Synthesis of galactofuranose-containing disaccharides using thioimidoyl-type donors

Article abstract of DOI:10.1016/j.carres.2006.10.001

Four galactofuranose-containing disaccharides have been prepared utilising various thioimidates [Galf-SC({double bond, long}NR)XR′] and suitably protected acceptors as key precursors. We observed that the efficiency of the coupling reactions was particula

Full text of DOI:10.1016/j.carres.2006.10.001

Carbohydrate Research 341 (2006) 2759–2768
Synthesis of galactofuranose-containing disaccharides using
thioimidoyl-type donors
ꢀ
*
`
Ronan Euzen, Vincent Ferrieres and Daniel Plusquellec
Ecole Nationale Supe´rieure de Chimie de Rennes, UMR CNRS 6226, Avenue du Ge´ne´ral Leclerc, F-35700 Rennes, France
Received 21 July 2006; received in revised form 28 September 2006; accepted 2 October 2006
Available online 7 October 2006
Abstract—Four galactofuranose-containing disaccharides have been prepared utilising various thioimidates [Galf-SC(@NR)XR0]
and suitably protected acceptors as key precursors. We observed that the efficiency of the coupling reactions was particularly depen-
dent on the aglycon present on the furanosyl donor when copper(II) ions were used as the promoter, and that activation could be
correlated with the nature of the third heteroatom, X.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Keywords: Galactofuranose-containing disaccharides; Glycosylation reactions; Thioimidates
1. Introduction
for the increasing attention that has been paid to new
chemical syntheses of oligofuranosides of varying com-
plexities.6–15
In mammalian biomolecules, hexoses are exclusively
found in the pyranose form. Nevertheless, numerous ef-
forts in the structural elucidation of compounds isolated
from various microorganisms have clearly established
the presence of hexofuranosides in some oligo- and
polysaccharides.1–4 Consequently, it is legitimate to
wonder about the gain for such organisms to biosynthe-
size hexofuranose-containing glycoconjugates consider-
ing that furanosides are generally less stable than their
pyranosidic counterparts.5 Moreover, many of these
microorganisms are pathogenic; however, the role of
hexofuranosides in the development of diseases and/or
the way in which they interact with receptors are not
yet clearly elucidated.6 Because mammalian cells are de-
void of the ability to biosynthesize and/or to metabolize
glycofuranosides, such molecules are of great interest to
increase knowledge relating to their biological and phys-
icochemical roles, and the biological mechanisms con-
nected with them. This observation favourably argues
Among hexoses present in a furanose form, three nat-
ural conjugates have been found containing D-glucose
(D-Glc)16–18 and one with D-mannose (D-Man).19 How-
ever, D-fucose (D-Fuc),20,21 (not L as in mammalian
cells) and D-galactose (D-Gal) are the most widely dis-
tributed. The D-Galf residue (Table 1) can be connected
either to pyranosyl units, that is, D-Manp,22,23 D-
Glcp,24,25 D-Galp,21 N-acetyl-D-galactosamine (D-Galp-
NAc)26–28 or glucosamine (D-GlcpNAc),29 L-rhamnose
(L-Rhap),30–32 D-fructose (D-Frup)33 or to galactofur-
anosyl entities.29,34,35,23 Among these natural products,
most of the Galf units are connected with their aglycon
through b-D-(1!6) or b-D-(1!5) linkages (Table 1).
In this context, and considering our interest in the
preparation of hexofuranose-containing derivatives,9–11
disaccharides 1–4 were selected as targets of biological
interest (Fig. 1). The sequence b-D-Galf-(1!6)-a-D-Glcp
as in 1 is a subunit of the O-antigenic glycoside pro-
duced by the K-12 strain of Escherichia coli,24 a bacteria
that has been commonly used for genetic studies. The
second saccharidic structure b-D-Galf-(1!6)-a-D-Manp,
which alternatively presents furanosyl and pyranosyl
entities (2) was designed as a building block for further
*
Corresponding author. Tel.: +33 223 238 058; fax: +33 223 238
046; e-mail: vincent.ferrieres@ensc-rennes.fr
ꢀ Present address: University of Bern, Department of Chemistry and
Biochemistry, Freiestrasse 3, 3012 Bern, Switzerland.
0008-6215/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2006.10.001

2760
R. Euzen et al. / Carbohydrate Research 341 (2006) 2759–2768
Table 1. Selection of microorganisms biosynthesizing oligofuranosides
R
OH
O
O
β-D-Galf-(1→R
OH
OH
OH
6)-D-Galf
5)-D-Galf
Arachniotus
Bionectria
Hypocrea
Aspergillusa
Actinobacillusa
Aspergillusa
Byssochlamys
Cryphonectriaa
Eupenicillium
Hypocrea
Arachniotus
Bionectria
Cryphonectriaa
Calonectria
Chaetosartorya
Discula
Discula
Fusarium
Gibberella
Myrothecium
Penicillium
Talaromyces
Mycobacteriuma
Renibacteriuma
Renibacteriuma
Trichoderma
Gymanascella
Mycobacteriuma
Penicillium
Nectria
Talaromyces
Trichoderma
6)-D-Glcp
6)-D-Manp
Escherichia coli K12
Aspergillusa
Epidermophyton
Paraccidioidesa
Trichophyton
Microsporum
Rocella decipens
a These microorganisms are potentially pathogenic.
OMe
SPh
OH
HO
HO
BnO
O
BnO
O
O
OBn
OH
OH
O
O
O
1
2
β-D-Gal
f
-(1→6)-α-D-Glcp
β-D-Gal
f
-(1→6)-α-D-Manp
OH
OH
OH
OH
OH
from
from
Escherichia coli (K-12 strain)
OH
Paracoccidiodes brasiliensis
OH
O
OH
O
OC8H17
OC8H17
OH
OH
O
OH
OH
OH
O
O
HO
O
OH
3
4
OH
OH
β-D-Gal
f
-(1→6)-β-D-Gal
from
Mycobacterium tuberculosis
f
β-D-Gal
f
-(1→5)-β-D-Galf
OH
OH
OH
from
Mycobacterium tuberculosis
Figure 1. Structure of targeted galactofuranose-containing disaccharides 1–4.
elongation to yield oligosaccharides found in Paracoc-
cidioides brasiliensis.36 This pathogenic fungus is known
to affect the lungs, lymphoid and mucocutaneaous
tissues of rural farm workers in South and Central
America. The glycosidic parts of digalactofuranosides
3 (b-D-Galf-(1!6)-a-D-Galf) and 4 (b-D-Galf-(1!5)-a-
D-Galf) are found in the cell wall that ensures the
membrane stability of Mycobacterium tuberculosis,30
the lethal bacteria responsible for the death of more
than 2 million people each year.
ation reactions based on the use of galactofuranosyl thio-
imidates as furanosyl donors. This family of donors
belongs to the thioglycoside-type donors but they poten-
tially differ from more standard thioglycosides because
they can be activated by ‘remote activation’ (Fig. 2).37,38
The first outcome of this mechanistic feature is that
some differences in the reactivity of the donors can be
connected with the structure of the aglycon and/or the
nature of the promoter. This is the reason why we have
selected thioimidoyl furanosides 5–8 as donors (Fig. 3).
