Anionic Additions to Glycosyl Iodides: Highly Stereoselective Syntheses of β C-, N-, and O-Glycosides

Article abstract of DOI:10.1021/jo970922t

Classically, glycosyl halides are activated as glycosyl donors by metal chelation under KoenigsKnorr or Helferich conditions. These reactions often proceed through oxonium formation, and the stereochemical outcome is dictated by the anorneric effect and/or the nature of the protecting group on the C2 hydroxyl. Alternatively, glycosyl halides may undergo direct displacement of the halide by an incoming nucleophile in an S_N2 mechanism. The latter reaction is far less common, and before this study it was primarily performed with glycosyl bromides. Having recently shown that both α and βglycosyl iodides could be efficiently generated, we embarked upon an investigation of nucleophilic additions to glycosyl iodides. The studies reported herein show that additions of stabilized anions to α-glycosyl iodides proceed with inversion of stereochemistry to give β-glycosides, even in the absence of a C2 participatory group. Glucosyl, galactosyl, and mannosyl iodides were studied, and the combined results indicate that the reactivity of 2,3,4,6-tetra-O-benzyl-α-D-galactosyl iodide > 2,3,4,6-tetra-O-benzyl-α-D-glucosyl iodide > 2,3,4,6-tetra-O-benzyl-α-D-mannosyl iodide. Both the glucosy] and galactosyl iodides are susceptible to E-2 elimination when treated with highly basic anions. In contrast, the mannosyl iodide undergoes substitution to give the 1,2 cis configuration. The overall sequence involves reaction of an anorneric acetate with trimethylsilyl iodide with in vacua removal of the resulting trimethylsilyl acetate. The iodide is then treated with a nucleophile without further characterization. A variety of nucleophiles were stereoselectively added to the glycosyl halides providing β-, C-, N-, and O-glycosides.

Full text of DOI:10.1021/jo970922t

J . Org. Chem. 1997, 62, 6961-6967
6961
An ion ic Ad d ition s to Glycosyl Iod id es: High ly Ster eoselective
Syn th eses of â C-, N-, a n d O-Glycosid es1
J acquelyn Gervay* and Michael J . Hadd
Department of Chemistry, The University of Arizona, Tucson, Arizona 85721
Received May 23, 1997X
Classically, glycosyl halides are activated as glycosyl donors by metal chelation under Koenigs-
Knorr or Helferich conditions. These reactions often proceed through oxonium formation, and the
stereochemical outcome is dictated by the anomeric effect and/or the nature of the protecting group
on the C2 hydroxyl. Alternatively, glycosyl halides may undergo direct displacement of the halide
by an incoming nucleophile in an SN2 mechanism. The latter reaction is far less common, and
before this study it was primarily performed with glycosyl bromides. Having recently shown that
both R and â glycosyl iodides could be efficiently generated, we embarked upon an investigation of
nucleophilic additions to glycosyl iodides. The studies reported herein show that additions of
stabilized anions to R-glycosyl iodides proceed with inversion of stereochemistry to give â-glycosides,
even in the absence of a C2 participatory group. Glucosyl, galactosyl, and mannosyl iodides were
studied, and the combined results indicate that the reactivity of 2,3,4,6-tetra-O-benzyl-R-D-galactosyl
iodide > 2,3,4,6-tetra-O-benzyl-R-D-glucosyl iodide > 2,3,4,6-tetra-O-benzyl-R-D-mannosyl iodide.
Both the glucosyl and galactosyl iodides are susceptible to E-2 elimination when treated with highly
basic anions. In contrast, the mannosyl iodide undergoes substitution to give the 1,2 cis
configuration. The overall sequence involves reaction of an anomeric acetate with trimethylsilyl
iodide with in vacuo removal of the resulting trimethylsilyl acetate. The iodide is then treated
with a nucleophile without further characterization. A variety of nucleophiles were stereoselectively
added to the glycosyl halides providing â-, C-, N-, and O-glycosides.
Sch em e 1
In tr od u ction
Glycosyl halides are most commonly employed in
glycosylation methods involving activation of the halide
through metal chelation.2 The stereochemical outcome
of the glycosylation may be dictated by a combination of
factors including neighboring group participation,3 sol-
vent participation,4 and the metal catalyst.5 An alterna-
tive glycosylation strategy involves reaction of glycosyl
halides under basic conditions. These reactions typically
proceed through nucleophilic displacement of the ano-
meric halide in an SN2 fashion, or by generation of an
oxonium ion that is subsequently trapped by the nucleo-
phile.6
One advantage of SN2 displacement reactions is their
potential for stereospecificity. In the absence of neigh-
boring group participation, the R-glycosyl halide would
be expected to give the â-glycoside and vice versa.
Furthermore, the reactions often proceed under relatively
neutral conditions. For example, treatment of a glycosyl
bromide with an alcohol in the presence of 2,6-lutidine
results in glycoside formation, albeit in low to moderate
yields.7 The yields and reaction rates can be increased
by in situ generation of the more reactive â-glycosyl
halide, with subsequent formation of the R-glycoside.
Glycosyl iodides have also been generated from other
glycosyl donors in situ, and they show greater reactivity
toward nucleophilic displacement under neutral condi-
tions.8 â-Glycosides generally result from anionic addi-
tions to R-glycosyl bromides, but the yields are variable
and often elimination of HBr and formation of the glycal
is observed.9
Recently we reported on the stereoselective generation
of glycosyl iodides from anomeric acetates.10 In the
course of those investigations we discovered that reaction
of 2,3,4,6-tetra-O-benzyl-R-D-glucopyranosyl acetate 1
with trimethylsilyl iodide results in the initial formation
of the â-glycosyl iodide 2, whereas the â-anomeric acetate
3 is converted into the R-iodide 4 (Scheme 1). These
X Abstract published in Advance ACS Abstracts, September 15, 1997.
(1) Presented at the 213th ACS National Meeting, San Francisco,
CA, CARB 100.
(2) Helferich, B.; Gootz, R. Ber. 1929, 62, 2791. De´jter-J uszynski,
M.; Flowers, H. M. Carbohydr. Res. 1971, 18, 219 and references cited
therein.
