Stereoselective synthesis of α- And β-glycofuranosyl amides by traceless ligation of glycofuranosyl azides

Article abstract of DOI:10.1002/chem.201200309

A highly stereoselective synthesis of α- or β-glycofuranosyl amides based on the traceless Staudinger ligation of glycofuranosyl azides of the galacto, ribo, and arabino series with 2-diphenylphosphanyl-phenyl esters has been developed. Both α- and β-isomers can be obtained with excellent selectivity from a common, easily available precursor. The process does not depend on the anomeric configuration of the starting azide but appears to be controlled by the C2 configuration and by the protection/deprotection state of the substrates. A mechanistic interpretation of the results, supported by ³¹P NMR experiments, is offered and merged with our previous mechanistic analysis of pyranosyl azide ligation reactions. Copyright

Full text of DOI:10.1002/chem.201200309

FULL PAPER
DOI: 10.1002/chem.201200309
Stereoselective Synthesis of a- and b-Glycofuranosyl Amides by Traceless
Ligation of Glycofuranosyl Azides
[
a]
Filippo Nisic, Gaetano Speciale, and Anna Bernardi*
Abstract: A highly stereoselective syn-
thesis of a- or b-glycofuranosyl amides
based on the traceless Staudinger liga-
tion of glycofuranosyl azides of the gal-
acto, ribo, and arabino series with 2-di-
phenylphosphanyl-phenyl esters has
been developed. Both a- and b-isomers
can be obtained with excellent selectiv-
ity from a common, easily available
precursor. The process does not
depend on the anomeric configuration
of the starting azide but appears to be
controlled by the C2 configuration and
by the protection/deprotection state of
the substrates. A mechanistic interpre-
tation of the results, supported by
3
1
P NMR experiments, is offered and
Keywords: azides · carbohydrates ·
merged with our previous mechanistic
analysis of pyranosyl azide ligation re-
actions.
glycomimetics
·
neo-glycoconju-
gates · Staudinger ligation · stereo-
selective synthesis
Introduction
The most widely employed method for the synthesis of
glycosyl amides is the condensation of carboxylic acid deri-
ACHTUNGTRENNUNG
[16]
N-Glycosyl amides and their structural mimics are currently
under intense scrutiny as potential effectors of carbohy-
drate-binding proteins, useful both as potential drugs and
for clarifying the biological roles of oligosaccharides.
major synthetic effort is under way to synthesize so-called
neo-glycoconjugates”, in which a sugar ring is connected
through a nitrogen atom to a carbon chain, often to a natural
[17–20]
azides.
Because glycopyranosylamines rapidly equili-
[
1]
A
brate to the most stable b-anomer, all of the approaches
that make use of isolated amine intermediates afford b-gly-
copyranosyl amides. An alternative methodology attempts
to avoid anomeric equilibration by reducing glycosyl azides
in the presence of acylating agents, most often employing
phosphines for the reduction step (Staudinger reduc-
“
[2–11]
amino acid, using an unnatural linkage.
Natural N-
[12]
linked glycopeptides are almost invariably b-linked. The
unnatural, a-linked isomers have attracted some attention
as glycopeptide mimics, since they may be stable to hydro-
lytic enzymes and may be used for in vivo applications. It
has been shown that the stereochemistry of the carbohy-
drate–peptide linkage has critical and unique conformation-
[21–26]
tion).
However, Staudinger intermediates (aza-ylides or
iminophosphoranes) are also subject to anomeric isomeriza-
tion, which for glycopyranoses is biased toward the b-anom-
ers. Thus, while b-glycosyl amides can be prepared, anomeri-
zation remains a significant problem in the synthesis of the
a-anomers.
[11]
[13]
al effects on glycopeptide structures.
Additionally, we
have shown that a-N-linked glycopyranosyl amides in the
gluco, galacto, and fuco series maintain the normal pyranose
conformation of the monosaccharide, which represents an
A limited number of methods are available for the syn-
thesis of a-glycosyl amides. Most of them require two steps
and have been described for a limited number of sub-
[11a]
[10,27,28]
important feature for their use as sugar mimics.
Indeed,
strates.
Our group has reported a method based on
[11a,28,29]
a small group of a-fucosyl amides have shown good affinity
the traceless Staudinger ligation of glycosyl azides,
for the carbohydrate recognition domain (CRD) of fucose-
binding proteins, such as DC-SIGN and PA-II lectin.
using functionalized phosphines 1 (2-diphenylphosphanyl-
[14]
[15]
[30]
phenyl esters) originally described by Bertozzi and Kies-
[31]
sling.
The phosphines used are modified to include an
acylating agent, which in 1 is a phenolic ester. Reduction of
the starting azide is followed by fast intramolecular trapping
of the reduction intermediates, which results in direct forma-
tion of an amide bond. We have shown that, depending on
the sugar protecting groups, this approach can prevent epi-
merization of glycopyranoses and allows retention of config-
[
a] Dr. F. Nisic, G. Speciale, Prof. Dr. A. Bernardi
Universitꢀ degli Studi di Milano
Dipartimento di Chimica Organica e Industriale,
via Venezian 21, 20133 Milano (Italy)
E-mail: anna.bernardi@unimi.it
[11,28]
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201200309: Synthesis and char-
acterization of furanosyl azides 9–14 and of b-ribo- and b-arabino-
pyranosyl azides 18–21; configurational assignment of the ribo- and
arabino-furanosyl amides 22a–25a; characterization of oxazoline 48;
uration at the anomeric carbon (Scheme 1).
In particu-
lar, we have identified reaction conditions that allow the
clean and stereoconservative ligation of unprotected glyco-
pyranosyl azides in good to moderate yields, affording a- or
b-glycopyranosyl amides depending on the configuration of
1
13
H and C NMR spectra of all new compounds.
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

Scheme 1. Staudinger ligation of gluco-pyranosyl azides with 1a (from
ref. [11a]).
Scheme 2. Stereoselective synthesis of a- and b-Galf amides 7a–g and
8a–f (from ref. [39]). Reagents and conditions: a) 708C, 4 h, 98:2 N,N-di-
methylacetamide/1,3-dimethyltetrahydro-2(1H)pyrimidinone
DMPU); b) H O, 2 h, 70 8C; c) MeO Na (0.05m), MeOH.
2
(DMA/
À
+
[29]
the starting material (Scheme 1).
Starting from a-glyco-
pyranosyl azides, a by-product of this reaction is the glyco-
furanosyl isomer derived from a ring-contraction process.
For instance, treatment of a-glucopyranosyl azide 2 with
phosphine 1a afforded a-N-valeroyl-glucopyranosyl amide
at odds with those obtained with a-pyranosyl azides, which
undergo ligation without inversion of configuration when
deprotected and anomerize to b-amides when acetylated
(Scheme 1). Therefore, after our preliminary report, we fo-
cused on examining the full scope of this reaction and on
understanding how its stereochemical course is determined.
Our results are reported herein. We first ascertained that
the stereochemical outcome of the process does not depend
on the anomeric configuration of the starting azide by syn-
thesizing the (previously unknown) a-epimer of 5 and study-
ing its ligation with 1a. We then examined the ligation of
ribo- and arabino-furanosyl azides, which allowed us to es-
tablish that the stereochemistry of the reaction is controlled
by the C2 configuration of the sugar. Finally, based on these
3
a and a-N-valeroyl-glucofuranosyl amide 4a. Their ratio
depended mainly on the reaction temperature, with the
amount of the furanose form increasing at higher tempera-
[29]
tures. Although the amount of furanose product could be
kept below 5–10%, this finding drew our attention towards
this atypical class of compounds. In particular, the furanose
[32]
form of d-glucose is known to be unstable and it is rarely
[33]
[34]
isolated, usually as a bis-acetonide or a boronate.
On the other hand, other hexofuranoses play important
[35]
roles in biological systems. For instance, galactofuranosyl
Galf) residues are constituents of the galactan polymer in
(
3
1
mycobacterial cell walls. The Galf units are created by bac-
terial mutases, which convert UDP-galactopyranose into
UDP-galactofuranose, and are essential for the viability of
results and on P NMR spectroscopic data, we suggest
a mechanistic interpretation of the process and merge it
with our previous mechanistic analysis of pyranosyl azide li-
gation reactions.
[36]
mycobacteria. Mammals do not synthesize hexofuranoses,
and therefore molecules of this class are emerging as poten-
tial candidates for therapeutic development. UDP-galacto-
pyranose mutase inhibitors have been reported and shown
Results
[37]
to block the growth of Mycobacterium tuberculosis.
Simple galactofuranosyl and arabinofuranosyl derivatives at
low micromolar concentrations have been shown to inhibit
bacterial growth by a mechanism that has not yet been fully
Synthesis of starting materials: The structures of the glyco-
furanosyl azides 5, 6, and 9–14 used in this work are collect-
ed in Figure 1. The 1,2-trans-furanosyl azides in the galacto
[38]
[40]
elucidated. We have initially focused our attention on gal-
actofuranosyl amides and have recently reported prelimina-
ry results showing that a- or b-Galf amides 7a–g and 8a–f
can be synthesized with high stereoselectivity by traceless
Staudinger ligation starting from unprotected b-galactofura-
nosyl azide 5 or tetra-O-acetyl-b-galactofuranosyl azide 6,
(6), ribo (10 ), and arabino (12) series were synthesized
[41,42]
starting from the corresponding acetates
under Paul-
senꢂs conditions (TMSN /SnCl ). The anomeric configura-
3
4
tion of the compounds was confirmed by NOE experiments
(CDCl ), which showed correlations of H1 with H3 in 5, H4
3
in 10, and H5 in 12. Zemplen deacetylation (cat. MeONa in
MeOH) afforded the unprotected b-galactofuranosyl azide
5, b-ribofuranosyl azide 9, and a-arabinofuranosyl 11 with-
out loss of configurational integrity.