These compounds differ from each other in (i) the nature
of the X group (X = NH or S, the latter having ‘softer’
These four disaccharides were also chosen as model
compounds to probe the structural effects in glycosyl-

R. Euzen et al. / Carbohydrate Research 341 (2006) 2759–2768
2761
X
X
OAc
O
N
OAc
O
OAc
O
S
S
OR
N
ROH
M(OTf)n
Mn+
Glycosylation
reaction
Remote
activation
OAc
OAc
OAc
OAc
OAc
OAc
OAc
OAc
OAc
Figure 2. Expected activation of thioimidates.
S
S
HN
N
OAc
O
OAc
O
OAc
O
OAc
O
N
N
S
N
S
S
S
N
OAc
OAc
OAc
OAc
OAc
OAc
OAc
OAc
OAc
OAc
OAc
OAc
5
6
7
8
Pyrimidinyl
derivative
Benzimidazolyl
derivative
Benzothiazolyl
derivative
Thiazolinyl
derivative
Figure 3. Structure of donors 5–8.
basic properties compared to the amino group, as based
upon the HSAB principle), (ii) the basicity of the hetero-
cycle and (iii) the size of the aglycon.
diastereocontrol of the Ferrier reaction. Better ratios
favouring the expected b-anomer were obtained with a
nitrogen atom (a/b = 0:1) instead of a sulfur (1:4.2 <
a/b < 1:6.7). Better selectivities were nevertheless
observed by reducing the reaction time but this modifi-
cation also resulted in lower yields. However, owing to
the presence of the 2-O-participating group, this para-
meter was not detrimental for further glycosylation
reactions.
On the another hand, acceptors 10–13 (Fig. 4) are
known compounds and were prepared as previously
described (10,5,47 11,48 12,49 1350). Nevertheless, an
interesting optimization for the synthesis of 12 was per-
formed. The preparation of the latter began (Scheme 2)
with the regioselective tritylation of an anomeric mix-
ture of the well-known octyl galactofuranoside
14a,b51,52 by trityl chloride (TrCl) in the presence of
N,N-dimethyl-4-aminopyridine (DMAP). Unfortu-
nately, a disappointing reaction was observed under
these conditions (yields lower than 5%). To overcome
this limitation, the desired etherification was performed
Moreover, because thioimidates are likely to be acti-
vated not directly at the anomeric sulfur atom, orthogo-
nality with thioethyl or thiophenyl glycosides can be
expected.39,40 We describe herein the synthesis of natu-
rally occurring galactofuranose-containing disaccha-
rides and our evaluation of four galactofuranosyl
thioimidates in reactions with acceptors of different ste-
ric accessibilities, for example, primary versus secondary
hydroxyl groups or furanosyl versus pyranosyl entities.
2. Results and discussion
To prepare 1,2-trans furanosides 1–4, the b-directing ef-
fect was achieved through the assistance of a 2-O-acetyl
protecting group on the galactosyl donor. Consequently,
the required galactofuranosides were obtained from per-
acetylated galactofuranose (9)41 according to a Ferrier
reaction42 catalyzed by the boron trifluoride-etherate
complex. Excess catalyst was used to shift the thione/
thiol equilibrium towards the thiol form and to limit
the formation of the N-glycoside as a by-product.43 Do-
nors 5–744 were thus obtained in good to excellent yields
(Scheme 1). This synthesis was also extended to the new
derivative 8,43 because the thiazolinyl group was used as
the versatile leaving group in 1,2-cis glycosidation reac-
tions of pyranosyl donors.39,45,46 In the preparation of
compounds 5–8, we observed that the nature of the third
heteroatom X of the thioimidoyl residue influenced the
5
a
OAc
O
b
6
OAc
c
7α,β
8α,β
d
OAc
OAc
OAc
9
Scheme 1. Synthesis of thioimidates 5–8. Reagents and conditions: (a)
(99%, a/b = 0:1);44 (b) (91%, a/b = 0:1);44 (c) (83%, a/b = 1:6.7);44 (d)
2-mercaptothiazoline, BF3ÆOEt2 (93%, a/b = 1:4.2).

2762
R. Euzen et al. / Carbohydrate Research 341 (2006) 2759–2768
OC8H17
OC8H17
OBn
O
OBn
O
OBn
O
HO
BnO
BnO
HO
BnO
BnO
O
OBn
OBn
OBn
OBn
OH
OMe
SPh
OH
OBn
13
10
11
12
Figure 4. Structures of required acceptors 10–13.
disappearance of the donor, thus showing its activation,
with the Lewis acids trimethylsilyl trifluoromethanesulf-
onate (TMSOTf) and copper(II) trifluoromethanesulfo-
nate. These results also showed the significant influence
of the nature of the leaving group because the thiazol-
inyl-type donors 7 and 8 (entries 8–11, X = S) demon-
strated very poor activity in the glycosylation of 10
while the desired disaccharide 17 was isolated in good
58% and 67% yield starting from the pyrimidinyl and
R1
R2
OR2
OC8H17
O
14β
15a
15b
16
H
MMTr
Tr
H
H
a
b
H
2OR2
OR
c
Tr
Bn
Bn
OR1
d
12
H
Scheme 2. Optimized synthesis of acceptor 12. Reagents: (a) MMTrCl,
DMAP (64%); (b) TrCl–DMAP salt (80%); (c) BnBr, NaH (99%); (d)
AcOH, H2O (78%).
benzimidazolinyl derivatives
6
(X = NH) and
5
(X = N), respectively (entries 2, 4, 5 and 7). Moreover,
the most efficient coupling was obtained with the more
thiophilic copper(II) salt (entries 1/2, 6/7 and 10/11);
adding only one molar equivalent of this promoter re-
sulted in a significantly decreased yield (entries 2/3). It
is also interesting to note that the modification of the
initial equimolar donor/acceptor ratio, whatever the
sense, had a positive influence on the yield of the reac-
tion as the desired coupling product 17 was isolated,
after deacetylation under Zemplen transesterification,
in ꢀ75% yield (entries 4 and 5). Further removal of
the benzyl groups by hydrogenolysis afforded the target
disaccharide 1 in 97% yield.
