharides are mainly dependent on the exo-anomeric effect, the (1f2)-bonded disaccharides
are strongly dependent on the anomeric configuration at C1, and the (1f3)- and (1f4)-linked
disaccharides are mainly dependent on the substituents and the solvent. The population of the gt
rotamer decreases as nonbonded interactions increase but increases as the exo-anomeric effect
becomes greater, as well as in the presence of intramolecular hydrogen bonding to the endocyclic
oxygen O5′. Comparison of the hydroxymethyl rotational preferences between our model disac-
charides revealed a dependence on the glycosidic linkage type. Thus the population of the gg and
gt rotamers decreases/increases from (1f2)- (â series), to (1f6)-, to (1f2)- (R series), to (1f4)-,
and to (1f3)-bonded disaccharides respectively, while the tg rotamer population remains almost
constant (around 20%), except for the (1f3)- and (1f4)-linked disaccharides with the intramolecular
hydrogen bonding to O5′, where this population decreases to 10%.
In tr od u ction
monosaccharides, as the factors governing their rotamer
populations are still not fully understood.
Carbohydrates are compounds of tremendous impor-
tance in nature due to their biological functions, many
of them being involved in recognition events. The in-
volvement of these molecules in biological events, either
alone or covalently linked to proteins or lipids, has led
to a new research area at the chemistry-biology inter-
face, called glycoscience or glycobiology. To understand
these biological events from a molecular point of view,
not only their three-dimensional structure but also their
conformational preferences in solution must be known.
Great difficulty may be encountered in determining the
conformation of an oligosaccharide, because of the flex-
ibility of the glycosidic linkages and the rotation of the
hydroxymethyl and other pendant groups. Although
many theoretical and experimental studies on the rota-
tional preferences of the hydroxymethyl group have been
carried out,1-24 most of them were performed with
The conformation of a disaccharide in solution depends
mainly on the rotations about its glycosidic linkage;
(8) Hoffmann, M.; Rychlewski, J . J . Am. Chem. Soc. 2001, 123, 2308.
(9) Spieser, S. A. H.; Kuik, J . A. v.; Kroon-Batenburg, L. M. J .;
Kroon, J . Carbohydr. Res. 1999, 322, 264.
(10) Senderowitz, H.; Parish, C.; Still, W. C. J . Am. Chem. Soc. 1996,
118, 2078.
(11) (a) Liu, H.-W.; Nakanishi, K. J . Am. Chem. Soc. 1981, 103, 5591.
(b) Liu, H. W.; Nakanishi, K. J . Am. Chem. Soc. 1982, 104, 1178.
(12) Nishida, Y.; Ohrui, H.; Meguro, H. Tetrahedron Lett. 1984, 25,
1575.
(13) Ohrui, H.; Nishida, Y.; Watanabe, M.; Hori, H.; Meguro, H.
Tetrahedron Lett. 1985, 26, 3251.
(14) DeVries, N.; Buck, H. M. Carbohydr. Res. 1987, 165, 1.
(15) Ohrui, H.; Nishida, Y.; Higuchi, H.; Hori, H.; Meguro, H. Can.
J . Chem. 1987, 65, 1145.
(16) Nishida, Y.; Hori, H.; Ohrui, H.; Meguro, H. J . Carbohydr.
Chem. 1988, 7, 239.
(17) Hori, H.; Nishida, Y.; Ohrui, H.; Meguro, H.; Uzawa, J .
Tetrahedron Lett. 1988, 29, 4457.
(18) Nishida, Y.; Hori, H.; Ohrui, H.; Meguro, H. Uzawa, J .; Reimer,
D.; Sinnwell, V.; Paulsen, H. Tetrahedron Lett. 1988, 29, 4461.
(19) Poppe, L. J . Am. Chem. Soc. 1993, 115, 8421.
(20) Barrows, S. E.; Storer, J . W.; Cramer, C. J .; French, A. D.;
Truhlar, D. G. J . Comput. Chem. 1998, 19, 1111.
(21) Yamada, H.; Harada, T.; Takahashi, T. Tetrahedron Lett. 1995,
36, 3185.
(1) Review: Bock, K.; Duus, J . J . Carbohydr. Chem. 1994, 13, 513.
(2) Tvarosˇka, I.; Taravel, F. R., Utille, J . P.; Carver, J . P. Carbohydr.
Res. 2002, 337, 353.
(3) Kirschner, K. N.; Woods, R. J . Proc. Natl. Acad. Sci. U.S.A. 2001,
98, 10541.
(4) Molteni, C.; Parrinello, M. J . Am. Chem. Soc. 1998, 120, 2168.
(5) Brown, J . W.; Wladkowski B. D. J . Am. Chem. Soc. 1996, 118,
1190.
(22) Hori, H.; Nishida, Y.; Ohrui, H.; Meguro, H. J . Carbohydr.
Chem. 1990, 9, 601.
(6) Tvarosˇka, I.; Carver, J . P. J . Phys. Chem. B 1997, 101, 2992.
(7) Rockwell, G. D.; Grindley, T. B. J . Am. Chem. Soc. 1998, 120,
10953.
(23) De Bruyn, A.; Anteunis, M. Carbohydr. Res. 1976, 47, 311.
(24) J ansson, P.-E.; Kenne, L.; Kolare, I. Carbohydr. Res. 1994, 257,
163.
10.1021/jo026913o CCC: $25.00 © 2003 American Chemical Society
Published on Web 05/13/2003
J . Org. Chem. 2003, 68, 4615-4630
4615

Roe¨n et al.
CHART 1
of the hydroxymethyl group also depends on the anomeric
configuration.30,31
Since in our previous studies NMR and CD proved to
be a good experimental tandem for this type of confor-
mational study27-31 and a general experimental study of
the rotational dependence of the hydroxymethyl group
on the glycosidic linkage has not been carried out so far,
our aim in this work is to study the rotational preferences
of hydroxymethyl groups in disaccharide derivatives in
solution and to provide experimental evidence on all the
factors affecting the rotamer population around the C5-
C6 bond for each of the glycosidic linkages analyzed.
Because disaccharides are repeating units in oligo- and
polysaccharides, knowledge of the factors governing their
conformational preferences can be of great help in
determining the conformation of oligosaccharides.
The present study performed with methyl O-(2,3,4,6-
t et r a k is-O-(p-br om oben zoyl)-â-D-glu copyr a n osyl)-
(1f2)-, (1f3)-, (1f4)-, and (1f6)-â-D- and -R-D-glucopy-
ranosides demonstrates the existence of a rotational
population dependence of the hydroxymethyl group in the
glucopyranosyl residue ΙΙ on the glycosidic linkage type,
as well as on the structural nature of the substituents
in residue Ι. Chemical transformations of these disac-
charides and CD and NMR analyses in protic and aprotic
solvents show how the rotamer population depends on
stereoelectronic effects (for all linkage types), on non-
bonded interactions (for 1f2, 1f3, and 1f4 disaccha-
rides), and on the intramolecular hydrogen bonding to
O5′ (only for disaccharides having 1f3 and 1f4 link-
ages). In addition, the (1f2)-bonded disaccharides exhibit
a strong dependence on the anomeric configuration of the
methoxyl group at C1.
F IGURE 1. (Top) Torsion angles φ and ψ, around the
glycosidic linkages, and torsion angles ω around the C5-C6
bonds. (Bottom) Newman projections of the gg (ω ) -60°), gt
(ω ) 60°), and tg (ω ) 180°) rotamers around the C5-C6 bond.
therefore, the relative orientations of saccharide units are
expressed in terms of the glycosidic linkage torsion angles
φ (O5′-C1′-O-Cx) and Ψ (C1′-O-Cx-C(x-1)), for a
1-x linkage. Besides NMR25 and X-ray diffraction, mo-
lecular modeling of carbohydrates has been devised as
an important tool for structural studies of these com-
pounds, which permits evaluation of the range of attain-
able conformations in terms of the potential energy at
each point specified by a pair of φ and Ψ.26 In addition
to the torsion angles φ and ψ, a third torsion angle ω
(O5-C5-C6-O6) needs to be considered when the hy-
droxymethyl group is involved in the linkage (see Figure
1). This torsion angle is also used to describe the
conformation of unsubstituted hydroxymethyl groups.
The conformation of the hydroxymethyl group around the
C5-C6 bond is generally described by means of the
populations of the gauche-gauche (gg), gauche-trans
(gt), and trans-gauche (tg) rotamers (see Figure 1). The
first descriptor indicates the torsional relationship be-
tween O6 and O5, and the second is that between O6
and C4.
Our research in this field has confirmed the existence,
on the basis of CD and NMR data, of a rotational
population dependence of the hydroxymethyl group in
gluco-27,28 and galactopyranosides29 on the aglycon and
its absolute configuration, revealing a clear correlation
between the rotamer distributions and the stereoelec-
tronic exo-anomeric effect. Furthermore, low-temperature
CD measurements with the alkyl galactopyranosides
confirmed that the most stable rotamer is the gt and not
the tg, as previously reported from CD results.11 Ad-
ditionally, a comparative study between anomers of alkyl
glucopyranosides revealed that the rotational population
Resu lts a n d Discu ssion
Syn t h esis. The model disaccharides used in the
present spectroscopic work contain exciton-coupled chro-
mophores, namely, p-bromobenzoate esters, to facilitate
the analyses by CD and, in addition, because these
groups affect the proton and carbon resonances where
they are located, leading therefore to less crowded NMR
spectra, allowing the coupling constants under study to
be measured accurately by means of a first-order NMR
analysis (ABX instead of ABC spin system). They were
synthesized (Schemes 1-4) in moderate to good yields
by coupling different glucosyl acceptors obtained from
commercially available methyl â-D-glucopyranoside (1)
with the same glucosyl donor: 2,3,4,6-tetrakis-O-(p-
(25) Duus, J . Ø.; Gotfredsen, C. H.; Bock, K. Chem. Rev. 2000, 100,
4589.
(26) Imberty, A.; Pe´rez, S. Chem. Rev. 2000, 100, 4567.
(27) Va´zquez, J . T. Recent Applications of Circular Dichroic to
Carbohydrate Conformational Analysis and Direct Determination of
Drug Levels. In The Biology-Chemistry Interface; Cooper, R., Snyder,
J . K., Eds.; Marcel Dekker: New York, 1999.
(28) Morales, E. Q.; Padro´n, J . I.; Trujillo, M.; Va´zquez, J . T. J . Org.
Chem. 1995, 60, 2537.
(29) Padro´n, J . I.; Morales, E. Q.; Va´zquez, J . T. J . Org. Chem. 1998,
63, 8247.
(30) Padro´n, J . I.; Va´zquez, J . T. Chirality 1997, 9, 626.
(31) Padro´n, J . I.; Va´zquez, J . T. Tetrahedron: Asymmetry 1998, 9,
613.
4616 J . Org. Chem., Vol. 68, No. 12, 2003

Hydroxymethyl Rotamers in Disaccharides
SCHEME 1. Syn th esis of Mod el (1f6)-Lin k ed
SCHEME 3. Syn th esis of Mod el (1f4)-Lin k ed
Disa cch a r id esa
Disa cch a r id esa
a
t
Conditions: (a) Bu(Ph)2SiCl, imidazole, dry DMF; (b) BnBr,
NaH, dry DMF; (c) (nBu)4NF‚3H2O, dry THF; (d) AgOTf, TMU,
dry CH2Cl2, -40 °C; (e) anhydrous FeCl3, dry CH2Cl2, 0 °C; (f)
Ac2O/Py; (g) anhydrous FeCl3, dry CH2Cl2; (h) p-TsOH, CH2Cl2/
MeOH (1:1).
SCHEME 2. Syn th esis of Mod el (1f2)-Lin k ed
a
Conditions: (a) PhCH(OCH3)2, p-TsOH, dry DMF, 50 °C,
Disa cch a r id esa
vacuum; (b) Ac2O/Py; (c) Na(CN)BH3, CF3CO2H, dry THF; (d)
AgOTf, sym-collidine, PhCH3/CH3NO2 (1:1), -40 °C; (e) anhydrous
FeCl3, dry CH2Cl2, 0 °C; (f) Ac2O/Py; (g) p-TsOH, CH2Cl2/MeOH
(1:1); (h) anhydrous FeCl3, dry CH2Cl2.
SCHEME 4. Syn th esis of Mod el (1f3)-Lin k ed
Disa cch a r id esa
a
a
Conditions: (a) AgOTf, dry PhCH3, -40 °C; (b) anhydrous
Conditions: (a) AgOTf, TMU, CH2Cl2, -40 °C; (b) (n-Bu)4NF,
FeCl3, dry CH2Cl2, 0 °C; (c) Ac2O/Py; (d) anhydrous FeCl3, dry
pyridinium chlorhydrate, dry THF; (c) Ac2O/Py; (d) anhydrous
CH2Cl2; (e) p-TsOH, CH2Cl2/MeOH (1:1).
FeCl3, dry CH2Cl2; (e) p-TsOH, CH2Cl2/MeOH (1:1).
silver triflate (AgOTf) as catalyst and 1,1,3,3-tetra-
methylurea (TMU) as proton acceptor in CH2Cl2 at -40
°C,32c the â-(1f6)-bonded disaccharide 6 was obtained
with a 92% yield. This compound was debenzylated34 with
anhydrous FeCl3 in dry CH2Cl2 at 0 °C to give compound
7 (76%), which was acetylated to obtain the model
â-(1f6)-linked disaccharide 8. See below for the synthesis
of the R-anomers 9 and 10.
The â-(1f2)-linked disaccharide 12 (Scheme 2) was
obtained in moderate yield (52%) by coupling the glucosyl
donor 2 and the glucosyl acceptor 11 under the modified
Koenings-Knorr method32b and taking advantage of the
bulkiness of the protecting group present in 11, which
bromobenzoyl)-R-D-glucopyranosyl bromide (2)28 by means
of modified Koenigs-Knorr methods.32
For the synthesis of the â-(1f6)-linked disaccharides,
the glucosyl acceptor 5 (Scheme 1) was obtained in three
steps by protection of the primary hydroxyl group with
tBu(Ph)2SiCl and imidazole in DMF,33 then perbenzyla-
tion of the secondary hydroxyl groups with benzyl
bromide and sodium hydride in DMF, and finally depro-
tection of the silyl group with tetra-n-butylammonium
fluoride (TBAF) in THF. By coupling the resulting
glucosyl acceptor (5) with the glucosyl donor (2) and using
(32) (a) Hanessian, S.; Banoub, J . Carbohydr. Res. 1977, 53, C13.
(b) Banoub, J .; Bundle, D. R. Can. J . Chem. 1979, 57, 2091. (c) Garegg,
P. J .; Norberg, T. Acta Chem. Scand. 1979, B33, 116. (d) Preparative
Carbohydrate Chemistry; Hanessian, S., Ed.; Marcel Dekker: New
York, 1997.
(34) (a) Park, M. H.; Takeda, R.; Nakanishi, K. Tetrahedron Lett.
1987, 28, 3823. (b) Ikemoto, N.; Kim, O. K.; Lo, L. C.; Satyanarayana,
V.; Chang, M.; Nakanishi, K. Tetrahedron Lett. 1992, 33, 4295. (c)
Padro´n, J . I.; Va´zquez, J . T. Tetrahedron: Asymmetry 1995, 6, 857.
(33) Wood, W. W.; Rashid, A. Tetrahedron Lett. 1987, 28, 1933.
J . Org. Chem, Vol. 68, No. 12, 2003 4617

