C2-amidoglycosylation. Scope and mechanism of nitrogen transfer

Article abstract of DOI:10.1021/ja026281n

A one-pot C2-amidoglycosylation reaction for the synthesis of 2-N-acyl-2-deoxy-β-pyranosides from glycals is described. Glycal donors activated by the reagent combination of thianthrene-5-oxide (11) and Tf₂O, followed by treatment with an amide nucleophile and a glycosyl acceptor, lead to the formation of various C2-amidoglycoconjugates. Both the C2-nitrogen transfer and the glycosidic bond formation proceed stereoselectively, allowing for the introduction of both natural and nonnatural amide functionalities at C2 with concomitant anomeric bond formation in a one-pot procedure. Tracking of the reaction by low-temperature NMR spectroscopy employing ¹⁵N- and ¹⁸O-isotope labels suggests a mechanism involving the formation of the C2-sulfonium glycosyl imidate 39 as well as oxazoline 37 as key intermediates in this novel oxidative glycosylation process.

Full text of DOI:10.1021/ja026281n

Published on Web 07/27/2002
C2-Amidoglycosylation. Scope and Mechanism of Nitrogen
Transfer
Jing Liu and David Y. Gin*
Contribution from the Department of Chemistry, UniVersity of Illinois at Urbana-Champaign,
Urbana, Illinois 61801
Received March 22, 2002
Abstract: A one-pot C2-amidoglycosylation reaction for the synthesis of 2-N-acyl-2-deoxy-â-pyranosides
from glycals is described. Glycal donors activated by the reagent combination of thianthrene-5-oxide (11)
and Tf2O, followed by treatment with an amide nucleophile and a glycosyl acceptor, lead to the formation
of various C2-amidoglycoconjugates. Both the C2-nitrogen transfer and the glycosidic bond formation
proceed stereoselectively, allowing for the introduction of both natural and nonnatural amide functionalities
at C2 with concomitant anomeric bond formation in a one-pot procedure. Tracking of the reaction by low-
temperature NMR spectroscopy employing 15N- and 18O-isotope labels suggests a mechanism involving
the formation of the C2-sulfonium glycosyl imidate 39 as well as oxazoline 37 as key intermediates in this
novel oxidative glycosylation process.
Scheme 1
Introduction
C2-Azaglycosides are ubiquitous building blocks in biologi-
cally important glycoconjugates including glycoproteins, pep-
tidoglycans, glycolipids, and glycosaminoglycans. These gly-
coconjugates are involved in critical biological functions1 such
as cell communication, cell adhesion, inflammatory response,
and immune response. However, access to significant quantities
of structurally well-defined, homogeneous complex oligosac-
charides for glycobiological exploration has relied on chemical2
or chemoenzymatic syntheses3 because the biosynthesis of
complex carbohydrates is not under direct transcriptional control.
In this context, the synthetic incorporation of C2-azaglycosyl
residues in the preparation of oligosaccharides and glycocon-
jugates is particularly challenging in that selective C2-N-
functionalization as well as glycosidic bond formation with
appropriate glycosyl acceptors are required for carbohydrate
assembly.4
Glycal donors5 are often employed as versatile starting
materials in the synthesis of C2-azaglycosides by way of C2-
N-functionalization accompanied by glycosidic bond formation.
Over the last few decades, a variety of methods have been
developed for the nitrogen ([N]) transfer to glycals (1 f 2,
Scheme 1). In this context, numerous N-functionalities have been
introduced at the C2-position including oxidized precursors to
amines, such as azide,6 hydrazine,7 triazene,8 or the nitro group,9
as well as protected amino-functionalities such as amides,10
carbamates,11 sulfonamides,12 or phosphoramides.13 However,
the majority of naturally occurring C2-azasugars are N-acetyl-
ated, and this highlights the importance of this class of
carbohydrate residue in glycoconjugate synthesis. While all of
the above-mentioned methods constitute efficient syntheses of
C2-azasugars, the direct installation of the naturally occurring
C2-acetamido group onto glycal donors in conjunction with
glycosidic coupling remains a challenge (Scheme 1; 1 f 2, [N]
) AcNH). To date, the only report on the direct transfer of the
native acetamido group to the C2-position of glycals employs
a Cr(II)-mediated N-chloroacetamide addition to glycals,11e
proceeding with modest yields. As a result, the preparation of
C2-acetamidoglycosides from glycal donors typically requires
multistep syntheses involving the distinct transformations of
(6) Lemieux, R. U.; Ratcliffe, R. M. Can. J. Chem. 1979, 57, 1244-1251.
(7) (a) Leblanc, Y.; Fitzsimmons, B. J.; Springer, J. P.; Rokach, J. J. Am. Chem.
Soc. 1989, 111, 2995-3000. (b) Grondin, R.; LeBlanc, Y.; Hoogsteen, K.
Tetrahedron Lett. 1991, 32, 5021-5024.
(8) Weng, M.; Geyer, A.; Friemel, A.; Jochims, J. C.; Lutz, M. J. Prakt. Chem.
2000, 342, 486-493.
(9) Winterfeld, G. A.; Schmidt, R. R. Angew. Chem., Int. Ed. 2001, 40, 2654-
2657 and references therein.
* To whom correspondence should be addressed. E-mail: gin@scs.
uiuc.edu.
(1) (a) Dwek, R. A. Chem. ReV. 1996, 96, 683-720. (b) Bertozzi, C. R.;
Kiessling, L. L. Science 2001, 291, 2357-2364. (c) Helenius, A.; Aebi,
M. Science 2001, 291, 2364-2369.
(2) (a) Toshima, K.; Tatsuta, K. Chem. ReV. 1993, 93, 1503-1531. (b) Schmidt,
R. R.; Kinzy, W. AdV. Carbohydr. Chem. Biochem. 1994, 50, 21-123. (c)
Davis, B. G. J. Chem. Soc., Perkin Trans. 1 1999, 3215-3237. (d) Davis,
B. G. J. Chem. Soc., Perkin Trans. 1 2000, 2137-2160.
(3) Kren, V.; Thiem, J. Chem. Soc. ReV. 1997, 26, 463-473.
(4) Banoub, J.; Boullanger, P.; Lafont, D. Chem. ReV. 1992, 92, 1167-1195.
(5) Danishefsky, S. J.; Bilodeau, M. T. Angew. Chem., Int. Ed. Engl. 1996,
35, 1380-1419.
(10) (a) Donohoe, T. J.; Blades, K.; Helliwell, M. Chem. Commun. 1999, 1733-
1734. (b) Du Bois, J.; Tomooka, C. S.; Hong, J.; Carreira, E. M. J. Am.
Chem. Soc. 1997, 119, 3179-3180.