This preliminary study allowed us to select the 2-thio-
pyrimidinyl aglycon as the most appropriate leaving
group in O-glycosylation reactions activated by cop-
per(II) triflate. The use of the thiopyrmidinyl donor
was further successfully extended to the preparation of
the b-D-Galf-(1!6)-a-D-Manp disaccharide 2, using thio-
mannopyranoside 11, which bears a primary hydroxyl,
as the acceptor (Scheme 3). The donor 5 thus smoothly
reacted with 11 and afforded, after deacetylation and
chromatographic purification, the desired disaccharide
2 in 60% yield.
Subsequently, the preparation of the disaccharides 3
and 4, fragments of the glycosidic skeleton of the galac-
tan produced by M. tuberculosis, required the use of gly-
cosyl acceptors in a furanose form. These compounds
are well known to be very sensitive to conformational
changes not only because the five-membered ring is
more flexible than the pyranosyl counterpart6 but also
because the side arm is less constrained, thus increasing
the number of possible rotamers. Consequently, this
molecular parameter may influence the progress of the
glycosylation reaction. Indeed, modest yields were
obtained for the synthesis of protected disaccharides
20 (30%) and 21 (14%) (Scheme 4) and the yields could
from the pure 1,2-trans-furanoside 14b, which could be
selectively crystallized from 14a,b in diethyl ether.52
To explain the apparent inhibition of the Williamson
etherification by the 1,2-cis-anomer, we assumed that
the standard tritylation reagent was not reactive enough,
and more importantly, that the nucleophilic base
DMAP could engage in competitive reactions, that is,
(i) nucleophilic activation of TrCl and (ii) possible
deprotonation of the more acidic and more hindered
2-hydroxyl function in the 1,2-cis galactoside 14a. Such
a property, which proved to be of particular interest in
selective 2-O-protection,53 was suspected to be a factor
that disfavoured the protection of the primary hydroxyl.
On this basis, the introduction of the trityl group was
first attempted starting from the pure 14b and the more
reactive p-methoxyphenyldiphenyl chloride (MMTrCl).
The target product 15a was then isolated in an increased
64% yield after only 4 h at room temperature and chro-
matographic purification. Secondly, the use of the elec-
trophilic salt previously prepared from trityl chloride
and DMAP54 also gave increased performance because
after 24 h at reflux in dichloromethane, the triol 15b
was obtained in a good and reproducible 80% yield on
a multi-gram scale (20 g). Subsequent benzylation of
15b under standard conditions followed by detritylation
in aqueous acetic acid afforded the desired compound 12
in 77% overall yield. This approach is highly competitive
with those proposed by Plusquellec et al.49 or Reynolds
and co-workers.50
For subsequent glycosidation of donors 5–8, we were
first interested in synthesizing the b-D-Galf-(1!6)-a-D-
Glcp disaccharide residue found in E. coli (K-12 strain)
and particularly focused our attention on the nature of
the promoter (Table 2). To ensure selective activation
of the thioimidoyl donor, no halonium ion was used
as promoter. After experimentation, we observed the

R. Euzen et al. / Carbohydrate Research 341 (2006) 2759–2768
Table 2. Model glycosidic coupling for selection of the most appropriate donor
2763
OMe
X
OR
O
RO
RO
OAc
O
N
S
HO
BnO
O
1) Promoter
OH
+
O
BnO
O
2) NaOMe, MeOH
OBn
OAc
OMe
OAc
R
OAc
5-8
10
OH
OH
OH
17 Bn
H2, Pd(OAc)2
(97%)
1
H
Entry
Donor
Aglycon
Donor/acceptor
Promoter (equiv)
17 (overall yield, %)
1
2
3
4
5
5
5
5
5
5
1:1
1:1
1:1
1:2
2:1
TMSOTf (2)
Cu(OTf)2 (2)
Cu(OTf)2 (1)
Cu(OTf)2 (2)
Cu(OTf)2 (2)
18
67
23
74
77
N
S
N
N
HN
6
7
6
6
1:1
1:1
TMSOTf (2)
Cu(OTf)2 (2)
nd
58
S
S
S
8
9
7
7
1:1
1:1
TMSOTf (2)
Cu(OTf)2 (2)
nd
nd
N
N
S
S
10
11
8
8
1:1
1:1
TMSOTf (2)
Cu(OTf)2 (2)
18
24
nd: not determined.
SPh
5
a
b
BnO
O
BnO
O
OR
OBn
a
O
C8H17
C8H17
5 + 11
OR
OBn
O
O
OR
R
O
O
R
OR
OR
19 Ac
b
OH
2
H
OR
O
OBn
OR
OH
O
BnO
Scheme 3. Orthogonal glycosylation. Reagents: (a) Cu(OTf)2; (b)
O
O
NaOMe, MeOH (60% for two steps).
OH
21
OH
OH
20 Bn
3
c
OH
OH
OH
H
not be improved even with a more standard promoter
such as NIS/TMSOTf. A final debenzylation performed
in a hydrogen atmosphere catalyzed by palladium(II)
acetate afforded 3 in good yield.
As expected, the results presented here suggested that
the reactions were dependant on the nature of the elec-
trophilic and nucleophilic partners and also on the pro-
moter. We first observed a better reactivity for primary
hydroxyl groups and showed that the higher yields of
coupling were obtained with pyranosyl acceptors versus
furanosyl ones. Secondly, the glycosylation could be
achieved due to activation of the donor, that is, interac-
Scheme 4. Synthesis of digalactofuranosides 3 and 21. Reagents: (a) (i)
12, Cu(OTf)2; (ii) NaOMe, MeOH (30% for two steps); (b) 13,
Cu(OTf)2; (ii) NaOMe, MeOH (14% for two steps); (c) H2, Pd(OAc)2
(3: 67% for two steps).