Roe¨n et al.
deeply decreased the reactivity of the free hydroxyl group
at C3. This acceptor was obtained in 66% yield from 1
by treatment with 1,3-dichloro-1,1,3,3-tetraisopropyl-
disiloxane (TIPDSCl2) in dry pyridine.35 The disaccharide
12 was deprotected with TBAF and pyridinium chlorhy-
drate in dry THF to provide the model disaccharide 13
(92%), which was acetylated to give its corresponding
triacetate 14.
The glucopyranosyl acceptor 19, used for the synthesis
of the model â-(1f4)-linked disaccharides, was obtained
with a good yield in three steps (Scheme 3). Benzylide-
nation of methyl â-D-glucopyranoside (1) with benzalde-
hyde dimethyl acetal and p-TsOH, acetylation of the
resulting 4,6-O-benzylidene 17, and regioselective open-
ing of the benzylidene group in the 2,3-di-O-acetyl-4,6-
O-benzylidene derivative 18 with sodium cyanoborohy-
dride and trifluoroacetic acid36 afforded the desired
acceptor 19. The Koenigs-Knorr coupling reaction be-
tween 19 and 2 using as solvent a mixture of toluene and
nitromethane (1:1)32a led to the disaccharide 20 with a
73% yield. Subsequent debenzylation with anhydrous
FeCl3 in CH2Cl2 at 0 °C gave compound 21 (95% yield),
which was acetylated to give the triacetyl derivative 22.
Finally, chemoselective deprotection37 of the acetyl groups
against the benzoyl groups, under catalytic acid condi-
tions (p-TsOH‚H2O) in a mixture of MeOH/CH2Cl2 (1:1),
led to a good yield (77%) of the triol 23.
F IGURE 2. Illustration of the configurational correlation and
the H NMR anisotropic chemical shifts for alkyl â-D-glucopy-
ranosides.
1
three clear cross-peaks between the anomeric proton H1′
and H3′, H5′, and HX. The anomeric configurations were
assigned in each case by measuring the coupling constant
between H1 and H2 for each glucopyranosidic ring
(CDCl3, doublet; â-configuration 7.4-8.1 Hz, R-configu-
ration 3.3-3.8 Hz) and confirmed by the chemical shift
in 13C NMR (â-configuration 100.2-104.8 Hz; R-config-
uration 96.3-99.4 Hz).39 Infrared spectra of the triol
disaccharides 23 and 28 were recorded in anhydrous
acetonitrile over a range of concentrations in order to
confirm the existence of intramolecular hydrogen bond-
ing. Since no spectral changes were observed over the
concentration range used, the two sharp bands observed
around 3630 and 3540 cm-1 were assigned, respectively,
to free OH and to intramolecular hydrogen bonding OH-
stretching vibrations.40
In addition,because the tetra-O-benzoyl-â-glucosylation
of alcohols induces dramatic shifts in the aglycon 1H
NMR peaks,28-30,41,42 some proton signals in residue Ι
were shielded or deshielded, depending on whether the
corresponding protons are located anti or syn, respec-
tively, to the endocyclic oxygen O5, giving rise to further
structural information (Figure 2). Thus, for example,
protons located syn to the benzoyl group at C2′ in our
model (1f6)-linked disaccharides (Me, H1) exhibited
chemical shifts at upper field (∆δ -), while H6S located
near O5′ exhibited them at lower field (∆δ +). On the
other hand, for the (1f2)-linked disaccharides the signals
of the Me group and H1 were located at lower field (∆δ
+) and H3 at upper field. In addition, substantial
shieldings were observed for the acetyl group located at
C3, with chemical shifts at upper field, up to δ 1.45 in
the case of compound 16. Similarly, for the (1f4)-linked
disaccharides shielded (∆δ -) and deshielded (∆δ +)
chemical shifts were observed for protons H6R and H3,
respectively. In the case of the (1f3)-linked disaccha-
rides, deshielded chemical shifts were observed for H4.
In the case of â-(1f3)-bonded disaccharides (Scheme
4), partial benzylation of the 4,6-O-benzylidene derivative
17, with 1.1 equiv of benzyl bromide and sodium hydride
in DMF, gave a mixture of compounds having the benzyl
group at C-2 or C-3 (1.2/1 ratio). These were separated
by chromatography, a low yield (23%) of compound 26
being isolated. Then, the glucopyranoside 26 was coupled
with the glucosyl donor (2) to obtain the disaccharide 27
(45%), which by treatment with anhydrous FeCl3 at room
temperature led directly to the desired triol 28 (85%).
Acetylation of 28 produced the model (1f3)-linked di-
saccharide 29.
The synthesis of the disaccharides having an R con-
figuration at the glycosidic linkage supporting the meth-
oxyl group, compounds 9, 15, 24, and 30 (Schemes 1-4),
were obtained in high yields from the corresponding
â-anomers, compounds 8, 14, 22, and 29, by treatment
with anhydrous FeCl3 in CH2Cl2 at room temperature.34
Deprotection of the acetyl groups37 under catalytic acid
conditions (p-TsOH‚H2O) in a mixture of MeOH/CH2Cl2
(1:1) led finally to the triols 10, 16, 25, and 31 with good
yields.
1
The H NMR signals of the prochiral protons at C6,
H6R, and H6S were differentiated according to the data
in the literature,1,12,13,16 namely, on the basis of their
chemical shifts and coupling constants (accuracy (0.1
Hz). In general, for the D-gluco-series saccharides, H6R
proton signals appear at a higher field than H6S signals
(δH6S > δH6R) and J H5,H6R coupling constants have higher
values than J H5,H6S. Different types of Karplus equations
Ch a r a cter iza tion a n d Sp ectr oscop ic An a lysis. All
of these compounds38 were characterized on the basis of
their one- (1H and 13C) and two-dimensional (COSY-G,
HMQC, and T-ROESY) NMR spectra. The type of glyco-
sidic linkage of model compounds was confirmed by
means of the T-ROESY experiments, i.e., by observing
(35) (a) Verdegaal, C. H. M.; J ansse, P. L.; de Rooij, J . F. M.; van
Boom, J . H. Tetrahedron Lett. 1980, 21, 1571. (b) van Boeckel, C. A.
A.; van Boom, J . H. Tetrahedron 1985, 41, 4545.
(36) J ohansson, R.; Samuelsson, B. J . Chem. Soc., Perkin Trans. 1
1984, 2371.
(37) Gonza´lez, A. G.; Brouard, I.; Leo´n, F.; Padro´n, J . I.; Bermejo,
J . Tetrahedron Lett. 2001, 42, 3187.
(38) The nomenclature is given as proposed by the IUPAC-IUBMB
J oint Commission on Biochemical Nomenclature; http://www.chem-
.qmul.ac.uk/iupac/2carb/.
(39) (a) Bock, K.; Lundt, I.; Pedersen, C. Tetrahedron Lett. 1973,
13, 1037. (b) Usui, T.; Yamaoka, N.; Matsuda, K.; Tuzimura, K.
Sugiyama, H.; Seto, S. J . Chem. Soc., Perkin Trans. 1 1973, 2425.
(40) Nakanishi, K.; Solomon, P. H. Infrared Absorption Spectroscopy;
Holden-Day, Inc.: Oakland, 1977.
(41) Trujillo, M.; Morales, E. Q.; Va´zquez, J . T. J . Org. Chem. 1994,
59, 6637.
(42) Nukada, T.; Berces, A.; Whitfield, D. M. J . Org. Chem. 1999,
64, 9030.
4618 J . Org. Chem., Vol. 68, No. 12, 2003

Hydroxymethyl Rotamers in Disaccharides
have been proposed16,43-45 to calculate the rotamer popu-
3
lations of the hydroxymethyl group from the J H5,H6
coupling constants, those reported by Nishida et al.16 and
by Haasnoot et al.43 being widely used. These approaches
have also been applied to disaccharides.13,17,46,47 In a
comparative study of the various strategies used for
obtaining these values, Bock and Duus1 were unable to
decide which method gives the best results, although they
suggested the use of that of Haasnoot et al.43 In addition,
they showed that although the values of the coupling
constants reported by Nishida et al.16 or those by Manor
et al.44 do not give negative populations, unlike the other
types of Karplus equations, these values could result in
an overestimation of the population of gg relative to gt
and commented that the values reported by Manor et al.44
might not be appropriate as they are based on coupling
in strained five-membered rings. Rockwell and Grindley7
almost eliminate in monosaccharides the problem of
negative populations by using the values of the limiting
coupling constants from nonstaggered geometries calcu-
lated by MM3. Serianni and co-workers45 have just
F IGURE 3. 2/6, 3/6, and 4/6 pairwise interactions involving
the chromophore at the 6 position in each of the three stable
rotamers (gg, gt, and tg) for the glucopyranosyl system.
proposed new limiting values for J H5,H6R and J H5,H6S
,
the 2/6, the 3/6, and the 4/6 pairwise interactions that
involve the chromophore at the 6 position. Because the
2/3 and 3/4 pairwise interactions possess equal magni-
tudes and opposite signs and no ring distortion has been
observed for these model compounds, the CD contribution
of these interactions to the total CD spectrum is nil.
Therefore, the CD spectra of these compounds come from
the pairwise interactions involving the chromophore at
position 6, and because there are three main rotamers
around the C5-C6 bond, the gg, gt, and tg rotamers, the
2/6, 3/6, and 4/6 pairwise interactions must be considered
as nine in order to interpret the CD spectral differences
correctly (see Figure 3).28,30,31 Note that the net CD
contribution for the gg rotamer for the 2/6, 3/6, and 4/6
interactions is positive, whereas those for the gt and tg
rotamers are negative and nil, respectively. The ampli-
tude (A value) of split CD Cotton effects is defined as A
) ∆ꢀ1 - ∆ꢀ2 where ∆ꢀ1 and ∆ꢀ2 are the intensities of the
first and second Cotton effects.51
based on J -couplings computed from density functional
theory (DFT), and applied them to several mono- and
oligosaccharides. The calculations yielded a more ac-
curate representation of the rotameric populations in
solution and positive values for the tg rotamer population
in all cases. Therefore, we have chosen this last improved
approach for the study of the rotamer populations of our
sixteen disaccharides,48 because the alternative of using
different values of the limiting coupling constants ob-
tained by MM calculations in the set of equations for each
case would introduce another variable into the present
comparative study.
CD An a lysis. Since all model disaccharides contain
exciton-coupled chromophores,49 namely, p-bromoben-
zoates, UV and CD spectroscopy was also used to
characterize these compounds. The intramolecular charge-
transfer band was around 245 nm in the UV, and the
exciton Cotton effects were around 251 and 234 nm in
the CD spectra. In addition, it is well-known that the CD
spectrum of a chromophorically 2,3,4,6-tetra-substituted
glucopyranosyl system is composed of six pairwise inter-
actions:50 three having constant intensity and sign, the
positive 2/3, the negative 3/4, and the nil 2/4 pairwise
interactions; and three with variable intensity and sign,
In accordance with the exciton chirality method,49 the
amplitude of split Cotton effects depends on the inter-
chromophoric distance and the dihedral angle. Because
the amplitude is inversely proportional to the square of
the interchromophoric distance, the 4/6 pairwise interac-
tion contributes more significantly to the observed spec-
tra than the 3/6 or 2/6 pairwise interactions,17,18 and for
the same reason, the gg rotamer CD contribution is
stronger than for the gt rotamer. These points must be
considered in order to understand the positive exciton
coupling observed in our model disaccharides, including
those compounds where the gt rotamer, having a net
negative CD contribution, has the greatest population.
Con for m a tion a l An a lysis of th e Disa cch a r id es.
(43) Haasnoot, C. A. G.; De Leeuw, F. A. A. M.; Altona, C.
Tetrahedron 1980, 36, 2783.
(44) Manor, P. C.; Saenger, W.; Davies, D. B.; J ankowski, K.;
Rabczenko, A. Biochim. Biophys. Acta 1974, 340, 472.
(45) Stenutz, R.; Carmichael, I.; Widmalm, G.; Serianni, A. S. J .
Org. Chem. 2002, 67, 949.
(46) Ohrui, H.; Nishida, Y.; Itoh, H.; Meguro, H. J . Org. Chem. 1991,
56, 1726.
(47) (a) Bock, K.; Refn, S. Acta Chem. Scand. 1987, B 41, 469. (b)
Bock, K.; Pedersen, H. Acta Chem. Scand. 1988, B 42, 190.
(48) Software for the calculation of rotameric populations is available
upon request from E. Q. Morales, Instituto de Investigaciones Quı´mi-
cas, Centro de Investigaciones Cientı´ficas Isla de la Cartuja, c/ Americo
Vespucio s/n, 41092 Seville, Spain. E-mail: ezequiel@cica.es.
(49) For a monograph on exciton CD spectroscopy, see: (a) Harada,
N.; Nakanishi, K. Circular Dichroic Spectroscopy Exciton Coupling in
Organic Stereochemistry; University Science Books: California, 1983.
(b) Nakanishi, K.; Berova, N. The Exciton Chirality Method in Circular
Dichroism, Principles and Applications; Nakanishi, K., Berova, N.,
Woody, R. W., Eds.; VCH Publishers: New York, 1994.
(50) (a) Wiesler, W. T.; Va´zquez, J . T.; Nakanishi, K. J . Am. Chem.
Soc. 1986, 108, 6811. (b) Wiesler, W. T.; Va´zquez, J . T.; Nakanishi, K.
J . Am. Chem. Soc. 1987, 109, 5586.
1
Gen er a l. The values of the H NMR coupling constants
for the prochiral protons at C6′ indicated that in general
gg is the most stable rotamer for the hydroxymethyl
group (residue ΙΙ) in the model disaccharides. For com-
(51) Occasionally the presence of a background ellipticity alters the
intensity of the Cotton effects at short wavelengths. For this reason,
the intensities of the second Cotton effects and the amplitudes (A
values) of the CD spectra of our model compounds may not be precise;
the intensities of the first Cotton effects are thus more accurate for
comparative analysis.
J . Org. Chem, Vol. 68, No. 12, 2003 4619

Roe¨n et al.
these NMR and CD values match those of the monosac-
charide methyl 2,3,4,6-tetrakis-O-(p-bromobenzoyl)-â-D-
glucopyranoside (J H5,H6R ) 4.7, J H5,H6S ) 3.4, in CDCl3;
A ) 29.9, CH3CN; A ) 27.4, CH3OH),28 corroborating this
conclusion. Thus, a general rotamer population of Pgg:
Pgt:Ptg = 45:35:20 can be established for these (1f6)-
bonded disaccharides in chloroform.
In addition, Table 1 shows for both series of anomers
a weak dependence between the solvent and the rotamer
populations; that is, the polar solvents CD3CN and CD3-
OD induce slightly greater/smaller gg/gt populations (Pgg:
Pgt:Ptg = 50:30:20) than the less polar solvent CDCl3 (for
example, see compound 7, entries 2 (CDCl3), 6 (CD3CN),
and 10 (CD3OD)).
F IGURE 4. Perspective view of methyl 6-O-(â-D-glucopyra-
nosyl)-â-D-glucopyranoside.
Independently of the solvent, compounds with the
â-configuration showed slightly more gg and less gt
rotamers than those compounds with the R-configuration.
To confirm this tendency, low-temperature CD measure-
ments of compounds 7 and 10 were performed in metha-
nol. Thus, striking increases were seen in the A values,
from 26.8 to 48.2 in the case of compound 7 and from
26.0 to 42.8 for compound 10, which can only be explained
by a rise in the population of the gg rotamer (with a net
positive CD contribution) and a drop in the population
of the gt rotamer (with a negative CD contribution).
Therefore, low-temperature CD data confirm that the
most stable rotamer is the gg and that those (1f6)-
bonded disaccharides having the â anomeric configura-
tion at C1 led to tiny but appreciable increases/decreases
in gg/gt rotamer populations.
(1f2)-Lin k ed Disa cch a r id es. As can be observed in
Table 2, the rotamer populations obtained for the model
(1f2)-linked disaccharides mainly depend on the ano-
meric configuration of the methoxyl group. In all cases
but one (compound 16 in CDCl3, entry 16), gg was the
most stable rotamer. In addition, compounds having the
methoxyl group in the R-configuration showed larger gt
populations, smaller tg populations, and smaller or
identical gg populations than the corresponding com-
pounds with a â-configuration. Thus, for example, the
populations obtained for compound 16 (entry 16) are Pgg:
Pgt:Ptg ) 42:43:15, whereas those obtained for its â-ano-
mer 13 (entry 13) are 50:29:21. For those compounds
having the â-anomeric configuration, these rotamer
population differences point to the existence of a non-
bonded interaction between the methoxyl group and the
hydroxymethyl group in its gt rotamer and therefore to
a decrease in its population. ROESY experiments per-
formed with disaccharide 14 showed the typical strong
cross-peak between the anomeric proton H1′ and H2,
confirming the 1f2 glycosidic union, and also interresi-
due cross-peaks between the methoxyl group and H1′,
H5′ and the prochiral proton H6′S. This shows the
closeness of their disposition and confirms the nonbonded
interaction (Figure 5).
pounds 16, 20, and 25 (in CDCl3) and 23, 28, and 31 (in
CDCl3 and CD3CN), gt was the most stable rotamer.
Furthermore, an excellent agreement was observed
between the J H5′,H6′ coupling constants in different sol-
vents and the magnitudes obtained by CD. Thus, as the
J H5′,H6′R and the J H5′,H6′S coupling constants increase and
decrease, respectively, the Cotton effects of the split CD
spectra are reduced. Note that an increase in J H5′,H6′R
means an increase in the population of the gt rotamer
(see Figure 1), which has a net negative CD contribution
(see Figure 3), explaining therefore a decrease in the CD
spectrum.
(1f6)-Lin k ed Disa cch a r id es. The T-ROESY experi-
ment performed with the (1f6)-linked disaccharide 7
showed a strong cross-peak for the anomeric H1′ proton
with the prochiral H6R proton and a weak one with H6S.
A strong cross-peak between H5 and H6S was also
detected, but no cross-peak at all was found between H5
and H6R. These experimental data, together with the
chemical shifts and coupling constants of the prochiral
H6 protons (H6S, δ 4.19, J H5,H6S ) 1.9 Hz; H6R, δ 3.71,
J H5,H6R ) 6.9 Hz), revealed that the most stable rotamer
for the hydroxymethyl group involved in the â intergly-
cosidic bond (C6) is gt, with a calculated rotational
population of Pgg:Pgt:Ptg ) 30:65:5, based on the experi-
mental coupling constants (Figure 4). Further confirma-
tion of the fact that the gt rotamer at C6 is the most
stable comes from the observed magnitudes and signs of
the chemical shift changes on glucosylation compound
5.41 A chemical shift difference of ∆δ7-5 ) -0.33 ppm,
induced by the benzoyl group at C2′, was observed for
the methoxyl group, a ∆δ7-5 ) +0.30 ppm for the H6S,
induced by the endocyclic glucopyranoside oxygen (O5′),
and a ∆δ7-5 ) -0.02 ppm for the H6R. Moreover, these
results are completely in agreement with other studies
where two major conformations were found, the gt (66%)
and gg (34%) for â-gentiobiose (â-D-Glcp-(1f6)-â-D-
Glcp),19 and the gt (69%) and gg (31%) for its per-O-
benzoyl derivative.13
The spectroscopic data from the rotational population
study of the model â-(1f6)-bonded disaccharides 6-10,
along with the calculated population of the gg, gt, and tg
rotamers, is summarized in Table 1. For all of these
(1f6)-linked disaccharides, gg was the most stable
rotamer for the hydroxymethyl group at C6′. The values
of the 1H NMR coupling constants for the prochiral
protons at C6′, as well as the CD A values, showed only
minor differences, and therefore negligible variations can
be expected in their rotameric populations. Furthermore,
As occurred with the (1f6)-bonded disaccharides, an
increase in solvent polarity led to an increase/decrease
in the population of the gg/gt rotamers, respectively
(Table 2). This result can be checked by comparing the
coupling constants of prochiral H6 protons or the calcu-
lated rotamer populations of compound 16 (entries 16 and
20).
Compound 13 (R ) H) has a very low solubility in the
solvents employed in the present study (CHCl3, CH3OH,
4620 J . Org. Chem., Vol. 68, No. 12, 2003

