Studies on the stereoselective synthesis of C-allyl glycosides

Article abstract of DOI:10.1021/ol801710s

(Equation Presented) An investigation was carried out to explore the use of sulfoxide donors as common precursors to stereoisomeric C-glycoconjugates of glycoprotein and glycolipid tumor antigens. A study focusing on the effects of reaction conditions and substrate structure on the stereoselectivity of allylation was carried out. Although conditions were realized to selectively afford α-allylation products in good to excellent yields, the search for conditions favoring β-selectivity proved less successful.

Full text of DOI:10.1021/ol801710s

ORGANIC
LETTERS
2008
Vol. 10, No. 21
4727-4730
Studies on the Stereoselective
Synthesis of C-Allyl Glycosides
Glenn J. McGarvey,* Christopher A. LeClair, and Bahar A. Schmidtmann
Department of Chemistry, UniVersity of Virginia, CharlottesVille, Virginia 22901
gjm@Virginia.edu
Received August 5, 2008
ABSTRACT
An investigation was carried out to explore the use of sulfoxide donors as common precursors to stereoisomeric C-glycoconjugates of
glycoprotein and glycolipid tumor antigens. A study focusing on the effects of reaction conditions and substrate structure on the stereoselectivity
of allylation was carried out. Although conditions were realized to selectively afford r-allylation products in good to excellent yields, the
search for conditions favoring ꢀ-selectivity proved less successful.
Studies examining the pivotal roles of carbohydrates in
biological processes, including promising applications to the
treatment of disease,1 have stimulated considerable effort to
make available natural and unnatural glycoconjugates by
chemical synthesis.2 It was with such objectives in mind that
we have examined the preparation of C-glycoamino acid
derivatives of tumor-associated carbohydrate antigens for use
in cotranslational synthesis of stable C-glycopeptides.3,4 This
strategy calls for the preparation of a C-allyl carbohydrate
core that is subsequently elaborated into a selected oligosac-
charide structure, then elaborated into the C-linked amino
acid via olefin metathesis.5 With our attention initially
directed toward cancer-linked mucin glycopeptide antigens,
the requisite R-C-allyl galactosamine core (1, Figure 1) was
(1) For reviews on glycobiology and disease: (a) Varki, A. Glycobiology
1993, 3, 97–130. (b) Dwek, R. A. Chem. ReV. 1996, 96, 683–720. (c)
Brockhausen, I.; Schutzbach, J.; Kuhns, W. Acta Anat. 1998, 161, 36–78.
(d) DeMarco, M. L.; Woods, R. J. Glycobiology 2008, 18, 426–440.
(2) For recent reviews addressing synthetic studies on biological
carbohydrates: (a) Warren, D. J.; Geng, X.; Danishefsky, S. J. Top. Curr.
Chem. 2007, 267, 109–141. (b) Galonic, D. P.; Gin, D. Y. Nature 2007,
446, 1000–1007. (c) Liakatos, A.; Kunz, H. Curr. Opin. Mol. Ther. 2007,
9, 35–44. (d) Haase, C.; Seitz, O. Top. Curr. Chem. 2007, 265, 1–36. (e)
Specker, D.; Wittmann, V. Top. Curr. Chem. 2007, 267, 65–107. (f) Buskas,
T.; Ingale, S.; Boons, G.-J. Glycobiology 2006, 16, 113R–136R. (g)
Czlapkinski, J. L.; Bertozzi, C. R. Curr. Opin. Chem. Biol. 2006, 10, 645–
651. (h) Macmillan, D.; Daines, A. M. Curr. Med. Chem. 2003, 10, 2733–
2773.
Figure 1. Target structures for C-linked cancer antigens.
concisely realized through radical-mediated allylation of per-
acetylated galactosamine chloride.6 Anticipating the applica-
tion of this strategy toward the preparation of C-analogs of
glycolipid cancer antigens,7 our attention was turned to the
(3) McGarvey, G. J.; Benedum, T. E.; Schmidtmann, F. W. Org. Lett.
2002, 4, 3591–3594
(4) Review of C-neoglycopeptides: Dondoni, A.; Marra, A. Chem. ReV.
2000, 100, 4395–4421
.
(5) Schmidtmann, F. W.; Benedum, T. E.; McGarvey, G. J. Tetrahedron
Lett. 2005, 46, 4677–4681.
.
(6) Ciu, J.; Horton, D. Carbohydr. Res. 1998, 309, 319–330.
10.1021/ol801710s CCC: $40.75
Published on Web 10/01/2008
2008 American Chemical Society