(3) Schmidt, R. R. Angew. Chem., Int. Ed. Engl. 1986, 25, 212.
Paulsen, H. Chem. Soc. Rev. 1984, 13, 15.
(4) Schmidt, R. R.; Rucker, E. Tetrahedron Lett. 1980, 21, 1421.
Schmidt, R. R.; Behrendt, M.; Toepfer, A. Synlett 1990, 694.
(5) For a recent example see: Lichtenthaler, F. W.; Kohler, B.
Carbohydr. Res. 1994, 258, 77.
(6) Lemieux, R. U.; Hendriks, K. B.; Stick, R. V.; J ames, K. J . Am.
Chem. Soc. 1975, 97, 4056.
(8) Schmid, U.; Waldemann, H. Tetrahedron Lett. 1996, 3837.
Hashimoto, S.-i; Honda, T. Ikegami, S. Tetrahedron Lett. 1990, 4769.
(9) Somsa´k, L.; Papp, E.; Batta, Gy.; Farkas, I. Carbohydr. Res.
1991, 211, 173.
(10) Gervay, J .; Nguyen, T. N.; Hadd, M. J . Carbohydr. Res. 1997,
300, 119.
(7) Kronzer, F. J .; Schuerch, C. Carbohydr. Res. 1974, 34, 71.
S0022-3263(97)00922-5 CCC: $14.00 © 1997 American Chemical Society

6962 J . Org. Chem., Vol. 62, No. 20, 1997
Gervay and Hadd
Sch em e 2
reactions proceed in quantitative yield with the only
byproduct being trimethylsilyl acetate which can be
removed in vacuo. Although the intermediacy of 2 and
4 had been reported,7,8 prior to our work these glycosyl
iodides had not been isolated and/or characterized.
In view of the often less than satisfactory reactivity of
glycosyl bromides in SN2 reactions, we decided to inves-
tigate anionic additions to anomeric iodides. All prior
reactions, wherein the glycosyl iodide was generated in
situ, resulted in preferential R-glycoside formation, pre-
sumably through nucleophilic displacement of the â-io-
dide or an SN1 mechanism. At the outset we were most
interested in being able to achieve â-glycoside formation,
since it is most difficult to achieve in the absence of
neighboring group participation. Reported herein is the
stereoselective addition of C-, N-, and O-based anions to
R-glycosyl iodides.
it appears that in situ anomerization is occurring and
the â-iodide is preferentially undergoing attack, although
here again an SN1 mechanism cannot be ruled out.
Interesting the galactosyl iodide 6 is much more reactive
toward malonate displacement. The analogous reaction
was complete within 1 h at room temperature, and a 1:10
R:â ratio of C-glycosides 7a b was obtained, suggesting
that in situ anomerization and/or oxonium formation do
not effectively compete in this reaction (Scheme 3). We
considered the possibility that the C-glycosides them-
selves could be undergoing anomerization via â-elimina-
tion followed by conjugate addition (Scheme 4). In order
to rule this mechanism out, 7a and 7b were separated
and resubjected to the reaction conditions and no erosion
of stereochemistry was observed.
The addition of TMSCN to anomeric acetates to give
pyranosyl cyanides has been reported,12 but to our
knowledge SN2 displacement of anomeric halides by
cyanide has not. In our investigations we first treated 4
with 5 equiv of tetrabutylammonium cyanide in THF at
room temperature (Scheme 5). After 10 min TLC showed
complete disappearance of starting material and the
formation of two products. Isolation of the products
revealed that the major product resulted from E-2
elimination of the glycosyl iodide to give the glycal 8, in
addition to a 32% yield of the â-glucosyl cyanide 9. A
number of different conditions were tried in order to
minimize elimination, including lower reaction temper-
atures and utilizing fewer equivalents of tetrabutylam-
monium cyanide, but the glycal was consistently the
major product obtained. We reasoned that the mannosyl
iodide should not be susceptible to E-2 elimination.
Therefore we prepared the R-mannosyl iodide 11 by
reaction of R-2,3,4,6-tetra-O-benzylmannosyl acetate 10
with TMSI in CH2Cl2 at 0 °C for 40 min. The solvent
was evaporated, and the mannosyl iodide was diluted in
THF before adding 5 equiv of tetrabutylammonium
cyanide. After 2 h the iodide was no longer present and
a single product was observed on TLC. The crude NMR
showed mainly one product, but the presence of other
minor products could not be ruled out since the benzylic
protons obscured the region where their resonances
would appear. Column chromatography afforded a 55%
yield of 1-deoxy-2,3,4,6-tetra-O-benzyl-â-D-mannosyl cya-
nide (12). This is a particularly important result since
Resu lts a n d Discu ssion
In 1974 Hanessian reported that glycosyl bromides
undergo reaction with sodio diethylmalonate to give a 3:1
â:R ratio of the C-glycoside.11 The reaction was per-
formed at room temperature for 40 h in diethyl malonate
as a solvent, and 10 equiv of base was added. Upon
addition of 10 equiv of tetrabutylammonium bromide, the
R-selectivity increased to give a 1:3 â:R ratio. One
possible mechanism that explains these results suggests
that the R-anomeric bromide initially undergoes SN2
attack by the malonate to give the â-C-glycoside. As the
concentration of bromide ion increases so does the
possibility of in situ anomerization to the â-anomeric
bromide which is subsequently displaced by the malonate
to give the R-C-glycoside. Addition of tetrabutylammo-
nium bromide increases the rate of in situ anomerization
relative to direct displacement of the R bromide by
malonate thereby increasing the yield of the R-glycoside
(Scheme 2). However it is possible that substitution via
an SN1 mechanism is a competing process.
Reaction of 4 and 1.3 equiv of diethyl malonate that
had previously been treated with sodium hexamethyl-
disilazane (NaHMDS) in the presence 15-crown-5 re-
sulted in the preferential formation of the R-C-glycoside
5a (5:1 R:â). The reaction was complete after 5 h at room
temperature in THF. This is a relatively slow reaction
(vide infra), suggesting that the R-iodide is not particu-
larly susceptible to attack by the malonate anion. Rather,
(12) Lo´pez, M.-T. G.; De Las Heras, F. G.; Fe´lix, A. S. J . Carbohydr.