Hydroxymethyl Rotamers in Disaccharides
TABLE 1. J H5′,H6′ Cou p lin g Con sta n ts, Ca lcu la ted Rota m er ic P op u la tion s (%) a r ou n d th e C5′-C6′ Bon d (Resid u e ΙΙ),
a n d CD Da ta for th e Mod el (1f6)-Lin k ed Disa cch a r id es 6-10
MeO
(C1)
∆ꢀ (nm)
251/234
A value
-80 °C
no.
compd
R
solventa
J H5′,H6′R
J H5′,H6′S
Pgg
Pgt
Ptg
A value
1
2
3
4
5
6
7
8
9
6
7
8
9
10
7
8
9
10
7
Bn
H
Ac
Ac
H
â
â
â
R
R
â
â
R
R
â
R
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CH3CN
CH3CN
CH3CN
CH3CN
CH3OH
CH3OH
4.7
4.4
4.7
4.7
4.6
4.1
4.3
4.0
4.3
4.0
4.3
3.4
3.3
3.1
3.3
3.2
3.2
3.3
3.4
3.2
3.3
3.4
45
48
46
45
47
52
50
52
50
53
49
34
31
36
35
34
29
30
26
31
27
30
21
21
18
20
19
19
20
22
19
20
21
H
21.8/-6.3
20.3/-7.4
21.4/-6.5
20.4/-7.0
20.2/-6.6
17.4/-8.6
28.1
27.7
27.9
27.4
26.8
26.0
Ac
Ac
H
H
H
10
11
48.2
42.8
10
a
Deuterated solvent for NMR analysis.
TABLE 2. J H5′,H6′ Cou p lin g Con sta n ts, Ca lcu la ted Rota m er ic P op u la tion s (%) a r ou n d th e C5′-C6′ Bon d (Resid u e ΙΙ),
a n d CD Da ta for th e Mod el (1f2)-Lin k ed Disa cch a r id es 12-16
MeO
(C1)
∆ꢀ (nm)
no.
compd
R
solventa
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CH3CN
CH3CN
CH3CN
CH3CN
CH3OH
CH3OH
J H5′,H6′R
J H5′,H6′S
Pgg
Pgt
Ptg
251/234
A value
12
13
14
15
16
17
18
19
20
21
22
12
13
14
15
16
13
14
15
16
13
16
-
H
Ac
Ac
H
â
â
â
R
R
â
â
R
R
â
R
4.4
4.2
4.3
4.8
5.3
3.0
3.4
3.6
3.0
2.8
50
50
47
46
42
33
29
29
37
43
17
21
24
17
15
b
H
23.9/-6.9
26.5/-7.9
19.3/-6.2
19.4/-6.7
21.5/-7.5
20.6/-7.1
30.8
34.4
25.5
26.1
29.0
27.7
Ac
Ac
H
H
H
3.7
4.4
4.3
3.4
2.9
3.0
55
51
51
23
33
32
22
16
17
4.5
3.2
48
33
19
a
b
Deuterated solvent for NMR analysis. CDCl3 /CD3OD 5%.
structural nature of the substituents located syn, but not
anti, to the endocyclic oxygen (O5′) and the rotamer
populations (Table 3). Thus, compounds 20-22, having
the same substituents in the syn disposition (acetyl) but
different in the anti (R ) Bn, H, Ac), exhibited very
similar rotamer populations in CDCl3 (entries 23-25),
whereas the unprotected compound 23 (R ) H) showed
an increase in the gt and a decrease in the tg rotamer
population in this solvent (entry 26). The same behavior,
either in CDCl3 or CD3CN, occurs when the methoxyl
group has the R-configuration; for example, the popula-
tions observed for the triacetate 24 (entry 27) Pgg:Pgt:Ptg
) 48:34:18 changed to 41:51:8 in the triol 25 (entry 28).
CD data (CH3CN) support these rotational populations;
the lower A value found for compound 25 (18.5, entry 32)
than for 24 (23.8, entry 31) means less gg and more gt.
F IGURE 5. Perspective view of methyl 2-O-(â-D-glucopyra-
nosyl)-â-D-glucopyranoside.
and CH3CN), making the NMR spectra difficult to obtain.
Unlike the other disaccharides synthesized in this work,
this one presents completely different polarity at each
end of the glucopyranosidic rings, which explains this
physical property. However, it was possible to obtain the
corresponding CD data in CH3OH and CH3CN as a result
of the extreme sensitivity of this technique, which allows
very dilute solutions to be analyzed. Thus, disaccharide
13 exhibited A values of 30.8 (CH3CN) and 29.0 (CH3-
OH), higher in magnitude than those obtained for its
R-anomer 16 (A values 26.1 and 27.7, respectively).
Therefore, these higher A values obtained for the â-ano-
mers are in total agreement with higher gg (net positive
CD contribution) and lower gt (net negative) populations
for these disaccharides compared to their corresponding
R-anomers.
Although the above rotamer population differences
could be explained by nonbonded interactions, analysis
of the populations obtained in different solvents revealed
a new feature. On changing to the protic solvent CD3-
OD, the unprotected disaccharides 23 and 25 showed
similar gg, decreased gt, and increased tg populations
compared to those obtained when aprotic solvents were
used (compare entries 26 and 33, or 28 and 34). CD also
detected the different behavior of compounds 23 and 25
with the protic nature of the solvent. They exhibited A
values of +18.9 and +18.5 in acetonitrile while showing
+23.4 and +22.1 in methanol, these higher intensities
being in agreement with decreased gt. Therefore, all of
these spectroscopic data point to two different effects
probably acting simultaneously on the rotamer popula-
(1f4)-Lin k ed Disa cch a r id es. Analysis of NMR data
for the model â-anomers showed a dependence on the
J . Org. Chem, Vol. 68, No. 12, 2003 4621

Roe¨n et al.
TABLE 3. J H5′,H6′ Cou p lin g Con sta n ts, Ca lcu la ted Rota m er ic P op u la tion s (%) a r ou n d th e C5′-C6′ Bon d (Resid u e ΙΙ),
a n d CD Da ta for th e Mod el (1f4)-Lin k ed Disa cch a r id es 20-25
MeO
(C1)
∆ꢀ (nm)
251/234
no.
compd
R
solventa
J H5′,H6′R
J H5′,H6′S
Pgg
Pgt
Ptg
A value
23
24
25
26
27
28
29
30
31
32
33
34
20
21
22
23
24
25
22
23
24
25
23
25
Bn
H
Ac
H
Ac
H
Ac
H
â
â
â
â
R
R
â
â
R
R
â
R
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CH3CN
CH3CN
CH3CN
CH3CN
CH3OH
CH3OH
5.3
5.2
5.0
5.7
4.6
5.7
4.8
5.4
4.6
5.3
5.1
4.8
2.9
3.0
3.2
2.5
3.1
2.2
3.1
2.6
3.0
1.9
3.2
3.1
41
42
43
39
48
41
45
42
48
48
41
45
43
41
38
49
34
51
37
45
35
47
40
37
16
17
19
12
18
8
18
13
17
5
17.3/-6.3
14.7/-4.2
16.2/-7.6
14.3/-4.2
18.8/-4.6
16.0/-6.1
23.6
18.9
23.8
18.5
23.4
22.1
Ac
H
H
19
18
H
a
Deuterated solvent for NMR analysis.
tions of (1f4)-bonded disaccharides, namely, nonbonded
interactions and intramolecular hydrogen bonding. The
observed differences between compounds 22 and 23 in
acetonitrile could be explained by a nonbonded interac-
tion between the acetyl group at C3 and the hydroxy-
methyl group in compound 22 (R ) Ac), absent in
compound 23 (R ) H). However, the differences observed
for compound 23 by changing the protic nature of the
solvent point to the hydrogen bond (O3H‚‚‚O5′, see below)
favoring the gt population.
Infrared spectra of the triol disaccharide 23 carried out
in anhydrous acetonitrile at different concentrations
confirmed the existence of intramolecular hydrogen
bonding. In addition, a simple experiment performed by
NMR confirmed this along with its influence on the
F IGURE 6. Lowest energy conformation of model methyl 4-O-
(2,3,4,6-tetrakis-O-acetyl-â-D-glucopyranosyl)-â-D-glucopyrano-
side.
approximately 8 kcal‚mol-1 coming from the former
hydrogen bond. The distance between the oxygen atoms
involved in the seven-membered O3H‚‚‚O5′ hydrogen
bond was found to be 2.67 Å and its angle was 157°, and
the distance for the nine-membered hydrogen bond was
3.09 Å with an angle of 161°. An almost identical
structure, but without the O6H‚‚‚O2′ hydrogen bond, was
found when the calculations were performed using a high
dielectric constant (ꢀ ) 80), proving the strength of the
O3H‚‚‚O5′ hydrogen bond. When this structure (gt rota-
mer at C6′) was minimized with the hydroxymethyl in
the gg disposition, an increase of 1.6 kcal‚mol-1 (ꢀ ) 80)
was obtained. In general, use of force fields does not lead
to a good correlation for the rotamer populations of the
hydroxymethyl group in solution, but recent studies with
monosaccharides2-4 have shown that a correct prediction
of the rotamer populations in solution could only be
obtained when solvent contributions were included in the
calculations, a rather complicated study to perform with
our disaccharides. In any case, the structures obtained
for these two rotamers included the O3H‚‚‚O5′ hydrogen
bond, which supports the CD and NMR data interpreta-
tion given in terms of the protic nature of the solvent.
(1f3)-Lin k ed Disa cch a r id es. Analysis of the cou-
pling constants of the prochiral protons at C6′, or their
calculated rotamer populations, clearly reveals (Table 4)
a relationship with the structure of these (1f3)-linked
disaccharides, independently of the solvent. Thus, for
example, acetylation of the disaccharide 28, having a
ratio Pgg:Pgt:Ptg ) 32:55:13 (entry 36), led to the triacety-
lated disaccharide 29, which showed very different values
of 44:36:20 (entry 37). Furthermore, CD data show a total
agreement with the NMR data, compound 28 (entry 40)
1
rotamer population. The H NMR spectrum of compound
23 in CD3CN showed the three corresponding peaks of
the hydroxylic protons, which were characterized by 2D-
NMR experiments. These peaks gradually disappeared
by successively adding a few microliters of CD3OD into
the NMR tube. After the first addition, the NMR signal
from the hydroxylic proton at C2 disappeared completely,
that from the one at C6 diminished, and that correspond-
ing to the hydroxyl group at C3 remained almost unal-
tered, having the highest downfield chemical shift.
Successive additions led first to a pronounced drop in the
signal from the hydroxylic proton at C6 and to a lesser
decrease of that at C3 and then to the complete disap-
pearance of these signals. In addition, the J H5′,H6′R
coupling constant decreased from 3.4 to 3.2 Hz during
these additions. The results of this NMR experiment are
completely in concordance with the increase in the gt
rotamer in the presence of the intramolecular hydrogen
bond (O3H‚‚‚O5′).
MM calculations52 were performed using methyl 4-O-
(2,3,4,6-tetra-O-acetyl-â-D-glucopyranosyl)-â-D-glucopyra-
noside as model to test for hydrogen bonding. The
structure with lowest energy (Figure 6) found using the
default dielectric constant (ꢀ ) 1.5) exhibited two in-
tramolecular hydrogen bonds: (O3H‚‚‚O5′) and (O6H‚‚‚
O2′), the total stabilization energy being ca. 9.3 kcal‚mol-1
,
(52) The MMX force field, including the extra term for hydrogen
bonding, was used to perform the molecular mechanics calculations.
PCMODEL (v. 7.0). Serena Sofware. Since the conformations of
acetates are similar to those of benzoates, and calculations do not need
the more complex π-minimization, acetates were used instead of the
more complicated per-p-bromobenzoates.
4622 J . Org. Chem., Vol. 68, No. 12, 2003