realization of the ꢀ-C-allyl lactose core (2) structure of these
antigens. Implementation of this strategy required us to
examine the means by which stereoselective C-allylation may
be realized to establish the desired stable anomeric stereo-
chemistry prior to subsequent elaboration.8,9
Scheme 2. Stereochemical Course of Glycosylation
We chose D-glucose as an easily available model upon
which to explore methods of stereoselective C-allylation.
Conveniently, both of the target isomers could be delivered
using established means. Specifically, the R-isomer may be
obtained through ionic allylation of the inexpensive peracetyl
derivative of D-glucose with allyl-SiMe3,10 whereas the
ꢀ-isomer can be realized through stereoelectronically guided
reduction of a lactol intermediate derived from allyl anion
addition to benzyl protected lactone 4 (R ) Bn, Scheme 1).11
intimate ion pair could afford the R-allylation product, while
isomerization of the intimate ion pair would presumably
favor the more stable R-isomer15 to offer a means to the
ꢀ-allyl product. It was anticipated that adjustment in the
sequence of donor activation and introduction of the nucleo-
phile could affect the interception of this equilibrium and
influence the stereochemical course of the reaction.
Scheme 1. Preparation of R- and ꢀ-C-Allyl Glucose
Table 1. Effect of Reaction Conditions and Protecting Group on
Steroselectivity
The modest yield for the R-isomer, the rather lengthy route
leading to the ꢀ-isomer, as well as limitations these routes
impose on protecting groups prompted us to explore more
convenient and efficient means to these stereoisomeric
C-glycosides.
It was with this in mind that we speculated that both R-C-
allyl and ꢀ-C-allyl isomers might be accessed by stereodi-
vergent allylation of the stable thioglycoside intermediate
3. The seminal studies of Crich and co-workers12 provided
us a working hypothesis by which anticipated allylation
reaction may be viewed as resulting from a series of
equilibrating ion pairs (Scheme 2).13 It was suggested that
the initially formed oxonium ion14 or ꢀ-isomer of a resulting
(7) Hakamori, S. Acta Anatomica 1998, 161, 79–90.
(8) For reviews on approaches to the synthesis of C-glycosides: (a) Du,
Y.; Linhardt, R. J. Tetrahedron 1998, 54, 9913–9959. (b) Lee, D. Y. W.;
He, M. S. Curr. Top. Med. Chem. 2005, 5, 1333–1350. (c) Yuan, X. J.;
Linhardt, R. J. Curr. Top. Med. Chem. 2005, 5, 1393–1430. (d) Liu, L.;
a Reaction Conditions: A: sulfoxide (1.0 equiv), allyl-M (2.5 equiv), Tf2O
(1.1 equiv), DTBMP, -78 °C to rt. B: sulfoxide (1.0 equiv), Tf2O (1.1
equiv), DTBMP, -78 °C, 10 min; allyl-M (2.5 equiv), -78 °C to rt. C:
allyl-M (2.5 equiv), Tf2O (1.1 equiv), DTBMP, -78 °C.; sulfoxide (1.0
equiv), -78 °C to rt. b Yields are of chromatographically purified products.
c R/ꢀ ratio determined through NMR observation of the C-2 1H of the allyl
group in the crude reaction mixture.18 d Reaction carried out in the absence
of DTBMP.
McKee, M.; Postema, M. H. D. Curr. Org. Chem. 2001, 5, 1133–1167
.
(9) For a previous approach to stereoisomeric C-glycosides: Shao, H.;
Wang, Z.; Lacroix, E.; Wu, S. H.; Jennings, H. J.; Zou, W. J. Am. Chem.
Soc. 2002, 124, 2130–2131
.
(10) Horton, D.; Miyake, T Carbohydr. Res. 1988, 184, 221–229.
(11) Brenna, E.; Fuganti, C.; Grasselli, P.; Serra, S.; Zambotti, S.
Chem.-Eur. J. 2002, 8, 1872–1878.
(12) For a thorough review: Crich, D.; Lim, L. B. L. Org. React. 2004,
64, 115–251, and references cited therein.
4728
Org. Lett., Vol. 10, No. 21, 2008