(11) (a) Kozlowska-Gramsz, E.; Descotes, G. Tetrahedron Lett. 1981, 22, 563-
566. (b) Kan, C.; Long, C. M.; Paul, M.; Ring, C. M.; Tully, S. E.; Rojas,
C. M. Org. Lett. 2001, 3, 381-384. (c) Levites-Agababa, E.; Menhaji, E.;
Perlson, L. N.; Rojas, C. M. Org. Lett. 2002, 4, 863-865. (d) Nicolaou,
K. C.; Baran, P. S.; Zhong, Y.-L.; Vega, J. A. Angew. Chem., Int. Ed.
2000, 39, 2525-2529. (e) Driguez, H.; Vermes, J.-P.; Lessard, J. Can. J.
Chem. 1978, 56, 119-130.
(12) Griffith, D. A.; Danishefsky, S. J. J. Am. Chem. Soc. 1990, 112, 5811-
5819.
(13) Lafont, D.; Descotes, G. Carbohydr. Res. 1988, 175, 35-48.
9
10.1021/ja026281n CCC: $22.00 © 2002 American Chemical Society
J. AM. CHEM. SOC. 2002, 124, 9789-9797
9789

A R T I C L E S
Liu and Gin
Scheme 2
Scheme 3
nitrogen transfer, glycosylation, and finally manipulation of the
nitrogen protective group.
1 at C2, affording a pyranosyl intermediate 4 that incorporates
a sulfonium moiety at C2, as well as an oxocarbenium
functionality at C1. Introduction of a suitable N-nucleophile at
C2 (with expulsion of Aryl2S) in conjunction with an appropriate
glycosyl acceptor (NuH) at C1 would result in a C2-azaglyco-
sylation reaction in which nitrogen transfer to C2, as well as
anomeric bond formation, are accomplished. In this context,
amide substrates (R′CONH2) would be suitable N-nucleophiles
as they possess two nucleophilic sites (O- and N-) that could
substitute at both C1 and C2 of the putative activated glycosyl
intermediate 4. Ideally, regiospecific nitrogen transfer to C2 and
O-substitution at C1 would result in the formation of oxazoline
5, an intermediate known to function as a glycosyl donor for
anomeric bond formation.18 Moreover, the successful use of
amide nucleophiles (R′CONH2) has the potential to accomplish
the direct transfer of an acetamido group to glycals and provide
the native C2-acetamidoglycoconjugates 6 (R′ ) Me) without
the need for additional protective group exchanges or C2-
functional group interconversions.
Recently, we have established a novel glycal activation
method employing the reagent combination of a diaryl sulfoxide
and triflic anhydride. Several glycosylation protocols have since
been developed with this reagent combination,14 including the
C2-acetamidoglycosylation of glycals.15 In this method, glycal
activation, nitrogen transfer to C2, glycosidic bond formation,
as well as installation of the native N-acetyl functionality are
all accomplished in one reaction vessel with good yield and
excellent diastereoselectivity. Here we report the development
of this methodology, establishment of the reaction scope, as
well as studies toward elucidating the mechanism of the C2-
amidoglycosylation process.
Results and Discussion
C2-Amidoglycosylation. We have established that sulfonium
reagents, generated in situ from the combination of a diaryl
sulfoxide and triflic anhydride,16 can activate a variety of
carbohydrate donors for glycosidic bond formation. With this
reagent combination, glycosyl hemiacetals can function as
effective donors for dehydrative glycosylation,17 a process
amenable to the construction of a variety of glycoconjugates.
More recently, the use of this reagent combination has also been
applied to the electrophilic activation of glycal donors, resulting
in an oxidative C2-hydroxyglycosylation,14 in which a hydroxyl
functionality is stereoselectively installed at the C2-position of
glycal with concomitant glycosylation of a nucleophilic acceptor.
On the basis of the established utility of these sulfonium
triflate reagents as electrophilic oxidants, a new method for the
synthesis of C2-azaglycosides was explored, in which an
appropriate N-functionality is installed at the C2-position of
glycals (Scheme 2). It was envisioned that a sulfonium triflate
species 3, generated in situ from a diaryl sulfoxide and triflic
anhydride, would activate the enol ether functionality of glycal
Initial explorations to establish the method focused on the
use of diphenyl sulfoxide and triflic anhydride as the glycal
oxidant. For example, tri-O-benzyl-D-glucal (7, Scheme 3) was
activated with this reagent combination at -78 °C, followed
by introduction of N-(TMS)acetamide as the nitrogen transfer
reagent and Hu¨nig’s base as a TMSOTf scavenger. The reaction
was warmed to 23 °C, and the glycosyl acceptor 2-propanol
and an acid promoter, Amberlyst-15, were added. Isopropyl 2-N-
acetylamino-3,4,6-tri-O-benzyl-2-deoxy-â-D-glucopyranoside (8)
was isolated in 15% yield (Scheme 3), thus demonstrating the
feasibility of a one-pot C2-acetamidoglycosylation reaction with
glycal donors. Although the desired C2-acetamidoglycoside 8
was obtained in only modest yield, several key observations
offer insight into the reaction pathway, and thus present avenues
for reaction optimization. First, the bicyclic oxazoline intermedi-
ate 9 (Scheme 3) could be detected and isolated in 18% yield
prior to the addition of the glycosyl acceptor. This is consistent
with the proposed pathway in Scheme 2 in which regioselective
C2-N and C1-O bond formations occur to generate oxazoline
5 as an in situ glycosyl donor. Second, diphenyl sulfide (17%)
was isolated as a byproduct of the reaction, which is consistent
with nitrogen transfer to the C2-position by the expulsion of
(14) (a) Di Bussolo, V.; Kim Y.-J.; Gin, D. Y. J. Am. Chem. Soc. 1998, 120,
13515-13516. (b) Kim, J.-Y.; Di Bussolo, V.; Gin, D. Y. Org. Lett. 2001,
3, 303-306. (c) Kim, Y.-J.; Gin, D. Y. Org. Lett. 2001, 3, 1801-1804.
(15) Di Bussolo, V.; Liu, J.; Huffman, L. G., Jr.; Gin, D. Y. Angew. Chem., Int.
Ed. 2000, 39, 204-207.
(16) For a recent review on Tf2O, see: Baraznenok, I. L.; Nenajdenko, V. G.;
Balenkova, E. S. Tetrahedron 2000, 56, 3077-3119.
(17) (a) Garcia, B. A.; Poole, J. L.; Gin, D. Y. J. Am. Chem. Soc. 1997, 119,
7597-7598. (b) Garcia, B. A.; Gin, D. Y. J. Am. Chem. Soc. 2000, 122,
4269-4279. (c) Garcia, B. A.; Gin, D. Y. Org. Lett. 2000, 2, 2135-2138.
(d) Nguyen, H. M.; Poole, J. L.; Gin, D. Y. Angew. Chem., Int. Ed. 2001,
40, 414-417. (e) For dehydrative glycosylation employing sulfides and
Tf2O, see: Nguyen, H. M.; Chen, Y.; Duron, S. G.; Gin, D. Y. J. Am.