tion between the aglycon and the reactive catalytic spe-
cies, Cu2+. On the basis of the pKa of the free aglycons
(Table 3), the activation process could not be simply as-
cribed to this parameter because donors 5 and 8b have
two heterocycles with similar acidity but a significantly

2764
R. Euzen et al. / Carbohydrate Research 341 (2006) 2759–2768
Table 3. Selected physicochemical properties of donors 5–8
thiofuranoside 5 in O-glycosylation reactions with sac-
charide acceptors. Second, it was also very surprising
to notice the low reactivity of the disarmed donor 8
belonging to the thiazolinyl family, although such do-
nors were presented as an interesting alternative to stan-
dard glycosylation using either armed or disarmed
thioglycopyranosides as an activator.40,55 As a conse-
quence, the nature of the interactions between the acti-
vating cation and the thioimidoyl function was
suspected to be responsible for these data. Indeed, in
the present study, the borderline Lewis acidic cupric cat-
ions, according to the hard and soft acid/base concept,
could interact with three different centres: the exocyclic
sulfur atom, the nitrogen atom and the X group
(X = S or NH) of the heterocycle. A computational ap-
proach of the relative charges on atoms possibly in-
volved in the glycosylation process gave the values
summarized in Table 4. The resulting data showed that
similar values were obtained for all donors 5–8 for S-1,
N-1 and C-1. On the assumption that thiophenylglyco-
sides are not activated by cupric ions, it was expected
that the exocyclic sulfur S-1 was not involved in the acti-
vation process, thus corroborating a remote activation
mechanism. The main difference was however obtained
for the X atom, because the charge on S-2 in donors 7
and 8 was significantly lower than that calculated for
N-2 in 5 and 6. We then hypothesized that a stronger
interaction between S-2 and Cu2+ was not able to really
activate the furanosyl donor but may be also responsible
for the observed lack of reactivity using 7 and 8.
a
Entry Donor Heterocycle
pKa
dH-1
dC-1
(ppm) (ppm)
N
1
2
5
6
1.65b
6.47
87.5
89.2
S
S
N
N
HN
0.31b 5.94
ꢁ3.17b 6.27
1.28b 5.39
S
S
3
7b
8b
89.3
88.4
N
N
S
S
4
a The pKa values correspond to the unglycosylated heterocycles.
b The pKa values were obtained from ‘calculated properties’ for each
compound in SciFinder Scholar v.2006.
different reactivity (Table 2, entries 2 and 11). On the
other hand, Hanessian et al. could correlate the high gly-
cosylating ability of MOP hexopyransides to upfield and
downfield chemical shifts for 1H and 13C, respectively.37
Unfortunately, such a correlation could not explain the
results observed with thioimidates 5–8 (Table 3).
To rationalize our results, two main sets of experi-
ments, already published, have also to be considered.
First, by comparison with the phosphorylation of thio-
imidates of the same series44 for which we had con-
cluded better suitability between phosphoric acid and
the benzimidazolyl derivative, it would have been diffi-
cult to predict the higher efficiency of the 2-pyrimidinyl
To probe this hypothesis, two supplementary reac-
tions were further performed (Scheme 5). Thus, the
Cu2+-activated glycosylation of n-octanol was carried
out using either thiazolinyl derivative 8 in the presence
of a large excess of copper(II) triflate or donor 5 in the
presence of diphenylsulfide as a scavenger of cupric cat-
Table 4. Selective data for galactofuranosyl donors 5–8
Entry
Donor
Heterocycle
C-1 anomeric
0.365
S-1
N-1
X
2
N
1
1
1
5
ꢁ0.462
ꢁ0.218
ꢁ0.230 (N-2)
S
N
2
HN
2
3
4
6
0.348
0.368
0.368
ꢁ0.430
ꢁ0.405
ꢁ0.396
ꢁ0.274
ꢁ0.203
ꢁ0.210
ꢁ0.173 (N-2)
ꢁ0.394 (S-2)
ꢁ0.385 (S-2)
1
N
1
S
2
S
1
7b
8b
1
S
N
2
S
1
N
1
S

2,3,4-tri-O-benzyl-a-D
-glucopyranoside (10) was ob-
R. Euzen et al. / Carbohydrate Research 341 (2006) 2759–2768
2765
N
tained as described by Reitz et al.56 Reactions were
monitored by TLC (E. Merck 60 F254 Silica Gel non
activated plates) and compounds were visualized by
UV-absorption and/or charring with a 5% solution of
H2SO4 in EtOH. Preparative chromatography was con-
ducted on Geduran Si 60 (40–63 lm) Silica Gel. Melting
points were determined on a Reichert microscope and
are uncorrected. Optical rotations were measured on a
Perkin–Elmer 341 polarimeter. 1H, 13C, HMBC,
HMQC and COSY NMR spectra were recorded at the
ENSCR on a Bruker ARX 400 spectrometer at
400 MHz for 1H and 100 MHz for 13C analyses. 1D
and 2D spectra of unprotected disaccharides were re-
corded on a Bruker Avance 500 spectrometer at
C8H17OH
Cu(OTf)2 (2 equiv.)
Ph2S
OAc
O
S
N
(traces)
5
OC8H17
OAc
O
OAc
OAc
OAc
OAc
O
S
OAc
OAc
N
S
C8H17OH
Cu(OTf)2 (3 equiv.)
(20%)
OAc
22
OAc
OAc
8
OAc
Scheme 5. Influence of additive diphenylsulfide or excess Cu(OTf)2 on
glycosidation of 5 and 8, respectively.
1
500 MHz for H and 125 MHz for 13C, by the ‘Service
ions. The resulting peracetylated octyl galactofuranoside
22 was thus isolated in moderate 20% yield, close to that
obtained for 17 under similar conditions. However, the
use of diphenylsulfide really inhibited the desired cou-
pling since only traces of 22 were observed by monitor-
ing the reaction by thin layer chromatography. This
result therefore support the notion that a sulfur atom
present on the heterocycle could modulate the activation
´
de Spectrometrie de RMN Haut Champ du Centre
´
Regional de Mesures Physiques de l’Ouest (CRMPO,
Rennes, France)’. Chemical shifts are given in d units
measured from the solvent signal. The HRMS were per-
formed at the CRMPO with a MS/MS ZabSpec TOF
Micromass by ESI in positive mode. Microanalyses were
performed by the ‘Service de Microanalyse de l’ICSN
(Gif sur Yvette, France)’ and by the CRMPO.
of thioimidates by Cu2+
.
In conclusion, the synthesis of relevant galactofura-
nose-containing disaccharides was achieved through
the remote activation of four furanosyl thioimidates
using copper(II) triflate as an activator. The large hetero-
geneousness of the results was first connected with the
size of the glycosyl acceptors (pyranosyl vs furanosyl)
and was secondly ascribed to the presence on the hetero-
cyclic aglycon of an endocyclic sulfur atom or amino
group. Better coordination of the cupric cations by this
sulfur than by a nitrogen atom manifested in a lack of
reactivity of the thiazole-like donors. Further studies
are currently underway to investigate the potential
orthogonality of these families of thioimidates in the
presence of various promoters.
