Anomeric phosphorodithioates as novel glycosylating agents

Full text of DOI:10.1021/jo981552r

9150
J . Org. Chem. 1998, 63, 9150-9151
of glycosyl acceptors at ambient temperature, including acid-
sensitive glycal acceptors.
An om er ic P h osp h or od ith ioa tes a s Novel
Glycosyla tin g Agen ts
To date, all protocols for the synthesis of anomeric
phosphate-based glycosyl donors have relied on the phos-
phitylation or phosphorylation of an anomeric hydroxyl
group following protection and deprotection protocols. The
preparation of differentially protected monosaccharide gly-
cosyl donors requires lengthy procedures in many cases.
Glycals, on the other hand, allow for the facile differential
protection of the hydroxyl functionalities and have been
shown to be versatile starting materials for the synthesis of
oligosaccharides and natural products.13 Until now, the
formation of â-glucosidic linkages involving glycal acceptors
was only possible with thioethyl glycosides while the cou-
pling conditions required for all other currently available
glycosyl donors were too harsh. Ideally, glycals could be
converted into stable glycosyl donors that could be selectively
activated at room temperature to fashion glycosidic linkages
with a variety of glycosyl donors including glycals to allow
for the repetition of the process. Anomeric dithiophosphates
fulfill these requirements.
Conversion of glycals to anomeric dithiophosphates was
achieved by epoxidation of the glycal double bond of tribenzyl
glucal 1 with dimethyldioxirane (DMDO) to furnish the 1,2-
anhydrosugar (Scheme 1). Opening of the epoxide with
diethyl dithiophosphate furnished a 1:1 mixture of R and â
anomeric phosphorodithioates (2 + 3) in 82% yield, and
purification by silica column chromatography allowed for
separation of phosphorodithioates 2 and 3. Anomeric phos-
phorodithioates are stable compounds that may be stored
for several weeks at room temperature without decomposi-
tion. Purification of anomeric phosphorodithioates by silica
column chromatography is not a problem.
Obadiah J . Plante and Peter H. Seeberger*
Department of Chemistry, Massachusetts Institute of
Technology, Cambridge, Massachusetts 02139
Received August 5, 1998
Complex glycoconjugates carry detailed structural infor-
mation that mediates a variety of biologically important
events, including inflammation, immune response, and
tumor metastasis at the level of cell-cell interactions.1 The
need for synthetic access to complex oligosaccharides and
glycoconjugates has led to increased interest in the develop-
ment of novel and powerful glycosylation reactions. There-
fore, a wide range of anomeric groups, including anomeric
trichloroacetimidates, sulfoxides, pentenyl glycosides, fluo-
rides, and thioethyl glycosides, have been explored for their
use as glycosyl donors.2 In biosynthesis, glycosyl transferases
utilize sugar-nucleotides such as UDP-glucose as substrates
in glycosylation reactions where pyrophosphate acts as an
anomeric leaving group.3 Surprisingly, until recently phos-
phate-based glycosyl donors have received relatively little
attention. Glycosyl phosphites4 as well as glycosyl phos-
phates5 and other phosphate analogues, including
dimethylphosphinothioates,6 phosphorimidates,7 and phos-
phoramidimidates,8 have been explored for their use as
glycosylating agents. To create novel phosphate analogues
of altered reactivity, the replacement of the oxygen bridging
C1 and phosphorus would be particularly desirable since this
change is expected to most profoundly influence the reactiv-
ity of the anomeric functionality toward different activators.
2-Deoxy-
Next, the introduction of different protecting groups on
the C2-hydroxyl group was investigated. Participating acetyl
and pivaloyl ester groups could be installed readily on C2.