Chem. Soc. 2001, 123, 8766-8772.
(18) Warren, C. D.; Jeanloz, R. W.; Strecker, G. Carbohydr. Res. 1981, 92,
85-101.
9
9790 J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002

C2-Amidoglycosylation
A R T I C L E S
Scheme 4
Chart 1. a
sulfide from 4. Finally, the major carbohydrate byproduct of
the reaction was found to be the C2-vinyl sulfonium triflate 10
(42%), likely formed via the elimination of the H2-proton in
the putative C2-sulfonium glycosyl intermediate 4 (Scheme 2).
To further develop this C2-amidoglycosylation reaction for
generalized carbohydrate synthesis, optimization of the reaction
focused on minimizing the presumed H2-elimination pathway
that produces 10. To this end, a variety of diaryl sulfoxide
reagents were screened to generate a putative C2-sulfonium
intermediate 4 (Scheme 2) with attenuated H2-acidity. Thian-
threne-5-oxide (11, Scheme 4), easily prepared from the
oxidation of commercially available thianthrene, was found to
provide an enhanced yield of the C2-amidoglycoside 8, presum-
ably a result of the ability of the remote sulfur atom in 11 to
stabilize the C2-sulfonium and C1-oxocarbenium functionalities
in 4 (vide infra). The aim of minimizing the formation of the
vinyl sulfonium byproduct 10 also led to the screening of a
series of tertiary amines, including several trialkylamines,
pyridines, and anilines, to find the appropriate reagent to serve
as an effective yet mild proton and trimethylsilyl scavenger in
the N-transfer process. Of these, N,N-diethylaniline, an effective
acid scavenger, was found to give optimal yields of the C2-
acetamidoglycoside 8 (73%) from tri-O-benzyl-D-glucal (7),
when 11 was employed as the sulfoxide reagent (Scheme 4).19
Using this reaction protocol, we applied the C2-acetami-
doglycosylation reaction to the synthesis of a wide variety of
C2-acetamidoglycosides (12-22, Chart 1), in which glycal acti-
vation, transfer of the native N-acetylamino functionality, and
glycosidic bond formation are all accomplished in a one-pot
procedure. The acetamidoglycosylation reaction is compatible
with glycal substrates bearing a variety of protective groups,
including alkyl ethers, silyl ethers, and esters. A wide variety
of glycosyl acceptors can be glycosylated, including simple pri-
mary and secondary alcohols (12-14, 22), primary and second-
ary hydroxyls within carbohydrates (15-17), as well as hydroxyl
functionalities on selectively protected R-amino acids (21).
Nitrogen nucleophiles such as azide are also glycosylated to
generate glycosyl azides (18-20), although a solvent exchange
from CH2Cl2 to DMF is required in the oxazoline ring-opening
stage of the transformation to ensure adequate solubility of
NaN3.20 It is worth noting that the glycal enol ether functionality
can be selectively activated with the reagent combination of
11‚Tf2O in the presence of simple alkenes, as demonstrated in
the reaction with 6-O-allyl-3,4-di-O-benzy-R--D-glucal to afford
a Key: (a) Amberlyst-15 employed as the acid promoter unless otherwise
stated; (b) amounts of reagents employed: 11, 2 equiv; Tf2O, 2 equiv;
TMSNHAc, 3 equiv; PhNEt2, 4 equiv; NuH, 3 equiv; acid, 2 equiv; (c)
glycosyl acceptor ) (+)-3-dihydrocholesterol; (d) camphorsulfonic acid
employed as acid promoter; (e) glycosyl acceptor ) NaN3; glycosylation
carried out in DMF with Cu(OTf)2 as the acid promoter; (f) dr ) 6:1 (â-
gluco:R-manno diastereomers); (g) dr ) 12:1 (â-gluco:R-manno diastere-
omers).
17. In addition, galactal donors such as tri-O-benzyl-D-galactal
provide the corresponding C2-acetamidogalactopyranosides (20,
22). In all of the C2-acetamidoglycosylation reactions, the C2-
aza-glycoconjugates are derived from acid-mediated ring open-
ing of a bicyclic oxazoline intermediate such as 5. These
substrates have been established as glycosyl donors in the
presence of strongly acidic media.18 However, in the context
of developing the C2-amidoglycosylation reaction, we have
found that sulfonic acids (i.e., Amberlyst-15, camphorsulfonic
acid) in dichloromethane, or Cu(OTf)2 in DMF, can function
as relatively mild acid promoters for glycosylations with
oxazoline donors.
The installation of the native acetamido functionality in the
C2-azaglycosylation reaction provides direct access to the most
abundant form of C2-azasugars. The versatility of this novel
nitrogen transfer reaction can be extended to C2-azaglycosides
bearing different C2-amido functionalities (Chart 2). For
example, N-(TMS)benzamide can be employed as the initial
N-nucleophile to afford C2-benzamidoglycosides (24, 25). In
addition, nonsilylated amides such as 2-benzyloxyacetamide can
(19) Acetamidoglycosylation with 7 employing Ph2SO instead of 11 under
otherwise identical conditions afforded 8 in 41% yield.
(20) The minor R-manno diastereomers of 18 and 19 were also detected, but
the reactions are highly selective for the â-gluco isomer. The minor
quantities of R-manno diastereomers of the other glycoconjugates in Chart
1 could not be readily isolated and characterized because of their
exceedingly small quantities.
9
J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002 9791

A R T I C L E S
Liu and Gin
Chart 2. a
Scheme 5
a Key: (a) Amberlyst-15 employed as the acid promoter unless otherwise
stated; (b) amounts of reagents employed: 11, 2 equiv; Tf2O, 2 equiv; amide,
2-3 equiv; PhNEt2, 4 equiv; NuH, 1.5-2 equiv; acid, 2 equiv; (c)
camphorsulfonic acid employed as acid promoter; (d) glycosyl acceptor )
(+)-3-dihydrocholesterol; (e) glycosyl acceptor ) NaN3; glycosylation
carried out in DMF with Cu(OTf)2 as the acid promoter.
generates the glycosyl N-acylaziridine 36, which then undergoes
rearrangement24 to afford the observed oxazoline 37. Finally,
acid-mediated ring opening with the appropriate nucleophilic
acceptor (Nu-H) yields the C2-amidoglycoside 38. In an
alternate pathway for N-transfer to C2 (Path B), initial O-
addition of the amide nucleophile 34 to C1 of the oxocarbenium
species 33 would generate the R-glycosyl imidate 39. Subse-
quent intramolecular displacement of thianthrene from C2 by
the imidate nitrogen directly affords the observed oxazoline 37,
the precursor to the C2-amidoglycoside product 38.