Chem. 1987, 6, 273. de las Heras, F. G.; Ferna´ndez-Reza, P. J . Chem.
Soc., Perkin Trans. 1 1982, 903.
(11) Hanessian, S.; Pernet, A. G. Can. J . Chem. 1974, 52, 1266.

Anionic Additions to Glycosyl Iodides
J . Org. Chem., Vol. 62, No. 20, 1997 6963
Sch em e 3
Sch em e 4
Sch em e 5
the cis 1,2 linkage in mannose is among the most difficult
to achieve.13 Furthermore, it indicates that the mannosyl
iodide is less reactive than the glucosyl iodide since
reaction of 4 was complete in 10 min rather than 2 h.
The combined malonate and cyanide data suggest that
galactosyl iodide is more reactive than glucosyl iodide,
which is more reactive than mannosyl iodide.
We next looked at the possibility of addition of nitrogen-
based nucleophiles. N-Glycosides are important natural
products most commonly found as asparagine conju-
gates.14 As a model for asparagine addition, acetamide
was deprotonated with sodium hydride in acetone and
added to 4 (Scheme 6). Unfortunately 8 was the only
detectable product, presumably resulting from E-2 elimi-
nation. The more stabilized potassium phthalimide in
the presence of 18-crown-6 did displace the iodide with
inversion to give 13 in 66% yield. Similarly, addition of
tetrabutylammonium azide to 4 gave the â-glucosyl azide
(14) in 92% yield within 5 min at room temperature.
Inspection of the crude NMR spectra from these reactions
showed no detectable R-products, however the â-phthal-
imido derivative does undergo anomerization upon stand-
ing. SN2 reactions of glycosyl chlorides and bromides
with azide anion are known to occur even in the presence
of a participatory group at C2.15 For example, reaction
of peracetylated mannosyl bromide with sodium azide in
HMPA at 90 °C for 1 h gave a quantitative yield of the
1,2 cis glycosyl azide. Inversion of stereochemistry is also
observed in the addition of NaN3 to 1-bromoglucopyra-
nosyl cyanides.16 The earlier studies combined with our
results demonstrate that glycosyl halides are useful
(13) Hodosi, G.; Kova´c, P. J . Am. Chem. Soc. 1997, 119, 2335. Crich,
D.; Sun, S. J . Org. Chem. 1996, 61, 4506 and references cited therein.
(14) Montreuil, J . Adv. Carbohydr. Chem. Biochem. 1980, 37, 157.
Paulsen, H. Angew. Chem. 1990, 102, 851.
(15) Gyorgydea´k, Z.; Paulsen, H. Liebigs Ann. Chem. 1977, 1987.
(16) Praly, J .-F.; Di Stefano, C.; Descotes, G.; Faure, R.; Somsa´k,
L.; Eperjesi, I. Tetrahedron Lett. 1995, 3329.

6964 J . Org. Chem., Vol. 62, No. 20, 1997
Gervay and Hadd
Sch em e 6
Sch em e 7
precursors to glycosyl azides and that the glycosyl iodides
are more reactive than the corresponding bromides.
Furthemore, within the detectable limits of NMR, these
reactions are stereospecific. Glycosyl azides are impor-
tant intermediates in route to N-linked glycopeptides17
and peptidosaccharides.18
Glycosyl iodides are known to undergo O-glycosylation
under neutral conditions to give the R-glycoside as the
major product, but this was only reported with in situ
generation of the iodide.7 If SN2 displacement is the
mechanism of reaction, then it can be concluded that the
â-iodide underwent attack in these reactions, and one
would expect the R-iodide to give the â-glycoside. Inter-
estingly when 4 was treated with 2 equiv of methanol in
benzene at room temperature, the reaction proceeded
slowly, giving the R-methyl glycoside (15) in 70% yield
after 45 h (Scheme 7). Similar to the reaction of 4 with
sodio diethyl malonate, the glucosyl iodide showed rela-
tively low reactivity, and the formation of the R-glucoside
is consistent with the in situ anomerization or SN1
mechanisms invoked in the C-glycosylation experiment
(Scheme 1). The reactivity of the galactosyl iodide toward
neutral displacement was also studied. Consistent with
earlier studies, it reacted faster than the glucosyl iodide
(after 23 h TLC showed no further conversion of the
iodide). Both anomers of the methyl galactoside were
formed (16a b) in a 1.2:1 R:â ratio.
We next explored the possibility of effecting displace-
ment by the reaction of an alkoxide anion. In the event
1,2:3,4-diisopropylidenegalactose was treated with NaH-
MDS in THF and added to 6 at room temperature. After
3 h the iodide was no longer present on TLC, and the
only product obtained was the glycal 17, an unpleasant
reminder that E-2 elimination occurs with highly basic
nucleophiles. However, the more stabilized phenoxide
anion did undergo reaction with 4 to give the â-phenyl
glycoside (18) in 61% yield. Carboxylate anions also add
to the iodides in a highly efficient manner. Acetate anion,
pivalate anion, and stearate anion19 all undergo reaction
with 4 in the presence of tetrabutylmmonium hydrogen
sulfate in less than 5 min at room temperature to give
virtually quantitative yields of the â-glycosides 19, 20,
and 21, respectively. â-D-Stearyl glucopyranoside is a
member of a group of naturally occurring plant hormones
called Brassins.20 Inspection of the crude NMR spectra
from these reactions showed no evidence of R-product
formation supporting the contention that addition is
stereospecific.
The combined O-glycosylation results suggest that
neither neutral alcoholic additions nor basic alkoxide
additions to benzyl protected glucosyl and galactosyl
iodides are likely to lead to efficient syntheses of â-O-
alkyl glycosides. However â-O-aryl and â-O-acyl glyco-
sides are formed in a high-yielding and highly stereose-
lective process.