Table 2. Effect of Substrate on Stereoselectivity
b
a Conditions A: see Table 1, entry 2; Conditions B: see Table 1, entry 6. R/ꢀ ratio determined through NMR observation of the C-2 H of the allyl
1
group in the crude reaction mixture.18
In an effort to focus on the details attending the intercep-
tion of the cationic intermediates, we chose to explore the
potential of pre- and postactivation of an anomeric sulfoxide
with respect to the presence of the nucleophile16 using readily
available perbenzyl glucose derivative 5 (R ) Bn, Table 1).17
Initial attempts to activate the sulfoxide using Tf2O in the
presence of allyl-SiMe3 resulted in only a modest yield of
We next chose to examine the effect of protecting groups
on promoting ꢀ-selectivity to this allylation reaction. Entries
7-18 describe the application of Conditions A and B to a
variety of protected D-glucose sulfoxides. It is noteworthy
that silyl protection gave the only the R-anomer regardless
of the conditions applied (entries 7 and 8). Ester protection
proved problematic, however. The acetate protected acceptor
yielded only acetal 6, evidence that the desired neighboring
participation took place but with an undesirable capture of
the nucleophile (entries 9 and 10).20 Benzoyl and pivolyl
protection proved only slightly better, affording low yields
of allyl anomers with modest to poor selectivity (entries
11-14), while trichloroacetate protection (TCA) led only
to decomposition (entries 15 and 16). The benzylidene
protected substrate offered no advantages over substrate 5
(R ) Bn), affording the R-anomer in modest to good yields
regardless of the conditions applied (entries 17 and 18).21
products favoring the R-product (16:1, Entry 1).18
A
significant improvement in yield was observed when 2,6-
di-tert-butyl-4-methylpyridine (DTBMP) 1.2 equiv was
introduced to act as an acid scavenger to cleanly afford the
R-product (entry 2).19 In an effort to access the ꢀ-isomer,
activation of the sulfoxide was carried out prior to introduc-
tion of the allyl nucleophile (Conditions B) to once again
afford the R-isomer in good yield (entry 3). Conditions of
inverse addition were also examined (Conditions C) to
similarly afford the R-isomer, but in diminished yield (entry
4).
Armed with these results, attention was turned to the
stereoselective C-allylation of carbohydrate substrates other
than D-glucose (Table 2). Using the conditions found best
for the diastereoselective allylation of D-glucose (Table 1,
entries 2 and 6), other perbenzylated donors were examined
(Table 2). The galactose and mannose donors followed the
trend noted for glucose, affording cleanly the R-anomer using
conditions A with allyl-TMS (entries 1 and 3) and showing
significant shift toward the ꢀ-anomer under conditions B with
allyl-SnBu3 (entries 2 and 4). The fucose donor, on the other
hand, proved poorly stereoselective under either set of
reaction conditions (entries 5 and 6). Finally, perbenzylated
lactose was examined in an effort to advance our efforts toward
C-allyl glycosides of the glycolipid cancer antigens. While the
In an effort to promote interception of the more stable,
and presumably less reactive, R-isomer of the intimate ion
pair (Scheme 2), we chose to examine a more reactive
nucleophile in the form of allyl-SnBu3. In the event,
application of Conditions A using the allylstannane led to
significantly diminished R-selectivity (entry 5). Activation
of the sulfoxide prior to introduction of the nucleophile
(Conditions B) led to a good yield of product favoring the
ꢀ-isomer (entry 6). Attempts to improve ꢀ-selectivity by
varying the amount of DTBMP and changing the solvent
(EtCN) proved ineffective. Nonetheless, conditions have been
identified to selectively realize either the R- or ꢀ-isomer of
this substrate in good to excellent yields (highlighted entries
2 and 6).
Org. Lett., Vol. 10, No. 21, 2008
4729