3.2. Thiazolin-2-yl 2,3,5,6-tetra-O-acetyl-1-thio-D-
galactofuranoside (8)
To a solution of 1,2,3,5,6-penta-O-acetyl-D-galactofura-
nose (9)41 (2.00 g, 5.12 mmol) in CH2Cl2 (82 mL) were
added 2-mercaptothiazoline (1.83 g, 15.37 mmol) and
BF3ÆOEt2 (5.8 mL, 46.08 mmol). The solution was stir-
red at rt for 24 h and was diluted with CH2Cl2
(80 mL). The reaction mixture was washed with a satd
aq NaHCO3 (3 · 50 mL) and with H2O (3 · 100 mL).
The aq layers thus obtained were extracted with CH2Cl2
(3 · 50 mL) and the combined organic layers were dried
(MgSO4) and finally concentrated. Purification of the
residue by column chromatography (9:1, CH2Cl2–
EtOAc) as eluent gave 8 in anomeric mixture (a/b =
1:4.2) as a colourless oil (2.15 g, 93%): TLC (9:1,
CH2Cl2–EtOAc): Rf = 0.3; 8a: 1H NMR (CDCl3): d
6.40 (d, 1H, J1,2 = 5.1 Hz, H-1), 5.51 (dd, 1H,
J2,3 = 3.3 Hz, H-2), 5.33 (dt, 1H, J4,5 = 7,6 Hz,
J5,6a = J5,6b = 4,3 Hz, H-5), 5.23 (dd, 1H, J3,4 = 3.8 Hz,
H-3), 4.38–4.11 (m, 5H, H-4, H-5a0, H-5b0, H-6a, H-
3. Experimental
3.1. General methods
All reactions were performed at rt under N2 atmosphere.
Pyrimidin-2-yl 2,3,5,6-tetra-O-acetyl-1-thio-b-D-galacto-
furanoside (5),44 benzimidazol-2-yl 2,3,5,6-tetra-O-acet-
yl-1-thio-b-D-galactofuranoside (6),44 benzothiazol-
2-yl 2,3,5,6-tetra-O-acetyl-1-thio-b-D-galactofuranoside
(7)44 and octyl 2,3,5-tri-O-benzyl-b-D-galactofuranoside
(12)49 were synthesized as previously described. Octyl
2,3,6-tri-O-benzyl-b-D-galactofuranoside (13) was ob-
tained as described by Pathak et al.50 Phenyl 2,3,4-tri-
3
6b), 3.40 (t, 2H, J = 8.0 Hz, H-4a0, H-4b0), 2.15, 2.13,
2.09, 2.04 (4s, 12H, CH3CO); 13C NMR (CDCl3): d
170.6, 170.1, 169.6, 169.3 (C@O), 163.3 (C-20), 85.9 (C-
1), 81.7 (C-4), 76.4 (C-2), 76.2 (C-3), 69.5 (C-5), 64.3
(C-50), 62.6 (C-6), 35.6 (C-40), 21.0, 20.8, 20.7, 20.6
1
(CH3CO); 8b: H NMR (CDCl3): d 6.21–6.20 (m, 1H,
H-1), 5.39 (dt, 1H, J4,5 = 7.4 Hz, J5,6a = J5ꢁ6b = 4.1 Hz,
H-5), 5.28 (t, 1H, J1,2 = J2,3 = 1.8 Hz, H-2), 5.10 (ddd,
1H, J3,4 = 4.8 Hz, H-3), 4.38–4.35 (m, 1H, H-4), 4.36
(dd, 1H, J6a,6b = 12.0 Hz, H-6a), 4.28, 4.19 (2dt, 2H,
O-benzyl-1-thio-a-D-mannopyranoside (11) was synthe-
48
´
sized as reported by Martın-Lomas et al. Methyl

2766
R. Euzen et al. / Carbohydrate Research 341 (2006) 2759–2768
3
2J = 14.8 Hz, J = 7.9 Hz, H-5a0, H-5b0), 4.18 (dd, 1H,
H-6b), 3.40 (t, 2H, H-4a0, H-4b0), 2.12, 2.12, 2.11, 2.05
(4s, 12H, CH3CO); 13C NMR (CDCl3): d 170.6, 170.1,
169.7, 169.4 (C@O), 162.8 (C-20), 88.4 (C-1), 81.7
(C-4), 80.9 (C-2), 76.4 (C-3), 69.2 (C-5), 64.3 (C-50),
62.6 (C-6), 35.4 (C-40), 20.9, 20.8, 20.8, 20.7 (CH3CO);
HRMS Calcd for C17H23NaNO9S2 [M+Na]+:
472.0712. Found: 472.0712; Anal. Calcd for
C17H23NO9S2: C, 45.43; H, 5.16; N, 3.12. Found: C,
45.09; H, 5.15; N, 3.16.
3.62–3.57 (m, 3H, H-6b, H-6a0, H-6b0), 3.54 (dd, 1H,
H-2), 3.51 (dd, 1H, J4,5 = 9.9 Hz, H-4), 3.38 (s, 3H,
OCH3); 13C NMR (CD3OD): d 140.1, 139.7, 139.6
(Cipso), 129.5, 129.4, 129.3, 129.2, 129.0, 128.9, 128.7,
128.6 (C6H5), 109.8 (C-10), 99.1 (C-1), 84.3 (C-40), 83.2
(C-20), 83.0 (C-3), 81.4 (C-2), 79.0 (C-4), 78.6 (C-30),
76.5, 76.0, 74.0 (OCH2Ph), 72.3 (C-50), 71.5 (C-5), 67.6
(C-6), 64.6 (C-60), 55.6 (OCH3); HRMS Calcd for
C34H42KO11 [M+K]+: 665.2364. Found: 665.2365.