Con clu sion s
Reported herein are the first studies of anion additions
to R-glycosyl iodides. The results of our studies are
(19) For leading references, see: Mandava, N.; Mitchell, J . W. Chem.
Ind. (London) 1972, 23, 930. Pfander, H.; Laderach, M. Carbohydr.
Res. 1982, 99, 175. Pfeffer, P. E.; Rothman, E. S.; Moore, G. G. J . Org.
Chem. 1976, 41, 2925.
(20) In the course of these experiments we observed that after
complete conversion of the iodide the acetate began to reappear on
TLC, suggesting that a nucleophilic acetate species was displacing the
iodide. This problem was alleviated by removing trimethylsilyl acetate
from the reaction by azeotrope with toluene.
(17) Gyorgydea´k, Z.; Szila´gy, L.; Paulsen, H. J . Carbohydr. Chem.
1993, 12, 139.
(18) Sabesan, S. Tetrahedron Lett. 1997, 38, 3127.

Anionic Additions to Glycosyl Iodides
J . Org. Chem., Vol. 62, No. 20, 1997 6965
Ta ble 1. Com p iled Resu lts fr om Nu cleop h ilic Ad d ition s to Glycosyl Iod id es
nucleophile
% yielda
sodio diethyl malonate 66
glycosyl iodide
R:â ratio
4
5
4
10
4
4
4
4
6
5:1
1:10
sodio diethyl malonate
NBu4CN
58
32b
only â observed
only â observed
only glycal observed
only â observed
only â observed
only R observed
1.2:1
NBu4CN
55
sodium acetamide
0
potassium phthalimide
66
NBu4N3
92
MeOH
MeOH
70
41c
6
4
4
sodium 1,2,3,4-diisopropylidine galactosyl alkoxide
sodium phenoxide
sodium acetate
0
61
90
only glycal observed
only â observed
1:7
4
4
sodium pivalate
sodium stearate
90
only â observed
>5:95
quantitative
a
In all reactions (except where noted) crude NMR and TLC indicated clean conversion of starting material, the isolated yields may
b
reflect error in weighing out the oily starting material and unavoidable loss during purification. The glycal resulting from E-2 elimination
was the major product. c Unoptimized yield.
mmol), and 15-crown-5 (129 µL, 0.65 mmol) in 5 mL of THF.
The reaction mixture was let stir for 5 h, after which the
solvent was removed in vacuo and the resulting oil chromato-
graphed using 3:1 hexanes:ethyl acetate to yield 257 mg (66%)
of a 5.1:1 (R:â) mixture of anomers. â-anomer: 1H NMR (250
MHz, C6D6) δ 3.23 (s, 3H), 3.29 (s, 3H), 3.39-3.43 (m, 1H),
3.63-3.70 (m, 3H), 3.78 (t, 1H, J ) 9.3 Hz), 3.93 (t, 1H, J )
9.5 Hz), 4.01 (d, 1H, J ) 5.4 Hz, CH), 4.23 (dd, 1H, J ) 5.3,
9.8 Hz, H-1), 4.38 (d, 1H, J ) 12.3 Hz), 4.54 (d, 1H, J ) 12.0
Hz), 4.60 (d, 1H, J ) 11.4 Hz), 4.72-4.82 (m, 3H), 4.87 (d, 1H,
J ) 11.3 Hz), 5.07 (d,1H, J ) 11.1 Hz), 7.04-7.36 (m, 20H);
13C NMR (250 MHz, C6D6) δ 51.9, 54.4, 68.9, 73.4, 74.7, 74.8,
75.5, 77.6, 78.7, 79.8, 80.2, 87.7, 127.6, 127.8, 128, 128.4, 139.0,
165.4; HRFABMS calc for C39H42O9 654.2829, found 655.2903
for (M + H)+. R-Anomer: 1H NMR (250 MHz, C6D6) δ 3.10 (s,
3H), 3.39 (s, 3H), 3.65-3.85 (m, 6H), 4.0-4.10 (m, 1H), 4.31-
4.55 (m, 7H), 4.67 (d, 1H, J ) 11.4 Hz), 4.76-4.83 (m, 2H),
5.30 (dd, 1H, J ) 5.2, 10.5 Hz, H-1), 7.05-7.32 (m, 20H); 13C
NMR (250 MHz, C6D6) δ 51.9, 52.2, 52.3, 69.7,73.3, 73.5, 73.8,
74.5, 74.7, 75.1, 78.0, 79.8, 82.0, 127.6, 128, 128.2, 128.4, 128.5,
166.6; HRFABMS calc for C39H42O9 654.2829, found 655.2928
(M + H)+.
Dim eth yl 2-(2,3,4,6-tetr a -O-ben zyl-â-D-ga la ctop yr a n o-
syl)m a lon a te (7a /b). To a solution of 1-O-acetyl-2,3,4,6-tetra-
O-benzyl-D-galactopyranoside (170 mg, 0.29 mmol) in 3 mL of
CH2Cl2 cooled to 0 °C was added 45 µL (0.32 mmol) of
iodotrimethylsilane and the solution let sit for 30 min. The
solvent was then removed in vacuo, and then 2 mL of THF
was added and the resulting solution was transferred to an
already stirring solution of dimethyl malonate (71 µL, 0.44
mmol), sodium hexamethyldisilazane (380 µL, 1 M solution
in THF, 0.38 mmol), and 15-crown-5 (75 µL, 0.38 mmol) in 5
mL of THF. The reaction mixture was stirred for 1 h at room
temperature and then concentrated. The residue was diluted
in a small amount of methylene chloride, loaded onto a silica
gel column, and subsequently eluted with 3:1 hexanes:ethyl
acetate to yield 110 mg (58%) of a 10:1 (â:R) mixture:
â-Anomer: 1H NMR (250 MHz, C6D6) δ 3.16 (s, 3H), 3.30 (s,
3H), 3.43 (dd, 1H, J ) 8.7, 2.6 Hz), 3.54-3.72 (m, 3H), 3.83
(d, 1H, J ) 2.6 Hz), 4.02 (d, 1H, J ) 5.7 Hz), 4.20 (d, 1H, J )
11.8 Hz), 4.23-4.48 (m, 6H), 4.53 (d, 1H, J ) 11.5 Hz), 4.62
(d, 1H, J ) 11.3 Hz), 4.92 (d, 1H, J ) 11.5 Hz), 5.09 (d, 1H, J
) 11.3 Hz), 7.0-7.39 (m, 20H); 13C NMR (250 MHz, C6D6) δ
51.9, 55.2, 69.0, 72.0, 73.5, 74.3, 74.8, 74.9, 77.0, 77.8, 78.0,
85.5, 127.4, 127.6, 127.7, 127.8, 127.9, 128.0, 128.1, 128.4,
128.5, 128.6, 138.7, 138.8, 139.5, 167.3; HRFABMS calc for
C39H42O9 654.2829, found 655.2899 (M + H)+.
