while in the D-series the converse is true and mannose must
be used as a starting material. The direct synthesis of â-D-
mannopyranosides, both in solution and on the solid phase,
is straightforward provided that the donor is protected by
the 4,6-O-benzylidene moiety or by a surrogate such as the
4,6-O-polystyrylboronate group.6 The problem of the â-D-
rhamnopyranosides therefore reduces to that of the reductive
regioselective cleavage of the 4,6-O-benzylidene group to
the 6-deoxy system. Here, we introduce the 4,6-O-[R-(2-(2-
iodophenyl)ethylthiocarbonyl)benzylidene] group as a sur-
rogate for the 4,6-O-benzylidene group that is capable of
enforcing high â-selectivity in direct mannosylations and
which is subsequently cleaved in a single step to the â-D-
rhamnopyranoside system.
itself be stable to the glycosylation conditions; thus, 2-(2-
iodophenyl)ethylthio esters13 became the precursors of
choice.
Reaction of 16 with 214 gave the acetals 3 and 4 as a 3/1
mixture in 85% yield, favoring the less polar isomer with
the axial carbomethoxy group.15 Treatment with a reagent
formed from 2-(2-iodophenyl)ethylthiol15 and AlMe3 in
toluene at reflux then afforded the separable thioesters 5 and
6 (Scheme 1).16 It is possible to perform all of the ensuing
Scheme 1. Preparation of Glycosyl Donors
The cleavage of 4,6-O-benzylidene-protected sugars to
their 4-O-benzoyl-6-bromo-6-deoxy congeners by N-bromo-
succinimide is an established reaction in carbohydrate
chemistry.7,8 Unfortunately, the reaction is not suitable for
use in the presence of benzyl and allyl-type protecting groups
due to the competing cleavage of these groups, which
significantly reduces yields and complicates isolation.9 This
incompatibility precludes the use of the NBS cleavage in
the synthesis of â-D-rhamnosides from the corresponding
â-D-mannnosides as ethers are the protecting groups of choice
for the 2- and 3-hydroxy groups in direct â-mannosylation.
A further disadvantage of the NBS reaction is the obvious
need to remove the bromide atom in a subsequent reduction.
As recognized from the outset by Hanessian,7a-c the actual
cleavage of the 6-C-O bond in the NBS cleavage reaction
can be envisaged as proceeding by either a radical or an ionic
fragmentation. Roberts developed an alternative based on
his catalysis of radical chain reactions by thiols, which leads
directly to the 6-deoxy system, and in doing so demonstrated
the key fragmentation to be radical in nature.10 The contra-
thermodynamic regioselectivity in these radical fragmenta-
tions was explained on the basis of MO calculations as
arising from the less strained transition state for cleavage of
the 6-C-O bond.11,12 Unfortunately, in our hands, even the
very mild Roberts’ conditions proved to be incompatible with
the presence of benzyl ethers. We were therefore led to
contemplate alternative entries into the key R-benzylidene
radical and ultimately turned to cyclic acetals derived from
benzoylformic acid with the R-benzylidene radical generation
through decarbonylation. The need to minimize transforma-
tions after glycosylation indicated that the radical precursor
chemistry with the isomeric mixture, but to simplify the
spectra at the stage of the glycosylation reaction, we have
worked with pure isomers. The action of N-benzene-
sulfinyl piperidine (BSP),17,18 2,4,6-tri-tert-butylpyrimidine
(TTBP),18,19 and triflic anhydride on 5 at -60 °C in
dichloromethane followed by addition of methyl 2,3-O-
isopropylidene-R-L-rhamnopyranoside as acceptor gave 7 in
94% yield as a single anomer. It is noteworthy that the BSP/
Tf2O combination cleanly activates the thioglycoside toward
coupling but leaves the thioester unchanged. Three further
coupling reactions were similarly conducted, each providing
the desired â-mannoside in excellent yield and selectivity
(Scheme 2, Table 1).
Scheme 2. Stereoselective Formation of Mannosides
(5) Synthesis of L-rhamnosides: Crich, D.; Picione, J. Org. Lett. 2003,
5, 781.
(6) Crich, D.; Smith, M. J. Am. Chem. Soc. 2002, 124, 8867.
(7) (a) Hanessian, S.; Plessas, N. R. J. Org. Chem. 1969, 34, 1035. (b)
Hanessian, S.; Plessas, N. R. J. Org. Chem. 1969, 34, 1045. (c) Hanessian,
S.; Plessas, N. R. J. Org. Chem. 1969, 34, 1053. (d) Hullar, T. L.; Siskin,
S. B. J. Org. Chem. 1970, 35, 225. (e) Chana, J. S.; Collins, P. M.; Farnia,
F.; Peacock, D. J. J. Chem. Soc., Chem. Commun. 1988, 94.
(8) Ionic cleavage: Binkley, R. W.; Goewey, G. S.; Johnston, J. C. J.
Org. Chem. 1984, 49, 992.
(9) Reported cleavage of benylidene acetals in the presence of benzyl
ethers: Liotta, L. J.; Dombi, K. L.; Kelley, S. A.; Targontsidis, S.; Morin,
A. M. Tetrahedron Lett. 1997, 38, 7833.
(10) Roberts, B. P.; Smits, T. M. Tetrahedron Lett. 2001, 42, 3663.
(11) Fielding, A. J.; Franchi, P.; Roberts, B. P.; Smits, T. M. J. Chem.
Soc., Perkin Trans. 2 2002, 155.
Radical fragmentation was achieved by dropwise addition
of tributyltin hydride and AIBN to the substrates in toluene
(13) Crich, D.; Yao, Q. J. Org. Chem. 1996, 61, 3566.
(14) Chan, T. H.; Brook, M. A.; Chaly, T. Synthesis 1983, 203.
(15) Stereochemistry of 3 and 4 was assigned by parallels in the 13C
NMR chemical shifts and the relative polarities with the pyruvate acetals.
Garegg, P. J.; Janesson, P.-E.; Lindberg, B.; Lindh, F.; Lonngren, J.;
Kvarnstrom, I.; Nimmich, W. Carbohydr. Res. 1980, 78, 127 and references
therein.
(12) Other contrathermodynamic radical ring openings and a similar
explanation: Ziegler, F. E.; Zheng, Z. L. J. Org. Chem. 1990, 55, 1416.
(16) Gennari, C.; Carcano, M.; Donghi, M.; Mongelli, N.; Vanotti, E.;
Vulpetti, A. J. Org. Chem. 1997, 62, 4746.
2190
Org. Lett., Vol. 5, No. 12, 2003