Hydroxymethyl Rotamers in Disaccharides
TABLE 4. J H5′,H6′ Cou p lin g Con sta n ts, Ca lcu la ted Rota m er ic P op u la tion s (%) a r ou n d th e C5′-C6′ Bon d (Resid u e II),
a n d CD Da ta for th e Mod el (1f3)-Lin k ed Disa cch a r id es 27-31
MeO
(C1)
∆ꢀ (nm)
251/234
no.
compd
R
solventa
J H5′,H6′R
J H5′,H6′S
Pgg
Pgt
Ptg
A value
35
36
37
38
39
40
41
42
43
44
45
27
28
29
30
31
28
29
30
31
28
31
â
â
â
R
R
â
â
R
R
â
R
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CH3CN
CH3CN
CH3CN
CH3CN
CH3OH
CH3OH
4.5
6.3
4.8
4.9
6.4
5.7
4.8
3.6
2.6
3.3
3.1
2.4
2.3
3.2
45
32
44
44
32
41
45
31
55
36
38
57
50
36
24
13
20
18
11
9
H
Ac
Ac
H
H
14.4/-4.4
15.4/-7.0
14.2/-6.7
13.1/-4.5
18.1/-6.4
16.5/-6.3
18.8
22.4
20.9
17.6
24.5
22.8
Ac
Ac
H
H
H
19
5.8
5.0
5.0
2.3
3.2
3.4
40
43
41
51
38
38
9
19
21
a
Deuterated solvent for NMR analysis.
constant of 1.5 or 80, respectively. The distance between
the oxygen atoms involved in this hydrogen bond was
found to be 2.65 Å and the angle O4H‚‚‚O5′ was 153°.
Therefore, this result agrees with the observed experi-
mental dependence of this type of disaccharide on the
protic (CH3OH) or aprotic (CHCl3 and CH3CN) nature of
the solvent. The existence of the above-mentioned in-
tramolecular hydrogen bond was observed in a series of
studies based on crystallographic data of laminarabiose
and laminarabioside derivatives.53 The negative Ψ [C1′-
O-C3-H(C3)] dihedral angle of -47° obtained in our
calculations is in excellent agreement with the fact that
when such intramolecular hydrogen bonds are formed
this dihedral angle assumes a negative value [for methyl
â-D-laminarabioside (methyl â-D-Glcp-(1f6)-â-D-Glcp) Ψ
[C1′-O-C3-H(C3)] ) -52°].53b The introduction of acetyl
groups prevents the formation of the intramolecular
hydrogen bond, giving rise to a small positive value (for
methyl hepta-O-acetyl-â-D-laminarabioside, Ψ [C1′-O-
F IGURE 7. Lowest energy conformation of model methyl 3-O-
(2,3,4,6-tetrakis-O-acetyl-â-D-glucopyranosyl)-â-D-glucopyrano-
side.
exhibiting an A value (+18.8) smaller than that of 29
(+22.4) (entry 41) as consequence of its somewhat smaller
gg and much larger gt populations. These results are
clearly a consequence of nonbonded interactions between
the acetyl group at C4 and the hydroxymethyl group in
residue ΙΙ in its gt rotamer. This reduced gt rotamer
population was also obtained for compound 27 (entry 35),
which exhibits nonbonded interactions between this
rotamer and the benzylidene group, confirming the above
relationship.
In addition, rotamer population comparison between
the corresponding â and R epimers reveals no significant
differences, independently of the solvent. In chloroform
it was found that Pgg:Pgt:Ptg = 32:56:12 (R ) H) and Pgg:
Pgt:Ptg = 44:37:19 (R ) Ac), whereas in acetonitrile Pgg:
Pgt:Ptg = 40:50:10 (R ) H) with the same rotamer
populations as in chloroform (45:36:19) for R ) Ac.
CD and NMR data for the unprotected disaccharides
28 and 31 in methanol showed similar gg, decreased gt,
and increased tg populations compared to those in aprotic
solvents (compare entries 40 and 44, and 43 and 45). As
with the (1f4)-bonded disaccharides, this result points
to the fact that the intramolecular hydrogen bond be-
tween the hydroxyl group at C4 and the endocyclic
oxygen O5′ led to an increase in gt and a decrease in the
tg rotamer populations. The gt rotamer even had the
highest population in CDCl3 and CD3CN.
53c
C3-H(C3)] ) +5°).
Com p a r a t ive Con for m a t ion a l An a lysis of t h e
Disa cch a r id es. The previous rotational studies per-
formed with chiral and nonchiral alkyl glycopyrano-
sides28-31 demonstrated the existence of a rotamer dis-
tribution dependence of the hydroxymethyl group on the
structure of the aglycon. For alkyl â-D-glucopyranoside
derivatives, it was concluded that the population of the
gg rotamer decreased and gt increased as the pKa of the
bonded alcohol (aglycon) increased, whereas the tg popu-
lation remained almost constant. This relationship was
explained by the stereoelectronic exo-anomeric effect,9,54
which grows with increasing ease for charge delocaliza-
tion from the aglycon to the anomeric carbon.55,56 Fur-
thermore, the magnitude of this stereoelectronic effect
(53) (a) Kajiwara, K.; Miyamoto, T. Progress in Structural Charac-
terization of Functional Polysaccharides in Polysaccharides; Dumitriu,
S., Ed.; Marcel Dekker: New York, 1998. (b) Noguchi, K.; Okuyama,
K.; Kitamura, S.; Takeo, K. Carbohydr. Res. 1992, 237, 33 (c) Ikegami,
M.; Noguchi, K.; Okuyama, K.; Kitamura, S.; Takeo, K.; Ohno, S.
Carbohydr. Res. 1994, 253, 29.
(54) (a) Thatcher, G. R. J . Anomeric and Associated Stereoelectronic
Effects. Scope and Controversy in The Anomeric Effect and Associated
Stereoelectronic Effects; Thatcher, G. R. J ., Ed.; ACS Symposium Series
539; American Chemical Society: Washington, DC, 1993. (b) J uaristi,
E.; Cuevas, G. The Anomeric Effect in New Directions in Organic and
Biological Chemistry; Rees, C. W., Ed.; CRC Press: Boca Raton, FL,
1995.
The molecular structure of the most stable conformer
obtained by MM calculations52 with the model (1f3)-
bonded disaccharide shows (Figure 7) the existence of the
intramolecular hydrogen bond O4H‚‚‚O5′, with a stabi-
lization energy of 7.4 or 6.6 kcal‚mol-1 using a dielectric
(55) Tvarosˇka, I.; Koza´r, T. J . Am. Chem. Soc. 1980, 102, 6929.
(56) Praly, J . P.; Lemieux, R. U. Can. J . Chem. 1987, 65, 213.
J . Org. Chem, Vol. 68, No. 12, 2003 4623

Roe¨n et al.
F IGURE 10. CD spectra comparison of model methyl â-D-
Glcp-(1fx)-R-D-Glcp (R ) H, in CH3CN): 1f6 (10, solid line),
1f2 (16, dashed line), 1f4 (25, dotted line), and 1f3 (31,
dashed dotted line).
F IGURE 8. CD spectra comparison of model methyl â-D-Glcp-
(1fx)-â-D-Glcp (R ) H, in CH3CN): 1f2 (13, dashed line),
1f6 (7, solid line), 1f4 (23, dotted line), and 1f3 (28, dashed
dotted line).
F IGURE 11. CD spectra comparison of model methyl â-D-
Glcp-(1fx)-R-D-Glcp (R ) Ac, in CH3CN): 1f6 (9, solid line),
1f2 (15, dashed line), 1f4 (24, dotted line), and 1f3 (30,
dashed dotted line).
F IGURE 9. CD spectra comparison of model methyl â-D-Glcp-
(1fx)-â-D-Glcp (R ) Ac, in CH3CN): 1f2 (14, dashed line),
1f6 (8, solid line), 1f4 (22, dotted line), and 1f3 (29, dashed
dotted line).
depends on the acetal group bond lengths,57 since linear
correlations between the two C-O bond lengths at the
acetal center and the pKa of ROH have been determined
in glucopyranosides.
bution) and lower gt (net negative) populations for the
former disaccharides. On the other hand, Figures 10 and
11 (R series) show higher first Cotton effect amplitudes
for the (1f6)-bonded disaccharides than those obtained
from the (1f2)-bonded disaccharides.
On the basis of the above relationship between the
stereoelectronic exo-anomeric effect and the rotamer
distribution, higher gg and lower gt populations must be
expected for the model (1f6)-bonded disaccharides than
for the other disaccharides, as a consequence of their type
of linkage (through a primary hydroxyl group). However,
comparison of the J H5′,H6′R and J H5′,H6′S coupling constants
and CD data for the (1f2)- and (1f6)-linked disaccha-
rides (R ) H, Ac) with the methoxyl group in the â
anomeric configuration reveal higher gg and tg and lower
gt populations for the former compounds, whereas the
opposite behavior was found when the methoxyl group
is in the R configuration. As can be observed in Figures
8 and 9 (â series) the amplitude of the first Cotton effect
is higher for the (1f2)-linked disaccharides (dashed lines)
than for the (1f6)-linked disaccharides (solid lines), in
total agreement with higher gg (net positive CD contri-
While the rotamer populations and CD data were
almost independent of the substitution (R ) H, Ac) and
of the anomeric configuration for the (1f6)-linked di-
saccharides, those obtained for the (1f2)-linked disac-
charides showed a strong dependence on the anomeric
configuration of the methoxyl group. This behavior can
be explained by the fact that for the former disaccharides
the two glucopyranosidic rings are far apart, and there-
fore nonbonded interactions between them play almost
no role. However, this is not the case for the (1f2)-
bonded disaccharides. As we have already mentioned, the
(1f2)-bonded disaccharides with the methoxyl group in
the â anomeric configuration show a strong nonbonded
interaction between the methoxyl group and the hy-
droxymethyl group in its gt rotamer, leading to larger
gg and smaller gt rotamer populations. For those cases
where the methoxyl group is in the R anomeric configu-
ration, such nonbonded interaction is reduced and there-
fore larger gt populations are obtained for these com-
pounds than for the (1f6)-linked disaccharides.
(57) (a) J ones, P. G.; Kirby A. J . J . Chem. Soc., Chem. Commun.
1979, 288. (b) Briggs, A. J .; Glenn, R.; J ones, P. G.; Kirby, A. J .;
Ramaswamy, P. J . Am. Chem. Soc. 1984, 106, 6200.
4624 J . Org. Chem., Vol. 68, No. 12, 2003

Hydroxymethyl Rotamers in Disaccharides
TABLE 5. Gen er a l Aver a ge Rota m er ic P op u la tion s (%)
a r ou n d th e C5′-C6′ Bon d (Resid u e ΙΙ) for th e Mod el
Disa cch a r id es (Meth yl â-D-Glcp(p-Br -Bz)-(1fx)-â/r-D-Glcp)
Larger gg and smaller gt populations should be ex-
pected for the (1f2)-linked disaccharides than for the
(1f6)-bonded disaccharides, due to the close proximity
between the two glucopyranosidic rings. Although that
behavior was found in the case of the disaccharides
having the methoxyl group in the â anomeric configura-
tion, it was not the case for the R anomers. This last
result can be explained by a greater delocalization of the
nonbonding electron pair of the interglycosidic oxygen
(O1′) toward the anomeric carbon (C1′) in the (1f2)-
bonded disaccharides, than in the (1f6)-linked disac-
charides, as a consequence of their type of linkage
(through a secondary hydroxyl group); in other words,
by a greater exo-anomeric effect for the (1f2)-linked
disaccharides.
CD and NMR data of the (1f3)- and (1f4)-bonded
disaccharides showed a similar conformational behavior
in general but quite different from that of the (1f6)- and
(1f2)-bonded disaccharides. Their higher J H5′,H6′R and
lower J H5′,H6′S coupling constants lead to greater popula-
tions of the gt rotamer and to smaller populations of gg
and tg than those obtained for the (1f6)- or (1f2)-linked
disaccharides. Furthermore, the smaller magnitudes of
the Cotton effects for the model (1f4)- and (1f3)-bonded
disaccharides (see Figures 8-11) than for the (1f6)- and
(1f2)-bonded disaccharides are in total agreement with
the above conclusion obtained by NMR. This general
behavior can be explained by a greater exo-anomeric
effect for these disaccharides, since steric interaction
considerations should normally lead to the opposite
behavior.
In addition, although this behavior was independent
of the substitution (Ac, H) and anomeric configuration
of the methoxyl group, different rotamer populations
were obtained for the (1f4)- and (1f3)-linked disaccha-
rides, depending apparently on the substitution (Ac, H).
Thus in Figures 8-11 (CH3CN) the unprotected (1f4)-
and (1f3)-linked disaccharides (R ) H) show Cotton
effects of smaller magnitudes than their corresponding
triacetylated disaccharides. Higher nonbonded interac-
tions between the acetyl groups and the hydroxymethyl
group for the acetylated derivatives could explain these
results; however, they cannot give a satisfactory explana-
tion for the observed change in rotamer distribution when
the unprotected disaccharides (R ) H) are measured in
a protic solvent (MeOH), giving rotamer populations
similar to those of the corresponding triacetylated deriva-
tives (see Tables 3 and 4). Therefore, all of these data
point to the intramolecular hydrogen bond in the unpro-
tected (1f3)- and (1f4)-bonded disaccharides (R ) H)
favoring the population of the gt rotamer at the expense
of the tg. Finally, the slightly smaller gg and greater gt
rotamer populations obtained for the (1f3)-linked dis-
accharides than for the (1f4)-bonded disaccharides (R
) H) could be due to a higher value of the exo-anomeric
effect and/or a hydrogen bond more favorable to the
former disaccharides.
solvent
CDCl3
linkage
MeO (C1)
R
Pgg
Pgt
Ptg
1f2
1f6
1f4
1f2
1f4
1f3
1f4
1f3
â
H, Ac
H, Ac
Ac
H, Ac
Ac
50
45
45
45
40
40
40
35
30
35
35
40
40
40
50
55
20
20
20
15
20
20
10
10
R and â
R
R
â
R and â
R and â
R and â
Ac
H
H
CD3CN
1f2
1f6
1f2
1f4
1f3
1f4
1f3
â
H, Ac
H, Ac
H, Ac
Ac
Ac
H
55
50
50
45
45
45
40
25
30
35
35
35
45
50
20
20
15
20
20
10
10
R and â
R
R and â
R and â
R and â
R and â
H
CD3OD
1f2
1f6
1f2
1f4
1f4
1f3
â
H
H
H
H
H
H
55a
50
45
45
40
40
25a
30
35
35
40
40
20a
20
20
20
20
20
R and â
R
R
â
R and â
a
Estimated values.
3
determined by analyzing the 3J H5′,H6′R and J H5′,H6′S values
and CD spectral data in different solvents. Additionally,
for each type of disaccharide the dependence of the
hydroxymethyl rotamer population on (i) the substitu-
ents, (ii) the anomeric configuration of the methoxyl
group, and (iii) the protic or aprotic nature of the solvent
has been analyzed.
The general hydroxymethyl rotational preferences
around the C5′-C6′ bond (residue ΙΙ) in chloroform,
acetonitrile, and methanol for the model disaccharides
methyl â-D-Glcp-(1fx)-â/R-D-Glcp are shown in Table 5.
According to each specific case, nonbonded interactions,
stereoelectronic effects and/or hydrogen bonds are re-
sponsible for the observed rotamer differences.
The observed rotamer populations for the (1f6)-linked
disaccharides are similar to those obtained for the
corresponding monosaccharide methyl 2,3,4,6-tetrakis-
O-(p-bromobenzoyl)-â-D-glucopyranoside and therefore
depend mainly on the exo-anomeric effect. Whereas their
rotamer populations are almost independent of the
substituents, of the anomeric configuration of the meth-
oxyl group, and of the nature of the solvent, this is not
the case with the (1f2)-bonded disaccharides. The hy-
droxymethyl rotamer population for these latter disac-
charides exhibits a clear relationship with the anomeric
configuration of the methoxyl group, as consequence of
nonbonded interactions between this group and the
hydroxymethyl group in the â-anomers, which obviously
have smaller gt populations, independently of the solvent
used. In addition, the exo-anomeric effect dependence of
the populations of the hydroxymethyl group in these
(1f2)-linked disaccharides is evident by comparing the
J H5′,H6′ coupling constants and CD values of (1f6)- and
(1f2)-bonded disaccharides (R-anomers).
Con clu sion s
The glycosidic linkage dependence of the rotational
populations of the hydroxymethyl group of methyl glu-
copyranosyl-glucopyranoside derivatives having 1f2,
1f3, 1f4, and 1f6 â glycosidic linkages has been
In sharp contrast to the (1f6)- and (1f2)-linked
disaccharides, the (1f3)- and (1f4)-linked disaccharides
show a dependence on both the substituents and the
protic nature of the solvent for those derivatives having
J . Org. Chem, Vol. 68, No. 12, 2003 4625

