Rapid synthesis of oligosaccharides using an anomeric fluorous silyl protecting group.

Article abstract of DOI:10.1039/b311448a

A fluorous silyl ether tag as protecting group at the anomeric position of sugar acceptors allows rapid synthesis of oligosaccharides by reducing the purification procedures to a simple and fast fluorous solid-phase extraction.

Full text of DOI:10.1039/b311448a

Rapid synthesis of oligosaccharides using an anomeric fluorous silyl
protecting group†
Leonardo Manzoni*
C.N.R. – Istituto di Scienze e Tecnologie Molecolari, Via Venezian 21, I-20133, Milano, Italy.
E-mail: leonardo.manzoni@istm.cnr.it; Fax: +39-0250314072; Tel: +39-0250314088
Received (in Cambridge, UK) 18th September 2003, Accepted 15th October 2003
First published as an Advance Article on the web 29th October 2003
A fluorous silyl ether tag as protecting group at the anomeric
position of sugar acceptors allows rapid synthesis of
oligosaccharides by reducing the purification procedures to
a simple and fast fluorous solid-phase extraction.
acceptor should enable the efficient synthesis of branched and
long-chain oligosaccharides. The glycosyl acceptor bound to
the fluorous tag can be coupled with the glycosyl donor to afford
the fluorous disaccharide which in principle should be separable
from non-fluorous material, such as excess donor, by solid-
phase extraction without the need for column chromatography.
Thus, the reaction mixture could be passed through fluorous
silica gel in a two-stage extraction which would first remove the
non-tagged products and then elute the fluorous-tagged one.
The fluorous tag is removed at the end of the synthetic sequence
to give the desired oligosaccharide. In this scheme, unreacted
glycosyl aceptor, which is fluorous tagged, cannot be easily
separated by fluorous extraction from the tagged reaction
product. However, the reaction could be monitored, e.g. by
MALDI MS, and driven to completion by multiple cycles,
similar to the practice of solid-phase synthesis.
To put this scheme into practice, the fluorous protecting
group must be easy to prepare and to introduce on the sugar,
capable of withstanding the typical reaction conditions required
for protection/deprotection of the acceptors and for the
glycosylation reactions, and finally it must be easily removed
from the final oligosaccharide. Given the peculiar reactivity of
the anomeric oxygen, esters were deemed unsuitable, and we
focused our attention on the use of silyl ethers. Among the
known lightly fluorinated silyl halides,9 tert-butyl-phenyl-
1H,1H,2H,2H-heptadecafluorodecyloxysilyl bromide6c was
chosen because it is easily synthesized and relatively stable to
acidic and basic conditions.10
Carbohydrates on cell surfaces play important roles in bio-
logical processes such as cell–cell interaction, cell adhesion,
and immunogenic recognition.1 It would be of great importance
for the development of glycobiology to have easy access to pure
synthetic oligosaccharides. However, the synthesis of oligo-
saccharides is a lengthy process that typically involves iterative
sequences of protecting group removal and glycosylation steps
starting from suitably protected and activated glycosyl accep-
tors and donors. In classical conditions, each of these steps
requires a chromatographic purification of mixtures that can be
rather complex. Solid-phase synthesis of oligosaccharides has
been extensively studied2 as a means to speed up this process.
However, solid-phase synthesis suffers from some disadvan-
tages, such as difficulties in monitoring the reaction process and
in characterizing the reaction products by NMR analysis, or
mass spectrometry. The use of a soluble PEG-based support3
can overcome some of these disadvantages, however the
reactions still could not be monitored by TLC and the
intermediates would not be easily purified from other PEG-
linked byproducts. Furthermore, in the NMR spectra the signals
due to the PEG protons extensively overlap with the region
typical of carbohydrate protons, thus complicating the charac-
terization of the compounds.
Recently, fluorous chemistry has been developed for use in
several fields such as combinatorial chemistry, parallel synthe-
sis, and the discovery of catalysts.4 This technology is very
attractive for strategic separation of reaction mixtures since
fluorous-tagged compounds can be quickly separated from non-
tagged compounds in binary liquid–liquid and solid–liquid
extractions.4d A large number of fluorine atoms can be required
to induce tagged molecules to partition into a fluorous solvent.
Such highly fluorinated molecules can have little or no
solubility in organic solvents, and therefore finding suitable
reaction conditions for this material is not trivial. On the other
hand, the recently introduced technique of fluorous solid-phase
extraction (SPE) is proving far superior to liquid–liquid
extractions for compounds with fewer fluorine atoms.5 Highly
fluorinated acetal, silyl, benzyl, and novel acyl protecting
groups for hydroxy functions have already been reported6–8 or
are commercially available.
The tert-butyl-phenyl-1H,1H,2H,2H-heptadecafluorodecy-
loxysilyl bromide was synthesized as described in the liter-
ature6c and the fluorous tag was attached to the anomeric
hydroxy group of the glucosamine derivative 1 (Scheme 2) by
using imidazole and DMAP to give the fluorous compound 4 in
85% yield. This silyl ether proved to be sufficiently robust to
withstand the classical reaction conditions of sugar chemistry,
as shown by the synthesis of the Galb1,3-GlcNAc disaccharide
8 reported in Scheme 2.
This scheme employs many of the reagents, protecting-group
manipulations, and glycosylation conditions that are standard in
the field. Removal of the acetyl groups from 4 under Zemplén
conditions11 followed by treatment of the crude product with
benzaldehyde dimethylacetal in presence of CSA afforded the
fluorous glycosyl acceptor 6. Due to the limited number of
fluorine atoms on the tag, 5 and 6 maintain a normal
We would like to report here the application of a known acid-
and base-stable silyl protecting group developed by Wipf and
co-workers6c to fluorous oligosaccharide synthesis.
Our aim is to exploit a fluorous silyl protecting group
appended at the anomeric position of glycosyl acceptors in a
growing oligosaccharide chain to avoid the tedious, expensive,
and time consuming flash chromatography purification of the
reaction product after every step. Our concept of oligosacchar-
ide synthesis using a fluorous tag is shown in Scheme 1.
Introduction of the tag at the anomeric position of the glycosyl
† Electronic supplementary information (ESI) available: experimental data.
Scheme 1 Concept of oligosaccharide synthesis using a fluorous tag at the
See http://www.rsc.org/suppdata/cc/b3/b311448a/
anomeric position of the glycosyl acceptor.
2930
CHEM. COMMUN., 2003, 2930–2931
This journal is © The Royal Society of Chemistry 2003