be used as the nitrogen transfer agent to install the biologically
relevant N-glycolamido substituent at C2 with concomitant
formation of the glycosidic bond (26-29).21 Finally, the use of
pent-4-enamide as the initial N-nucleophile provides the C2-
N-pentenoylglycosides 30 and 31, whose C2-amido functionality
can be deprotected to expose the C2-amino functionality.22
Probing the Mechanism of Nitrogen Transfer. The reaction
pathway proposed for N-transfer in the C2-amidoglycosylation
reaction involves glycal activation, oxazoline formation, and
glycosylation (Scheme 5). In this process, the combination of
Tf2O and thianthrene-5-oxide (11) is likely to generate thian-
threne bis(triflate) (32) in situ.23 Electrophilic activation of the
glucal donor 1 at C2 generates the C1-oxocarbenium-C2-
sulfonium intermediate 33. Nitrogen transfer can then proceed
via introduction of an unprotected or N-silylated amide 34 and
the acid/TMS scavenger PhNEt2, leading to initial N-glycosy-
lation by 33 (Path A) to form the glycosyl amide intermediate
35. Subsequent intramolecular displacement of thianthrene from
C2 by the amide nitrogen in 35 with concomitant N-desilylation
Several initial observations are consistent with the reaction
pathways outlined in Scheme 5. First, the oxazoline intermedi-
ates 9, 40-42 (Chart 3) can be isolated in good yields if the
C2-amidoglycosylation process is intercepted immediately prior
(23) While it is reasonable that the combination of 11 and Tf2O would generate
the sulfonium triflate species 32, our current investigations do not exclude
the possibility of this species existing/reacting as the σ-sulfurane, the
thianthrene sulfur dication, or the thianthrene radical cation. For the
preparation and reactions of sulfonium triflate: (a) Hendrickson, J. B.;
Schwartzman, S. M. Tetrahedron Lett. 1975, 16, 273-276. (b) Nenajdenko,
V. G.; Vertelezkij, P. V.; Gridnev, I. D.; Shevchenko, N. E.; Balenkova,
E. S. Tetrahedron 1997, 53, 8173-8180. σ-Sulfurane: (c) Martin, J. C.;
Perozzi, E. F. Science 1976, 191, 154-159. Thianthrene dication and
thianthrene radical cation: (d) Shine, H. J.; Bandlish, B. K.; Mani, S. R.;
Padilla, A. G. J. Org. Chem. 1979, 44, 915-917. (e) Shine, H. J.; Bae, D.
H.; Mansurul Hoque, A. K. M.; Kajstura, A.; Lee, W. K.; Shaw, R. W.;
Soroka, M.; Engel, P. S.; Keys, D. E. Phosphorus Sulfur Relat. Elem. 1985,
23, 111-141. (f) Lee, W. K.; Liu, B.; Park, C. W.; Shine, H. J.; Guzman-
Jimenez, I. Y.; Whitmire, K. H. J. Org. Chem. 1999, 64, 9206-9210.
(24) Nishiguchi, T.; Tochio, H.; Nabeya, A.; Iwakura, Y. J. Am. Chem. Soc.
1969, 91, 5835-5841.
(21) The use of nonsilylated amides to form 26-31 suggests that the principal
role of the TMS group in N-(TMS)acetamide and N-(TMS)benzamide is
to enhance the solubility of these amide reagents.
(22) Madsen, R.; Roberts, C.; Fraser-Reid, B. J. Org. Chem. 1995, 60, 7920-
7926.
9
9792 J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002

C2-Amidoglycosylation
A R T I C L E S
Chart 3. a
Scheme 6
a Key: (a) dr ) 14:1 (gluco:manno diastereomers); (b) dr ) 17:1 (gluco:
manno diastereomers).
to the introduction of the nucleophilic acceptor. Second, the
C2-thianthrene sulfonium glycal 43 is isolated as a minor
byproduct (<15%) in glycosylations where 7 serves as the glycal
donor. Formation of 43 is likely a result of the unproductive
elimination of H2 from the putative C2-sulfonium species 33,
a pathway that is notably lessened with the use of thianthrene-
5-oxide as the sulfoxide reagent as opposed to Ph2SO (Scheme
3). The advantageous use of thianthrene-5-oxide in this capacity
is presumably a result of added stabilization of a putative C2-
sulfonium intermediate such as 33. For example, the distal sulfur
atom in the C2-sulfonium moiety in 33 can be envisioned to
participate in attenuating the H2-acidity of 33, either by
resonance stabilization (33i) of the C2-sulfur cation or by direct
participation in the stabilization of the C1-oxocarbenium
functionality (33ii).
On the basis of these initial observations, the formation of
oxazoline 37 from 1 is consistent with nucleophilic attack by
the amide nucleophile at C1 and C2 of an activated glycosyl
intermediate such as 33, resulting in displacement of thianthrene
as depicted in Scheme 5. However, the mechanism of N-transfer
to C2 remains unclear, as it can proceed either via the glycosyl
amide and aziridine intermediates (33 f 35 f 36 f 37, Path
A) or via a C2-sulfonium glycosyl imidate intermediate (33 f
39 f 37, Path B).
the two possible pathways, efforts focused on the identification
of key intermediates such as [35/36/39] and determination of
their C1/C2-heteroatom connectivity. To this end, both the 13C1-
and the 13C2-labeled tri-O-benzyl-D-glucals (7a/b) were pre-
pared25 and employed as glycosyl donors in the C2-amidogly-
cosylation using 15N-labeled N-(TMS)benzamide as the amide
nucleophile (Schemes 6 and 7).
In the first series of low-temperature NMR experiments
(Scheme 6), 13C1-labeled tri-O-benzyl-D-glucal (7a) was treated
with thianthrene-5-oxide (11, 2 equiv) and Tf2O (2 equiv) at
-60 °C. Unfortunately, 13C NMR detection of a carbohydrate
species such as 44a at this temperature was not possible owing
to the presence of several distinct 13C1-resonances indicating a
complex mixture of intermediates. However, upon introduction
of 15N-labeled N-(TMS)benzamide (99% 15N-incorporation) with
N,N-diethylaniline and warming to -20 °C, a principal carbo-
hydrate species (tentatively assigned as either the imidate 45a,
the amide 46a, or the aziridine 47a) was detected, characterized
by its 13C1-resonance (δ ) 86.707 ppm), existing as a doublet
with a 15N-13C coupling value of 3.58 Hz. Upon warming of
the reaction to 20 °C, formation of the oxazoline 41a was
observed, revealing only a singlet 13C1-resonance at δ 100.759
ppm. The presence of a 13C coupling constant in the intermediate
[45a/46a/47a] suggests a structure that incorporates a 13C1-
15N linkage such as that in the glycosyl amide 46a or aziridine
47a. However, it is worth noting that the presence of the J13C-15N
coupling constant does not preclude the possibility of the
Detection and Characterization of Reactive Glycosyl
Intermediates. A key difference in these two pathways (A vs
B, Scheme 5) is the sequence of heteroatom transfer to both
C1 and C2 of the glycal donor. In Path A, a C1-N linkage is
formed prior to oxazoline formation; in Path B, a C1-O linkage
is formed prior to oxazoline formation. To distinguish between
(25) Shull, B. K.; Wu, Z.; Koreeda, M. J. Carbohydr. Chem. 1996, 15, 955-
964.