Roe¨n et al.
(1.0 equiv per acetate) was added. The resulting mixture was
stirred at room temperature or at 40 °C, and the reaction
monitored by TLC. After the reaction was complete, the
mixture was extracted with CH2Cl2, and the organic phase
washed with aqueous NaHCO3, dried over anhydrous Na2SO4,
and evaporated in a vacuum. The product was purified by flash
column chromatography.
2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-r-D-glu cop yr a n o-
syl Br om id e (2). This compound was prepared from D-(+)-
glucopyranose in two steps as described in ref 28.
free hydroxyl groups. Therefore, the rotational behavior
of (1f3)- and (1f4)-bonded disaccharides is a conse-
quence of a combination of effects: nonbonded interac-
tions, the stereoelectronic exo-anomeric effect, and the
existence, respectively, of O4H‚‚‚O5′ or O3H‚‚‚O5′ in-
tramolecular hydrogen bonding. Thus, the population of
the gt rotamer decreases as the nonbonded interactions
increase, whereas it increases as the exo-anomeric effect
increases and also when intramolecular hydrogen bond-
ing is present.
Correlations of the NMR and CD data to the types of
glycosidic linkage led to the following general information
with regard to the rotational population of the hy-
droxymethyl group: Pgg (1f2)- (â series) > (1f6)f
(1f2)- (R series) > (1f4)f (1f3)-linked disaccharides;
Pgt (1f2)- (â series) < (1f6)- < (1f2)- (R series) < (1f4)-
< (1f3)-linked disaccharides; while the Ptg (around 20%)
is almost independent of glycosidic linkage type, solvent,
and substitution, except with those disaccharides having
the intramolecular hydrogen bonding to O5′, the (1f3)-
and (1f4)-bonded disaccharides in aprotic solvents, when
Ptg drops to 10%.
Meth yl 6-O-(ter t-Bu tyld ip h en ylsilyl)-â-D-glu cop yr a n o-
sid e (3). To a solution of methyl â-D-glucopyranoside (1.0 g,
5.15 mmol) in dry DMF (5 mL) were added sequentially
t
imidazole (771 mg, 11.33 mmol) and Bu(Ph)2SiCl (1.47 mL,
5.66 mmol). After 24 h, the solution was diluted with ether,
washed with H2O and saturated aqueous NH4Cl, and dried
over MgSO4. It was then concentrated and chromatographed
(flash column, n-hexane/EtOAc, 3:7) to provide compound 3
(2.21 g, 99% yield): TLC Rf ) 0.69 (EtOAc/MeOH, 7:3);
[R]2D5 ) -33.1 (c 1.25, CHCl3); 1H NMR (CDCl3) δ 7.69 (d, J )
7.9 Hz, 4H), 7.39 (m, 6H), 4.19 (d, J ) 7.7 Hz, 1H), 3.92 (d, J
) 4.9 Hz, 2H), 3.61 (t, J ) 8.8, 1H), 3.57 (t, J ) 8.8, 1H), 3.50
(s, 3H), 3.41 (m, 1H), 3.37 (t, J ) 8.1 Hz, 1H), 1.05 (s, 9H); 13
C
NMR (CDCl3) δ 135.6-127.6, 103.3, 76.4, 75.0, 73.5, 71.6, 64.5,
56.8, 26.7, 19.2. Anal. Calcd for C23H32O6Si: C, 63.86; H, 7.46.
Found: C, 63.87; H, 7.44.
Met h yl 2,3,4-Tr i-O-b en zyl-6-O-(ter t-b u t yld ip h en ylsi-
lyl)- â-D-glu cop yr a n osid e (4). To a solution of the glucopy-
ranoside 3 (2.19 g, 5.06 mmol) in dry DMF (10 mL) were added
sodium hydride 80% dispersion in mineral oil (557 mg, 18.21
mmol) and benzyl bromide (2.71 mL, 22.77 mmol), and the
reaction was stirred overnight. Then, the solution was diluted
with ether, washed (H2O and NH4Cl), dried (MgSO4), and
concentrated. The residue was chromatographed on silica gel
(n-hexane/EtOAc, 9:1) to give 4 (1.42 g, 40% yield): TLC Rf )
Exp er im en ta l Section
Gen er a l. 1H NMR spectra were recorded at 400 and 500
MHz, and 13C NMR were recorded at 100 MHz, VTU 300.0
°K. Chemical shifts are reported in parts per million. The
residual solvent peak was used as an internal reference.
Optical rotations were measured on a digital polarimeter in a
1-dm cell. UV and CD spectra were recorded in the range 400-
200 nm by using 10-mm cells. Prior to measurement of CD
spectra all compounds were purified by HPLC by using a
µ-Porasil column, 300 × 7.8 mm i.d., 254 nm, and HPLC grade
n-hexane/EtOAc solvent systems. The concentrations of the
CD samples were ascertained for the UV spectra, using the
standard ꢀ values at 245 nm: tetrakis-(p-bromobenzoate)
76400.11
For analytical and preparative thin-layer chromatography,
silica gel ready-foils and glass-backed plates (1 mm) were used,
respectively, being developed with 254 nm UV light and/or
spraying with AcOH/H2O/H2SO4 (80:16:4) and heating at 150
°C. Column chromatography was performed using silica gel
(0.015-0.04 mm) and n-hexane/EtOAc solvent systems or
Sephadex LH-20 (CHCl3/MeOH/n-hexane, 1:1:2). All reagents
were obtained from commercial sources and used without
further purification. Solvents were dried and distilled before
use. All reactions were performed under a dry argon atmos-
phere.
0.74 (n-hexane/EtOAc, 8:2); [R]2D5 ) -0.8 (c 1.07, CHCl3); H
1
NMR (CDCl3) δ 7.82 (d, J ) 7.8 Hz, 2H), 7.76 (d, J ) 7.9 Hz,
2H), 7.48-7.31 (m, 19H), 7.25 (m, 2H), 5.01 (d, J ) 11.4 Hz,
1H), 4.99 (d, J ) 11.2 Hz, 1H), 4.95 (d, J ) 10.8 Hz, 1H), 4.88
(d, J ) 10.9 Hz, 1H), 4.79 (d, J ) 11.1 Hz, 1H), 4.75 (d, J )
10.8 Hz, 1H), 4.38 (d, J ) 7.7 Hz, 1H), 3.99 (brd, J ) 2.7 Hz,
2H), 3.82 (t, J ) 9.1 Hz, 1H), 3.72 (t, J ) 9.1 Hz, 1H), 3.64 (s,
3H), 3.52 (dd, J ) 7.9 and 9.1 Hz, 1H), 3.40 (dt, J ) 9.1 and
2.7 Hz, 1H), 1.12 (s, 9H); 13C NMR (CDCl3) δ 138.7-127.5,
104.5, 84.7, 82.6, 77.7, 75.8, 75.6, 75.1, 74.8, 62.6, 56.6, 26.8,
19.3. Anal. Calcd for C44H50O6Si: C, 75.18; H, 7.17. Found:
C, 75.15; H, 7.26.
Meth yl 2,3,4-Tr i-O-ben zyl-â-D-glu cop yr a n osid e (5). Tet-
rabutylammonium fluoride trihydrate (1.0 g, 3.42 mmol) was
added to a solution of 4 (1.2 g, 1.71 mmol) in dry THF at room
temperature. After 12 h, the reaction mixture was quenched
with H2O, diluted with CH2Cl2, and washed with saturated
aqueous NaCl. The organic layers were dried over MgSO4,
filtered, concentrated, and chromatographed on silica gel (n-
hexane/EtOAc, 6:4) to give 642 mg of 5 (81% yield): TLC Rf )
Gen er a l P r oced u r e for An om er iza tion a n d /or Deben -
zyla tion .34 To a stirred solution of the substrate (20-70 µmol)
in dry CH2Cl2 (20 mL/mmol) at the selected temperature under
dry argon was added 3 equiv of anhydrous FeCl3 (or 3 equiv
per benzyl ether group in case of debenzylation), and the
reaction left until the color of the reaction mixture changed to
brown. The reaction was quenched by addition of water (1 mL).
This mixture was stirred for 1 min and then extracted with
CH2Cl2 (25 mL). The combined organic layers were dried over
magnesium sulfate, and the solvent was removed under
reduced pressure. This crude reaction mixture was purified
by silica gel column chromatography. In the case of â-glyco-
pyranosides, debenzylation occurred with simultaneous ano-
merization or retention of the anomeric configuration, depend-
ing on the reaction conditions. While anomerization was
favored by increasing the equiv of FeCl3, retention of the
anomeric configuration was obtained by lowering the temper-
ature to 0 or -20 °C.
0.49 (n-hexane/EtOAc, 6:4); [R]2D5 ) +9.8 (c 0.48, CHCl3); H
1
NMR (CDCl3) δ 7.36-7.28 (m, 15H), 4.94 (d, J ) 10.8 Hz, 1H),
4.91 (d, J ) 10.7 Hz, 1H), 4.87 (d, J ) 10.9 Hz, 1H), 4.82 (d,
J ) 10.9 Hz, 1H), 4.72 (d, J ) 11.0 Hz, 1H), 4.65 (d, J ) 10.9
Hz, 1H), 4.37 (d, J ) 7.8 Hz, 1H), 3.89 (brdd, J ) 2.2 and 11.8
Hz, 1H), 3.73 (brdd, J ) 3.9 and 11.8 Hz, 1H), 3.68 (t, J ) 9.0
Hz, 1H), 3.58 (t, J ) 9.0 Hz, 1H), 3.58 (s, 3H), 3.41 (dd, J )
7.8 and 9.0 Hz, 1H), 3.38 (m, 1H); 13C NMR (CDCl3) δ 138.5-
127.6, 104.8, 84.4, 82.4, 77.5, 75.7, 75.1, 75.0, 74.8, 62.0, 57.3.
Anal. Calcd for C28H32O6: C, 72.39; H, 6.94. Found: C 72.34;
H, 6.93.
Meth yl 6-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu cop yr a n osyl)-2,3,4-t r i-O-b e n zyl-â-D-glu cop yr a n o-
sid e (6). To a solution of 5 (198 mg, 0.426 mmol) in dry CH2Cl2
(4 mL) at -40 °C were added TMU (51µL, 0.426 mmol), silver
triflate (219 mg, 0.852 mmol) and the glucosyl donor 2 (830
mg, 0.852 mmol). The reaction was quenched after 30 min by
Gen er a l P r oced u r e for Dea cetyla tion .37 The substrate
was dissolved in CH2Cl2/MeOH (9:1), and then p-TsOH‚H2O
4626 J . Org. Chem., Vol. 68, No. 12, 2003

Hydroxymethyl Rotamers in Disaccharides
adding a few drops of pyridine, filtered through Celite, and
concentrated. Then, the residue was purified by Sephadex LH-
20 (CHCl3/MeOH/n-hexane, 1:1:2) to give the disaccharide 6
(536 mg, 92% yield): TLC Rf ) 0.54 (n-hexane/EtOAc, 7:3);
[R]2D5 ) +10.6 (c 0.94, CHCl3); FAB-MS m/z 1381 (16, [M +
Na]+), 893 (4, [M + Na - C28H31O5]+), 183 (100, BrBz); 1H NMR
(CDCl3) δ 7.83 (d, J ) 8.6 Hz, 2H), 7.72 (m, 4H), 7.66 (d, J )
8.6 Hz, 2H), 7.52-7.43 (m, 8H), 7.26 (m, 13H), 7.12 (dd, J )
1.9 and 7.6 Hz, 2H), 5.78 (t, J ) 9.6 Hz, 1H), 5.61 (t, J ) 9.6
Hz, 1H), 5.51 (dd, J ) 7.9 and 9.6 Hz, 1H), 4.88 (d, J ) 7.7
Hz, 1H), 4.86 (d, J ) 10.6 Hz, 1H), 4.85 (d, J ) 11.1 Hz, 1H),
4.66 (t, J ) 12.0 Hz, 3H), 4.58 (dd, J ) 3.4 and 12.2 Hz, 1H),
4.48 (dd, J ) 4.7 and 12.2 Hz, 1H), 4.39 (d, J ) 11.1 Hz, 1H),
4.18 (d, J ) 7.7 Hz, 1H), 4.15 (dd, J ) 1.1 and 11.0 Hz, 1H),
4.05 (m, 1H), 3.72 (dd, J ) 5.9 and 11.0 Hz, 1H), 3.55 (t, J )
8.9 Hz, 1H), 3.44 (m, 1H), 3.36 (brs, 4H), 3.30 (dd, J ) 8.0 and
8.9 Hz, 1H); 13C NMR (CDCl3) δ 165.3, 165.1, 164.4, 164.2,
138.4, 137.9, 131.9-127.3, 104.5, 101.1, 84.4, 82.1, 77.7, 75.6
74.8, 74.7, 74.3, 73.1, 71.9, 71.8, 69.7, 68.7, 63.1, 56.8. Anal.
Calcd for C62H54O15Br4: C, 54.81; H, 4.01. Found: C 54.76; H,
4.33.
Meth yl 6-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu cop yr a n osyl)-â-D-glu cop yr a n osid e (7). Following the
general procedure for debenzylation, anhydrous FeCl3 (41 mg,
0.253 mmol) was added to a solution of disaccharide 6 (38 mg,
0.028 mmol) in dry CH2Cl2 (1.5 mL) at 0°C. The crude reaction
mixture was purified by chromatography on silica gel (CHCl3/
MeOH, 98:2) to give 7 (23.1 mg, 76% yield): TLC Rf ) 0.5
(CHCl3/MeOH 9:1); [R]25 ) +43.5 (c 0.92, CHCl3); 1H NMR
(CDCl3/CD3OD, 95:5) δD7.81 (d, J ) 8.5 Hz, 2H), 7.74 (d, J )
8.5 Hz, 2H), 7.67 (d, J ) 8.6 Hz, 2H), 7.62 (d, J ) 8.6 Hz, 2H),
7.51 (d, J ) 8.6 Hz, 2H), 7.48 (d, J ) 8.6 Hz, 2H), 7.45 (d, J )
8.6 Hz, 2H), 7.40 (d, J ) 8.6 Hz, 2H), 5.77 (t, J ) 9.6 Hz, 1H),
5.59 (t, J ) 9.6 Hz, 1H), 5.43 (dd, J ) 7.8 and 9.6 Hz, 1H),
4.92 (d, J ) 7.8 Hz, 1H), 4.60 (dd, J ) 3.3 and 12.2 Hz, 1H),
4.43 (dd, J ) 4.4 and 12.2 Hz, 1H), 4.19 (dd, J ) 1.9 and 11.2
Hz, 1H), 4.10 (m, 1H), 4.01 (d, J ) 7.8 Hz, 1H), 3.71 (dd, J )
6.9 and 11.2 Hz, 1H), 3.39 (ddd, J ) 1.9, 6.9 and 9.4 Hz, 1H),
3.33 (t, J ) 9.4 Hz, 1H), 3.25 (s, 3H), 3.23 (t, J ) 9.4 Hz, 1H),
3.14 (dd, J ) 7.8 and 9.4 Hz, 1H); 13C NMR (CDCl3/CD3OD,
95:5) δ 165.5, 165.1, 164.4 (2C), 131.8-127.2, 103.4, 101.3,
76.3, 75.0, 73.3, 73.0, 72.0, 71.8, 70.4, 69.6, 69.5, 62.9, 56.7.
Anal. Calcd for C41H36O15Br4: C, 45.25; H, 3.33. Found: C
45.28; H, 3.44.
J ) 8.6 Hz, 2H), 7.48 (d, J ) 8.6 Hz, 2H), 7.44 (d, J ) 8.6 Hz,
2H), 5.82 (t, J ) 9.6 Hz, 1H), 5.61 (t, J ) 9.6 Hz, 1H), 5.48
(dd, J ) 7.8 and 9.6 Hz, 1H), 5.39 (t, J ) 10.1 Hz, 1H), 4.86
(d, J ) 7.8 Hz, 1H), 4.79 (t, J ) 10.1 Hz, 1H), 4.70 (d, J ) 3.6
Hz, 1H), 4.65 (dd, J ) 3.6 and 10.1 Hz, 1H), 4.60 (d, J ) 3.3
and 12.2 Hz, 1H), 4.46 (dd, J ) 4.7 and 12.2 Hz, 1H), 4.11
(ddd, J ) 3.3, 4.7 and 9.6 Hz, 1H), 3.96 (dd, J ) 1.9 and 11.1
Hz, 1H), 3.89 (ddd, J ) 1.9, 6.8 and 10.1 Hz, 1H), 3.57 (dd, J
) 6.8 and 11.1 Hz, 1H), 3.10 (s, 3H), 2.03 (s, 3H), 1.97 (s, 3H),
1.95 (s, 3H); 13C NMR (CDCl3) δ 170.0, 169.9, 169.7, 165.3,
165.0, 164.4, 164.4, 131.9-127.3, 101.4, 96.3, 72.9, 71.9, 71.8,
70.7, 69.8, 69.7, 69.0, 68.5, 68.1, 63.0, 54.9, 20.7 (2x), 20.6.
Anal. Calcd for C47H42O18Br4: C, 46.48; H, 3.49. Found: C
46.51; H, 3.52.
Meth yl 6-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu cop yr a n osyl)-r-D-glu cop yr a n osid e (10). Following the
general procedure, deacetylation of 9 (5.0 mg, 4.3 µmol) led to
compound 10 (4.1 mg, 3.8 µmol) in 75% yield: TLC Rf ) 0.6
(CHCl3/MeOH, 9:1); [R]25 ) +58.1 (c 0.86, CHCl3); 1H NMR
D
(CDCl3) δ 7.85 (d, J ) 8.6 Hz, 2H), 7.78 (d, J ) 8.6 Hz, 2H),
7.71 (d, J ) 8.6 Hz, 2H), 7.66 (d, J ) 8.6 Hz, 2H), 7.54 (d, J )
8.6 Hz, 2H), 7.52 (d, J ) 8.6 Hz, 2H), 7.48 (d, J ) 8.6 Hz, 2H),
7.43 (d, J ) 8.6 Hz, 2H), 5.83 (t, J ) 9.6 Hz, 1H), 5.63 (t, J )
9.6 Hz, 1H), 5.50 (dd, J ) 7.8 and 9.6 Hz, 1H), 4.93 (d, J ) 7.8
Hz, 1H), 4.63 (dd, J ) 3.2 and 12.2 Hz, 1H), 4.59 (d, J ) 3.8
Hz, 1H), 4.48 (dd, J ) 4.6 and 12.2 Hz, 1H), 4.15 (m, 2H), 3.80
(dd, J ) 5.6 and 11.0 Hz, 1H), 3.65 (m, 1H), 3.59 (t, J ) 9.2
Hz, 1H), 3.35 (m, 2H), 3.21 (s, 3H); 13C NMR (CDCl3) δ 165.5
(2x), 165.1, 164.4, 131.9-127.3, 101.5, 99.0, 74.7, 73.0, 72.1,
72.0, 71.9, 70.3, 69.7 (2x), 69.2, 63.0, 55.2. Anal. Calcd for
C
41H36O15Br4: C, 45.25; H, 3.33. Found: C 45.29; H, 3.40.
Meth yl 4,6-Di-O-(1,1,3,3-tetr aisopr opyldisiloxan yliden e)-
â-D-glu cop yr a n osid e (11). To a solution of methyl â-D-
glucopyranoside (2.0 g, 10.29 mmol) in dry pyridine (15 mL)
was added 1,3-dichloro-1,1,3,3-tetraisopropyldisiloxane (TIP-
DSCl2) (4.8 mL, 15.43 mmol). After 2 h, the pyridine was
distilled and the resulting oil was purified by flash chroma-
tography (EtOAc) to give 11 (2.95 g, 66% yield): TLC Rf )
0.70 (CHCl3/MeOH, 9:1); [R]25 ) -27.2 (c 2.73, CHCl3); 1H
D
NMR (CDCl3) δ 4.17 (d, J ) 7.7 Hz, 1H), 4.09 (brd, J ) 12.4
Hz, 1H), 3.96 (brd, J ) 12.4 Hz, 1H), 3.81 (t, J ) 9.1 Hz, 1H),
3.58 (t, J ) 9.1 Hz, 1H), 3.54 (s, 3H), 3.35 (t, J ) 8.5 Hz, 1H),
3.18 (brd, J ) 9.3 Hz, 1H), 3.04 (br s, 1H), 2.80 (br s, 1H),
1.10-1.02 (m, 28H); 13C NMR (CDCl3) δ 104.0, 76.6, 76.2, 73.9,
69.2, 60.8, 57.3, 17.4, 17.3, 17.3, 17.2, 17.2, 17.0, 13.6, 13.2,
13.1, 12.5. Anal. Calcd for C19H40O7Si2 C, 52.26; H, 9.23.
Found: C 52.11; H, 9.42.
Meth yl 6-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu copyr an osyl)-2,3,4-tr i-O-acetyl-â-D-glu copyr an oside (8).
Acetylation of the disaccharide 7 (15 mg, 0.014 mmol) with 2
mL of acetic anhydride/pyridine (1:1) led, after chromatogra-
phy on silica gel (n-hexane/EtOAc, 6:4), to compound 8 (15.8
Meth yl 2-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu cop yr a n osyl)-4,6-d i-O-(1,1,3,3-tetr a isop r op yld isiloxa -
n ylid en e)-â-D-glu cop yr a n osid e (12). To a stirred solution
of compound 11 (336 mg, 0.77 mmol) and TMU (92 µL, 0.77
mmol) in dry CH2Cl2 (7.7 mL) at -40 °C were added silver
triflate (395 mg, 1.54 mmol) and the glucosyl donor 2 (1.5 g,
1.54 mmol). After being stirred 40 min, the reaction mixture
was quenched with a few drops of water and filtered through
a bed of Celite with CH2Cl2. The filtrate was evaporated under
diminished pressure, and the residue was chromatographed
(n-hexane/EtOAc 9:1) to give the disaccharide 12 (503 mg, 52%
mg, 95% yield): TLC Rf ) 0.45 (n-hexane/EtOAc, 6:4); [R]25
)
D
1
+16.6 (c 1.00 CHCl3); H NMR (CDCl3) δ 7.85 (d, J ) 8.4 Hz,
2H), 7.78 (d, J ) 8.4 Hz, 2H), 7.71 (d, J ) 8.5 Hz, 2H), 7.65 (d,
J ) 8.5 Hz, 2H), 7.57 (d, J ) 8.6 Hz, 2H), 7.52 (d, J ) 8.5 Hz,
2H), 7.48 (d, J ) 8.5 Hz, 2H), 7.43 (d, J ) 8.6 Hz, 2H), 5.80 (t,
J ) 9.6 Hz, 1H), 5.60 (t, J ) 9.6 Hz, 1H), 5.46 (dd, J ) 7.8 and
9.6 Hz, 1H), 5.12 (t, J ) 9.5 Hz, 1H), 4.90 (d, J ) 7.8 Hz, 1H),
4.81 (m, 2H), 4.60 (d, J ) 2.9 and 12.2 Hz, 1H), 4.46 (dd, J )
4.8 and 12.2 Hz, 1H), 4.22 (d, J ) 7.9 Hz, 1H), 4.11 (m, 1H),
3.96 (brd, J ) 8.3 Hz, 1H), 3.63 (brd, J ) 8.3 Hz, 2H), 3.16 (s,
3H), 2.00 (s, 3H), 1.96 (s, 6H); 13C NMR (CDCl3) δ 170.1, 169.6,
169.3, 165.3, 165.0, 164.4, 164.3, 131.9-127.3, 101.2, 101.2,
73.2, 73.0, 72.6, 72.0, 71.8, 71.2, 69.6, 69.1, 68.6, 63.0, 56.5,
20.6, 20.5 (2C). Anal. Calcd for C47H42O18Br4: C, 46.48; H, 3.49.
Found: C 46.50; H, 3.46.
yield): TLC Rf ) 0.41 (n-hexane/EtOAc, 7.5/2.5); [R]D25
)
+22.1 (c 1.07, CHCl3); FAB-MS m/z MS 1352 (0.1, [M + Na]+),
894 (0.05, [M - C19H39O7Si2]+), 183 (100, BrBz); 1H NMR
(CDCl3) δ 7.85-7.43 (16H), 5.81 (t, J ) 9.6 Hz, 1H), 5.65 (t, J
) 9.6 Hz, 1H), 5.50 (dd, J ) 7.9 and 9.6 Hz, 1H), 5.17 (d, J )
7.9 Hz, 1H), 4.63 (dd, J ) 3.0 and 12.2 Hz, 1H), 4.48 (dd, J )
4.4 and 12.2 Hz, 1H), 4.27 (d, J ) 7.6 Hz, 1H), 4.16 (m, 1H),
4.00 (brd, J ) 12.4 Hz, 1H), 3.91 (brd, J ) 12.3 Hz, 1H), 3.58
(t, J ) 9.1 Hz, 1H), 3.48 (s, 4H), 3.33 (t, J ) 8.3 Hz, 1H), 3.07
Meth yl 6-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu cop yr a n osyl)-2,3,4-t r i-O-a cet yl-r-D-glu cop yr a n osid e
(9). Compound 9 (14.0 mg, 11.2 µmol, 46% yield) was obtained
from its â-anomer 8 (30 mg, 0.025 mmol) following the general
procedure for anomerization: TLC Rf ) 0.45 (n-hexane/EtOAc,
6:4); [R]2D5 ) +37.6 (c 0.37 CHCl3); 1H NMR (CDCl3) δ 7.85 (d,
J ) 8.6 Hz, 2H), 7.80 (d, J ) 8.6 Hz, 2H), 7.70 (d, J ) 8.6 Hz,
2H), 7.66 (d, J ) 8.6 Hz, 2H), 7.56 (t, J ) 8.6 Hz, 2H), 7.53 (d,
(brd, J ) 9.1 Hz, 1H), 2.22 (br s, 1H), 1.06-0.78 (m, 28H); 13
C
NMR (CDCl3) δ 165.4, 165.1, 165.0, 164.5, 131.8-127.5, 103.2,
101.8, 82.7, 76.2, 75.3, 73.1, 72.8, 72.1, 69.8, 68.8, 63.3, 60.6,
57.4, 17.3, 17.2, 17.0, 16.9, 13.4, 13.2, 12.5, 12.4. Anal. Calcd
for C53H62O16Si2Br4: C, 47.83; H, 4.70. Found: C 47.87; H, 4.70.
J . Org. Chem, Vol. 68, No. 12, 2003 4627