carbohydrates, the fluorous tag strategy has the advantage of
allowing the use of silica gel TLC to monitor the reaction
process. Furthermore, the NMR signals originating from the
fluorous tag do not interfere with the carbohydrate region, thus
allowing an easy characterization of the tagged compounds. The
optimization of the glycosidation conditions, such as the study
of new silyl fluorous tags, and further applications to the
synthesis of several carbohydrates and glyconjugates are now in
progress.
Notes and references
1 (a) A. Varki, Glycobiology, 1993, 3, 97; (b) D. L. Blithe, Trends
Glycosci. Glycotechnol., 1993, 5, 81.
2 (a) S. Manabe and Y. Ito, J. Am. Chem. Soc., 2002, 124, 12638; (b) O.
J. Plante, E. R. Palmacci and P. H. Seeberger, Science, 2001, 291, 1523
and references therein (c) E. R. Palmacci, M. C. Hewitt and P. H.
Seeberger, Angew. Chem., Int. Ed., 2001, 40, 4433; (d) F. Roussel, M.
Takhi and R. R. Schmidt, J. Org. Chem., 2001, 66, 8540; (e) X. Wu, M.
Grathwohl and R. R. Schmidt, Org. Lett., 2001, 3, 747; (f) F. Roussel,
L. Knerr, M. Grathwohl and R. R. Schmidt, Org. Lett., 2000, 2, 3043; (g)
J. Rademann and R. R. Schmidt, Carbohydr. Res., 1995, 269, 217.
3 (a) E. Eicher, F. Yan, J. Sealy and D. M. Whitfield, Tetrahedron, 2001,
57, 6679; (b) H. Ando, S. Manabe, Y. Nakahara and Y. Ito, J. Am. Chem.
Soc., 2001, 123, 3848; (c) Y. Ito, O. Kanie and T. Ogawa, Angew.
Chem., Int. Ed. Engl., 1996, 35, 2510; (d) S. P. Douglas, D. M. Whitfield
and J. J. Krepinsky, J. Am. Chem. Soc., 1995, 117, 2116.
Scheme
2 a) tert-Butyl-phenyl-1H,1H,2H,2H-heptadecafluorodecylox-
4 (a) Z. Luo, Q. Zhang, Y. Oderatoshi and D. P. Curran, Science, 2001,
291, 1766; (b) A. G. M. Barret, D. C. Braddock, D. Catterick, D.
Chadwick, J. P. Henschke and R. M. McKinnell, Synlett, 2000, 847; (c)
D. P. Curran, Pure Appl. Chem., 2000, 72, 1649; (d) D. P. Curran,
Angew. Chem., Int. Ed., 1998, 37, 1175 and references therein (e) W.
Zhang, Tetrahedron, 2003, 59, 4475.
5 (a) D. P. Curran and Z. Luo, J. Am. Chem. Soc., 1999, 121, 9069; (b) Q.
Zhang, Z. Luo and D. P. Curran, J. Org. Chem., 2000, 65, 8866.
6 (a) P. Wipf and J. T. Reeves, Tetrahedron Lett., 1999, 40, 5139; (b) P.
Wipf and J. T. Reeves, Tetrahedron Lett., 1999, 40, 4649; (c) S. Röver
and P. Wipf, Tetrahedron Lett., 1999, 40, 5667; (d) P. Wipf, J. T.
Reeves, R. Balachandram, K. A. Giuliano, E. Hamel and B. W. Day, J.
Am. Chem. Soc., 2000, 122, 9391.
7 D. P. Curran, R. Ferritto and Y. Hua, Tetrahedron Lett., 1998, 39,
4937.
8 (a) T. Miura, Y. Hirose, M. Ohmae and T. Inazu, Org. Lett., 2001, 3,
3947; (b) T. Miura and T. Inazu, Tetrahedron Lett., 2003, 44, 1819; (c)
T. Miura, K. Goto, D. Hosaka and T. Inazu, Angew. Chem., Int. Ed.,
2003, 42, 2047.
ysilyl bromide, imidazole, DMAP, DCM, 85%; b) MeONa, MeOH, quant.;
c) benzaldehyde dimethylacetal, CSA, CH3CN, 75% after 2 cycles; d) 7,
TMSOTf, DCM, 78% after 2 cycles; e) TBAF, THF, 62%.
chromatographic behavior on standard silica gel, which allows
monitoring of the course of the reactions by TLC. After a first