Reaction of 2 and 3 with acetic anhydride and DMAP yielded
the C2-acetyl phosphorodithioates 4 (85%) and 5 (81%),
while reaction with pivaloyl chloride in the presence of
DMAP furnished glycosyl donors 6 and 7 in 83% and 80%
yield, respectively. The formation of C2 ether protecting
groups was unsuccessful. Benzylation of 2 with benzyl
bromide and sodium hydride resulted in the formation of a
single compound. Characterization of the reaction product
revealed that instead of the desired protected anomeric
phosphorodithioate, the benzyl C2-phosphorothioate thiogly-
coside 8 was obtained exclusively. Intramolecular nucleo-
philic attack of the C2 alkoxide on phosphorus displaced the
anomeric sulfur, which was then benzylated, thus providing
8. In the case of â-anomer 3, the yield for the formation of
the C2 phosphorothioate 9 was significantly lower (18%)
than for the formation of 8 (49%) and accompanied by
unidentified side products.
After having established a straightforward and high-
yielding synthetic route for the preparation of differentially
protected anomeric phosphorodithioates, a variety of activa-
tion protocols were screened for use with glycosyl phospho-
rodithioate donors. Ideally, glycosylation conditions would
allow for the selective formation of glycosidic linkages in high
yield at ambient temperature in solution and in solid support
synthesis paradigms. First, a number of thiophiles that had
been successfully employed as activators of phosphorodithio-
ate 2-deoxyglycosides were tested (Table 1). Silver salts such
as silver fluoride, requiring prolonged reaction times of 48
h,9 and iodonium perchlorate as well as N-iodosuccinimide
sugar glycosyl phosphorodithioate donors that contain a
bridging sulfur in place of oxygen have proven to be powerful
donors for the formation of 2-deoxy glycosidic linkages.9-11
While peracetylated phosphorodithioate glycosides have
been prepared, their reactivity stood in stark contrast to that
observed with phosphorodithioates in the 2-deoxy sugar
series, and no successful glycosylation reactions involving
these donors have been disclosed.12 We now report an
efficient, straightforward synthesis of differentially protected
anomeric phosphorodithioates from glycal precursors. These
stable glycosyl donors were applied to the selective and high-
yielding construction of â-glucosidic linkages with a variety
(1) Varki, A. Glycobiology 1993, 3, 97. Dwek, R. A. Chem. Rev. 1996, 96,
683. Lee, Y. C.; Lee, R. T. Acc. Chem. Res. 1995, 28, 321.
(2) For a review see: Toshima, K.; Tatsuta, K. Chem. Rev. 1993, 93, 1503.
(3) Leloir, L. F. Science 1971, 172, 1299. Kornfeld, R.; Kornfeld, S. Annu.
Rev. Biochem. 1985, 54, 631.
(4) Sim, M. M.; Kondo, H.; Wong, C.-H. J . Am. Chem. Soc. 1993, 115, 5,
2260. Kondo, H.; Aoki, S.; Ichikawa, Y.; Halcomb, R.; Ritzen, H.; Wong,
C.-H. J . Org. Chem. 1994, 59, 864. Martin, T. J .; Schmidt, R. R. Tetrahedron
Lett. 1992, 33, 6123. Watanabe, Y.; Nakamoto, C.; Yamamoto, T.; Ozaki, S.
Tetrahedron 1994, 50, 6523.
(5) Hashimoto, S.-I.; Honda, T.; Ikegami, S. J . Chem. Soc., Chem.
Commun. 1989, 685.
(6) Yamanoi, T.; Nakamura, K.; Sada, S.; Goto, M.; Furusawa, Y.;
Takano, M.; Fujioka, A.; Yanagihara, K.; Satoh, Y.; Hosokawa, H.; Inazu,
T. Bull. Chem. Soc. J pn. 1993, 66, 2617.
(7) Pan, S.; Li, H.; Hong, F.; Yu, B.; Zhao, K. Tetrahedron Lett. 1997, 38,
6139. Hashimoto, S.-I.; Sakamoto, H.; Honda, T.; Ikegami, S. Tetrahedron
Lett. 1997, 38, 5181. Hashimoto, S.-I.; Sakamoto, H.; Honda, T.; Abe, H.;
Nakamura, S.-I.; Ikegami, S. Tetrahedron Lett. 1997, 38, 8969.
(8) Chen, M.-J .; Ravindran, K.; Landry, D. W.; Zhao, K. Heterocycles 1997,
45, 1247.
(9) Bielawska, H.; Michalska, M. J . Carbohydr. Chem. 1991, 10, 107.