9
J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002 9793

A R T I C L E S
Liu and Gin
Scheme 7
intermediate generated prior to oxazoline formation. Efforts then
focused on the determination of C1-O connectivity in the
intermediate in question to distinguish between the structures
45 and 46. It is known that an 18O atom directly bound to carbon
induces a small but significant (0.01-0.05 ppm) upfield shift
in the 13C NMR signal relative to that of the 16O isotopomer.
This 18O-induced isotopic shift in 13C NMR was first predicted
by Jameson28 and later confirmed experimentally by Van Etten,29
Vederas,30 and Darensbourg.31 The technique has frequently
been used in kinetic studies as a continuous, nondestructive
assay method32 and has been employed to identify the point of
C-O bond cleavage in a series of hydrolysis reactions.33
To secure the presence/absence of a C1-O linkage in 45/
46, the activation of 13C1-tri-O-benzyl-D-glucal (7a) with 11‚
Tf2O, followed by introduction of 18O enriched N-(TMS)-
benzamide (47% 18O), was carried out (Scheme 8), monitored
by 13C NMR, and compared to the data obtained in the
analogous reaction with unlabeled N-(TMS)benzamide. When
47% 18O-labeled N-(TMS)benzamide was introduced to the
activated glycosyl species 44a, detection of the 13C1 signal of
the resulting intermediate in question revealed a species with
two 13C1-resonances (δ 86.726 ppm and δ 86.702 ppm) in
approximately a 53:47 ratio (Scheme 8), as opposed to the
appearance of only a single resonance when unlabeled N-(TMS)-
benzamide is employed. The upfield chemical shift perturbation
(∆δ ) 0.024 ppm) of 13C1 in this intermediate upon introduction
of 18O-enriched N-(TMS)benzamide is indicative of a C1-O
linkage, consistent with a glycosyl imidate structure 45a instead
of the glycosyl amide structure 46a.34 When the putative
intermediate 45a was allowed to warm to 20 °C to form
oxazoline 41a, a similar 13C1-18O chemical shift perturbation
16
18
(δC1( O) 100.740 ppm, δC1( O) 100.710 ppm, ∆δ ) 0.030 ppm)
was also observed as expected in the oxazoline 41a.
The analogous 18O-labeling experiment was subsequently
performed with 13C2-labeled tri-O-benzyl-D-glucal (7b, Scheme
9). The reaction of 7b with 11‚Tf2O and 18O-enriched N-(TMS)-
benzamide (47% 18O) revealed no 18O-induced isotopic 13C2-
shift perturbation in either the putative imidate intermediate 45b
or the oxazoline product 41b. This result is again consistent
with a nitrogen-transfer process originally outlined in Path B,
Scheme 5.
Having established the intermediacy of the glycosyl imidate
45 as a precursor to oxazoline formation and the likelihood of
the N-transfer process proceeding according to Path B, Scheme
5, it is crucial to establish the stereochemical configuration of
imidate 45 for the determination of facial selectivity in the initial
activation of glycal with 11‚Tf2O. Direct observation of the
glycosyl imidate structure 45a, in which a three-bond coupling
(3J13C-15N ) 3.58 Hz) could give rise to the observed doublet
resonance. Indeed, three-bond 15N-13C couplings of comparable
magnitude in imidate functionalities are not uncommon,26
wherein the magnitude of the coupling depends on orientation
of the nitrogen lone pair as well as on the conformation of the
substrate in question.27
Although this 13C1 NMR data does not allow for clear
distinction between the structures of 45, 46, and 47 as the initial
intermediate prior to oxazoline formation, further information
on its identity was obtained by performing an analogous 13C
NMR experiment with 13C2-labeled tri-O-benzyl-D-glucal (7b,
Scheme 7). Activation of 7b with 11‚Tf2O followed by
introduction of 15N-(TMS)benzamide led to the formation of
an intermediate with a singlet 13C2-resonance (δ 59.446 ppm),
prior to its conversion to the oxazoline 41b (δC2 66.043 ppm,
d, J ) 3.29 Hz) upon warming to 20 °C. The lack of 13C-15N
coupling in the C2 resonance of the intermediate in question
indicates the absence of a 13C2-15N linkage, thereby disqualify-
ing aziridine 47b as a viable structure for this species.
(28) Jameson, C. J. J. Chem. Phys. 1977, 66, 4983-4988.
(29) Risley, J. M.; Van Etten, R. L. J. Am. Chem. Soc. 1979, 101, 252-253.
(30) Vederas, J. C. J. Am. Chem. Soc. 1980, 102, 374-376.
(31) Darensbourg, D. J. J. Organomet. Chem. 1979, 174, C70-C76.
(32) (a) Mega, T. L.; Cortes, S.; Van Etten, R. L. J. Org. Chem. 1990, 55,
522-528. (b) Cortes, S. J.; Mega, T. L.; Van Etten, R. L. J. Org. Chem.
1991, 56, 943-947. (c) Risley, J. M.; Van Etten, R. L.; Uncuta, C.; Balaban,
A. T. J. Am. Chem. Soc. 1984, 106, 7836-7840. (d) Risley, J. M.; Van
Etten, R. L. J. Am. Chem. Soc. 1981, 103, 4389-4392.
(33) (a) Mega, T. L.; Van Etten, R. L. J. Am. Chem. Soc. 1988, 110, 6372-
6376. (b) Risley, J. M.; Kuo, F.; Van Etten, R. L. J. Am. Chem. Soc. 1983,
105, 1647-1652. (c) Parente, J. E.; Risley, J. M.; Van Etten, R. L. J. Am.
Chem. Soc. 1984, 106, 8156-8161. (d) Dessinges, A.; Castillon, S.; Olesker,
A.; Thang, T. T.; Lukacs, G. J. Am. Chem. Soc. 1984, 106, 450-451.
(34) The 18O isotope shifts at the â-carbon and γ-carbon have also been reported,
although magnitudes are much smaller (<0.01 ppm). (a) Moore, R. N.;
Diakur, J.; Nakashima, T. T.; McLaren, S. L.; Vederas, J. C. J. Chem.
Soc., Chem. Commun. 1981, 501-502. (b) Mega, T. L.; Van Etten, R. L.
J. Am. Chem. Soc. 1993, 115, 12056-12059.
On the basis of the previous 13C NMR investigation with 15N-
(TMS)benzamide, the glycosyl imidate 45 and the glycosyl
amide 46 emerge as possible structures for the initial glycosyl
(26) Levy, G. C.; Lichter, R. L. Nitrogen-15 Nuclear Magnetic Resonance
Spectroscopy; Wiley & Sons: New York, 1979; Chapter 4. For example,
in 15N-labeled methyl benzimidate, a three-bond 15N-13C coupling (3JC-N
) 1.9 Hz) between the nitrogen and methyl carbon is observed.