To the best of our knowledge, 1,2-cis a-galactofuranosyl
azides in unprotected (13) and tetra-O-acetylated form (14)
have never been described before. They were prepared from
[39]
respectively (Scheme 2).
These results offer a simple, stereodivergent way of syn-
thesizing glycofuranosyl amides of either anomeric configu-
ration starting from the same azide and controlling the pro-
cess through the absence/presence of acetyl protecting
groups. At first sight, however, these results are intriguingly
&
2
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Synthesis of a- and b-Glycofuranosyl Amides by Staudinger Ligation
FULL PAPER
value of the vicinal coupling constant of the anomeric
proton (J1,2 =8 Hz) and their pyranose structure by the
chemical shift for C4 (d=67.8 ppm).
2
-Diphenylphosphanyl-phenyl esters 1 were prepared as
[11,39]
previously described
nophenol.
by acylation of o-diphenylphosphi-
[46]
Figure 1. Furanosyl azides used in this work.
Ligation reactions: All ligation reactions were performed
using the protocol reported for b-galactofuranosyl azides 5
and 6 in our preliminary communication. In brief, a mix-
ture of unprotected azide and phosphine 1 (Scheme 2) in
DMA/DMPU (98:2) was stirred
1
,2,3,5,6-penta-O-tert-butyldimethylsilyl-b-galactofuranose
[39]
15 (Scheme 3), which was recently described by Baldoni and
[43]
[43]
Marino.
This was transformed into iodide 16
by re-
for 4 h at 708C, then water was
added and, after an additional
2
h at the same temperature,
the solvent was evaporated and
the product was isolated by ex-
traction with water. Diastereo-
meric ratios were determined
Scheme 3. Synthesis of a-d-Galf azides 13 and 14 from galactofuranosyl iodide 16. Reagents and conditions: for the crude products by
a) TMSI, CH Cl , 08C, 30 min (see ref. [42]); nBu NN , diisopropylethylamine (DIPEA); c) TBAF; d) Ac O/
pyridine. TBS=tert-butyldimethylsilyl.
1
2
2
4
3
2
H NMR (CD OD) and the re-
3
sulting amides were purified by
flash chromatography. The
same conditions were adopted
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
action with trimethylsilyl iodide (TMSI; 1.2 equiv) in anhy-
for the ligation of tetra-O-acetyl azides. Again, the crude re-
1
drous CH Cl at 08C for 30 min. When TLC showed com-
action mixtures were analyzed by H NMR (CDCl ) to es-
2
2
3
plete consumption of 15, tetrabutylammonium azide
nBu NN ) was added and the mixture was stirred overnight
tablish the isomer distribution, but, to simplify purification
of the reaction products from phosphine oxide, the tetra-O-
acetyl amides were not isolated but were deacetylated to
afford the corresponding unprotected amides. The latter
were finally isolated and purified by flash chromatography
on silica gel. The deprotection step was found to be critical
to preserve configurational integrity at the anomeric carbon.
While a 0.05m solution of MeONa in MeOH and short con-
tact times (45 min) could be used successfully, more concen-
trated (0.1m) solutions caused anomeric epimerization. This
process most probably occurs by MeONa-induced deproto-
nation of the amide nitrogen (see below) and can be sup-
pressed by using a milder base for the deprotection. No epi-
(
4
3
at room temperature. The reaction afforded the 2,3,4,6-
tetra-O-tert-butyldimethylsilyl-a-galactofuranosyl azide 17 in
good yield and with high stereoselectivity, the anomeric con-
figuration of which was confirmed by the observation of
a NOESY correlation between H1 and H4 (Scheme 3). Re-
moval of the TBS groups from 17 was accomplished with
tetrabutylammonium fluoride (TBAF) in THF and afforded
unprotected a-galactofuranosyl azide 13. The anomeric con-
figuration of 13 was also confirmed by NOESY experiments,
which showed a clear correlation between protons H1 and
H4, and by the chemical shift for C4 of 84.5 ppm, which is
[47]
diagnostic of the furanose form. Acetylation of 13 (Ac O in
merization was observed using MeNH in EtOH.
2
2
pyridine, Scheme 3) afforded 2,3,5,6-tetra-O-acetyl-a-galac-
tofuranosyl azide 14. A NOESY contact between H1 and
The reactions of b-galactofuranosyl azides 5 and 6 were
studied first (Scheme 2). The structures and anomeric con-
figurations of 7a and 8a were determined by NMR, as de-
scribed above for the starting azides and fully detailed in
H4 (CDCl ) was also observed in this compound.
3
For comparison purposes and to fully characterize the li-
[39]
gation products, the corresponding b-d-ribopyranosyl
the preliminary communication.
In all cases, b-anomers
[44]
[45]
azides
18 and 19 and b-l-arabinopyranosyl azides
20
were obtained from tetra-O-acetyl azide 6, with retention of
the anomeric configuration, whereas a-anomers were ob-
tained from unprotected azide 5, which appears to undergo
inversion of its anomeric configuration during the reaction.
This anomeric inversion must derive from a ring-opening
process after the azide reduction step. We will discuss our
mechanistic hypothesis for this intriguing process in the fol-
lowing section. For the moment, we note only that ring-
opening should actually be favored for the O-acetylated sub-
strate, thus we initially found these results rather puzzling.
In an effort to better correlate the steric course of the re-
action with the structure of the starting azides, we submitted
and 21 (Figure 2) were also synthesized (from the corre-
sponding acetates under Paulsenꢂs conditions) and reacted
with 1. The b-configuration of 18–21 was supported by the
Figure 2. Pentapyranosyl azides (b-d-ribopyranosyl azides 18 and 19 and
b-l-arabinopyranosyl azides 20 and 21) used for comparison purposes.
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

A. Bernardi et al.
tion, whereas O-acetyl-glycofu-
AHCTUNGTRENNUNG
1
ers.
Reaction mechanism
Unprotected furanosyl azides:
The ligation of unprotected 1,2-
trans furanosyl azides 5, 9, and
1
1 with phosphine 1a afforded
glycofuranosyl amides with the
opposite 1,2-cis anomeric con-
figuration relative to the start-
Scheme 4. Ligation of a-galactofuranosyl azides 13 and 14 with phosphine 1a. Reagents and conditions:
a) 708C, 4 h, DMA/DMPU 98:2; b) H O, 2 h, 70 8C; c) MeO Na (0.05m), MeOH.
2
À
+
the a-galactofuranosyl azides 13 and 14 to Staudinger liga-
tion with 1a (Scheme 4). Surprisingly, the unprotected azide
13 underwent a stereoconservative process yielding the a-
amide 7a, whereas inversion of the anomeric configuration
was observed for the tetra-O-acetyl substrate 14. In other
words, regardless of their anomeric configurations, both the
a- and b-unprotected galactofuranosyl azides 5 and 13 gave
a-amides (1,2-cis), whereas both the a- and b-tetra-O-
acetyl-galactofuranosyl azides 6 and 14 gave b-amides (1,2-
trans).
To further explore the scope and mechanism of this re-
ACHTUNGTRENNUNGa ction, the ribofuranosyl azides 9 and 10 and the arabinofur-
anosyl azides 11 and 12 (Scheme 5) were subjected to trace-
less ligation with compound 1a under the same conditions
as used above. All of these azides feature a 1,2-trans config-
uration, with opposite configuration at C2 in the ribo and
arabino series. These reactions also afforded 1,2-cis amides
(
1
22a and 24a) from the unprotected azides (9 and 11) and
,2-trans amides (23a and 25a) from the corresponding O-
[48]
acetyl compounds (10 and 12) (Scheme 5). Thus, these ex-
periments established that the stereochemical outcome of
the process is controlled by the C2 configuration, as well as
by the protection state of the monosaccharide.
To fully confirm the characterization of 22a–25a, the cor-
responding pyranosyl azides 18–21 were also subjected to li-
gation with 1a (Scheme 6). As expected for b-pyranosyl azi-
Scheme 5. Staudinger ligation of 1a with b-ribofuranosyl azides 9 and 10
and a-arabinofuranosyl azides 11 and 12. Reagents and conditions:
a) 708C, 4 h, DMA/DMPU 98:2; b) H O, 2 h, 70 8C; c) MeNH (0.046m),
2 2
[
11a]
des,
b-amides (26a and 27a) were obtained in all cases,
EtOH.
as established by the vicinal coupling constant of the anome-
ric protons (26a J1,2 =8 Hz; 27a J1,2 =8.8 Hz) and the chemi-
cal shift of C4 (d=68.8 ppm).
The studies detailed above allowed us to fully describe
the steric course of the ligation of glycofuranosyl azides and
to compare it with the available data for the same reaction
of the pyranose isomers. In the pyranose series (Scheme 1
and Scheme 6), b-azides always give b-amides irrespective
of the protection state of the sugar hydroxy groups; a-azides
undergo inversion of their anomeric configuration if O-ace-
tylated and preserve it if unprotected (Scheme 1). In the fu-
ACHTUNGTRENNUNGr anose series (Schemes 2, 4, and 5), irrespective of the
anomeric configuration of the starting material, unprotected
azides (5, 9, 11, and 13) afford amides of 1,2-cis configura-
Scheme 6. Staudinger ligation of b-d-ribo- and b-l-arabinopyranosyl
azides with 1a. Reagents and conditions: a) 1a, 708C, 4 h, DMA/DMPU
98:2; b) H O, 2 h, 70 8C; c) MeO Na (0.05m), MeOH.