R-anomer could be obtained cleanly using conditions A with
allyl-SiMe3, the desired ꢀ-isomer was not available using the
alternative reaction conditions (entries 7 and 8). These results
would seem to serve to illustrate how structural differences can
affect the stereoselectivity of the incipient cation intermediates
in this glycosyl transfer reactions.14
In summary, it has been found that thioglycosides, via their
corresponding sulfoxides, offer convenient access to R-C-
allyl glycosides that are useful intermediates for further
elaboration into a variety of C-glycosides.22 Unfortunately,
access to the ꢀ-anomers of these C-glycosides through
manipulation of both reaction conditions and donor structure
has shown limited success to date. Realization of C-analogs
of ꢀ-linked glycoconjugates will require alternative ap-
proaches, and we will report on our efforts in this regard in
due course.
Acknowledgment. We thank the National Institutes of
Health (GM58510) and the University of Virginia for support
of this work.
Supporting Information Available: Experimental details
and physical data are available for the new compounds
described herein. This material is available free of charge
via the Internet at http://pubs.acs.org.
OL801710S
(13) For discussions regarding the basis for stereoselection in sulfur-
mediated glycosylations: (a) Crich, D.; Lim, L. B. L. Org. React. 2004, 64,
115–251. (b) Crich, D.; Sun, S. X. J. Org. Chem. 1996, 61, 4506–4507. (c)
Crich, D.; Sun, S. X. J. Org. Chem. 1997, 62, 1198–1199. (d) Crich, D.;
Sun, S. X. J. Am. Chem. Soc. 1998, 120, 435–436. (e) Crich, D.; Smith, M.
J. Am. Chem. Soc. 2001, 123, 9015–9020. (f) Crich, D.; Sun, S. X. J. Am.
Chem. Soc. 1997, 119, 11217–11223. (g) Crich, D.; Chandrasekera, N. S.
Angew. Chem., Int. Ed. 2004, 43, 5386–5389.
(17) Compound 5 was prepared by the methods given in: (a) Xia, J.;
Alderfer, L.; Piskorz, D. R.; Locke, R. D.; Matta, K. L. Carbohydr. Res.
2000, 328, 147–163. (b) Khiar, N.; Martin-Lomas, M. J. Org. Chem. 1995,
60, 7017–7021.
(18) Product identities were confirmed by chromatographic separation
of the isomers and comparison with pure allylated products prepared by
independent synthesis (see Supporting Information).
(19) It is surmised that DTBMP is scavenging extraneous protons in
the reaction mixture, though the presence of this base has also been noted
to effect the stereoselectivity of glycosylations.13b
(20) Whitfield, D. M.; Nukata, T. Carbohydr. Res. 2007, 342, 1291–
1304.
(14) For a comprehensive discussion of the current understanding of
the mechanism of cationic glycosylation: Bennet, A. J.; Kitos, T. E. J.Chem.
Soc., Perkin Trans. 2 2002, 1207–1222.
(15) NMR experiments have shown this to be the favored anomeric
stereochemistry: (a) Crich, D.; Sun, S. X. J. Am. Chem. Soc. 1997, 119,
11217–11223. (b) Crich, D.; Chandrasekera, N. S. Angew. Chem., Int. Ed.
2004, 43, 5386–5389.
(21) For a discussion addressing the use of cyclic protecting groups,
see: (a) Crich, D.; Subramanian, V.; Hutton, T. K. Tetrahedron 2007, 63,
5042–5049. (b) Litiens, R. E. J. N.; van den Bos, J. L.; Codee, J. D. C.;
Overkleeft, H. S.; van der Marel, G. Carbohydr. Res. 2007, 342, 419–429.
(22) McGarvey, G. J.; Schmidtmann, F. W.; Benedum, T. E.; Kizer, D.
E Tetrahedron Lett. 2003, 44, 3775–3779.
(16) For an earlier example of analogous sulfoxide-mediated allylation,
see: Berkowitz, D. B.; Choi, S.; Bhuniya, D.; Shoemaker, R. Org. Lett.
2000, 2, 1149–1152.
4730
Org. Lett., Vol. 10, No. 21, 2008