compiled in Table 1. Glucosyl, galactosyl, and mannosyl
iodides were studied, and C-, N-, and O-glycosides were
formed. The reactions proceed most efficiently with the
additions of stabilized anions, and the mechanism of
reaction is believed to be SN2 displacement of the R-iodide
to give the â-glycoside. Treatment of the iodides with
highly basic anions results in elimination to give the
corresponding glycal, with the exception of mannosyl
iodide which is not susceptible to E-2 elimination. The
iodides are displaced by neutral alcohols, but there is
erosion of â-selectivity, presumably due to in situ ano-
merization to the more reactive â-glycosyl iodide which
undergoes SN2 reaction to give the R-glycoside.
Overall these studies offer highly efficient and stereo-
selective syntheses of â-C-, N-, and O-glycosides from
precursors that do not require a C2 participatory group.
Furthermore the starting acetates are readily available,
and the iodide is quantitatively generated with in vacuo
removal of the resulting trimethylsilyl acetate. The
glycosyl iodides are used without further purification in
subsequent nucleophilic addition reactions. Therefore
these methods provide a one-pot procedure for trans-
forming anomeric acetates into a variety of important
precursors to both natural and unnatural products.
Exp er im en ta l Section
Starting materials and reagents purchased from suppliers
were used without further purification. Chemicals were
obtained from the following suppliers: trimethylsilyl iodide
and tetra-O-benzylglucopyranose, Fluka; tetrabutylammonium
cyanide, tetrabutylammonium azide,21 and sodium hexameth-
yldisilazane, Aldrich. Solvents were dried by distillation prior
to use. Dichloromethane and toluene were dried over calcium
hydride, and tetrahydrofuran was dried over sodium/ben-
zophenone. Chromatography was performed using silica gel
60 (230-400 mesh ASTM). Mass spectrometry was preformed
by the University of Minnesota Mass Spectrometry Service and
the University of Arizona Mass Spectrometry Facility.
Dim eth yl 2-(2,3,4,6-Tetr a -O-ben zyl-r,â-D-glu cop yr a n o-
syl)m a lon a tes (5a /b). To a solution of 1-O-acetyl-2,3,4,6-
tetra-O-benzyl-D-glucopyranoside (344 mg, 0.59 mmol) in 3 mL
of CH2Cl2 cooled to 0 °C was added 93 µL (0.65 mmol) of
trimethylsilyl iodide and the reaction mixture let sit for 30
min. The solvent was then removed in vacuo, and 2 mL of
toluene was added and again removed in vacuo. The resulting
oil was diluted in 3 mL of THF and added to a previously
stirring solution of dimethyl malonate (88 µL, 0.77 mmol),
sodium hexamethyldisilazane (650 µL of a 1 M in THF, 0.65
1-Deoxy-2,3,4,6-tetr a-O-ben zyl-â-D-glu copyr an osyl Cya-
n id e (9). To a solution of 1-O-acetyl-2,3,4,6-tetra-O-benzyl-
D-glucopyranoside (53.9 mg, 0.093 mmol) in 1 mL of CH2Cl2
cooled to 0 °C was added 14.5 µL (0.10 mmol) of trimethylsilyl
iodide and the reaction mixture let sit for 30 min. The solvent
was then removed in vacuo, and 1 mL of toluene was added
and again removed in vacuo. The resulting oil was diluted in
2 mL of THF and added to a previously stirring solution of
(21) Brandstrom, A.; Lamm, B.; Palmertz, I. Acta Chem. Scand. B
1974, 6, 28.

6966 J . Org. Chem., Vol. 62, No. 20, 1997
Gervay and Hadd
tetrabutylammonium cyanide (124 mg, 0.46 mmol) containing
3 Å molecular sieves. After 10 min, the solvent was removed
in vacuo, and the resulting material was chromatographed
using 3:1 hexanes:ethyl acetate as eluent to yield 16.5 mg
(32%) of the C-glycoside. 1H spectra matched that of the
previously reported compound.12
Azid o 1-Deoxy-2,3,4,6-tetr a -O-ben zyl-â-D-glu cop yr a n o-
sid e (14). To a solution of acetyl 2,3,4,6-tetra-O-benzyl-D-
glucopyranoside (179 mg, 0.31 mmol) in 3 mL of CH2Cl2 cooled
to 0 °C was added 48 µL (0.39 mmol) of trimethylsilyl iodide
and the reaction mixture let sit for 30 min. The solvent was
removed in vacuo, and 2 mL of toluene was added and again
removed in vacuo. The resulting oil was diluted in 2 mL of
THF and transferred to a previously stirring solution of
tetrabutylammonium azide (437 mg, 1.53 mmol) in 3 mL of
THF containing 3 Å molecular sieves. After 5 min TLC (3:1
hexanes:ethyl acetate) showed complete disappearance of the
iodide, the solvent was removed under vacuum, and the
resulting oil was subjected to flash column chromatography
using 6:1 hexanes:ethyl acetate as the eluent to yield 160 mg
(92%) of the â-azide: [R]25D +28.4° (19.0, CDCl3); 1H NMR (250
MHz, C6D6) δ 3.16-3.25 (m, 1H), 3.30 (t, 1H, J ) 8.7 Hz), 3.48
(t, 1H, J ) 9.0 Hz), 3.55-3.61 (m, 2H), 3.68 (t, 1 H, J ) 9.2
Hz), 4.24 (d, 1H, J ) 8.5 Hz, H-1), 4.32 (d, 1H, J ) 12.2 Hz),
4.42 (d, 1H, J ) 8.5 Hz), 4.55 (d, 1H, J ) 11.2 Hz), 4.61 (d,
1H, J ) 11.1 Hz), 4.75 (d, 2H, J ) 11.4Hz), 4.81-4.90 (m, 2H),
7.0-7.40 (m, 20H); 13C NMR (250 MHz, C6D6) δ 68.8, 73.7,
75.1, 75.6, 77.7, 82.1, 85.2, 90.3, 127.7, 127.8, 128.0, 128.1,
128.4, 128.5, 128.6, 128.7, 138.7, 138.8, 139.1, 139.3; HR-
FABMS calcd for C34H35O5N3 565.2576, found 564.2539 (M -
H)+.