2
À
+
&
4
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Synthesis of a- and b-Glycofuranosyl Amides by Staudinger Ligation
FULL PAPER
ing azide precursor (Schemes 2
and 5). This anomeric epimeri-
zation must derive from a ring-
opening process after the azide
reduction step.
mechanism for the reaction of
with b-galactofuranosyl
A
possible
1
a
azide 5 is shown in Scheme 7.
Upon azide reduction, imino-
phosphorane 28 (Scheme 7) is
formed, ring-opening of which
affords phosphinimine 29. From
this open-chain intermediate,
the ring-closure step could be
biased by the unprotected hy-
droxyl group at C2, which can
trap the phosphorus atom of 29
as the cyclic oxazaphospholane
intermediate 30, thus enforcing,
after ring-closure, formation of
the 1,2-cis isomer 31. In other
words, the anomeric carbon of
the sugar is blocked by chela-
tion of the phosphorus in inter-
mediate 30, such that only a- Scheme 7. Proposed mechanism for the formation of 1,2-cis-amides from unprotected glycofuranosyl azides.
amides can be obtained upon
ring-closure. Intramolecular acyl transfer then affords 32,
which is finally hydrolyzed to the 1,2-cis amide 7a with over-
all inversion of the anomeric configuration.
This proposal implies that ligation of unprotected furano-
syl azides, irrespective of the anomeric configuration of the
azide, will enforce a 1,2-cis configuration in the resulting
amides by coordination of the free OH group in the 2-posi-
tion with the P atom of the imi-
of a 1,2-cis sugar ring is enforced. In a pyranose ring, this
implies formation of the a-anomer of a pyranosyliminophos-
phorane (33, Scheme 8), which is sterically destabilized.
Hence, the process is entirely funneled through the 1,2-cis
furanose structure, via intermediate 31.
nophosphorane. Thus, the ste-
reochemical course of the re-
ACHTUNGTRENNUNGa ction is controlled by the C2
configuration of the sugar,
which explains the anomeriza-
tion observed upon reaction of
1
a with unprotected a-arabino-
furanosyl azide 11 and b-ribo-
furanosyl azide 9 (Scheme 5),
as well as the retention of
anomeric configuration ob-
served upon ligation of unpro-
tected a-galactofuranosyl azide
13 (Scheme 4).
Additionally, this mechanistic
proposal explains the puzzling
observation of a furanose ring-
opening event that occurs with-
out causing ring-expansion to
the normally favored pyranose
structure. Indeed, once oxygen–
phosphorus coordination has
occurred to give 30, formation Scheme 8. Furanose-to-pyranose isomerization is disfavored by formation of oxazaphospholane 30.
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

A. Bernardi et al.
O-Acetyl furanosyl azides: Given the known propensity of
O-acetyl glycopyranosyl iminophosphoranes to undergo
Discussion
[11]
ring-opening and anomeric equilibration,
it was rather
The mechanism of the Staudinger reaction has been exten-
[53–55]
surprising to observe a totally stereoconservative process in
the ligation of 1,2-trans O-acetyl-glycofuranosyl azides, such
as 6 (b-tetra-O-acetyl-galactofuranosyl azide, Scheme 2) or
sively studied by computational
and experimental meth-
[26b,56]
ods.
The numbers of intermediates and possible reac-
tion pathways connecting them to each other and to the
starting and final products is large and the potential-energy
surface is extremely complex. Nucleophilic attack of the
phosphine on an azide generates a phosphazide (phospho-
triazadiene) intermediate (34), which decomposes in a pro-
10 and 12 (b-tri-O-acetyl-ribofuranosyl azide and a-tri-O-
acetyl-arabinofuranosyl azide, Scheme 5). Literature data in-
dicate small energetic differences for the a- and b-isomers
[32]
of furanoses and hence ring-opening would be expected
to produce a mixture of anomeric amides. For instance, re-
duction of 1,2-trans-b-tri-O-benzoyl-ribofuranosyl azide and
in situ acylation with DCC and Cbz-glycine was reported to
cess accelerated by polar solvents, to release N and an imi-
nophosphorane (35, Scheme 9).
2
[49]
afford a 2:1 mixture of 1,2-cis:1,2-trans amides. Indeed, li-
gation of 1,2-cis a-tetra-O-acetyl-galactofuranosyl azide 14
with 1a (Scheme 4) affords a 4:1 mixture of b- (1,2-trans)
and a- (1,2-cis) isomeric amides 8a and 7a, which must orig-
inate from a ring-opening event. The partial inversion of
anomeric configuration observed in this reaction strongly
suggests that when ring-opening occurs in these furanose
substrates, the subsequent ring-closure is biased towards the
formation of 1,2-trans-amides, presumably for steric reasons,
amplified by the steric bulk of the P substituents.
Scheme 9. Schematic mechanism of the classic Staudinger reaction.
3
1
P NMR spectroscopy: Evidence for the formation of O–P
coordination in the ligation of unprotected azides was gath-
31
ered by P NMR spectroscopy in [D ]DMF. Initial experi-
Iminophosphoranes are relatively stable species and have
7
[53]
ments performed using 5 and 1a were disappointing: disap-
pearance of the phosphine signal at d=À14.9 ppm was ob-
served, but this was not accompanied by accumulation of
the relevant intermediates. Similar observations were report-
ed by Raines et al. and Bertozzi et al. in NMR studies
of traceless ligation reactions with different phosphines.
been isolated in a number of instances. In particular, gly-
cosyl iminophosphoranes from O-acetyl-protected sugars
are rather stable, and many have been isolated and charac-
[
28b,57]
terized by NMR spectroscopy.
of azides with functionalized phosphines has recently been
The Staudinger ligation
[50]
[51]
[58]
reviewed. Mechanistic investigations of the traceless Stau-
dinger ligation with phosphine 36 and with (diphenylphos-
[51]
However, when galactofuranosyl azide 5 was added to Ph P
3
[50,59]
at 708C, four new P signals appeared at d=20.4, 13.9,
À48.3, and À48.5 ppm, all correlated to a proton in the
phino)methanethiol 37
(Figure 3) have been reported.
3
1
1
anomeric region of the P- H-HMBC spectrum, and were
found to persist for over 1 h, before addition of water trans-
formed all phosphorus intermediates into triphenyl phos-
phine oxide (d= +26.2 ppm). The two signals at d=
13.9 ppm and 20.4 ppm can be assigned to the b- and a-imi-
nophosphoranes, respectively. The b-signal is formed first
and slowly decreases with time, while two new signals
appear at d=À48.5 and À48.3 ppm. These values are consis-
tent with a large upfield shift caused by coordination of the
oxygen to the phosphorus atom and can most probably be
assigned to two stereoisomers of the postulated oxazaphos-
pholane. On the contrary, when tetra-O-acetyl-galactofura-
Figure 3. Phosphines 36 and 37.
In previous studies in our laboratory, we have examined
the stereochemical outcomes of ligations of a-tetra-O-
[11,28]
[11]
[29]
benzyl,
a-tetra-O-acetyl, and unprotected a-glyco-
pyranosyl azides with phosphine 1a, and we proposed some
mechanistic hypotheses to interpret the resulting observa-
tions. Reduction–acylation of a-tetra-O-acetyl-glycopyrano-
syl azides (Scheme 1) was found to be a non-stereoconserva-
tive process and b-glycopyranosyl amides were consistently
obtained. These results are in agreement with literature re-
ports obtained by the Kiessling group using dialkylphosphi-
nosyl azide 6 was treated with Ph P under the same condi-
3
tions, only the signal of the b-iminophosphorane was ob-
served at d=12.1 ppm, together with that of triphenyl phos-
[52]
phine oxide. Thus, these data support the mechanistic hy-
pothesis outlined in Schemes 7 and 8.
[60]
no-(borane)methanethioesters and acetylated sugars.
We speculated that a-tetra-O-acetyl-glycopyranosyl imi-
nophosphoranes may be deactivated toward acyl transfer by
&
6
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Synthesis of a- and b-Glycofuranosyl Amides by Staudinger Ligation
FULL PAPER
the electron-withdrawing effect of acetyl groups, which de-
localizes the negative charge on the nitrogen atom and
allows anomeric equilibration to occur before the acyl-trans-
fer step (Scheme 10, k >k ). For tetra-O-benzyl and un-
er intermediates must also undergo a ring-opening process,
at least to some extent. This, however, does not lead to a-
b equilibration, but rather to ring-contraction, which is most
unusual for pyranoses in the gluco configuration.
eq
ac
Closer examination of the ligation reaction of unprotected
furanosyl azides has now revealed a more complex mecha-
nistic picture. First, and perhaps most strikingly, the reac-
tions of pyranosyl and furanosyl azides appear, at first sight,
to follow opposite stereochemical pathways. In fact, 1,2-
trans O-acetyl-furanosyl azides of the galacto-, ribo-, and
arabino- series (6, 10, and 12) do not isomerize upon liga-
tion with 1, whereas the same furanosyl azides do isomerize
when they are unprotected (compounds 5, 10, and 12, re-
spectively). This behavior, exemplified in Scheme 2 for b-
galactofuranosyl azides 6 and 5, is exactly the opposite of
that described for pyranoses in Scheme 1.