cycle of acetal formation, TLC and MALDI-TOF analysis of the
reaction crude revealed a small quantity of non reacted starting
material 5. This could have been removed by standard flash
chromatography on regular silica gel. However, as a proof of
principle, the reaction mixture was subjected to SPE on fluorous
silica gel. Elution with 80% MeOH–H2O removed all non-
fluorous material, and the fluorous tagged compounds were
eluted from the column using pure MeOH. The mixture was
resubmitted to the acetal formation. This second cycle con-
sumed all the starting material and 6 was isolated in 75%
yield.
9 (a) D. P. Curran, Angew. Chem., Int. Ed., 1998, 37, 1174; (b) I. T.
Horvát, Acc. Chem. Res., 1998, 31, 641; (c) A. Studer, S. Halidida, R.
Ferritto, S.-Y. Kim, P. Jeger, P. Wipf and D. P. Curran, Science, 1997,
275, 823; (d) A. Studer, P. Jeger, P. Wipf and D. P. Curran, J. Org.
Chem., 1997, 62, 2917.
10 The commercially available fluorous silyl chloride 2 yields a relatively
unstable silyl ether, which cannot survive, for instance, the methoxide
catalyzed deacetylation conditions. Compounds tagged at the anomeric
position with 2 could nonetheless be separated from untagged material
as described in this paper. For instance, the fully protected fluorous silyl
glucosamine 3 could easily be separated on fluorous silica gel from
tetra-O-acetylglucose by first eluting the non-tagged compound with
80% MeOH–H2O, followed by pure MeOH that cleanly removed the
tagged product 3 from the column.
The disaccharide 8 was obtained by reaction of the pure
acceptor 6 with the glycosyl donor 712 (3 equiv.) in the presence
of TMSOTf (0.1 equiv.) in CH2Cl2. Again, the course of the
reaction could be monitored by TLC on standard plates. The
fluorous-tagged material could be isolated from by-products by
eluting first with 80% MeOH–H2O (to remove the nontagged
products) and then with MeOH (to remove the tagged products),
and the reaction was driven to completion by a second cycle of
glycosylation with 2 equivalents of the glycosyl donor 7.
Finally, the fluorous silyl protecting group of 8 was removed
by TBAF to afford crude 9, which was passed through fluorous
silica gel, eluting with MeOH–H2O.
In conclusion, the use of a fluorous tag as protecting group at
the anomeric position of a glycosyl acceptor allowed rapid
synthesis of a disaccharide by a fluorous solid-phase extraction
purification. The disaccharide 9 was obtained in 36% overall
yield from 4 (four steps) without silica-gel chromatographic
purification. Each synthetic intermediate could be easily
purified and characterized by NMR, mass spectroscopy
(MALDI-TOF), and TLC on standard silica gel plates. This is a
major advantage over classical solid-phase synthesis condi-
tions, and allows rapid optimization of the reaction conditions
for each synthetic step. Compared to reaction of PEG-supported
11 G. Zemplén, Ber. Dtsch. Chem. Ges., 1927, 60, 1555.
12 R. R. Schmidt, J. Michel and M. Roos, Liebigs Ann. Chem., 1984,
1343.
CHEM. COMMUN., 2003, 2930–2931
2931