3.5. Phenyl b-D-galactofuranosyl-(1!6)-2,3,4-tri-O-
benzyl-1-thio-a-D-mannopyranoside (2)
3.3. General procedure A: glycosylation reactions
To
a
solution of thioimidoyl galactofuranoside
This disaccharide was obtained according to the general
procedure A from the thiogalactofuranoside 5 (68 mg,
0.15 mmol) and the mannopyranosidic acceptor 11
(168 mg, 0.31 mmol). The acetylated intermediate 19
was purified by chromatography (4:1!3:2, light petro-
leum–EtOAc). The fractions containing this product
[TLC (4:1, light petroleum–EtOAc): Rf = 0.1] were
(0.23 mmol) in CH2Cl2 (10 mL/g) were successively
added the appropriate acceptor (0.12 mmol or
0.46 mmol; depending on the acceptor) and Cu(OTf)2
(163 mg, 0.45 mmol). The mixture was stirred at rt for
a given period and the reaction was quenched by addi-
tion of a few drops of Et3N. Insoluble materials were re-
moved by filtration and the filtrate was concentrated to
dryness. The residue thus obtained was then purified by
chromatography on silica gel. The fractions containing
the target protected disaccharide were pooled, concen-
trated under reduced pressure and diluted in dry MeOH
(20 mL). A 0.1 M solution of NaOMe in MeOH
(2.3 mL, 0.23 mmol) was then added and the mixture
was stirred at rt overnight. The reaction was quenched
by adding a few drops of AcOH and the solution was
concentrated under reduced pressure. The partially
deprotected disaccharide was finally recovered by chro-
matographic purification.
´
pooled, concentrated and subjected to Zemplen deacet-
ylation. The disaccharide 2 thus generated was purified
by flash chromatography (19:1, CH2Cl2–MeOH). This
compound was then isolated as a colourless oil (65 mg,
20
60%): ½aꢂD +30.2 (c 1.0, MeOH); TLC (19:1, CH2Cl2–
MeOH): Rf = 0.2; 1H NMR (CD3OD): d 7.48–7.45,
7.39–7.26 (2 m, 20H, C6H5), 5.47 (d, 1H, J1,2 = 1.5 Hz,
H-1), 4.91–4.86, 4.70–4.54 (2 m, 6H, OCH2Ph), 4.87–
4.86 (m, 1H, H-10), 4.20 (ddd, 1H, J5,6b = 5.6 Hz,
J5,6a = 1.8 Hz, H-5), 4.05 (dd, 1H, J2,3 = 3.0 Hz, H-2),
4.00–3.94 (m, 2H, H-20, H-30), 3.96–3.92 (m, 2H, H-40,
H-6a), 3.92 (dd, 1H, J4,5 = 9.6 Hz, H-4), 3.82 (dd,
0
1H, J3,4 = 9.2 Hz, H-3), 3.70 (ddd, 1H, J5 ;6a0
¼
J50;6b ¼ 6:1 Hz, J4 ;5 ¼ 3:0 Hz, H-50), 3.64 (dd, 1H,
J6a,6b = 11.2 Hz, H-6b), 3.60–3.57 (m, 2H, H-6a0, H-
6b0); 13C NMR (CD3OD): d 139.8, 139.4, 139.3 (Cipso
OCH2Ph), 135.4 (Cipso SPh), 133.3, 130.2, 129.5, 129.4,
129.2, 129.2, 129.0, 128.9, 128.8, 128.8 (C6H5), 109.6
(C-10), 87.3 (C-1), 84.7 (C-40), 83.0 (C-20), 81.0 (C-3),
79.0 (C-30), 77.4 (C-2), 76.1, 73.2, 72.9 (OCH2Ph), 75.9
(C-4), 73.7 (C-5), 72.4 (C-50), 67.5 (C-6), 64.7 (C-60);
HRMS Calcd for C39H44NaO10S [M+Na]+: 727.2553.
Found: 723.2550.
0
0
0
3.4. Methyl b-D-galactofuranosyl-(1!6)-2,3,4-tri-O-
benzyl-a-D-glucopyranoside (17)
This compound was produced according to the general
procedure A from the pyrimidin-2-yl galactofuranoside
5 (226 mg, 0.51 mmol) and from the glucopyranoside
acceptor 10 (119 mg, 0.26 mmol). After stirring for 2 h
and work-up, the resulting disaccharide was purified
by chromatography using 3:2, light petroleum–EtOAc
as eluant [TLC (3:2, light petroleum–EtOAc):
Rf = 0.4]. The mixture thus obtained was deacetylated
as described. The partially deprotected disaccharide 17
was purified by chromatography (19:1, CH2Cl2–MeOH)
and was then isolated as a white solid (124 mg, 77%): mp
3.6. Octyl b-D-galactofuranosyl-(1!6)-2,3,5-tri-O-benz-
yl-b-D-galactofuranoside (20)
20
98 ꢁC (CH2Cl2–MeOH); ½aꢂD ꢁ1.0 (c 1.1, MeOH); TLC
The disaccharide 20 was synthesized according the gen-
eral procedure A from thioimidoyl galactofuranoside 5
(181 mg, 0.41 mmol) and the galactofuranosidic accep-
tor 12 (459 mg, 0.82 mmol). The acetylated intermediate
was then purified by flash chromatography (4:1!3:2,
light petroleum–EtOAc) [TLC (4:1, light petroleum–
EtOAc): Rf = 0.4]. The resulting crude oil was further
deacetylated as described. The partially benzylated
disaccharide 20 was purified by chromatography on sil-
(19:1, CH2Cl2–MeOH): Rf = 0.1; 1H NMR (CD3OD): d
7.38–7.24 (m, 15H, C6H5), 4.92–4.62 (m, 6H, OCH2Ph),
4.87 (d, 1H, J1 ;2 ¼ 1:8 Hz, H-10), 4.71 (d, 1H,
0
0
0
0
J1,2 = 3.6 Hz, H-1), 4.00 (dd, 1H, J3 ;4 ¼ 6:4 Hz,
J2 ;3 ¼ 0:8 Hz, H-30), 3.97 (dd, 1H, H-20), 3.94 (dd,
0
0
1H, J4 ;5 ¼ 3:1 Hz, H-40), 3.92 (dd, 1H, J6a,6b = 11.2 Hz,
J5,6a = 1.8 Hz, H-6a), 3.87 (dd, 1H, J2,3 = 9.4 Hz,
J3,4 = 9.2 Hz, H-3), 3.74–3.69 (m, 2H, H-5, H-50),
0
0

R. Euzen et al. / Carbohydrate Research 341 (2006) 2759–2768
2767
13
ica gel eluting with 19:1, CH2Cl2–MeOH, and was then
(s, 1H, OCH3); C NMR (D2O): d 107.9 (C-10), 99.3
(C-1), 82.6 (C-40), 81.0 (C-20), 76.6 (C-30), 73.0 (C-3),
71.1 (C-2), 70.7 (C-50), 70.6 (C-5), 69.5 (C-4), 66.7 (C-
6), 62.7 (C-60), 55.2 (OCH3); HRMS Calcd for
C13H24NaO11 [M+Na]+: 379.1216. Found: 379.1219.