Roe¨n et al.
Meth yl 2-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu cop yr a n osyl)-â-D-glu cop yr a n osid e (13). To a solution
of the disaccharide 12 (100 mg, 0.075 mmol) in dry THF (2
mL) were added pyridinium chlorhydrate (22 mg, 0.187 mmol)
and TBAF (375 µL, 1 M in THF), and the reaction was stirred
for 40 min. The solution was diluted with EtOAc, washed with
H2O and saturated aqueous NaCl, dried (MgSO4), and con-
centrated. The resulting oil was chromatographed (n-hexane/
EtOAc, 3:7) to provide compound 13 (75 mg, 92% yield): TLC
Rf ) 0.3 (CHCl3/MeOH, 9:1); [R]25 ) +33.1 (c 0.62, DMF);
FAB-MS m/z MS 1110 ([M + Na -DH]+), 894 ([M - C7H12O6]+),
183 (100, BrBz); 1H NMR (CDCl3/CD3OD, 95:5) δ 7.80 (d, J )
8.5 Hz, 2H), 7.76 (d, J ) 8.5 Hz, 2H), 7.67 (d, J ) 8.5 Hz, 2H),
7.63 (d, J ) 8.5 Hz, 2H), 7.50-7.40 (8H), 5.78 (t, J ) 9.6 Hz,
1H), 5.62 (t, J ) 9.6 Hz, 1H), 5.41 (dd, J ) 7.9 and 9.6 Hz,
1H), 5.19 (d, J ) 7.9 Hz, 1H), 4.61 (dd, J ) 3.4 and 12.2 Hz,
1H), 4.43 (dd, J ) 4.2 and 12.2 Hz, 1H), 4.28 (d, J ) 7.6 Hz,
1H), 4.12 (m, 1H), 3.78 (dd, J ) 3.1 and 12.1 Hz, 1H), 3.69
(dd, J ) 4.3 and 12.1 Hz, 1H), 3.44 (s, 3H), 3.37-3.31 (3H),
3.19 (m, 1H); 13C NMR (CDCl3/CD3OD, 95:5) δ 165.5, 165.2,
165.1, 164.5, 131.8-127.3, 102.6, 101.0, 81.6, 75.6, 75.2, 73.2,
72.7, 71.8, 69.8, 69.7, 63.2, 61.7, 57.2. Anal. Calcd for C41H36O15-
Br4: C, 45.25; H, 3.33. Found: C 45.24; H, 3.43.
Meth yl 2-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu cop yr a n osyl)-3,4,6-t r i-O-a cet yl-â-D-glu cop yr a n osid e
(14). Acetylation of 13 (47.1 mg, 0.043 mmol) with acetic
anhydride/pyridine (2 mL) at room temperature led to com-
pound 14 (53.1 mg, quantitative): TLC Rf ) 0.47 (n-hexane/
EtOAc, 6:4); [R]2D5 ) +42.2 (c 2.46, CHCl3); FAB-MS m/z MS
1237 (7, [M + Na]+), 183 (100, BrBz); 1H NMR (CDCl3) δ 7.84
(d, J ) 8.6 Hz, 2H), 7.74 (d, J ) 8.6 Hz, 2H), 7.69 (d, J ) 8.6
Hz, 2H), 7.63 (d, J ) 8.6 Hz, 2H), 7.53 (d, J ) 8.6 Hz, 4H),
7.47 (d, J ) 8.6 Hz, 2H), 7.42 (d, J ) 8.6 Hz, 2H), 5.75 (t, J )
9.6 Hz, 1H), 5.64 (t, J ) 9.6 Hz, 1H), 5.40 (dd, J ) 7.8 and 9.6
Hz, 1H), 5.13 (t, J ) 9.6 Hz, 1H), 5.11 (d, J ) 7.8 Hz, 1H),
4.93 (t, J ) 9.6 Hz, 1H), 4.64 (dd, J ) 3.6 and 12.2 Hz, 1H),
4.46 (dd, J ) 4.3 and 12.2 Hz, 1H), 4.45 (d, J ) 7.4 Hz, 1H),
4.23 (dd, J ) 4.7 and 12.3 Hz, 1H), 4.11 (m, 1H), 4.07 (dd, J )
2.2 and 12.3 Hz, 1H), 3.70 (dd, J ) 7.4 and 9.6 Hz, 1H), 3.63
(ddd, J ) 2.2, 4.7 and 9.9 Hz, 1H), 3.53 (s, 3H), 2.05 (s, 3H),
1.93 (s, 3H), 1.71 (s, 3H); 13C NMR (CDCl3) δ 170.5, 169.6,
169.6, 165.3, 165.0, 164.3, 164.2, 131.9-127.3, 102.5, 100.3,
78.7, 73.5, 73.3, 72.5, 71.7, 71.3, 69.6, 68.4, 63.0, 61.9, 57.1,
20.7, 20.5, 20.3. Anal. Calcd for C47H42O18Br4: C, 46.48; H,
3.49. Found: C 46.46; H, 3.43.
2H), 7.58 (d, J ) 8.5 Hz, 2H), 7.54 (d, J ) 8.6 Hz, 2H), 7.51 (d,
J ) 8.6 Hz, 2H), 7.45 (d, J ) 8.5 Hz, 2H), 5.84 (t, J ) 9.7 Hz,
1H), 5.60 (t, J ) 9.7 Hz, 1H), 5.50 (dd, J ) 8.0 and 9.7 Hz,
1H), 5.20 (d, J ) 8.0 Hz, 1H), 4.74 (d, J ) 3.3 Hz, 1H), 4.73
(dd, J ) 2.8 and 12.3 Hz, 1H), 4.39 (dd, J ) 5.3 and 12.3 Hz,
1H), 4.16 (m, 1H), 3.90 (t, J ) 9.7 Hz, 1H), 3.80 (m, 2H), 3.62
(m, 1H), 3.54 (dd, J ) 3.3 and 9.7 Hz, 1H), 3.52 (t, J ) 9.7 Hz,
1H), 3.30 (s, 3H); 13C NMR (CDCl3) δ 165.4, 165.1, 164.5, 164.5,
132.0-127.3, 101.0, 99.4, 81.0, 73.0, 72.2 (x 2), 71.7, 70.8, 70.5,
69.4, 62.8, 62.4, 55.3. Anal. Calcd for C41H36O15Br4: C, 45.25;
H, 3.33. Found: C 45.20; H, 3.51.
Meth yl 4,6-O-Ben zylid en e-â-D-glu cop yr a n osid e (17). A
solution of methyl â-D-glucopyranoside (998 mg, 5.13 mmol),
p-TsOH‚H2O (9.7 mg, 0.05 mmol), and benzaldehyde dimethyl-
acetal (1.16 mL, 7.73 mmol) in dry DMF (10 mL) was heated
in a rotavapor at 50 °C. After 2 h, the DMF was distilled off,
and then saturated aqueous NaHCO3 (10 mL), ether (10 mL)
and ice were added. After being stirred for 30 min, the mixture
was filtered, and the solid was washed with ether and cold
water, to give the benzylidene 17 (1.27 g, 87%): TLC Rf ) 0.62
(CHCl3/MeOH, 9:1); [R]25 ) -33.9 (c 0.59, CHCl3); 1H NMR
(CDCl3) δ 7.49 (m, 2H), D7.38 (m, 3H), 5.54 (s, 1H), 4.36 (dd, J
) 5.0 and 10.5 Hz, 1H), 4.33 (d, J ) 7.8 Hz, 1H), 3.83 (t, J )
9.1 Hz, 1H), 3.79 (t, J ) 10.2 Hz, 1H), 3.58 (s, 3H), 3.55 (t, J
) 9.2 Hz, 1H), 3.50 (t, J ) 9.2 Hz, 1H), 3.46 (m, 1H), 2.83 (br
s, 1H), 2.68 (br s, 1H); 13C NMR (CDCl3) δ 136.9, 129.3, 128.4,
126.3, 104.1, 101.9, 80.6, 74.6, 73.2, 68.7, 66.4, 57.5. Anal.
Calcd for C14H18O6: C, 59.57; H, 6.43. Found: C 59.66; H, 6.39.
Meth yl 2,3-Di-O-a cetyl-4,6-O-ben zylid en e-â-D-glu cop y-
r a n osid e (18). Acetylation of 17 (923.5 mg, 3.25 mmol) with
acetic anhydride/pyridine (1:1) led to compound 18 (3.05 mmol,
94% yield) after chromatography on silica gel (n-hexane/
EtOAc, 9:1): TLC Rf ) 0.51 (n-hexane/EtOAc, 7:3); [R]25
)
D
1
-97.9 (c 3.33, CHCl3); H NMR (CDCl3) δ 7.43 (m, 2H), 7.35
(m, 3H), 5.50 (s, 1H), 5.32 (t, J ) 9.3 Hz, 1H), 4.99 (dd, J )
7.8 and 9.3 Hz, 1H), 4.50 (d, J ) 7.8 Hz, 1H), 4.37 (dd, J ) 5.0
and 10.5 Hz, 1H), 3.80 (t, J ) 10.3 Hz, 1H), 3.69 (t, J ) 9.3
Hz, 1H), 3.53 (m, 1H), 3.51 (s, 3H), 2.06 (s, 3H), 2.04 (s, 3H);
13C NMR (CDCl3) δ 170.0, 169.6, 136.8, 129.0, 128.2, 126.1,
126.1, 102.2, 101.4, 78.3, 72.2, 71.8, 68.5, 66.2, 57.2, 20.7, 20.6.
Anal. Calcd for C18H22O8: C, 59.01; H, 6.05. Found: C 59.00;
H, 5.98.
Met h yl 2,3-Di-O-a cet yl-6-O-b en zyl-â-D-glu cop yr a n o-
sid e (19). To a solution of 18 (429.6 mg, 1.34 mmol) in dry
THF (10 mL) at 0 °C were added sodium cyanoborohydride
(420 mg, 6.68 mmol) and a solution of trifluoroacetic acid (1.03
mL, 13.37 mmol) in dry THF (8 mL). After being stirred 1 h
at 0 °C and 2 weeks at room temperature, the reaction mixture
was quenched with saturated aqueous NaHCO3, diluted with
CH2Cl2 (300 mL), and washed with saturated aqueous NaH-
CO3. The organic layer was dried over MgSO4, filtered, and
concentrated under reduced pressure. The residue was purified
by flash chromatography (n-hexane/EtOAc, 6:4) to provide the
benzyl alcohol 19 (360 mg, 73%): TLC Rf ) 0.44 (n-hexane);
[R]2D5 ) -29.3 (c 3.0, CHCl3); 1H NMR (CDCl3) δ 7.29 (m, 5H),
5.03 (t, J ) 9.4 Hz, 1H), 4.87 (dd, J ) 7.9 and 9.4 Hz, 1H),
4.59 (d, J ) 12.0 Hz, 1H), 4.54 (d, J ) 12.0 Hz, 1H), 4.37 (d,
J ) 7.9 Hz, 1H), 3.76 (brd, J ) 4.8 Hz, 2H), 3.70 (brt, J ) 9.4
Hz, 1H), 3.51 (m, 1H), 3.46 (s, 3H), 3.30 (br s, 1H), 2.04 (s,
3H), 2.01 (s, 3H); 13C NMR (CDCl3) δ 171.3, 169.7, 137.5,
129.0-127.8, 101.5, 75.6, 74.2, 73.7, 71.4, 70.5, 69.9, 56.8, 20.8,
20.7. Anal. Calcd for C28H24O8: C, 58.69; H, 6.57. Found: C
58.68; H, 6.52.
Meth yl 4-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu cop yr a n osyl)-2,3-d i-O-a cetyl-6-O-ben zyl-â-D-glu cop y-
r a n osid e (20). To a solution of acceptor 19 (163.4 mg, 0.44
mmol), silver triflate (226 mg, 0.88 mmol), and sym-collidine
(52 µL, 0.40 mmol) in dry toluene/nitromethane (1:1) was
added the glucosyl bromide 2 (859 mg, 0.88 mmol). After being
stirred for 75 min, a few drops of pyridine were added to the
reaction, which was then filtered through a bed of Celite and
concentrated under reduced pressure. The residue was purified
Meth yl 2-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu cop yr a n osyl)-3,4,6-t r i-O-a cet yl-r-D-glu cop yr a n osid e
(15). Following the general procedure, anomerization of 14
with anhydrous FeCl3 led to compound 15 (19.9 mg, 55%
yield): TLC Rf ) 0.48 (n-hexane/EtOAc, 6:4); [R]25 ) +12.6 (c
D
0.33, CHCl3); 1H NMR (CDCl3) δ 7.86 (d, J ) 8.5 Hz, 2H), 7.74
(d, J ) 8.5 Hz, 2H), 7.72 (d, J ) 8.5 Hz, 2H), 7.63 (d, J ) 8.5
Hz, 2H), 7.57 (d, J ) 8.5 Hz, 2H), 7.52 (d, J ) 8.5 Hz, 2H),
7.49 (d, J ) 8.5 Hz, 2H), 7.42 (d, J ) 8.5 Hz, 2H), 5.81 (t, J )
9.7 Hz, 1H), 5.62 (t, J ) 9.7 Hz, 1H), 5.47 (dd, J ) 8.0 and 9.7
Hz, 1H), 5.34 (t, J ) 9.7 Hz, 1H), 4.98 (d, J ) 3.4 Hz, 1H),
4.93 (t, J ) 9.7 Hz, 1H), 4.92 (d, J ) 8.0 Hz, 1H), 4.70 (dd, J
) 3.0 and 12.2 Hz, 1H), 4.39 (dd, J ) 4.8 and 12.2 Hz, 1H),
4.26 (dd, J ) 4.6 and 12.3 Hz, 1H), 4.09 (m, 1H), 4.03 (dd, J )
1.9 and 12.3 Hz, 1H), 3.94 (m, 1H), 3.74 (dd, J ) 3.5 and 10.0
Hz, 1H), 3.37 (s, 3H), 2.08 (s, 3H), 1.94 (s, 3H), 1.45 (s, 3H);
13C NMR (CDCl3) δ 170.6, 169.8, 169.3, 165.2, 165.1, 164.4,
163.9, 131.9-127.2, 102.0, 99.1, 78.4, 73.0, 72.0, 71.9, 71.0,
69.4, 68.7, 66.8, 62.6, 62.0, 55.6, 20.7, 20.5, 20.1. Anal. Calcd
for C47H42O18Br4: C, 46.48; H, 3.49. Found: C 46.45; H, 3.82.
Meth yl 2-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu cop yr a n osyl)-r-D-glu cop yr a n osid e (16). According to
the general procedure for deacetylation, 26 mg (0.021 mmol)
of 15 led to the desired disaccharide 16 (10.4 mg, 41% yield):
TLC Rf ) 0.42 (CHCl3/MeOH, 9:1); [R]25 ) +3.83 (c 0.43,
D
1
THF); H NMR (CDCl3) δ 7.88 (d, J ) 8.5 Hz, 2H), 7.79 (d, J
) 8.5 Hz, 2H), 7.74 (d, J ) 8.5 Hz, 2H), 7.67 (d, J ) 8.5 Hz,
4628 J . Org. Chem., Vol. 68, No. 12, 2003