Meth yl 2,3,4,6-Tetr a-O-ben zyl-r-D-glu copyr an oside (15).
To a solution of 2,3,4,6-tetra-O-benzyl glucosyl acetate (206
mg, 0.35 mmol) in 3 mL of methylene chloride was added 55
µL (0.39 mmol) of trimethylsilyl iodide and the reaction
mixture let sit for 30 min at 0 °C. The solvent was removed
in vacuo, and 2 mL of benzene was added and the resulting
solution added to an already stirring solution of methanol (29
µL, 0.70 mmol) and 2,6 di-tert-butylpyridine (172 µL, 0.87
mmol) in 5 mL of benzene and let sit for 45 h. The solvent
was removed in vacuo and the resulting oil put down a 6:1
column (hexanes:ethyl acetate) to yield 138 mg (70%) of the
R-anomer.6
Met h yl 2,3,4,6-Tet r a -O-b en zyl-r,â-D-ga la ct op yr a n o-
sid es (16a /b). To a solution of 2,3,4,6-tetra-O-benzylgalactosyl
acetate 158 mg (0.27 mmol) in 2 mL of methylene chloride was
added 43 µL (0.30 mmol) of trimethylsilyl iodide and the
reaction mixture let sit for 30 min at 0 °C. The solvent was
removed in vacuo, and 2 mL of benzene was added and again
removed in vacuo. An additional 2 mL of benzene was added
and the resulting solution added to an already stirring solution
of methanol (22 µL, 0.54 mmol) and 2,6 di-tert-butylpyridine
(132 µL, 0.58 mmol) in 2 mL of benzene and let sit for 42 h.
The solvent was removed in vacuo and the resulting oil put
down a 6:1 column (hexanes:ethyl acetate) to yield 61.5 mg
(41% unoptimized yield) of a 1.2:1 R:â mixture.6
P h en yl 2,3,4,6-Tetr a-O-ben zyl-â-D-glu copyr an oside (18).
To a solution of acetyl 2,3,4,6-tetra-O-benzyl-D-glucopyranoside
(303 mg, 0.52 mmol) in 4 mL of CH2Cl2 cooled to 0 °C was
added 81 µL (0.57 mmol) of trimethylsilyl iodide and the
reaction mixture let sit for 30 min. The solvent was removed
in vacuo, and 2 mL of toluene was added and again removed
in vacuo. The resulting oil was diluted in 3 mL of THF and
transferred into a previously stirring solution of phenol (74
mg, 0.78 mmol), sodium hexamethyldisilazane (624 µL of a 1
M solution in THF, 0.63 mmol), and 15-crown-5 (124 µL, 0.63
mmol) in 5 mL of THF. After 15 min TLC (3:1 hexanes:ethyl
acetate) showed complete disappearance of the iodide. The
solution was diluted in ether and washed twice with 1 M
NaOH and once with saturated brine and then dried over
sodium sulfate. The resulting oil was purified on a column of
Gen er a l P r oced u r e for Acetyla tion of th e 2,3,4,6-
Tetr a -O-ben zylp yr a n osid es (10). First, 3.0 g (5.55 mmol)
of 2,3,4,6-tetra-O-benzyl-D-glucopyranose was dissolved in 10
mL of CH2Cl2 and cooled to 0 °C; 2.24 mL (27.7 mmol) of
pyridine was then added followed by 1.57 mL (22.2 mmol) of
acetyl chloride. The solution was stirred for 4 h, then diluted
in CH2Cl2, extracted twice with 2 M H2SO4 followed by brine,
and dried over sodium sulfate. The crude oil was subjected
to flash column chromatography using hexanes:ethyl acetate
as eluent to yield 2.9 g (89%) as a clear oil, with the R-anomer
always as the major product (5-10:1, R:â). Characterization
of 2,3,4,6-tetra-O-benzyl-D-glucopyranosyl acetate,10 2,3,4,6-
tetra-O-benzyl-D-galactopyranosyl acetate,12 and 2,3,4,6-tetra-
O-benzyl-D-mannopyranosyl acetate is as follows. R-Anomer:
1H NMR (250 MHz, C6D6) δ 1.54 (s, 3H), 3.68-3.76 (m, 2H),
3.83 (dd, 1H, J ) 11.1, 4.3 Hz), 3.93-3.98 (dd, 1H, J ) 9.5 ,
3.0 Hz), 4.03-4.09 (m, 1H), 4.37-4.45 (m, 4H), 4.51-4.61 (m,
4H), 4.96 (d, 1H, J ) 11.3 Hz), 6.59 (d, 1H, H-1, J ) 1.9 Hz),
7.03 (m, 20H). â-Anomer: 1H NMR (250 MHz, C6D6) δ 1.62
(s, 3H), 3.48 (dd, 1H, J ) 9.3, 2.8 Hz), 3.50 (dq, 1H, J ) 9.5,
2.1 Hz), 3.66-3.80 (m, 3H), 4.25 (t, 1H, J ) 9.5 Hz), 4.32-
4.41(m, 3H), 4.53 (apparent d, 1H, J ) 11.4 Hz), 4.55 (apparent
d, 1H, J ) 12.0 Hz), 4.74-4.89 (m, 3H), 5.63 (s, 1H, H-1) 6.99-
7.48 (m, 20H).