(10) Michalska, M.; Michalski, J . Heterocycles 1989, 28, 1249 and
references therein.
(11) Laupichler, L.; Sajus, H.; Thiem, J . Synthesis 1992, 1133.
(12) Kudelska, W.; Michalska, M. Tetrahedron Lett. 1994, 35, 7459;
Kudelska, W.; Michalska, M. Synthesis 1995, 1539.
(13) Seeberger, P. H.; Bilodeau, M. T.; Danishefsky, S. J . Aldrichim. Acta
1997, 30, 75.
10.1021/jo981552r CCC: $15.00 © 1998 American Chemical Society
Published on Web 11/18/1998

Communications
J . Org. Chem., Vol. 63, No. 25, 1998 9151
Sch em e 1
Ta ble 1
Ta ble 2
activator
T (°C)
reaction time (h)
yield (%)
NIS
rt
20
1
20
10
10
10
16
<5
-
NIS/TfOH
I(coll)2ClO4
I(coll)2ClO4
AgF
rt
rt
15
<5
-
rt
rt
AgClO4
MeOTf
rt
35
70
0-rt
(NIS)11 were not successful in activating phosphorodithioate
glycosyl donors and resulted in slow and low-yielding
reactions. Activation with silver perchlorate succeeded in the
formation of glycosidic linkages but resulted in anomeric
mixtures.
Phosphorodithioate glycosides 4 and 5 equipped with a
C2-acetyl group failed to yield any of the desired disaccha-
ride under the activation conditions tested but instead
produced a mixture of side products. Activation of anomeric
phosphorodithioates 6 and 7 with methyl triflate produced
exclusively the â-glycosidic linkage by virtue of the partici-
pating pivaloyl ester functionality on C2.14 The yield as well
as the rate of reaction were indistinguishable when either
the R- or the â-phosphorodithioate was used. The coupling
of 6 and 7 with a variety of glycosyl acceptors revealed that
these novel glycosyl donors also selectively furnished the
â-glucosidic linkage with hindered acceptors in good yield
(Table 2). Acid-sensitive glycal acceptors could successfully
be accommodated in coupling reactions when a base such
as di-tert-butylpyridine was included without requiring
prolonged reaction times. The couplings to form â (1f3)
disaccharide 13, â (1f4) disaccharide 15, and â (1f6)
disaccharide 17 proceeded in good yields with complete
stereoselectivity. Other acceptors containing secondary hy-
droxyl moieties (e.g., 18) could also be coupled in good yields
using methyl triflate activation while reactions with tertiary
alcohol 20 as acceptor proceeded in low yield (18%).
In summary, we have developed a straightforward syn-
thesis of anomeric phosphorodithioates from glycal precur-
sors. This novel synthetic scheme can be employed in the
facile synthesis of differentially protected glycosyl donors.
We have demonstrated that these anomeric phosphorodithio-
ates can be activated by methyl triflate to serve as glycosyl
donors in reactions with a variety of glycosyl acceptors,
including acid-sensitive glycals. Currently, we are expanding
this concept to the formation of other glycosidic linkages in
solution and on the solid support. Other phosphate ana-
logues, which are now easily accessible from glycals using
the described synthetic protocols, are presently being evalu-
ated for their potential use as glycosyl donors.
Ack n ow led gm en t. Financial support from the Depart-
ment of Chemistry at the Massachusetts Institute of
Technology is gratefully acknowledged.
Su p p or tin g In for m a tion Ava ila ble: Detailed experimental
procedures and compound characterization data, including 1H, 13C,
and 31P NMR spectra for all compounds in this study (45 pages).
(14) For use of the Piv-group with other glycosyl donors, see: Kunz H.;
Harreus, A. Liebigs Ann. Chem. 1982, 42. Kahne, D.; Walker, S.; Cheng,
Y.; VanEugen, D. J . Am. Chem. Soc. 1989, 111, 6881. Seeberger, P. H.;
Eckhardt, M.; Gutteridge, C. A.; Danishefsky, S J . Am. Chem. Soc. 1997,
119, 10064.
J O981552R