The mechanistic path proposed in Scheme 7 can be used
to reconcile the apparently opposite behaviors of pyranosyl
and furanosyl azides in the ligation process. The full mecha-
nistic picture is shown in Scheme 11, using unprotected gal-
actosyl azides 5 and 13 as substrates. The reaction course is
dominated by the competition between acyl-transfer and
ring-opening processes, as suggested above. However, for
unprotected sugars, the behavior of the acyclic intermediate,
a phosphinimine, is more complex than previously discussed.
In fact, once the phosphinimine is formed, the phosphorus
atom is probably trapped by the C2 hydroxy group of the
Scheme 10. Anomeric equilibration versus acylation in the Staudinger re-
duction–acylation of a-glycopyranosyl azides.
protected derivatives (Scheme 1), a more localized negative
charge on the anomeric nitrogen leads to faster acyl transfer
(
k >k ) and blocks ring-opening and anomerization. The
ac eq
same explanation may be applied to the formation of a-
amides 3a from unprotected azide 2 (Scheme 1). However,
formation of variable quantities of glucofuranosyl amides 4a
upon reaction of 2 suggests that these unprotected Stauding-
Scheme 11. Full mechanistic pathway for the reaction of unprotected galactosyl azides.
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

A. Bernardi et al.
sugar in a five-membered oxazaphospholane ring, exempli-
and hence ultimately affords 4a. Indirect proof of this is
provided by the absence of pyranosyl amides in the ligation
products of furanosyl azides. Increasing the reaction temper-
ature favors the ring-opening process, thus explaining the
fied as 30 in Scheme 11. Formation of this cyclic intermedi-
3
1
ate is supported by the P NMR data reported above. Once
this cycle is formed, the ring-closure process can no longer
lead to reversion to b-isomers of the iminophosphorane and
only the left-hand side of Scheme 11 is still available. In par-
ticular, 30 can cyclize by attack of O4 at the anomeric
carbon to form the a-furanosyl iminophosphorane 31,
which, upon acyl transfer and hydrolysis, leads to the a-fura-
nosyl amides 7. An alternative pathway for 30 is represented
by attack of O5 at the anomeric carbon, leading to the a-
pyranosyl iminophosphorane 33, acyl transfer from which
would generate the a-galactopyranosyl amide 38. This path-
way does not appear to be followed, since no a-pyranosyl
amide is observed in the ligation of furanosyl azides 5 and
[29a]
temperature dependence of the 3a/4a ratio.
When the ligation is performed starting from an unpro-
tected b-pyranosyl azide, a faster acyl transfer is expected.
Thus, in the gluco series, a b-pyranosyl amide is the only re-
action product (Scheme 1). However, if ring-opening of the
b-pyranosyl iminophosphorane 40 (Scheme 11) can occur,
the reaction can again be funneled through the cyclic oxaza-
phospholane 31, thus providing a path for pyranose-to-a-
AHCTUNGTRENNUNG
[29b]
(Scheme 11).
Finally, the b-furanosyl iminophosphorane
1
3. However, the reverse pathway, leading from 33 to 30, is
28 (Scheme 11) must initially be formed upon reduction of
b-galactofuranosyl azide 5. This, however, appears to be in-
capable of direct acyl transfer but undergoes fast ring-open-
ing leading, via 30 and 31, to the a-furanosyl amide 7
(Scheme 11).
the source of a-furanosyl amide formation in the ligation of
a-pyranosyl azide 39 and of the corresponding azide in the
gluco series (2).
phosphoranes 33 is relatively slow, allowing for ring-opening
and phosphorus coordination by the 2-hydroxy group. The
latter event prevents a–b isomerization of the pyranose ring
to b-pyranosyl iminophosphorane 40, and instead redirects
the open sugar through the furanose pathway. Hence, attack
of O4 on the anomeric carbon leads to 31 and thus to the a-
[29,61]
Acyl transfer in the a-pyranosyl imino-
The results obtained concerning the ligation of pyranosyl
and furanosyl tetra-O-acetyl-glycosyl azides with phosphine
1a are summarized in Scheme 13. Ring-contraction or -ex-
pansion paths are not available to these fully O-acetylated
substrates, so ring-opening events can only be detected as
anomeric equilibration of the reaction products. Ligation of
tetra-O-acetyl-b-glucopyranosyl azide 45 proceeds unevent-
fully to afford the b-glucopyranosyl amide (Scheme 13a).
The same product is formed by ligation of the a-anomer 46
[61]
furanosyl amide by-products in both the galacto- (7) and
gluco- (4a, Scheme 1) series.
Thus, in this rather complex framework, the mechanism
for the reaction of a-glucopyranosyl azide 2 with 1a, as
shown in Scheme 1, can be reformulated as proposed in
Scheme 12. The a-iminophosphorane 43 formed by azide re-
duction can undergo either direct acyl transfer (to yield the
a-pyranosyl amide 3a) or ring-opening. This is probably not
an equilibrium reaction, because the resulting phosphini-
mine is blocked by O–P coordination to afford 44, which
strongly favors the formation of [5.5]-fused bicyclic systems
[11]
(Scheme 13b).
As previously discussed, complete inver-
sion of the anomeric configuration observed in this reaction
can be explained in terms of ring-opening of the intermedi-
ate iminophosphorane, facilitated by the electron-withdraw-
ing effect of the acetyl groups, which favors phosphinimine
formation. Similar behavior is observed for a-tetra-O-
acetyl-galactopyranosyl azide 47 (Scheme 13c), although the
anomeric inversion is not com-
plete and a 4:1 b:a mixture of
[61]
epimeric amides is formed.
This suggests that either acyl
transfer is faster for the a-gal-
actopyranosyl iminophosphor-
ane than for the corresponding
gluco derivative, or that the
anomeric equilibrium in the
galactopyranose series is less
shifted toward the b-isomer. A
similar 4:1 mixture of b- and a-
furanosyl amides (8a and 7a)
was formed upon ligation of a-
tetra-O-acetyl-galactofuranosyl
azide 14 with 1a (Scheme 13e).
The partial inversion of anome-
ric configuration observed in
this reaction suggests that when
ring-opening occurs for galacto-
furanose, subsequent ring-clo-
Scheme 12. Furanosyl amide formation in the ligation of unprotected pyranosyl azides.
&
8
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Synthesis of a- and b-Glycofuranosyl Amides by Staudinger Ligation
FULL PAPER
tational studies. From a synthet-
ic point of view, however, it is
worth noting that ligation with
1
is the only known process
that allows the stereoselective
reduction/acylation of furanosyl
[8g,49,62]
azides.
With these observations, we
have now begun to clarify the
complex mechanistic picture of
the Staudinger ligation of glyco-
syl azides, both as pyranose and
furanose isomers, to the point
where predictions can be made
concerning the stereochemistry
of the reaction on new sub-
strates and with new combina-
tions of reagents. Work is in
progress in our laboratory to
test hypotheses generated by
the models described above.
The glycofuranosyl amides de-
scribed in this work are novel
glycoconjugates, of potential in-
terest as bioactive compounds
or new materials. Further stud-
ies are in progress to analyze
the activity of galactofuranosyl
amides against Galf-containing
pathogens.
Scheme 13. Ligation of tetra-O-acetyl-glycosyl azides.
sure is biased towards the formation of 1,2-trans-amides,
presumably for steric reasons. The same effect appears to
allow a totally stereoconservative ligation of b-tetra-O-
acetyl-galactofuranosyl azide 6 (Scheme 13d), which affords
exclusively the corresponding b-tetra-O-acetyl-galactofura-
nosyl amide 8a.
Experimental Section
General: Solvents were dried by standard procedures: dichloromethane,
methanol, N,N-diisopropylethylamine, and triethylamine were dried over
calcium hydride; N,N-dimethylacetamide (DMA), 1,3-dimethyltetrahy-
dro-2(1H)pyrimidinone (DMPU), chloroform, and pyridine were dried
over activated molecular sieves. Reactions requiring anhydrous condi-
Conclusion
1
13
31
tions were performed under nitrogen. H, C, and P NMR spectra were
recorded at 400 MHz on a Bruker AVANCE-400 instrument. Chemical
We have studied the traceless Staudinger ligation of glyco-
furanosyl azides with phosphines 1 and have discovered
a stereoconvergent synthesis of glycofuranosyl amides that
is independent of the configuration of the starting azide. No-
tably, both a- and b-isomers can be obtained with excellent
selectivity from a common, easily available precursor. The
steric course of the reaction is controlled by the C2 configu-
ration of the sugar and by the protection state. Unprotected
glycofuranosyl azides consistently afford amides of 1,2-cis
configuration. The 1,2-trans counterpart is obtained starting
from the corresponding O-acetyl-glycofuranosyl azides. For
unprotected furanoses, the anomeric configuration appears
to be dictated by O–P coordination, as supported by
1
13
shifts (d) in H and C spectra are expressed in ppm relative to Me
4
Si as
an internal standard. Chemical shifts (d) in P spectra are expressed in
ppm relative to H PO as an internal standard. Signals are abbreviated as
3
1
3
4
s, singlet; brs, broad singlet; d, doublet; t, triplet; q, quartet; m, multiplet.
Mass spectra were obtained with a Bruker Esquire 3000 ion-trap appara-
tus (ESI) and an FT-ICR mass spectrometer with APEX II & Xmass
software (Bruker Daltonics) -4.7 Magnet. Optical rotations [a]
D
were
measured from solutions in a cell of pathlength 1 dm and capacity 1 mL
with a Perkin–Elmer 241 polarimeter. Thin-layer chromatography (TLC)
was carried out on pre-coated Merck F254 silica gel plates. Flash chroma-
tography (FC) was carried out on columns of Macherey–Nagel silica gel
6
0 (230–400 mesh). The syntheses of phosphines 1a–g and of b-galacto-
furanosyl azides 5 and 6 have been described in the preliminary commu-
[
39]
nication. The syntheses and characterizations of ribo- and arabino-fur-
anosyl azides 9–12 and of the corresponding pyranosyl isomers 18–21 are
described in the Supplementary Information.