(27) Breitmaier, E.; Voelter, W. Carbon-13 NMR Spectroscopy: High-Resolution
Methods and Applications in Organic Chemistry and Biochemistry, 3rd
ed.; VCH publishers: New York, 1987; pp 155-160.
9
9794 J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002

C2-Amidoglycosylation
A R T I C L E S
Scheme 9
3
1
Figure 1. JHH and JCH values for d21-45.
Scheme 8
H4 (3JH5-H4 ) 9.7 Hz) as well as H4 and H3 (3JH4-H3 ) 9.2
Hz), which are consistent with H3, H4, and H5 existing in
mutually trans-axial orientations. The relatively small coupling
constant between H3 and H2 (3JH3-H2 ) 3.6 Hz) is consistent
with an axial-equatorial coupling magnitude, suggesting an
equatorial orientation of H2. Although H1 is a broad singlet in
1
the H NMR spectrum at -20 °C, the multiplet pattern of H2
1
1
(dd) as well as H-homodecoupling and H-COSY data reveal
a small coupling between H1 and H2 (3JH2-H1 ) 1.9 Hz),
consistent with diequatorial orientation of both H1 and H2. The
orientation of H1 was further supported by the relatively large
anomeric 1H-13C coupling constant (1JC1-H1 ) 176 Hz),
consistent with an equatorial anomeric proton of a pyranoside
species in a 4C1 conformation.36 On the basis of these data, it is
likely that the structure of imidate 45 incorporates substituents
at C3, C4, and C5 in equatorial positions, with both the imidate
and the sulfonium substituents at C1 and C2, respectively, in
axial orientations, a structure consistent with initial axial (â)
approach of 11‚Tf2O onto the glycal donor (i.e., 1 f 33, Scheme
5).
3JH-H coupling constants in the pyranoside ring of imidate 45
at low temperature was not possible because of their resonances
overlapping with those of benzylic protons in the 1H NMR data,
3
rendering most of the JH-H coupling constants unresolvable.
Therefore, tri-O-d7-benzyl-D-glucal (d21-7) was prepared and
used to generate imidate d21-45 (Figure 1) so that all of the
pyranoside protons could be discerned by 1H NMR at -20 °C.35
The reaction of d21-7 to form d21-45 at -20 °C according to
the above-mentioned protocol allowed for the observation of
relatively large coupling constants (Figure 1) between H5 and
(35) The assignment of proton signals and coupling constants was confirmed
(36) Duus, J. Ø.; Gotfredsen, C. H.; Bock, K. Chem. ReV. 2000, 100, 4589-
by both 1H-COSY and 1H-homodecoupling experiments.
4614.
9
J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002 9795

A R T I C L E S
Liu and Gin
1
Taken collectively, all of the 13C and H NMR data thus
obtained are consistent with the reaction outlined in Scheme 5,
Path B for C2-amidoglycosylation, a process that involves: (1)
activation of the glycal donor by the 11‚Tf2O from the top (â)
face to give C2-sulfonium intermediate 33; (2) nucleophilic
attack of the amide oxygen at the C1 position of 33 to afford
C2-sulfonium glycosyl imidate 39; (3) intramolecular displace-
ment of thianthrene from C2 by imidate nitrogen at the C2-
position to provide oxazoline 37; and (4) oxazoline ring opening
by glycosyl acceptor under acidic conditions to yield the final
2-N-acylamino-2-deoxy-â-pyranoside 38.
(neat film): 3274, 2932, 2867, 1732, 1652, 1566, 1453, 1371, 1316,
1285, 1170, 1115, 1078, 1029, 985, 745, 697 cm-1. HRMS (FAB)+
m/z calcd for C54H81NO7Na (M + Na)+, 878.5911; found, 878.5914.
Method (B): Benzyl O-(2-N-Acetylamino-2-deoxy-3,4,6-tri-O-
benzyl-â-D-glucopyranosyl)-(1-4)-2,3-O-cyclohexylidene-r-L-rham-
nopyranoside (16). Trifluoromethanesulfonic anhydride (24 µL, 0.144
mmol, 2.0 equiv) was added to a solution of 3,4,6-tri-O-benzyl-D-glucal
(30 mg, 0.072 mmol, 1.0 equiv) and thianthrene-5-oxide (33.5 mg, 0.144
mmol, 2.0 equiv) in a mixture of chloroform and dichloromethane (2.4
mL; 3:1 v/v) at -78 °C. The reaction mixture was stirred at this
temperature for 12 min, then N,N-diethylaniline (45 µL, 0.29 mmol,
4.0 equiv) was added, followed by N-(TMS)acetamide (28 mg, 0.22
mmol, 3.0 equiv). The mixture was immediately warmed to 23 °C and
was stirred at this temperature for 40 min. A solution of benzyl 2,3-
O-cyclohexylidene-R-L-rhamnopyranoside (72 mg, 0.22 mmol, 3.0
equiv) in 1.0 mL of dichloromethane was added via cannula. Cam-
phorsulfonic acid (33.5 mg, 0.144 mmol, 2.0 equiv) was then added,
and the reaction was stirred for 40 h. The mixture was diluted with 40
mL of dichloromethane and washed with 40 mL of saturated aqueous
sodium bicarbonate solution and 40 mL of saturated aqueous sodium
chloride solution. The organic layer was dried over MgSO4 and
concentrated, and the residue was purified by silica gel flash column
chromatography (94:6 dichloromethane/ethyl acetate) to afford the
product (31 mg, 53% yield) as a white solid (mp 153-154 °C). Rf )
Conclusion
A one-pot C2-amidoglycosylation reaction for the synthesis
of 2-N-acylamino-2-deoxy-â-pyranosides from glycals was
developed. Both the C2-nitrogen transfer and the glycosidic bond
formation proceed stereoselectively, allowing for the introduc-
tion of both natural and nonnatural amide functionalities at C2
with concomitant anomeric bond formation in a one-pot
procedure. Tracking of the reaction by low-temperature NMR
spectroscopy employing 15N- and 18O-isotope labels suggests a
mechanism involving the formation of the C2-sulfonium gly-
cosyl imidate 39 as well as oxazoline 37 as key intermediates
in this oxidative glycosylation with glycals. These studies
provide critical insights into the pathway of nitrogen transfer
and should facilitate the development of this method for the
nitrogen transfer to electron-rich olefins in general.