Hydroxymethyl Rotamers in Disaccharides
by chromatography on Sephadex LH-20 (CHCl3/MeOH/n-
hexane, 1:1:2) to give the disaccharide 20 (408 mg, 73%
yield): TLC Rf ) 0.53 (n-hexane/EtOAc, 7:3); [R]25 ) +10.0 (c
1H), 3.52 (s, 3H), 3.45 (m, 1H), 3.38 (dd, J ) 7.8 and 9.1 Hz,
1H), 3.28 (brd, J ) 9.3 Hz, 1H); 13C NMR (CDCl3) δ 165.4,
164.9, 164.4, 164.2, 132.0-127.2, 103.4, 101.5, 80.2, 74.4, 74.0,
73.7, 72.9, 72.6, 71.7, 69.1, 62.5, 60.3, 57.4. Anal. Calcd for
2.5, CHCl3); FAB-MS m/z 1284 (24, [M + Na]+), D894 (3, [M -
1
18H21O8]+), 183 (100, BrBz); H NMR (CDCl3) δ 7.88 (d, J )
C
41H36O15Br4: C, 45.25; H, 3.33. Found: C 45.35; H, 3.62.
C
Meth yl 4-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
8.6 Hz, 2H), 7.72-7.41 (19 H), 5.61 (t, J ) 9.6 Hz, 1H), 5.49
(t, J ) 9.6 Hz, 1H), 5.37 (dd, J ) 8.0 and 9.6 Hz, 1H), 5.16 (t,
J ) 9.6 Hz, 1H), 4.90 (dd, J ) 8.0 and 9.6 Hz, 1H), 4.78 (d, J
) 12.2 Hz, 1H), 4.77 (d, J ) 8.0 Hz, 1H), 4.57 (dd, J ) 2.9 and
12.2 Hz, 1H), 4.41 (d, J ) 12.2 Hz, 1H), 4.36 (dd, J ) 5.3 and
12.2 Hz, 1H), 4.28 (d, J ) 8.0 Hz, 1H), 4.03 (t, J ) 9.6 Hz,
1H), 3.80 (ddd, J ) 2.9, 5.3 and 9.6 Hz, 1H), 3.64 (dd, J ) 2.8
and 11.3 Hz, 1H), 3.56 (brd, J ) 10.2 Hz, 1H), 3.44 (s, 3H),
3.31 (brd, J ) 9.8 Hz, 1H), 2.02 (s, 3H), 1.94 (s, 3H); 13C NMR
(CDCl3) δ 169.9, 169.5, 165.3, 164.9, 164.3, 163.8, 137.9, 132.0-
127.3, 101.6, 100.2, 75.1, 74.3, 73.7, 73.2, 72.6, 71.8, 71.7, 71.5,
glu cop yr a n osyl)-2,3,6-t r i-O-a cet yl-r-D-glu cop yr a n osid e
(24). According to the general procedure for anomerization,
compound 22 (24.5 mg, 0.02 mmol) was treated with anhy-
drous FeCl3 leading to the desired compound 24 (0.01 mmol,
50% yield): TLC Rf ) 0.40 (n-hexane/EtOAc, 6:4); [R]D25
)
+38.4 (c 0.31, CHCl3); 1H NMR (CDCl3) δ 7.86 (d, J ) 8.5 Hz,
2H), 7.81 (d, J ) 8.5 Hz, 2H), 7.68 (d, J ) 8.5 Hz, 2H), 7.62 (d,
J ) 8.5 Hz, 2H), 7.59 (d, J ) 8.5 Hz, 2H), 7.55 (d, J ) 8.5 Hz,
2H), 7.48 (d, J ) 8.5 Hz, 2H), 7.43 (d, J ) 8.5 Hz, 2H), 5.76 (t,
J ) 9.6 Hz, 1H), 5.60 (t, J ) 9.7 Hz, 1H), 5.47 (m, 2H), 4.81
(m, 2H), 4.80 (d, J ) 8.0 Hz, 1H), 4.57 (dd, J ) 3.1 and 12.3
Hz, 1H), 4.47 (dd, J ) 4.6 and 12.3 Hz, 1H), 4.21 (brd, J )
12.2 Hz, 1H), 4.16 (dd, J ) 3.6 and 12.2 Hz, 1H), 4.07 (m, 1H),
3.78 (m, 1H), 3.76 (t, J ) 10.1 Hz, 1H), 3.31 (s, 3H), 2.05 (s,
3H), 2.04 (s, 3H), 1.94 (s, 3H); 13C NMR (CDCl3) δ 170.4, 170.3,
169.5, 165.3, 164.9, 164.3, 163.9, 132.0-127.2, 100.9, 96.6, 76.7,
73.1, 71.9, 71.8, 70.7, 69.4, 69.3, 67.8, 62.8, 61.8, 55.3, 21.2,
20.8 (x2). Anal. Calcd for C47H42O18Br4: C, 46.48; H, 3.49.
Found: C 46.44; H, 3.64.
69.6, 67.0, 62.9, 56.9, 20.7, 20.6. Anal. Calcd for C52H46O17
Br4: C, 49.47; H, 3.67. Found: C 49.49; H, 3.54.
-
Meth yl 4-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu copyr an osyl)-2,3-di-O-acetyl-â-D-glu copyr an oside (21).
Debenzylation of the disaccharide 20 (20 mg, 0.016 mmol), in
dry CH2Cl2 (1.5 mL) and at 0 °C, with anhydrous FeCl3 (16
mg, 0.099 mmol) led to compound 21 (17.7 mg, 95% yield) after
chromatography on silica gel (n-hexane/EtOAc, 7:3): TLC Rf
) 0.26 (n-hexane/EtOAc, 6:4); [R]2D5 ) +4.7 (c 0.87, CHCl3); 1H
NMR (CDCl3) δ 7.88 (d, J ) 8.6 Hz, 2H), 7.81 (d, J ) 8.6 Hz,
2H), 7.69 (d, J ) 8.6 Hz, 2H), 7.63 (d, J ) 8.6 Hz, 2H), 7.62 (d,
J ) 8.5 Hz, 2H), 7.55 (d, J ) 8.6 Hz, 2H), 7.48 (d, J ) 8.6 Hz,
2H), 7.43 (d, J ) 8.6 Hz, 2H), 5.78 (t, J ) 9.7 Hz, 1H), 5.58 (t,
J ) 9.7 Hz, 1H), 5.46 (dd, J ) 8.0 and 9.7 Hz, 1H), 5.20 (t, J
) 9.6 Hz, 1H), 4.97 (d, J ) 8.0 Hz, 1H), 4.86 (dd, J ) 8.0 and
9.6 Hz, 1H), 4.62 (dd, J ) 3.0 and 12.2 Hz, 1H), 4.41 (dd, J )
5.2 and 12.2 Hz, 1H), 4.34 (d, J ) 8.0 Hz, 1H), 4.12 (ddd, J )
3.0, 5.2 and 9.7 Hz, 1H), 3.99 (t, J ) 9.6 Hz, 1H), 3.76 (brd, J
) 12.4 Hz, 1H), 3.68 (m, 1H), 3.44 (s, 3H), 3.26 (br d, J ) 9.7
Meth yl 4-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu cop yr a n osyl)-r-D-glu cop yr a n osid e (25). Following the
general procedure, deacetylation of 24 (11.0 mg, 9.0 µmol) with
p-TsOH‚H2O (8 mg, 0.041 mmol) led to compound 25 (7.8 µmol,
87% yield): TLC Rf ) 0.56 (CH2Cl2/MeOH, 9:1); [R]25 ) +33.9
D
1
(c 0.31, CHCl3); H NMR (CDCl3) δ 7.95 (d, J ) 8.5 Hz, 2H),
7.82 (d, J ) 8.5 Hz, 2H), 7.74 (d, J ) 8.5 Hz, 2H), 7.64 (d, J )
8.5 Hz, 2H), 7.58 (d, J ) 8.5 Hz, 2H), 7.55 (d, J ) 8.5 Hz, 2H),
7.52 (d, J ) 8.5 Hz, 2H), 7.44 (d, J ) 8.5 Hz, 2H), 5.82 (t, J )
9.7 Hz, 1H), 5.59 (t, J ) 9.7 Hz, 1H), 5.51 (dd, J ) 8.0 and 9.7
Hz, 1H), 4.98 (d, J ) 8.0 Hz, 1H), 4.76 (dd, J ) 2.2 and 12.3
Hz, 1H), 4.74 (d, J ) 3.7 Hz, 1H), 4.36 (dd, J ) 5.7 and 12.3
Hz, 1H), 4.22 (m, 1H), 3.88 (br s, 1H), 3.84 (t, J ) 9.1 Hz, 1H),
3.67 (t, J ) 9.1 Hz, 1H), 3.51 (m, 4H), 3.35 (s, 3H), 2.30 (br s,
1H), 1.63 (br s, 1H); 13C NMR (CDCl3) δ 165.4, 164.9, 164.4,
164.2, 132.0-127.2, 101.5, 99.1, 80.6, 72.9, 72.5, 72.3, 72.2,
Hz, 1H), 2.03 (s, 3H), 1.96 (s, 3H). Anal. Calcd for C45H40O17
Br4: C, 46.10; H, 3.44. Found: C 46.31; H, 3.47.
-
Meth yl 4-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu cop yr a n osyl)-2,3,6-t r i-O-a cet yl-â-D-glu cop yr a n osid e
(22). Compound 21 (12 mg, 0.010 mmol) was acetylated by
treatment with 2 mL of acetic anhydride/pyridine (1:1) at room
temperature to afford compound 22 (12.1 mg, 97% yield): TLC
Rf ) 0.64 (n-hexane/EtOAc, 6:4); [R]2D5 ) +10.2 (c 0.88,
71.8, 69.5, 69.2, 62.6, 60.4, 55.5. Anal. Calcd for C41H36O15
Br4: C, 45.25; H, 3.33. Found: C 45.34; H, 3.60.
-
1
Meth yl 2-O-Ben zyl-4,6-O-ben zylid en e-â-D-glu cop yr a -
n osid e (26). To a solution of 17 (677 mg, 2.38 mmol) in dry
DMF (2 mL), sodium hydride 80% dispersion oil (85.7 mg, 2.86
mmol) and benzyl bromide (310 µL, 2.61 mmol) were added.
After being stirred overnight, the reaction mixture was diluted
with CH2Cl2, washed with H2O and saturated aqueous NH4-
Cl, dried (MgSO4), and concentrated. The residue was chro-
matographed on silica gel (n-hexane/EtOAc, 9:1) to provide
compound 26 (205 mg, 23% yield): TLC Rf ) 0.30 (n-hexane/
EtOAc, 6:4); [R]2D5 ) -26.2 (c 0.94, CHCl3); 1H NMR (CDCl3) δ
7.46 (m, 2H), 7.37-7.27 (m, 8H), 5.48 (s, 1H), 4.88 (d, J ) 11.4
Hz, 1H), 4.72 (d, J ) 11.4 Hz, 1H), 4.38 (d, J ) 7.7 Hz, 1H),
4.31 (dd, J ) 5.0 and 10.4 Hz, 1H), 3.80 (t, J ) 9.3 Hz, 1H),
3.73 (t, J ) 10.2 Hz, 1H), 3.54 (s, 3H), 3.50 (t, J ) 9.3 Hz, 1H),
3.37 (m, 1H), 3.31 (dd, J ) 7.7 and 9.3 Hz, 1H), 3.15 (brs, 1H);
13C NMR (CDCl3) δ 138.3, 137.0, 128.9, 128.2, 128.1, 127.8,
127.6, 126.2, 104.7, 101.5, 81.9, 80.4, 74.6, 72.9, 68.5, 65.9, 57.2.
Anal. Calcd for C21H24O5: C, 70.77; H, 6.79. Found: C 70.78;
H, 6.82.
Meth yl 3-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu cop yr a n osyl)-2-O-b en zyl-4,6-O-b en zylid en e-â-D-glu -
cop yr a n osid e (27). To a solution of 26 (122 mg, 0.32 mmol)
and silver triflate (164 mg, 0.64 mmol) in dry toluene (4.0 mL),
the glucopyranosyl bromide 2 (635 mg, 0.64 mmol) was added.
After 30 min the reaction mixture was quenched by adding a
few drops of water and filtering through a Celite bed. Purifica-
tion by flash chromatography (n-hexane/EtOAc, 9:1) led to the
disaccharide 27 (182 mg, 45% yield): TLC Rf ) 0.37 (n-hexane/
EtOAc, 7:3); [R]2D5 ) +23.3 (c 1.26, CHCl3); 1H NMR (CDCl3) δ
CHCl3); H NMR (CDCl3) δ 7.87 (d, J ) 8.6 Hz, 2H), 7.80 (d,
J ) 8.6 Hz, 2H), 7.69 (d, J ) 8.6 Hz, 2H), 7.62 (t, J ) 8.6 Hz,
4H), 7.55 (d, J ) 8.5 Hz, 2H), 7.48 (d, J ) 8.5 Hz, 2H), 7.43 (d,
J ) 8.6 Hz, 2H), 5.76 (t, J ) 9.6 Hz, 1H), 5.58 (t, J ) 9.6 Hz,
1H), 5.45 (dd, J ) 8.0 and 9.6 Hz, 1H), 5.21 (t, J ) 9.5 Hz,
1H), 4.87 (dd, J ) 7.9 and 9.5 Hz, 1H), 4.79 (d, J ) 8.0 Hz,
1H), 4.59 (dd, J ) 3.2 and 12.3 Hz, 1H), 4.45 (dd, J ) 5.0 and
12.3 Hz, 1H), 4.31 (d, J ) 7.9 Hz, 1H), 4.27 (dd, J ) 1.6 and
12.2 Hz, 1H), 4.11 (dd, J ) 4.6 and 12.2 Hz, 1H), 4.07 (m, 1H),
3.81 (t, J ) 9.5 Hz, 1H), 3.49 (m, 1H), 3.42 (s, 3H), 2.02 (s,
6H), 1.96 (s, 3H); 13C NMR (CDCl3) δ 170.4, 169.8, 169.6, 165.3,
164.9, 164.4, 163.9, 132.0-127.3, 101.4, 100.9, 76.5, 73.1, 72.5,
72.4, 72.0, 72.0, 71.5, 69.4, 62.9, 61.8, 57.0, 20.7, 20.7, 20.6.
Anal. Calcd for C47H42O18Br4: C, 46.48; H, 3.49. Found: C
46.47; H, 3.64.
Meth yl 4-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu cop yr a n osyl)-â-D-glu cop yr a n osid e (23). Following the
general procedure for deacetylation, p-TsOH‚H2O was added
to a solution of disaccharide 22 (6.5 mg, 5.3 µmol) to give
compound 23 (4.5 mg, 77% yield): TLC Rf ) 0.65 (CHCl3/
MeOH, 9:1); [R]2D5 ) +32.0 (c 0.1, CHCl3); H NMR (CDCl3) δ
1
7.95 (d, J ) 8.5 Hz, 2H), 7.82 (d, J ) 8.5 Hz, 2H), 7.74 (d, J )
8.5 Hz, 2H), 7.64 (d, J ) 8.5 Hz, 2H), 7.59 (d, J ) 8.6 Hz, 2H),
7.51 (d, J ) 8.6 Hz, 2H), 7.52 (d, J ) 8.5 Hz, 2H), 7.44 (d, J )
8.5 Hz, 2H), 5.82 (t, J ) 9.8 Hz, 1H), 5.59 (t, J ) 9.8 Hz, 1H),
5.50 (dd, J ) 8.0 and 9.8 Hz, 1H), 4.99 (d, J ) 8.0 Hz, 1H),
4.78 (dd, J ) 2.5 and 12.3 Hz, 1H), 4.35 (dd, J ) 5.7 and 12.3
Hz, 1H), 4.24 (d, J ) 7.8 Hz, 1H), 4.22 (m, 1H), 3.96 (br s,
1H), 3.74 (t, J ) 9.1 Hz, 1H), 3.67 (t, J ) 9.1 Hz, 1H), 3.60 (m,
J . Org. Chem, Vol. 68, No. 12, 2003 4629