31
P NMR studies. A convincing explanation for the forma-
2
,3,5,6-Tetra-O-tert-butyldimethylsilyl-a-d-galactofuranosyl azide (17): A
tion of 1,2-trans products from O-acetyl-glycofuranosyl
azides could not be achieved based on the empirical results
collected here, but may now be addressed by future compu-
solution of 1,2,3,5,6-penta-O-tert-butyldimethylsilyl-b-d-galactofuranose
[43]
1
5
(200 mg, 0.26 mmol) in anhydrous CH
2 2
Cl (10 mL) was cooled to
08C and stirred for 10 min under nitrogen. Iodotrimethylsilane (1.2 equiv,
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
9
&
ÞÞ
These are not the final page numbers!

A. Bernardi et al.
0
0
.042 mL, 0.32 mmol) was then added, and the solution was stirred at
8C until TLC monitoring (hexane/AcOEt, 10:1) showed complete con-
The solvent was evaporated under reduced pressure, and the residue was
purified as indicated below for each compound.
version of 15 into two products (R
lution of tetrabutylammonium azide (0.5 m) was prepared.
f f
=0.70 and R =0.54). In parallel, a so-
General procedure for the deprotection of per-O-acetyl-glycosyl amides
[
63]
A 10m
Procedure A: A 0.1m solution of NaOMe in dry MeOH (0.5 equiv) was
added, at room temperature and under nitrogen, to a 0.1m solution of O-
acetyl-glycosyl amide (1 equiv) in dry MeOH. The mixture was stirred at
room temperature. After 45 min, TLC monitoring (hexane/AcOEt, 50:50,
sodium hydroxide solution (280 mL) was added to tetrabutylammonium
hydrogensulfate (383.3 mg, 1 mmol) in water (560 mL), then a solution of
sodium azide (147 mg, 2.3 mmol) in water (280 mL) was added and tetra-
butylammonium azide was extracted with dichloromethane (1 mL). The
organic layer was separated and the aqueous phase was extracted with
further dichloromethane (1 mL). The combined organic phases were
and CHCl /MeOH, 80:20) showed complete consumption of the starting
3
+
material and Amberlyst IRA 120 H was added. The mixture was stirred
for 30 min at pH 3. The resin was then filtered off and washed with
MeOH, and the solvent was removed under reduced pressure. The prod-
uct, isolated in quantitative yield, was used without further purification.
dried with Na
rabutylammonium azide as a white solid, which was redissolved in dry
CH Cl (2 mL). The solutions thus obtained were used directly in the fol-
2 4
SO and concentrated in vacuo at 408C to yield crude tet-
2
2
Procedure B: The glycosyl amide (1 equiv) was dissolved in a 4m solution
lowing step.
of MeNH
2
in EtOH (0.046m) and the mixture was stirred at room tem-
2
Tetrabutylammonium azide solution (1 mmol) and EtN CAHTUNGTENRUNG( iPr) (0.054 mL,
perature. After 2 h, TLC monitoring (hexane/AcOEt, 50:50, and CHCl
3
/
0
.032 mmol) were added to the galactofuranosyl iodide, and stirring was
MeOH, 90:10) showed complete consumption of the starting material.
Evaporation of the solvent in vacuo and co-evaporation with MeOH af-
forded the unprotected amide without further purification.
continued until TLC revealed consumption of the components at R
f
=
0
5
.70 and R
HCl, saturated aqueous NaHCO
SO ), and concentrated. The syrup obtained was purified by flash
column chromatography (hexane/AcOEt, 98:2) to afford 17 in 75%
f
=0.54. The solution was diluted with CH
2 2
Cl , washed with
%
3
solution, and water, dried
N-Pentanoyl-a-d-ribofuranosyl amide (22a): This compound was ob-
(
Na
2
4
tained by ligation of 9 with 1a and was purified by flash chromatography
1
(
CHCl
3
/MeOH, 80:20). Yield: 59%. H NMR (400 MHz, CD
3
OD, 258C):
1
yield. H NMR (400 MHz, CDCl
3
, 258C): d=5.11 (d, J1,2 =4 Hz, 1H; H-
), 4.24 (t, J2,3 =4.4, J3,4 =4.6 Hz, 1H; H-3), 3.99 (t, J1,2 =4, J2,3 =4.4 Hz,
H; H-2), 3.87 (t, J3,4 =4.6, J4,5 =4.4 Hz, 1H; H-4), 3.79 (ddd, J4,5 =4.4,
5,6 =6, J5,6’ =5.6 Hz, 1H; H-5), 3.66 (dd, J5,6 =6, J6,6’ =10 Hz, 1H; H-6),
.60 (dd, J5,6’ =5.6, J6,6’ =10 Hz, 1H; H-6’), 0.92–0.84 (m, 36H; SiC-
d=5.67 (d, J1,2 =4.8 Hz, 1H; H-1), 4.12–4.06 (m, 2H; H-2, H-3), 3.93–
1
1
J
3
1
.87 (q, J4,5 =3.2 Hz, J4,5’ =4.4 Hz, 1H; H-4), 3.69 (dd, J4,5 =3.2 Hz, J5,5’
2 Hz, 1H; H-5), 3.55 (dd, J4,5’ =4.4 Hz, J5,5’ =12 Hz, 1H; H-5’), 2.24 (t,
), 1.65–1.57 (m, 2H; CH ), 1.43–1.33 (m, 2H; CH ),
); C NMR (100 MHz, CD OD, 258C):
d=176.6, 84.2 (C-4), 81.5 (C-1), 72.6, 71.8, 62.9 (C-5), 36.9 (CH ), 28.8
CH ), 23.3 (CH ), 14.1 ppm (CH ); FT-ICR (ESI): calcd for C10
M+Na] 256.12632; found 256.12643.
,3,5-Triacetyl-N-pentanoyl-a-d-ribofuranosyl
10 equiv) and a catalytic amount of N,N-dimethylaminopyridine were
=
J=7.2 Hz, 2H; CH
.95 ppm (t, J=7.2 Hz, 3H; CH
2
2
2
3
13
0
3
3
1
3
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(CH
3 3 3 2 3
) ), 0.18–0.06 ppm (m, 24H; Si ACHTUNGTERNNUNG( CH ) ); C NMR (100 MHz, CDCl ,
2
2
58C): d=90.9 (C-1), 85.5 (C-4), 79.5 (C-2), 76.3 (C-3), 73.0 (C-5), 65.0
(
[
2
2
3
H19NO
5
(
C-6), 26.2, 26.1, 26.0, 25.9, À0.25 to À0.5 ppm; ESI-MS: m/z 684.4
+
+
[
M+Na] .
2
(
amide
(59):
2
Ac O
a-d-Galactofuranosyl azide (13): A solution of 17 (60 mg, 0.09 mmol,
equiv) in anhydrous THF dry (180 mL) was cooled to 08C and stirred
1
added to a 0.1m solution of substrate 22a (1 equiv) in dry pyridine at
room temperature. The mixture was stirred for 24 h and then concentrat-
ed in vacuo. The residue was dissolved in AcOEt, and this solution was
for 10 min under nitrogen. A 1m solution of tetrabutylammonium fluo-
ride in THF (450 mL, 0.45 mmol, 5 equiv) was added, and the mixture
was stirred at room temperature until TLC monitoring (hexane/AcOEt,
washed with aqueous 5% HCl, aqueous 5% NaHCO
ganic layer was dried over Na SO and concentrated to give the product
in quantitative yield. H NMR (400 MHz, CD OD, 258C): d=6.06 (d,
1,2 =4.8 Hz, 1H; H-1), 5.45 (t, J1,2 =4.8, J2,3 =5.6 Hz, 1H; H-2), 5.36 (t,
2,3 =5.6 Hz, 1H; H-3), 4.36–4.31 (m, 2H; H-4, H-5), 4.21 (dd, J4,5’ =4.8,
5,5’ =12 Hz, 1H; H-5’), 2.23 (t, J=7.2 Hz, 2H; CH ), 2.19 (s, 3H; OAc),
.15 (s, 3H; OAc), 2.12 (s, 3H; OAc), 1.67–1.59 (m, 2H; CH ), 1.44–1.35
); C NMR (100 MHz,
OD, 258C): d=176.1, 172.2, 171.5, 171.3, 80.8 (C-1), 79.2 (C-4), 72.9
), 29 (CH ), 23.4 (CH ), 20.7–20.5
); ESI-MS: m/z: 359.1 [M+Na] .
3
, and water. The or-
6
3
0:40, and CHCl /MeOH, 80:20) showed complete transformation of 17
2
4
into 13. The reaction mixture was diluted with water, washed with di-
chloromethane and AcOEt, and then concentrated. The syrup obtained
1
3
J
J
J
was purified by flash column chromatography (CHCl
3
/MeOH, 90:10).