25
0.07 (94:6 dichloromethane/ethyl acetate). [R]D ) -13.8 (c ) 0.55,
CHCl3). 1H NMR (500 MHz, CDCl3): δ 7.38-7.27 (m, 18H), 7.25-
7.23 (m, 2H), 5.52 (d, 1H, J ) 8.6 Hz), 5.05 (s, 1H), 4.83 (d, 1H, J )
11.4 Hz), 4.82 (d, 1H, J ) 10.9 Hz), 4.71 (d, 1H, J ) 8.5 Hz), 4.69 (d,
1H, J ) 11.1 Hz), 4.68 (d, 1H, J ) 12.2 Hz), 4.63 (d, 1H, J ) 12.2
Hz), 4.62 (d, 1H, J ) 10.8 Hz), 4.56 (d, 1H, J ) 12.2 Hz), 4.48 (d,
1H, J ) 11.7 Hz), 4.12 (d, 1H, J ) 4.9 Hz), 4.07 (dd, 1H, J ) 7.4, 5.7
Hz), 3.86 (dt, 1H, J ) 9.8, 8.5 Hz), 3.76 (dd, 1H, J ) 11.2, 4.3 Hz),
3.73 (dd, 1H, J ) 12.0, 6.2 Hz), 3.74-3.68 (m, 2H), 3.64 (dd, 1H, J
) 9.9, 8.7 Hz), 3.46 (dd, 1H, J ) 10.0, 7.4 Hz), 3.44 (ddd, 1H, J )
9.4, 4.2, 2.2 Hz), 1.88 (s, 3H), 1.70 (q, 1H, J ) 8.3 Hz), 1.65-1.47
(m, 7H), 1.41-1.36 (m, 2H), 1.31 (d, 3H, J ) 6.3 Hz). 13C NMR (101
MHz, CDCl3): 170.2, 138.3, 138.2, 138.0, 137.0, 128.5, 128.4, 128.3,
128.3, 128.1, 128.0, 127.9, 127.8, 127.7, 127.6, 127.5, 110.0, 101.7,
96.0, 82.3, 81.7, 78.2, 77.9, 75.7, 75.4, 74.8, 74.4, 73.5, 69.1, 68.8,
64.4, 55.6, 37.8, 35.6, 28.1, 24.9, 24.1, 23.6, 23.5, 17.4. FTIR (neat
film): 3272, 3030, 2934, 1655, 1562, 1496, 1452, 1370, 1310, 1099,
1067, 1027, 936, 735, 697 cm-1. HRMS (FAB)+ m/z calcd for C48H57-
NO10Na (M + Na)+, 830.3880; found, 830.3882.
Experimental Section
Typical Procedure for C2-Amidoglycosylation. Method (A): (+)-
3-Dihydrocholesterol 3,4-Di-O-benzyl-2-N-acetylamino-2-deoxy-6-
O-dimethylpropanoyl-â-D-glucopyranoside (14). Trifluoromethane-
sulfonic anhydride (28 µL, 0.17 mmol, 2.0 equiv) was added to a
solution of 3,4-di-O-benzyl-6-O-dimethylpropanoyl-D-glucal (34 mg,
0.083 mmol, 1.0 equiv) and thianthrene-5-oxide (38.5 mg, 0.17 mmol,
2.0 equiv) in a mixture of chloroform and dichloromethane (2.4 mL;
3:1 v/v) at -78 °C. The reaction mixture was stirred at this temperature
for 11 min, then N,N-diethylaniline (53 µL, 0.33 mmol, 4.0 equiv) was
added, followed by N-(TMS)acetamide (33 mg, 0.25 mmol, 3.0 equiv).
The mixture was immediately warmed to 23 °C and stirred at this
temperature for 30 min. A solution of (+)-3-dihydrocholesterol (97
mg, 0.25 mmol, 3.0 equiv) in 1.0 mL of dichloromethane was added
via cannula. Amberlyst-15 acidic resin (66 mg) was then added, and
the reaction was stirred for 30 h. The mixture was filtered and
concentrated under vacuum, and the residue was purified by silica gel
flash column chromatography (4:1 hexane/ethyl acetate) to afford the
product (52 mg, 73% yield) as a white solid (mp 207-208 °C). Rf )
0.12 (4:1 hexane/ethyl acetate). [R]D25 ) +45.4 (c ) 0.65, CHCl3). 1H
NMR (500 MHz, CDCl3): δ 7.35-7.28 (m, 10H), 5.59 (d, 1H, J )
7.5 Hz), 5.02 (d, 1H, J ) 7.8 Hz), 4.83 (d, 1H, J ) 10.0 Hz), 4.82 (d,
1H, J ) 11.7 Hz), 4.66 (d, 1H, J ) 11.3 Hz), 4.59 (d, 1H, J ) 11.1
Hz), 4.40 (dd, 1H, J ) 11.7, 2.4 Hz), 4.28 (dd, 1H, J ) 9.8, 8.7 Hz),
4.14 (dd, 1H, J ) 12.0, 6.6 Hz), 3.65 (ddd, 1H, J ) 9.4, 6.7, 2.5 Hz),
3.53 (tt, 1H, J ) 11.1, 5.0 Hz), 3.44 (t, 1H, J ) 9.4 Hz), 3.16 (dt, 1H,
J ) 9.8, 7.8 Hz), 1.95 (dt, 1H, J ) 12.6, 3.4 Hz), 1.86 (s, 3H), 1.82-
1.77 (m, 2H), 1.70-1.61 (m, 2H), 1.57-1.40 (m, 5H), 1.37-0.95 (m,
20H), 1.20 (s, 9H), 0.89 (d, 3H, J ) 6.5 Hz), 0.86 (d, 3H, J ) 6.6 Hz),
0.85 (d, 3H, J ) 6.7 Hz), 0.76 (s, 3H), 0.63 (s, 3H), 0.58 (td, 1H, J )
11.5, 4.2 Hz). 13C NMR (126 MHz, CDCl3): 178.2, 170.5, 138.3, 137.6,
129.2, 128.6, 128.5, 128.5, 128.0, 127.9, 127.9, 127.8, 98.0, 80.2, 79.3,
79.1, 74.9, 74.7, 72.8, 63.4, 58.1, 56.4, 56.2, 54.3, 44.7, 44.3, 42.5,
40.0, 38.8, 37.0, 36.1, 35.8, 35.5, 35.4, 34.5, 32.0, 29.4, 28.7, 28.2,
28.0, 27.2, 24.2, 23.8, 23.6, 22.8, 22.5, 21.2, 18.6, 12.2, 12.0. FTIR
Method (C): Azido 3,4,6-Tri-O-benzyl-2-N-acetylamino-2-deoxy-
â-D-glucopyranoside (19). Trifluoromethanesulfonic anhydride (40 µL,
0.24 mmol, 2.0 equiv) was added to a solution of 3,4,6-tri-O-benzyl-
D-glucal (50 mg, 0.12 mmol, 1.0 equiv) and thianthrene-5-oxide (56
mg, 0.24 mmol, 2.0 equiv) in a mixture of chloroform and dichlo-
romethane (2.4 mL; 3:1 v/v) at -78 °C. The reaction mixture was stirred
at this temperature for 12 min, then N,N-diethylaniline (76 µL, 0.48
mmol, 4.0 equiv) was added, followed by N-(TMS)acetamide (47 mg,
0.36 mmol, 3.0 equiv). The mixture was immediately warmed to 23
°C and stirred at this temperature for 1 h. The solvent was quickly
removed under vacuum, and N,N-dimethylformamide (2.0 mL), sodium
azide (78 mg, 1.20 mmol, 10.0 equiv), and copper(II) trifluoromethane-
sulfonate (87 mg, 0.24 mmol, 2.0 equiv) were added, and the reaction
was stirred for 48 h. The reaction mixture was diluted with water (80