Roe¨n et al.
7.75 (d, J ) 8.5 Hz, 2H), 7.65 (d, J ) 8.6 Hz, 2H), 7.62 (d, J )
8.6 Hz, 2H), 7.59 (d, J ) 8.6 Hz, 2H), 7.51-7.54 (m, 6H), 7.40
(d, J ) 8.5 Hz, 2H), 7.34-7.25 (m, 8H), 7.17 (m, 2H), 5.74 (t,
J ) 9.6 Hz, 1H), 5.59 (t, J ) 9.6 Hz, 1H), 5.57 (s, 1H), 5.54
(dd, J ) 7.9 and 9.6 Hz, 1H), 5.21 (d, J ) 7.9 Hz, 1H), 4.65 (d,
J ) 11.2 Hz, 1H), 4.45 (dd, J ) 3.6 and 11.9 Hz, 1H), 4.43 (d,
J ) 11.2 Hz, 1H), 4.34 (m, 1H), 4.33 (d, J ) 7.5 Hz, 1H), 4.26
(dd, J ) 4.5 and 11.9 Hz, 1H), 3.99 (t, J ) 9.3 Hz, 1H), 3.86
(m, 1H), 3.78 (t, J ) 10.3 Hz, 1H), 3.70 (t, J ) 9.3 Hz, 1H),
3.47 (s, 3H), 3.38 (m, 2H); 13C NMR (CDCl3) δ 165.2, 165.0,
164.4, 164.3, 138.2, 137.1, 131.8-126.0, 104.9, 101.3, 100.8,
81.8, 80.4, 79.2, 74.7, 73.3, 72.3, 71.6, 69.8, 68.7, 66.0, 63.1,
57.3. Anal. Calcd for C55H46O14Br4: C, 52.82; H, 3.71. Found:
C 52.88; H, 3.65.
Meth yl 3-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu cop yr a n osyl)-â-D-glu cop yr a n osid e (28). To a solution
of the disaccharide 27 (14 mg, 0.011 mmol) in dry CH2Cl2 (2
mL) at 0 °C, anhydrous FeCl3 (16 mg, 0.099 mmol) was added
to afford the triol 28 (10.2 mg, 85% yield): TLC Rf ) 0.4 (CH2-
Cl2/MeOH, 9:1); [R]25 ) -4.2 (c 0.12, CHCl3); 1H NMR (CDCl3)
δ 7.92 (d, J ) 8.6 HDz, 2H), 7.78 (d, J ) 8.6 Hz, 2H), 7.74 (d, J
) 8.6 Hz, 2H), 7.67 (d, J ) 8.6 Hz, 2H), 7.57 (d, J ) 8.6 Hz,
2H), 7.52 (d, J ) 8.5 Hz, 2H), 7.51 (d, J ) 8.6 Hz, 2H), 7.44 (d,
J ) 8.6 Hz, 2H), 5.85 (t, J ) 9.7 Hz, 1H), 5.57 (t, J ) 9.7 Hz,
1H), 5.50 (dd, J ) 8.0 and 9.7 Hz, 1H), 5.00 (d, J ) 8.0 Hz,
1H), 4.75 (dd, J ) 2.6 and 12.3 Hz, 1H), 4.34 (dd, J ) 6.3 and
12.3 Hz, 1H), 4.20 (m, 1H), 4.14 (d, J ) 7.8 Hz, 1H), 3.91 (dd,
J ) 3.4 and 11.8 Hz, 1H), 3.76 (dd, J ) 5.1 and 11.8 Hz, 1H),
3.72 (brs, 1H), 3.56 (t, J ) 8.9 Hz, 1H), 3.49 (m, 1H), 3.49 (s,
3H), 3.32 (m, 2H), 2.05 (brs, 1H), 1.65 (brs, 1H); 13C NMR
(CDCl3) δ 165.4, 165.0, 164.5 (x2), 132.0-127.2, 103.6, 102.5,
88.4, 75.2, 72.7 (x2), 72.6, 72.5, 71.8, 69.4 (x 2), 62.8, 62.7, 57.3.
Anal. Calcd for C41H36O15Br4: C, 45.25; H, 3.33. Found: C
45.24; H, 3.40.
101.4, 100.8, 78.7, 73.1, 72.5, 72.2, 71.9, 71.1, 69.7, 68.3, 63.1,
62.2, 56.5, 20.9, 20.8, 20.7. Anal. Calcd for C47H42O18Br4: C,
46.48; H, 3.49. Found: C 46.47; H, 3.80.
Meth yl 3-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu cop yr a n osyl)-2,4,6-t r i-O-a cet yl-r-D-glu cop yr a n osid e
(30). Compound 30 (13.4 mg, 0.011 mmol, 65% yield) was
obtained from its â-anomer 29 (21 mg, 0.017 mmol) following
the general procedure for anomerization: TLC Rf ) 0.36 (n-
hexane/EtOAc, 6:4); [R]2D5 ) +27.6 (c 0.37, CHCl3); 1H NMR
(CDCl3) δ 7.87 (d, J ) 8.5 Hz, 2H), 7.71-7.63 (m, 6H), 7.57 (d,
J ) 8.5 Hz, 2H), 7.51 (d, J ) 8.5 Hz, 2H), 7.48 (d, J ) 8.5 Hz,
2H), 7.44 (d, J ) 8.5 Hz, 2H), 5.80 (t, J ) 9.7 Hz, 1H), 5.60 (t,
J ) 9.7 Hz, 1H), 5.33 (dd, J ) 8.1 and 9.7 Hz, 1H), 5.00 (t, J
) 9.7 Hz, 1H), 4.99 (d, J ) 8.1 Hz, 1H), 4.85 (d, J ) 3.6 Hz,
1H), 4.71 (dd, J ) 3.6 and 10.0 Hz, 1H), 4.56 (dd, J ) 3.1 and
12.3 Hz, 1H), 4.49 (dd, J ) 4.9 and 12.3 Hz, 1H), 4.17 (t, J )
9.4 Hz, 1H), 4.11 (m, 3H), 3.79 (m, 1H), 3.36 (s, 3H), 2.08 (s,
3H), 1.97 (s, 3H), 1.96 (s, 3H); 13C NMR (CDCl3) δ 170.7, 169.2,
169.2, 165.3, 165.2, 164.3, 164.2, 131.8-127.2, 100.9, 96.4, 76.0,
73.2, 73.0, 72.4, 71.6, 69.6, 67.9, 67.4, 63.1, 62.1, 55.4, 20.8,
20.7, 20.6. Anal. Calcd for C47H42O18Br4: C, 46.48; H, 3.49.
Found: C 46.48; H, 3.74.
Meth yl 3-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu cop yr a n osyl)-r-D-glu cop yr a n osid e (31). This compound
(5.0 mg, 4.6 µmol) was obtained from 30 (8 mg, 6.6 µmol)
following the general procedure for deacetylation (70% yield):
TLC Rf ) 0.66 (CH2Cl2/MeOH, 9:1); [R]25 ) +44.2 (c 0.50,
D
1
CHCl3); H NMR (CDCl3) δ 7.93 (d, J ) 8.5 Hz, 2H), 7.81 (d,
J ) 8.5 Hz, 2H), 7.75 (d, J ) 8.5 Hz, 2H), 7.67 (d, J ) 8.5 Hz,
2H), 7.57 (d, J ) 8.5 Hz, 2H), 7.52 (d, J ) 8.5 Hz, 4H), 7.45 (d,
J ) 8.5 Hz, 2H), 5.83 (t, J ) 9.7 Hz, 1H), 5.57 (t, J ) 9.7 Hz,
1H), 5.49 (dd, J ) 8.0 and 9.7 Hz, 1H), 5.00 (d, J ) 8.0 Hz,
1H), 4.73 (dd, J ) 2.4 and 12.3 Hz, 1H), 4.68 (d, J ) 3.8 Hz,
1H), 4.35 (dd, J ) 6.4 and 12.3 Hz, 1H), 4.19 (m, 1H), 3.85
(dd, J ) 3.3 and 11.6 Hz, 1H), 3.76 (dd, J ) 4.5 and 11.6 Hz,
1H), 3.67 (t, J ) 8.9 Hz, 1H), 3.59 (m, 1H), 3.52 (m, 2H), 3.40
(s, 3H); 13C NMR (CDCl3) δ 165.4, 165.0, 164.6, 164.5, 132.0-
127.2, 102.4, 99.0, 87.4, 72.8, 72.5, 72.0, 70.9, 70.8, 69.4, 69.3,
62.9, 62.6, 55.3. Anal. Calcd for C41H36O15Br4: C, 45.25; H,
3.33. Found: C 45.28; H, 3.46.
Meth yl 3-O-(2,3,4,6-Tetr a k is-O-(p-br om oben zoyl)-â-D-
glu cop yr a n osyl)-2,4,6-t r i-O-a cet yl-â-D-glu cop yr a n osid e
(29). Acetylation of 28 (11.3 mg, 9.6 µmol) with acetic
anhydride/pyridine (1:1) led to compound 29 (11.5 mg) in 98%
yield: TLC Rf ) 0.26 (n-hexane/EtOAc, 6:4); [R]25 ) -2.0 (c
D
0.84, CHCl3); 1H NMR (CDCl3) δ 7.86 (d, J ) 8.5 Hz, 2H), 7.69
(d, J ) 7.3 Hz, 4H), 7.64 (d, J ) 8.6 Hz, 2H), 7.57 (d, J ) 8.6
Hz, 2H), 7.51 (d, J ) 8.5 Hz, 2H), 7.48 (d, J ) 8.6 Hz, 2H),
7.43 (d, J ) 8.6 Hz, 2H), 5.80 (t, J ) 9.7 Hz, 1H), 5.60 (t, J )
9.7 Hz, 1H), 5.33 (dd, J ) 7.9 and 9.7 Hz, 1H), 5.01 (t, J ) 9.3
Hz, 1H), 4.94 (d, J ) 7.9 Hz, 1H), 4.89 (dd, J ) 7.8 and 9.3
Hz, 1H), 4.57 (dd, J ) 3.3 and 12.2 Hz, 1H), 4.49 (dd, J ) 4.8
and 12.2 Hz, 1H), 4.25 (d, J ) 7.8 Hz, 1H), 4.17 (m, 2H), 4.11
(m, 1H), 3.95 (t, J ) 9.3 Hz, 1H), 3.58 (m, 1H), 3.38 (s, 3H),
2.07 (s, 3H), 1.97 (s, 3H), 1.96 (s, 3H); 13C NMR (CDCl3) δ
170.7, 169.1, 168.4, 165.3, 165.2, 164.4, 164.3, 131.9-127.2,
Ack n ow led gm en t. This research was supported by
the Ministerio de Ciencia y Tecnologı´a (Spain), through
grants BQU2000-0250 and 1FD97-0747-C04-01. The
authors are grateful to Dr. Ezequiel Q. Morales for his
help in the rotameric calculations.48 ARM thanks the
Banco Santander for a fellowship.
J O026913O
4630 J . Org. Chem., Vol. 68, No. 12, 2003