OD, 258C): d=5.17 (d, J1,2
.8 Hz, 1H; H-1), 4.08 (t, J2,3 =6.4, J3,4 =6 Hz, 1H; H-3), 4.04 (t, J1,2 =4.8,
2,3 =6.4 Hz, 1H; H-2), 3.77 (dd, J3,4 =6, J4,5 =4.4 Hz, 1H; H-4), 3.69
1
Quantitative yield. H NMR (400 MHz, CD
4
J
3
=
2
2
2
13
(
2 3
m, 2H; CH ), 0.99 ppm (t, J=7.2 Hz, 3H; CH
(
ddd, J4,5 =4.4, J5,6 =5.2, J5,6’ =6.8 Hz, 1H; H-5), 3.66 (dd, J5,6 =5.2, J6,6’
=
CD
3
1
1.2 Hz, 1H; H-6), 3.60 ppm (dd, J5,6’ =6.8, J6,6’ =11.2 Hz, 1H; H-6’);
(
(
C-3), 71.9 (C-2), 64.8 (C-5), 36.8 (CH
3ꢃOAc), 14.2 ppm (CH
2
2
2
1
3
C NMR (100 MHz, CD
3
OD, 258C): d=92.8 (C-1), 84.5 (C-4), 79 (C-2),
+
3
+
7
6.2 (C-3), 73.5 (C-5), 64.1 ppm (C-6); ESI-MS: m/z: 228.1 [M+Na] .
N-Pentanoyl-b-d-ribofuranosyl amide (23a): Reaction of 10 with 1a af-
forded tri-O-acetyl-N-pentanoyl-b-d-ribofuranosyl amide. After deacety-
2
,3,4,6-Tetra-O-acetyl-a-d-galactofuranosyl azide (14): Acetic anhydride
(
10 equiv) and a catalytic amount of N,N-dimethylaminopyridine were
lation (procedure B), the compound was purified by flash chromatogra-
added to a 0.1m solution of a-d-galactofuranosyl azide 13 (1 equiv) in
pyridine at room temperature. The mixture was stirred for 24 h and then
concentrated in vacuo. The residue was dissolved in AcOEt, and this so-
lution was washed with aqueous 5% HCl, aqueous 5% NaHCO
water. The organic layer was dried over Na SO and concentrated to
afford the product 14 in quantitative yield. H NMR (400 MHz, CDCl
1
phy (CHCl
3
/MeOH, 80:20). Yield: 55%. H NMR (400 MHz, CD
3
OD,
2
1
3
58C): d=5.38 (d, J1,2 =4.8 Hz, 1H; H-1), 4.06 (t, J2,3 =5.2, J3,4 =5.2 Hz,
H; H-3), 3.91 (t, J1,2 =4.8, J2,3 =5.2 Hz, 1H; H-2), 3.88–3.83 (q, J4,5
.6 Hz, J4,5’ =4.4 Hz, 1H; H-4), 3.70 (dd, J4,5 =3.6 Hz, J5,5’ =12 Hz, 1H; H-
=
3
, and
2
4
5), 3.61 (dd, J4,5’ =4.4 Hz, J5,5’ =12 Hz, 1H; H-5’), 2.23 (t, J=7.2 Hz, 2H;
CH ), 1.65–1.57 (m, 2H; CH ), 1.43–1.33 (m, 2H; CH ), 0.95 ppm (t, J=
1
3
,
2
2
2
1
3
2
1
J
6
4
58C): d=5.55 (d, J1,2 =5.6 Hz, 1H; H-1), 5.41 (t, J2,3 =6.8, J3,4 =6.4 Hz,
H; H-3), 5.28 (ddd, J4,5 =2, J5,6 =4.8, J5,6’ =6.4 Hz, 1H; H-5), 5.14 (t,
1,2 =5.6, J2,3 =6.8 Hz, 1H; H-2), 4.35 (dd, J5,6 =4.8, J6,6’ =12 Hz, 1H; H-
), 4.17 (dd, J5,6’ =6.4, J6,6’ =12 Hz, 1H; H-6’), 4.13 (dd, J3,4 =6, J4,5
.4 Hz, 1H; H-4), 2.13 (s, 3H; OAc), 2.10 (s, 3H; OAc), 2.08 (s, 3H;
3 3
7.2 Hz, 3H, CH ); C NMR (100 MHz, CD OD, 258C): d=176.8, 85.8
(C-1), 85.1 (C-4), 75.9 (C-2), 71.9 (C-3), 63.3 (C-5), 37.1 (CH ), 28.9
2
(CH ), 23.5 (CH ), 14.3 ppm (CH ); FT-ICR (ESI): calcd for C H NO
5
2
2
3
10 19
+
=
[M+Na] 256.12632; found 256.12627.
N-Pentanoyl-b-d-arabinofuranosyl amide (24a): The compound was ob-
1
3
OAc), 2.05 ppm (s, 3H; OAc); C NMR (100 MHz, CDCl
3
, 258C): d=
70.6, 170.2, 170.1, 169.9, 88.8 (C-1), 79.1 (C-4), 75.9 (C-2), 73.6 (C-3),
9.6 (C-5), 62.4 (C-6) 20.9–20.6 ppm (4ꢃOAc); ESI-MS: m/z: 396.1
tained by ligation of 11 with 1a and was purified by flash chromatogra-
1
6
1
phy (CHCl
3
/MeOH, 80:20). Yield: 57%. H NMR (400 MHz, CD
3
OD,
2
58C): d=5.72 (d, J1,2 =4.4 Hz, 1H; H-1), 4.01 (t, J2,3 =3.6 Hz, J3,4
=
+
[
M+Na] .
3
.6 Hz, 1H; H-3), 3.90 (dd, J1,2 =4.4 Hz, J2,3 =3.6 Hz, 1H; H-2), 3.76
General procedure for the stereoselective ligation of glycosyl azides:
Phosphine 1 (1.2 equiv) was added to a 0.1m solution of glycosyl azide
(ddd, J3,4 =3.6 Hz, J4,5 =3.6 Hz, J4,5’ =5.2 Hz, 1H; H-4), 3.70 (dd, J4,5 =
3.6 Hz, J5,5’ =11.6 Hz, 1H; H-5), 3.64 (dd, J4,5’ =5.2 Hz, J5,5’ =11.6 Hz, 1H;
H-5’), 2.28 (t, J=7.6 Hz, 2H; CH ), 1.65–1.57 (m, 2H; CH ), 1.43–1.33
(
1 equiv) in N,N-dimethylacetamide/DMPU (98:2) at room temperature.
2
2
1
3
The mixture was stirred for 4 h at 708C, then water was added, and the
2 3
(m, 2H; CH ), 0.94 ppm (t, J=7.6 Hz, 3H; CH ); C NMR (100 MHz,
resulting mixture was stirred for a further 2 h at the same temperature.
3
CD OD, 258C): d=176.8, 84.4 (C-4), 82.2 (C-1), 78.3 (C-3), 77.3 (C-2),
&
10
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Synthesis of a- and b-Glycofuranosyl Amides by Staudinger Ligation
FULL PAPER
6
3.5 (C-5), 36.9 (CH ), 28.9 (CH
2
2
), 23.5 (CH
2
), 14.3 ppm (CH
3
); FT-ICR
[7] a) T. K. Lindhorst, Nachr. Chem. Tech. Lab. 2010, 44, 1073; b) N.
Jayaraman, S. A. Negopodiev, J. F. Stoddart, Chem. Eur. J. 1997, 3,
+
(
ESI): calcd for C10H19NO [M+Na] 256.12632; found 256.12605.
5
1
193–1199.
N-Pentanoyl-a-d-arabinofuranosyl amide (25a): Reaction of 12 with 1a
afforded tri-O-acetyl-N-pentanoyl-a-d-arabinofuranosyl amide. After de-
[8] a) N. Tewari, V. K. Tiwari, R. C. Mishra, R. P. Tripathi, A. K. Srivas-
tava, R. Ahmad, R. Srivastava, B. S. Srivastava, Bioorg. Med. Chem.
2003, 11, 2911–2922; b) H. Paulsen, K. Todt, Adv. Carbohydr.
Chem. 1968, 23, 115–232; c) E. Truscheit, W. Frommer, B. Junge, L.
Mueller, D. D. Schmidt, W. Wingender, Angew. Chem. 1981, 93,
738–755; Angew. Chem. Int. Ed. Engl. 1981, 20, 744–761; d) S.
Inouye, T. Tsuruoka, T. Ito, T. Niida, Tetrahedron 1968, 24, 2125–
2144; e) P. B. Anzeveno, L. J. Creemer, J. K. Daniel, C. H. R. King,
P. S. Liu, J. Org. Chem. 1989, 54, 2539–2542; f) D. P. Gamblin, E. M.
Scanlan, B. G. Davis, Chem. Rev. 2009, 109, 131–163; g) G. J. van
der Heden van Noort, M. G. Van der Horst, H. S. Overkleeft, G. A.
Van der Marel, D. V. Filippov, J. Am. Chem. Soc. 2010, 132, 5236–
5240; h) R. J. Payne, C.-H. Wong, Chem. Commun. 2010, 21–43.
[9] a) S. Umezewa, T. Tsuchiya, in Aminoglycosides Antibiotics (Eds.:
H. Umezewa, R. I. Hooper), Springer, Berlin, 1982, pp. 37–110;
b) H. A. Kirst, in Burgerꢀs Medicinal Chemistry and Drug Discovery,
(Ed.: M. E. Wolff), Wiley, New York, 1996, pp. 463–525.
acetylation (procedure B), the compound was purified by flash chroma-
1
tography (CHCl
CD
H; H-3), 3.88 (t, J1,2 =4.8, J2,3 =5.2 Hz, 1H; H-2), 3.84 (q, J4,5 =3.6 Hz,
4,5’ =4.4 Hz, 1H; H-4), 3.67 (dd, J4,5 =3.6, J5,5’ =12 Hz, 1H; H-5), 3.55
dd, J4,5’ =4.4 Hz, J5,5’ =12 Hz, 1H; H-5’) 2.20 (t, J=7.6 Hz, 2H; CH ),
), 0.92 ppm (t, J=7.2 Hz,
OD, 258C): d=177.1, 85.7 (C-1), 85
), 28.9 (CH ), 23.5
); FT-ICR (ESI): calcd for C10 [M+Na]
56.12632; found 256.12638.