mL) and extracted four times with diethyl ether (20 mL). The combined
ether extract was then washed with saturated aqueous sodium chloride
solution (30 mL) and dried over sodium sulfate. The solution was
filtered and concentrated under vacuum, and the residue was purified
by silica gel flash column chromatography (3:1 benzene/ethyl acetate)
to afford the product (44 mg, 71% yield) as a white solid (mp 168-
25
169 °C). Rf ) 0.17 (3:1 benzene/ethyl acetate). [R]D ) +6.9 (c )
1
0.75, CHCl3). H NMR (500 MHz, CDCl3): δ 7.36-7.27 (m, 13H),
7.21-7.18 (m, 2H), 5.31 (d, 1H, J ) 7.7 Hz), 4.97 (d, 1H, J ) 8.9
9
9796 J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002

C2-Amidoglycosylation
A R T I C L E S
Hz), 4.84 (d, 1H, J ) 11.7 Hz), 4.79 (d, 1H, J ) 10.8 Hz), 4.64 (d,
1H, J ) 12.2 Hz), 4.62 (d, 1H, J ) 11.4 Hz), 4.59 (d, 1H, J ) 10.9
Hz), 4.56 (d, 1H, J ) 12.1 Hz), 3.99 (dd, 1H, J ) 10.1, 8.7 Hz), 3.77
(dd, 1H, J ) 11.0, 1.3 Hz), 3.74 (dd, 1H, J ) 11.0, 4.9 Hz), 3.68 (t,
1H, J ) 8.5 Hz), 3.61 (dt, 1H, J ) 9.6, 3.4 Hz), 3.40 (dt, 1H, J ) 9.4,
8.7 Hz), 1.84 (s, 3H). 13C NMR (126 MHz, CDCl3): 170.6, 138.2,
127.9, 137.8, 128.5, 128.4, 128.4, 128.0, 127.9, 127.8, 127.8, 127.7,
87.9, 80.4, 78.1, 77.0, 74.8, 74.7, 73.5, 68.3, 56.1, 23.4. FTIR (neat
film): 3261, 3088, 3031, 2915, 2870, 2113, 1651, 1556, 1496, 1452,
1375, 1319, 1247, 1113, 1087, 1072, 1027, 957, 749, 694 cm-1. HRMS
(FAB)+ m/z calcd for C29H32N4O5Na (M + Na)+, 539.2270; found,
539.2271. The corresponding manno-diastereomer, azido 3,4,6-tri-O-
benzyl-2-N-acetylamino-2-deoxy-R-D-mannopyranoside, was also iso-
lated in 6% yield (4 mg).
mg, 0.044 mmol, 2.0 equiv) in 0.7 mL of CDCl3 in a 10 mL conical
shape Schlenk flask at -60 °C. The reaction mixture was stirred at
this temperature for 10 min, then N,N-diethylaniline (14.0 µL, 0.088
mmol, 4.0 equiv) was added, followed by N-(TMS)benzamide (12.7
mg, 0.066 mmol, 3.0 equiv). The reaction mixture was stirred at -60
°C for 4 min and then transferred via cannula to a standard 5 mm NMR
tube cooled to -45 °C under argon atmosphere. The contents were
then briefly mixed at -45 °C by using a Fisher Vortex Genie 2
1
instrument. The reaction was gradually warmed to -20 °C, and H,
1
1H-COSY, and H-homodecoupling NMR data were acquired. The
1
reaction was then further slowly warmed to 23 °C during which H
NMR data were acquired at -15, -5, and 20 °C, respectively.
Acknowledgment. This research was supported by the
National Institute of Health (GM-58833) and Johnson &
Johnson. D.Y.G. is a Cottrell Scholar of Research Corporation.
NMR spectra were obtained in the Varian Oxford Instrument
Center for Excellence in NMR Laboratory, funded in part by
the W. M. Keck Foundation, the National Institutes of Health
(PHS 1 S10 RR10444), and the National Science Foundation
(NSF CHE 96-10502). The assistance of Dr. Vera Mainz of
the VOICE NMR Lab at the University of Illinois is gratefully
recognized. We thank Professor Anthony Serianni and Omicron
Biochemicals, Inc. for generous gifts of 13C1- and 13C2-labeled
D-glucose, and Mr. Larry Huffman, Jr. for the preparation of
13 and 17.
Representative Procedures for NMR Detection of Glycosyl
Intermediates. Method (A). A solution of C1-13C-tri-O-benzyl-D-glucal
(10.0 mg, 0.024 mmol, 1.0 equiv) and thianthrene-5-oxide (11.1 mg,
0.048 mmol, 2.0 equiv) in 0.6 mL of CDCl3 was transferred via cannula
to a standard 5 mm NMR tube under argon atmosphere. The reaction
was cooled to -60 °C, and 1H, 13C NMR data were acquired.
Trifluoromethanesulfonic anhydride (8.1 µL, 0.048 mmol, 2.0 equiv)
was added to the reaction at -60 °C, and the contents were mixed by
using a Fisher Vortex Genie 2 instrument. 1H and 13C NMR data were
acquired at -60 °C. N,N-Diethylaniline (15.2 µL, 0.096 mmol, 4.0
1
equiv) was then added, briefly mixed at -60 °C, and H, 13C NMR
data were acquired. A solution of 15N-N-(TMS)benzamide (14.0 mg,
0.072 mmol, 3.0 equiv) in 0.1 mL of CDCl3 was then added via syringe,
and the reaction was briefly mixed. 1H and 13C NMR data were acquired
1
at -60 °C. The reaction was gradually warmed, and H, 13C NMR
Supporting Information Available: Experimental procedures
and spectra (PDF). This material is available free of charge via
the Internet at http://pubs.acs.org.
data were acquired at -45, -20, -5, and 20 °C.
Method (B). Trifluoromethanesulfonic anhydride (7.4 µL, 0.044
mmol, 2.0 equiv) was added to a solution of 3,4,6-tri-O-d7-benzyl-D-
glucal (9.6 mg, 0.022 mmol, 1.0 equiv) and thianthrene-5-oxide (10.2
JA026281N
9
J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002 9797