1-Deoxy-2,3,4,6-t et r a -O-b en zyl-â-D-m a n n op yr a n osyl
Cya n id e (12). To a solution of acetyl 2,3,4,6-tetra-O-benzyl-
D-mannopyranoside (88 mg, 0.15 mmol) in 2 mL of CH2Cl2
cooled to 0 °C was added 24 µL (0.17 mmol) of trimethylsilyl
iodide and the reaction mixture let sit for 40 min. The solvent
was then removed in vacuo, and 1 mL of toluene was added
and again removed in vacuo. The resulting oil was diluted in
2 mL of THF and added to a previously stirring solution of
tetrabutylammonium cyanide (202 mg, 0.76 mmol) in 5 mL of
THF. After 2 h the solvent was removed in vacuo, and the
residue was chromatographed using 3:1 hexanes:ethyl acetate
as eluent to yield 46 mg (55%) of the C-glycoside: [R]25D -51.2°
(6.6, CDCl3); 1H NMR (250 MHz, CDCl3) δ 3.31-3.37 (m, 1H),
3.45 (dd, 1H, J ) 9.3, 2.3 Hz, H-3), 3.61-3.63 (m, 2H), 3.83 (t,
1H, J ) 9.3 Hz, H-4), 3.92 (d, 1H, J ) 2.3 Hz, H-2), 4.12 (s,
1H, H-1), 4.43-4.60 (m, 6H), 4.76 (d, 1H, J ) 10.8 Hz), 4.83
(d, 1H, J ) 11.5 Hz), 4.90 (d, 1H, J ) 11.5 Hz), 7.13-7.48 (m,
20H); 13C NMR (250 MHz, CDCl3) δ 67.4, 68.8, 72.6, 73.6, 73.9,
74.1, 74.6, 75.3, 80.5, 82.3, 127.6, 127.8, 127.9, 128.0, 128.3,
128.4, 128.5, 137.3, 137.6, 137.8; HRFABMS calc for C35H35O5N
549.2515, found 550.2614 (M + H)+.
P h t h a lim id o 1-Deoxy-2,3,4,6-t et r a -O-b en zyl-â-D-glu -
cop yr a n osid e (13). To a solution of acetyl 2,3,4,6-tetra-O-
benzyl-D-glucopyranoside (332 mg, 0.57 mmol) in 3 mL of
CH2Cl2 cooled to 0 °C was added 89 µL of trimethylsilyl iodide
and the reaction mixture let sit for 30 min. The solvent was
then removed in vacuo, and 2 mL of toluene was added and
removed in vacuo. The compound was diluted in 2 mL of THF
and added to an already stirring solution of potassium ph-
thalimide (159 mg, 0.86 mmol) and 18-crown-6 (226 mg, 0.86
mmol) in 7 mL of THF at room temperature. The reaction
was complete after 15 min, at which time the product was
diluted in methylene chloride and washed twice with 1 M KOH
and once with brine and then dried over sodium sulfate to yield
a yellow oil. The resulting oil was subjected to flash column
chromatography using 4:1 hexanes:ethyl acetate as eluent to
yield 252 mg (66%) of the title compound: 1H NMR (250 MHz,
C6D6) δ 3.41-3.50 (m, 1H), 3.56 (d, 1H, J ) 10.8 Hz), 3.63-
3.75 (m, 2H), 3.90 (t, 1H, J ) 9.5 Hz), 4.32 (d, 1H, J ) 12.0
Hz), 4.50-4.67 (m, 3H), 4.80-4.90 (m, 4H), 5.03 (t, 1H, J )
9.2Hz), 5.64 (d, 1H, J ) 9.4 Hz, H-1), 6.76 (m, 4H), 7.08 (m,
20H); 13C NMR (250 MHz, C6D6) δ 69.1, 73.7, 74.9, 75.0, 75.5,
78.1, 78.3, 79.8, 87.0, 123.3, 127.3, 127.4, 127.6, 127.8, 127.9,
128.0, 128.2, 128.4, 128.5, 133.6, 138.9, 139.3, 167.3; HR-
FABMS calcd for C42H39O7N 669.2726, found 670.2808 (M +
H)+.
silica gel using 6:1 hexanes:ethyl acetate as the eluent to yield
1
196 mg (61%) of the glycoside: [R]25 +42.3° (2.6, CDCl3); H
D
NMR (250 MHz, C6D6) δ 3.32-3.4 (m, 1H), 3.59-3.75 (m, 4H),
3.82 (dd, 1H, J ) 7.9, 8.8 Hz, H-2), 4.31 (d, 1H, J ) 12.3 Hz),
4.40 (d, 1H, J ) 12.1 Hz), 4.55 (d, 1H, J ) 11.3 Hz), 4.77-
4.88 (m, 3H), 4.96 (d, 1H, J ) 7.7 Hz, H-1), 5.00 (d, 1H, J )
11.4 Hz), 5.09 (d, 1H, J ) 11.3 Hz), 6.85-7.37 (m, 25H); 13C
NMR (250 MHz, C6D6) δ 69.2, 73.4, 74.9, 75.0, 75.5, 75.54, 82.4,
84.9, 102.1, 117.4, 122.8, 127.6, 127.7, 127.8, 127.9, 128.0,
128.4, 128.5, 128.8, 129.6, 138.8, 139.1, 139.4, 158.1; HR-
FABMS calc for C40H40O6 616.2825, found 639.2693 (M + Na)+.

Anionic Additions to Glycosyl Iodides
J . Org. Chem., Vol. 62, No. 20, 1997 6967
Acetyl 2,3,4,6-Tetr a -O-ben zyl-â-D-glu cop yr a n osid e (19).