3
/MeOH, 80:20). Yield: 53%. H NMR (400 MHz,
3
OD, 258C): d=5.39 (d, J1,2 =4.8 Hz, 1H; H-1), 4.03 (t, J2,3 =5.2 Hz,
1
J
(
2
1
3
.60–1.54 (m, 2H; CH
2
), 1.40–1.29 (m, 2H; CH
2
1
3
H; CH
3
); C NMR (100 MHz, CD
3
(
(
C-4), 81.5 (C-3), 77.1 (C-2), 63.1 (C-5), 37.1 (CH
CH ), 14.3 ppm (CH H19NO
5
2
2
+
2
3
2
N-Pentanoyl-b-d-ribopyranosyl amide (26a): a) The compound was ob-
tained by ligation of 18 with 1a and was purified by flash chromatogra-
phy (CHCl /MeOH, 80:20). Yield: 68%. b) Reaction of 19 with 1a af-
3
forded tri-O-acetyl-N-pentanoyl-a-d-ribopyranosyl amide. After deacety-
lation (procedure A), the compound was purified by flash chromatogra-
1
phy (CHCl
3
/MeOH, 80:20). Yield: 55%. H NMR (400 MHz, CD
3
OD,
[10] F. Damkaci, P. DeShong, J. Am. Chem. Soc. 2003, 125, 4408–4409.
[11] a) A. Bianchi, A. Bernardi, J. Org. Chem. 2006, 71, 4565–4577;
b) C. Colombo, A. Bernardi, Eur. J. Org. Chem. 2011, 3911–3919.
[12] S. Shibata, T. Takeda, Y. Natori, J. Biol. Chem. 1988, 263, 12483–
12485.
2
3
58C): d=5.15 (d, J1,2 =9.2 Hz, 1H; H-1), 4.07 (d, J2,3 =2.8 Hz, 1H; H-3),
.73–3.57 (m, 3H; H-4, H-5, H-5’), 3.39 (dd, J1,2 =9.2, J2,3 =2.8 Hz, 1H;
), 1.64–1.56 (m, 2H; CH ), 1.43–1.32
); C NMR (100 MHz,
OD, 258C): d=177.5, 78.0 (C-1), 72.4 (C-3), 71.3 (C-2), 68.8 (C-4),
5.7 (C-5), 37.0 (CH ), 28.9 (CH ), 23.5 (CH ), 14.3 ppm (CH ); FT-ICR
ESI): calcd for C10 [M+Na] 256.12632; found 256.12645.
H-2), 2.22 (t, J=7.6 Hz, 2H; CH
m, 2H; CH ), 0.94 ppm (t, J=7.6 Hz, 3H; CH
CD
2
2
1
3
(
2
3
3
[13] C. J. Bosques, S. M. Tschampel, R. Woods, B. Imperiali, J. Am.
Chem. Soc. 2004, 126, 8421–8425.
6
(
2
2
2
3
+
H19NO
5
[
14] a) G. Timpano, G. Tabarani, M. I. Anderluh, D. Invernizzi, F. Vasile,
D. Potenza, P. M. Nieto, J. Rojo, F. Fieschi, A. Bernardi, ChemBio-
Chem 2008, 9, 1921–1930; b) M. Andreini, D. Doknic, I. Sutkevi-
ciute, J. J. Reina, J. Duan, E. Chabrol, M. Thꢄpaut, E. Moroni, F.
Doro, L. Belvisi, J. Weiser, J. Rojo, F. Fieschi, A. Bernardi, Org.
Biomol. Chem. 2011, 9, 5778–5786.
[15] M. Andreini, M. I. Anderluh, A. Audfray, A. Bernardi, A. Imberty,
Carbohydr. Res. 2010, 345, 1400–1407.
[16] a) C. P. R. Hackenberger, M. K. OꢂReilly, B. Imperiali, J. Org.
Chem. 2005, 70, 3574–3578; b) J. van Ameijde, H. B. Albada,
R. M. J. Liskamp, J. Chem. Soc. Perkin Trans. 1 2002, 1042–1049;
c) C. M. Kaneshiro, K. A. Michael, Angew. Chem. 2006, 118, 1094–
N-Pentanoyl-b-l-arabinopyranosyl amide (27a): a) The compound was
obtained by ligation of 20 with 1a and was purified by flash chromatogra-
phy (CHCl /MeOH, 80:20). Yield: 70%. b) Reaction of 21 with 1a af-
3
forded tri-O-acetyl-N-pentanoyl-a-d-arabinopyranosyl amide. After de-
acetylation (procedure A), the compound was purified by flash chroma-
1
tography (CHCl
CD
H-5), 3.73–3.66 (m, 3H; H-2, H-4, H-5’), 2.34 (t, J=7.6 Hz, 2H; CH
3
/MeOH, 80:20). Yield: 57%. H NMR (400 MHz,
3
OD, 258C): d=4.90 (d, J1,2 =8 Hz, 1H; H-1), 3.95–3.88 (m, 2H; H-3,
),
), 1.03 ppm (t, J=7.6 Hz,
OD, 258C): d=177.2, 81.8 (C-1),
5.3 (C-2), 71.4 (C-4), 70.3 (C-3), 68.9 (C-5), 37.1 (CH ), 29.0 (CH ), 23.5
), 14.2 ppm (CH ); FT-ICR (ESI): calcd for C10 [M+Na]
56.12632; found 256.12621.
2
1
3
7
.74–1.67 (m, 2H; CH
2
), 1.53–1.42 (m, 2H; CH
2
1
3
H; CH
3
); C NMR (100 MHz, CD
3
2
2
+
(
CH
2
3
H19NO
5
1
098; Angew. Chem. Int. Ed. 2006, 45, 1077–1081.
17] P. R. Sridhar, K. R. Prabhu, S. Chandrasekaran, J. Org. Chem. 2003,
8, 5261–5264.
2
[
[
6
18] I. Matsuo, Y. Nakahara, Y. Ito, T. Nukada, T. Nakahara, T. Ogawa,
Bioorg. Med. Chem. 1995, 3, 1455–1463.
[
[
[
[
19] U. K. Saha, R. Roy, Tetrahedron Lett. 1995, 36, 3635–3638.
20] S. Sabesan, Tetrahedron Lett. 1997, 38, 3127–3130.
21] T. Inazu, K. Kobayashi, Synlett 1993, 869–870.
Acknowledgements
This work was supported by the Universitꢀ degli Studi di Milano
PUR2008). High-resolution mass spectra were recorded at the CIGA
Centro Interdipartimentale Grandi Apparecchiature, Universitꢀ degli
22] I. Bosch, P. Romea, F. Urpi, J. Villarrasa, Tetrahedron Lett. 1993, 34,
(
(
4
671–4674.
23] P. Boullanger, V. Maunier, D. Lafont, Carbohydr. Res. 2000, 324,
7–106.
24] J. P. Malkinson, R. A. Falconer, I. Toth, J. Org. Chem. 2000, 65,
249–5252.
[
[
[
[
Studi di Milano).
9
5
[
1] P. Norris, Curr. Appl. Phys. Curr. Topics Med. Chem. 2008, 8, 101–
13.
2] a) P. Cheshev, A. Marra, A. Dondoni, Org. Biomol. Chem. 2006, 4,
225–3227; b) M. Lo Conte, A. Chambery, A. Dondoni, Synlett
009, 16, 2679–2681.
3] D. M. M. Jaradat, H. Hamouda, C. P. R. Hackenberger, Eur. J. Org.
Chem. 2010, 5004–5009.
4] G. A. Ellestad, D. B. Cosulich, R. W. Broschard, J. H. Martin, M. P.
Kunstmann, G. O. Morton, J. E. Lancaster, W. Fulmor, F. M. Lovell,
J. Am. Chem. Soc. 1978, 100, 2515–2524.
25] N. Shangguan, S. Katukojvala, R. Greenberg, L. J. Williams, J. Am.
Chem. Soc. 2003, 125, 7754–7755.
26] a) H. Staudinger, J. Meyer, Helv. Chim. Acta 1919, 2, 635–646;
b) Y. G. Gololobov, L. F. Kasukhin, Tetrahedron 1992, 48, 1353–
1
[
3
2
1
407.
27] a) A. J. Ratcliffe, B. J. Fraser-Reid, J. Chem. Soc. Perkin Trans.
1989, 1805–1810; b) A. J. Ratcliffe, P. Konradsson, B. Fraser-Reid,
[
[
[
1
J. Am. Chem. Soc. 1990, 112, 5665–5667; c) A. J. Ratcliffe, P. Kon-
radsson, B. Fraser-Reid, Carbohydr. Res. 1991, 216, 323–335.
[
[
5] K. Dobashi, K. Nagaoka, Y. Watanobe, M. Nishida, M. Hamada, H.
Naganawa, T. Tacita, T. Takenchi, H. Umezawa, J. Antibiot. 1985,
[28] a) A. Bianchi, A. Bernardi, Tetrahedron Lett. 2004, 45, 2231–2234;
b) A. Bianchi, A. Russo, A. Bernardi, Tetrahedron: Asymmetry
2005, 16, 381–386.