20
isolated as a colourless oil (88 mg, 30%): ½aꢂD ꢁ81.3 (c
1.4, MeOH); TLC (19:1, CH2Cl2–MeOH): Rf = 0.2;
1H and 13C NMR spectra were identical to the reported
data;50 HRMS Calcd for C41H56NaO11 [M+Na]+:
747.3720. Found: 747.3721.
3.10. Octyl b-D-galactofuranosyl-(1!6)-b-D-galactofur-
3.7. Octyl b-D-galactofuranosyl-(1!5)-2,3,5-tri-O-benz-
yl-b-D-galactofuranoside (21)
anoside (3)
The unprotected disaccharide 3 could be obtained
according the general procedure B from the partially
benzylated disaccharide 21 (67 mg, 0.09 mmol). The tar-
get glycoconjugate 3 could then be isolated after chro-
matographic purification using 19:1, CH2Cl2–MeOH
This compound was obtained as described in procedure
A from the galactofuranosidic donor 5 (220 mg,
0.50 mmol) and the acceptor 13 (137 mg, 0.24 mmol).
The partially acetylated disaccharide was then isolated
by chromatographic purification (4:1, light petroleum–
EtOAc) [TLC (7:3, light petroleum–EtOAc): Rf = 0.3]
and the fractions containing this product were com-
bined, concentrated and subjected to Zemplen transeste-
rification. The disaccharide 21 was purified by
chromatography on silica gel (19:1, CH2Cl2–MeOH)
and could be finally isolated as a colourless oil (24 mg,
20
as eluent, as a colourless oil (28 mg, 67%): ½aꢂD +12.6
(c 1.5, H2O); TLC (19:1, CH2Cl2–MeOH): Rf = 0.3;
1H and 13C NMR spectra were identical to the reported
data;49,50 HRMS Calcd for C20H38NaO11 [M+Na]+:
477.2312. Found: 477.2292.
20
14%): ½aꢂD ꢁ60.9 (c 1.0, MeOH); TLC (19:1, CH2Cl2–
3.11. Calculations of the charges
MeOH): Rf = 0.2; 1H and 13C NMR spectra were
identical to the reported data;50 HRMS Calcd for
C41H56NaO11 [M+Na]+: 747.3720. Found: 747.3726.
Hyperchem Molecular Visualization and Simulation
software (Hypercube, Waterloo, Ontario, Canada, soft-
ware HyperChem 7.5) was used to calculate the charges
on the interesting atoms. The initial geometries of the
compounds 5–8 were built by means of standard para-
meters, applying an Amber molecular mechanics force
field and then optimized by TNDO geometrical optimi-
zation. The resulting charge values were not considered
individually but were rather used to make comparisons
between similar species. Selected data are summarized
in Table 4.
3.8. General procedure B: debenzylation reactions
AcOH (2.5 mL) and Pd(OAc)2 (5 mg) were added to a
solution of disaccharide (30 mg) in EtOAc (2.5 mL).
After stirring under hydrogen (1 atm) at rt for 72 h,
the mixture was filtered, concentrated under reduced
pressure and traces of AcOH were coevaporated with
toluene (5 · 5 mL). The totally unprotected disaccharide
was then isolated by flash chromatography on silica gel.
3.9. Methyl b-D-galactofuranosyl-(1!6)-a-D-glucopyr-
Acknowledgements
anoside (1)
We are grateful to the French Ministry of Education,
Research and Technology for a grant to R.E. We also
This compound was synthesized according to the gen-
eral procedure B from the benzylated disaccharide 17
(31 mg, 0.05 mmol). The target product 1 was then puri-
fied by chromatography on silica gel (3:3:1, EtOAc–
i-PrOH–H2O), and was isolated as a colourless hygro-
´
thank Martine Lefeuvre and Jean-Paul Guegan
(ENSCR) for recording NMR spectra. Finally, we are
grateful to Richard Daniellou (ENSCR) and Jean-Louis
Reymond (University of Bern) for helpful discussions.
20
scopic foam (17 mg, 97%): ½aꢂD +13.7 (c 0.3, H2O);
1
TLC (3:3:1, EtOAc–i-PrOH–H2O): Rf = 0.4; H NMR
(D2O): d 4.95 (d, 1H, J1 ;2 ¼ 1:8 Hz, H-10), 4.72
0
0
0
0
(d, 1H, J1,2 = 3.8 Hz, H-1), 4.03 (dd, 1H, J2 ;3
¼
References
3:9 Hz, H-20), 3.99 (dd, 1H, J3 ;4 ¼ 6:6 Hz, H-30), 3.93
0
0
(dd, 1H, J6a,6b = 11.0 Hz, J5,6a = 1.4 Hz, H-6a),
1. McConville, M. J.; Homans, S. W.; Thomas-Oates, J. E.;
Dell, A.; Bacic, A. J. Biol. Chem. 1990, 265, 7385–7394.
2. Sprott, G. D. J. Bioenerg. Biomembr. 1992, 24, 555–566.
3. Koga, Y.; Nishihara, M.; Morri, H.; Akagawa-Matsush-
ita, M. Microbiol. Rev. 1993, 57, 164–182.
4. de Lederkremer, R. M.; Colli, W. Glycobiology 1995, 5,
547–552.
5. Green, J. W. Adv. Carbohydr. Chem. 1966, 21, 95–142.
3.91 (dd, 1H, J4 ;5 ¼ 4:1 Hz, H-40), 3.75 (ddd, 1H,
0
0
J5 ;6b ¼ 7:3 Hz, J5 ;6a0 ¼ 4:3 Hz, H-50), 3.70 (ddd,
0
0
0
1H, J4,5 = 9.3 Hz, J5,6b = 5.4 Hz, H-5), 3.66 (dd, 1H,
H-6b), 3.64 (dd, 1H, J6a0;6b ¼ 11:7 Hz, H-6a0), 3.59
0
(dd, 1H, J2,3 = 9.8 Hz, J3,4 = 9.3 Hz, H-3), 3.58 (dd,
1H, H-6b0), 3.49 (dd, 1H, H-2), 3.37 (t, 1H, H-4), 3.34

2768
R. Euzen et al. / Carbohydrate Research 341 (2006) 2759–2768
31. Brennan, P. J.; Nikaido, H. Annu. Rev. Biochem. 1995, 64,
6. Joe, M.; Sun, D.; Taha, H.; Completo, G. C.; Croudace,
J. E.; Lammas, D. A.; Besra, G. S.; Lowary, T. L. J. Am.
Chem. Soc. 2006, 128, 5059–5072.