To a solution of acetyl 2,3,4,6-tetra-O-benzyl-D-glucopyranoside
(207 mg, 0.36 mmol) in 3 mL of CH2Cl2 cooled to 0 °C was
added 56 µL (0.39 mmol) of trimethylsilyl iodide and the
reaction mixture let sit for 35 min. The solvent was removed
in vacuo, and 2 mL of THF was added and the resulting
solution transferred into an already stirring solution of
anhydrous sodium acetate (88 mg, 1.07 mmol) and tetrabu-
tylammonium hydrogen sulfate (133 mg, 0.39 mmol) in 5 mL
of THF. The reaction mixture was stirred at room tempera-
ture for 1.5 h, at which point the solvent was removed in vacuo
and the resulting oil chromatographed using 6:1 hexanes:ethyl
acetate as eluent to yield 187 mg (90%) of a 7:1 mixture of â:R
anomers.12
P iva loyl 2,3,4,6-Tet r a -O-b en zyl-â-D-glu cop yr a n osid e
(20). Meth od A. To a solution of acetyl 2,3,4,6-tetra-O-benzyl-
D-glucopyranoside (126 mg, 0.22 mmol) in 3 mL of CH2Cl2
cooled to 0 °C was added 34 µL (0.24 mmol) of trimethylsilyl
iodide and the reaction mixture let sit for 30 min. The solvent
was removed in vacuo , and 2 mL of THF was added and the
resulting solution transferred into an already stirring solution
of pivalic acid (45mg, 0.43 mmol), tetrabutylammonium trif-
luoromethanesulfonate (127 mg, 0.32 mmol), and sodium
hexamethyldisilazane (324 µL pf a 1 M solution in THF, 0.32
mmol) in 5 mL of THF. The reaction mixture was stirred for
2 h at which point 15-crown-5 (43 µL, 0.22 mmol) was added
and the reaction mixture let stir for 19 h. The solvent was
removed in vacuo and the resulting oil chromatographed using
6:1 hexanes:ethyl acetate as eluent to yield 60 mg (45%) of
the glycoside.
Meth od B. To a solution of acetyl 2,3,4,6-tetra-O-benzyl-
D-glucopyranoside (263 mg, 0.45 mmol) in 3 mL of CH2Cl2
cooled to 0 °C was added 71 µL (0.497 mmol) of trimethylsilyl
iodide and the reaction mixture let sit for 30 min. The solvent
was then removed in vacuo followed by the addition of 2 mL
of toluene which again was removed in vacuo. Next, 2 mL of
THF was added, and the resulting solution was pipeted into
an already stirring solution of pivalic acid (230 mg, 2.3 mmol),
tetrabutylammonium hydrogen sulfate (306 mg, 0.90 mmol),
and sodium hexamethyldisilazane (1.80 µL of a 1 M solution
in THF, 1.80 mmol) in 5 mL of THF. The reaction mixture
was stirred for 20 min, at which point ethyl acetate was added
and the reaction mixture was washed with sodium bicarbonate
(twice) and brine. The ethyl acetate layer was dried over
sodium sulfate and removed in vacuo. The resulting oil was
chromatographed using 8:1 hexanes:ethyl acetate as eluent to
yield 256 mg (90%) of the â-glycoside with no detectable
R-product: [R]25 +98.9 (7.3, CDCl3); 1H NMR (250 MHz,
D
CDCl3) δ 1.20 (s, 9H), 3.51-3.73 (m, 6H), 4.46 (d, 1H, J ) 12.3
Hz), 4.52 (d, 1H, J ) 11.0 Hz), 4.58 (d, 1H, J ) 12.3 Hz), 4.68-
4.86 (m, 5H), 5.57 (d, 1H, J ) 7.9 Hz, H-1); 13C NMR (250
MHz, C6D6) δ 27.0, 38.7, 68.6, 73.4, 74.7, 74.9, 75.4, 76.0, 77.7,
81.5, 85.1, 94.8, 127.6, 127.7, 127.9, 128.0, 128.4, 128.5, 128.6,
138.9, 139.2; HRFABMS calc for C39H44O7 624.3068, found
623.3013 (M - H)+.
Stear yl 2,3,4,6-Tetr a-O-ben zyl-â-D-glu copyr an oside (21).
To a solution of 2,3,4,6-tetra-O-benzylglucosyl acetate (273 mg,
0.46 mmol) in 3 mL of methylene chloride was added 73 µL
(0.51 mmol) of trimethylsilyl iodide and the mixture let sit for
30 min at 0 °C. The solvent was removed in vacuo, and 2 mL
of toluene was added and again removed in vacuo followed by
the addition of 2 mL of THF. This solution was added to an
already stirring solution of sodium stearate (a mixture with
sodium palmitate) (430 mg, 1.41 mmol) and tetrabutylammo-
nium hydrogen sulfate (397 mg, 1.17 mmol) in 5 mL of THF.
Immediately upon addition, TLC analysis (6:1 hexanes:ethyl
acetate) showed complete disappearance of the iodide. The
solvent was removed in vacuo after 10 min and the resulting
oil purified using a 9:1 hexanes:ethyl acetate column to give a
quantative yield of a >95:5 mixture of â:R-glycosides, as a
mixture of lipids: 1H NMR (CDCl3) δ 1.23-1.31 (m), 1.63 (m),
1.90-1.96 (m), 2.55-2.63 (t, J ) 7.5 Hz), 3.87-4.13 (m), 4.75
(d, J ) 12 Hz), 4.86-4.94 (m), 5.10-5.32 (m), 6.06 (d, 1H, H-1,
J ) 7.9 Hz), 7.47-7.62 (m, 20H); HRFABMS for the stearate
derivative calc for C52H70O7 806.5121, found 829.5035 (M +
Na)+; HRFABMS for the palmitate derivative calc for C50H66O7
778.4808, found 801.4694 (M + Na)+. No mass peaks for the
corresponding acids were identified, but 1H NMR integration
proved difficult with the varying lipid components.
Ack n ow led gm en t. Financial support for this re-
search was gratefully received from The Arizona Dis-
ease Control Research Commisssion.
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra for
compounds lacking combustion analyses (5a /b, 7a /b, 12, 13,
14, 18, 20, 21) (9 pages). This material is contained in libraries
on microfiche, immediately follows this article in the microfilm
version of the journal, and can be ordered from the ACS; see
any current masthead page for ordering information.
J O970922T