[29] a) F. Nisic, A. Bernardi, Carbohydr. Res. 2008, 343, 1636–1643; b) F.
Nisic, M. Andreini, A. Bernardi, Eur. J. Org. Chem. 2009, 5744–
5751.
3
8, 1166–1170.
6] a) Y. Ichikawa, T. Nishiyama, M. Isobe, Synlett 2000, 1253–1256;
b) Y. Ichikawa, T. Nishiyama, M. J. Isobe, Org. Chem. 2001, 66,
4
200–4205.
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
11
&
ÞÞ
These are not the final page numbers!

A. Bernardi et al.
[
30] E. Saxon, J. I. Armstrong, C. R. Bertozzi, Org. Lett. 2000, 2, 2141–
143.
31] B. L. Nilsson, L. L. Kiessling, R. T. Raines, Org. Lett. 2000, 2, 1939–
941.
32] S. J. Angyal, Carbohydr. Res. 1994, 263, 1–11.
33] Z. Gyçrgydeꢀk, L. Szitagyi, H. Paulsen, J. Carbohydr. Chem. 1993,
[47] M. S. Drenichev, I. V. Kulikova, G. V. Bobkov, V. I. Tararov, S. N.
Mikhailov, Synthesis 2010, 22, 3827–3834.
[48] Configurational assignments were made on the basis of NMR spec-
troscopic data (NOE experiments) and by comparison with the cor-
responding pyranosides. Details are collected in the Supporting In-
formation.
2
[
1
[
[
1
2, 139–163.
[49] D. H. Boschelli, D. Powell, V. Sharky, M. F. Semmelhack, Tetrahe-
dron Lett. 1989, 30, 1599–1600.
[50] M. B. Soellner, B. L. Nilsson, R. T. Raines, J. Am. Chem. Soc. 2006,
128, 8820–8828.
[
[
34] J. C. Norrild, H. Eggert, J. Am. Chem. Soc. 1995, 117, 1479–1484.
35] P. Peltier, R. Euzen, R. Daniellou, C. Nugier-Chauvin, V. Ferrieres,
Carbohydr. Res. 2008, 343, 1897–1923.
[
[
36] R. M. Richards, T. L. Lowary, ChemBioChem 2009, 10, 1920–1938.
37] a) R. E. Lee, M. D. Smith, R. J. Nash, R. C. Griffiths, M. McNeil,
R. K. Grewal, W. X. Yan, G. S. Besra, P. J. Brennan, G. W. J. Fleet,
Tetrahedron Lett. 1997, 38, 6733–6736; b) K. Itoh, Z. S. Huang,
H. W. Liu, Org. Lett. 2007, 9, 879–882; c) J. Kovensky, M. McNeil,
P. Sinaꢅ, J. Org. Chem. 1999, 64, 6202–6205; d) A. Caravano, S. P.
Vincent, P. Sinaꢅ, Chem. Commun. 2004, 1216–1217; e) A. Carava-
no, D. Mengin-Lecreuix, J. M. Brondello, S. P. Vincent, P. Sinaꢅ,
Chem. Eur. J. 2003, 9, 5888–5898; f) W. D. Pan, C. Ansiaux, S. P.
Vincent, Tetrahedron Lett. 2007, 48, 4353–4356; g) M. R. Scherman,
Antimicrob. Agents Chemother. 2003, 47, 378–382; h) R. P. Tangalla-
pally, R. Yendapally, R. E. Lee, K. Hevener, V. C. Jones, A. J. M.
Lenaerts, M. R. McNeil, Y. H. Wang, S. Franzblau, R. E. Lee, J.
Med. Chem. 2004, 47, 5276–5283; i) M. Soltero-Higgin, E. E. Carl-
son, J. H. Phillips, L. L. Kiessling, J. Am. Chem. Soc. 2004, 126,
[51] F. L. Lin, H. Hoyt, H. van Halbeek, R. G. Bergman, C. Bertozzi, J.
Am. Chem. Soc. 2005, 127, 2686–2695.
[52] Direct formation of Ph PO in this experiment derives from aza-
3
Wittig reaction between the b-triphenyliminophosphorane and the
acetyl group in position 3 of the
sugar, leading to clean formation
of oxazoline 48, which was isolat-
ed in high yields at the end of
the reaction.
[53] C. Widauer, H. Grꢆtzmacher, I.
Shevchenko, V. Gramlich, Eur. J.
Inorg. Chem. 1999, 1659–1664
and references therein.
[54] M. Alajarꢇn, C. Conesa, H. S. Rzepa, J. Chem. Soc. Perkin Trans. 2
1999, 1811.
1
0532–10533; j) E. E. Carlson, J. F. May, L. L. Kiessling, Chem.
[55] W. Q. Tian, Y. A. Wang, J. Org. Chem. 2004, 69, 4299–4308.
[56] Y. G. Gololobov, I. N. Zhmurova, L. F. Kasukhin, Tetrahedron 1981,
37, 437–472.
[57] a) L. Kovꢀcs, E. ꢈsz, V. Domokos, W. Holzer, Z. Gyçrgydeꢀk, Tetra-
hedron 2001, 57, 4609–4621; b) L. Kovꢀcs, I. Pintꢄr, A. Messmer,
Carbohydr. Res. 1985, 141, 57–65; c) L. Kovꢀcs, I. Pintꢄr, A. Messm-
er, Carbohydr. Res. 1987, 166, 101–111.
[58] a) I. C. Schilling, N. Jung, M. Biskup, U. Schepers, S. Brꢉse, Chem.
Soc. Rev. 2011, 40, 4840–4871; b) S. S. van Berkel, M. B. van Eldijk,
J. C. M. van Hest, Angew. Chem. 2011, 123, 8968–8989; Angew.
Chem. Int. Ed. 2011, 50, 8806–8827.
[59] M. B. Soellner, A. Tam, R. T. Raines, J. Org. Chem. 2006, 71, 9824–
9830.
Biol. 2006, 13, 825–837; k) E. C. Dykhuizen, J. F. May, A. Tongpe-
nay, L. L. Kiessling, J. Am. Chem. Soc. 2008, 130, 6706–6707;
l) E. C. Dykhuizen, L. L. Kiessling, Org. Lett. 2009, 11, 193–196;
m) N. Veerapen, Y. Yuan, D. A. R. Sanders, B. M. Pinto, Carbohydr.
Res. 2004, 339, 2205–2217.
[
38] a) D. J. Owen, C. B. Davis, R. D. Hartnell, P. D. Madge, R. J. Thom-
son, A. K. J. Chong, R. L. Coppel, M. von Itzstein, Bioorg. Med.
Chem. Lett. 2007, 17, 2274–2277; b) C. B. Davis, R. D. Hartnell,
P. D. Madge, D. J. Owen, R. J. Thomson, A. K. J. Chong, R. L.
Coppel, M. von Itzstein, Carbohydr. Res. 2007, 342, 1773–1780.
39] F. Nisic, A. Bernardi, Carbohydr. Res. 2011, 346, 465–471.
40] A. Stimac, J. Kobe, Carbohydr. Res. 1992, 232, 359–365.
41] G. J. F. Chittenden, Carbohydr. Res. 1972, 25, 35–41.
[
[
[
[
[60] Y. He, R. J. Hinklin, J. Chang, L. L. Kiessling, Org. Lett. 2004, 6,
4479–4482.
42] B. L. Kam, J. L. Barascut, J. L. Imbach, Carbohydr. Res. 1979, 69,
1
35–142.
[61] F. Nisic, Ph.D. Thesis, Universitꢀ degli Studi di Milano (Italy), 2010.
[62] M. A. Bonache, F. Nuti, A. Le Chevalier Isaad, F. Real-Fꢄrnandez,
M. Chelli, P. Rovero, A. M. Papini, Tetrahedron Lett. 2009, 50,
4151–4153.
[
[
43] L. Baldoni, C. Marino, J. Org. Chem. 2009, 74, 1994–2003.
44] Z. Gyçrgydeꢀk, I. Ling, R. Bognꢀr, Liebigs Ann. Chem. 1983, 279–
2
89.
[
45] A. Bertho, Justus Liebigs Ann. Chem. 1949, 562, 229–239;
N. S. A. M. Khalil, Nucleos. Nucleot. Nucl. 2005, 24, 1277–1287.
46] a) T. B. Rauchfuss, Inorg. Chem. 1977, 16, 2966–2968; b) H. Luo, I.
Setyawati, S. J. Rettig, C. Orvig, Inorg. Chem. 1995, 34, 2287–2299.
[63] A. Brandstrom, B. Lamm, I. Palmertz, Acta Chem. Scand. Ser. B
1974, 28, 699–701.
[
Received: January 28, 2012
Published online: && &&, 0000
&
12
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Synthesis of a- and b-Glycofuranosyl Amides by Staudinger Ligation
FULL PAPER
Carbohydrates
F. Nisic, G. Speciale,
A. Bernardi* .................. &&&& — &&&&
Stereoselective Synthesis of a- and b-
Glycofuranosyl Amides by Traceless
Ligation of Glycofuranosyl Azides
Without a trace: A highly stereoselec-
tive synthesis of both a- and b-glyco-
furanosyl amides based on the trace-
less Staudinger ligation of glycofurano-
syl azides of the galacto, ribo, and ara-
bino series with 2-diphenylphosphanyl-
phenyl esters is reported (see scheme).
The process does not depend on the
anomeric configuration of the starting
azide but appears to be controlled by
the C2 configuration and by the pro-
tection/deprotection state of the sub-
strates.
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
13
&
ÞÞ
These are not the final page numbers!