7. Hederos, M.; Konradsson, P. J. Am. Chem. Soc. 2006,
128, 3414–3419.
29–63.
32. Czaja, J.; Jachymek, W.; Niedziela, T.; Lugowski, C.;
Aldova, E.; Kenne, L. Eur. J. Biochem. 2000, 267, 1672–
1679.
8. Gandolfi-Donadio, L.; Gallo-Rodriguez, C.; de Leder-
kremer, R. M. Can. J. Chem. 2006, 84, 486–491.
33. Zdorovenko, E. L.; Ovod, V. V.; Zatonsky, G. V.;
Shashkov, A. S.; Kocharova, N. A.; Knirel, Y. A.
Carbohydr. Res. 2001, 330, 505–510.
`
9. Gelin, M.; Ferrieres, V.; Lefeuvre, M.; Plusquellec, D.
Eur. J. Org. Chem. 2003, 1285–1293.
34. Sørum, U.; Robertsen, B.; Kenne, L. Carbohydr. Res.
1998, 306, 305–314.
`
10. Gelin, M.; Ferrieres, V.; Plusquellec, D. Eur. J. Org.
Chem. 2000, 1423–1431.
35. Perry, M. B.; MacLean, L. L. Carbohydr. Res. 2004, 339,
1399–1402.
`
11. Ferrieres, V.; Roussel, M.; Gelin, M.; Plusquellec, D. J.
Carbohydr. Chem. 2001, 20, 855–865.
12. McAuliffe, J. C.; Hindsgaul, O. J. Org. Chem. 1997, 62,
1234–1239.
13. Choudhury, A. K.; Roy, N. Carbohydr. Res. 1998, 308,
207–211.
36. Levery, S. B.; Toledo, M. S.; Suzuki, E.; Salyan, M. E. K.;
Hakomori, S. I.; Straus, A. H.; Takahashi, H. K. Biochem.
Biophys. Res. Commun. 1996, 222, 639–645.
37. Hanessian, S.; Bacquet, C.; Lehong, N. Carbohydr. Res.
1980, 80, C17–C22.
14. Marino, C.; Chiocconi, A.; Varela, O.; de Lederkremer,
R. M. Carbohydr. Res. 1998, 311, 183–189.
15. Sarkar, S. K.; Kumar, C. A.; Mukhopadhyay, B.; Roy, N.
J. Carbohydr. Chem. 1999, 18, 1121–1130.
16. Kerr, A.; Htay, H. Physiol. Plant Pathol. 1974, 4, 37–
44.
38. Mukaiyama, T.; Nakatsuta, T.; Shoda, S. I. Chem. Lett.
1979, 487–490.
39. De Meo, C.; Parker, O. Tetrahedron: Asymmetry 2005, 16,
303–307.
40. Pornsuriyasak, P.; Demchenko, A. V. Tetrahedron: Asym-
metry 2005, 16, 433–439.
`
17. Chaves das Neves, H. J.; Pais, M. S. S. Tetrahedron Lett.
1980, 21, 4387–4390.
18. Park, M. K.; H., P. J.; Shin, Y. G.; Kim, W. Y.; Lee, J. H.;
Kim, K. H. Planta Med. 1996, 62, 363–365.
41. Ferrieres, V.; Gelin, M.; Boulch, R.; Toupet, L.; Plus-
quellec, D. Carbohydr. Res. 1998, 314, 79–83.
42. Ferrier, R. J.; Furneaux, R. H. Carbohydr. Res. 1976, 52,
63–68.
´
´
´
´
´
`
19. Micovic, V. M.; Hranisavljevic, M.; Miljkovic-Stojanovic,
43. Ferrieres, V.; Blanchard, S.; Fischer, D.; Plusquellec, D.
J. Carbohydr. Res. 1969, 10, 525–533.
Bioorg. Med. Chem. Lett. 2002, 12, 3515–3518.
`
20. Gerwig, G. J.; Kamerling, J. P.; Vliegenthart, J. F. G.;
Morag-Morgenstern, E.; Lamed, R.; Bayer, E. A. Eur. J.
Biochem. 1992, 205, 799–808.
21. Pettit, G. R.; Xu, J. P.; Gingrich, D. E.; Williams, M. D.;
Doubek, D. L.; Chapuis, J. C.; Schmidt, J. M. Chem.
Commun. 1999, 915–916.
44. Euzen, R.; Ferrieres, V.; Plusquellec, D. J. Org. Chem.
2005, 70, 847–855.
45. Demchenko, A. V. Curr. Org. Chem. 2003, 7, 35–79.
46. Demchenko, A. V.; Malysheva, N. N.; De Meo, C. Org.
Lett. 2003, 5, 455–458.
47. Chang, C. W. T.; Chen, X. H.; Liu, H. W. J. Am. Chem.
Soc. 1998, 120, 9698–9699.
´
22. Domenech, J.; Prieto, A.; Barasoain, I.; Gomez-Miranda,
´
´
´
B.; Bernabe, M.; Leal, J. A. Microbiology 1999, 145, 2789–
2796.
48. Martın-Lomas, M.; Khiar, N.; Garcıa, S.; Koessler, J. L.;
Nieto, P. M.; Rademacher, T. W. Chem. Eur. J. 2000, 6,
3608–3621.
23. Molinaro, A.; Piscopo, V.; Lanzetta, R.; Parrilli, M.
Carbohydr. Res. 2002, 337, 1707–1713.
24. Stevenson, G.; Neal, B.; Liu, D.; Hobbs, M.; Packer, N.
H.; Batley, M.; Redmond, J. W.; Lindquist, L.; Reeves, P.
J. Bacteriol. 1994, 176, 4144–4156.
49. Velty, R.; Benvegnu, T.; Gelin, M.; Privat, E.; Plusquellec,
D. Carbohydr. Res. 1997, 299, 7–14.
50. Pathak, A. K.; Pathak, V.; Seitz, L.; Maddry, J. A.;
Gurcha, S. S.; Besra, G. S.; Suling, W. J.; Reynolds, R. C.
Bioorg. Med. Chem. 2001, 9, 3129–3143.
25. Germond, J. E.; Delley, M.; D’Amico, N.; Vincent, S. J. F.
Eur. J. Biochem. 2001, 268, 5149–5156.
`
51. Ferrieres, V.; Bertho, J. N.; Plusquellec, D. Tetrahedron
26. Linnerborg, M.; Weintraub, A.; Widmalm, G. Eur. J.
Biochem. 1999, 266, 460–466.
Lett. 1995, 36, 2749–2752.
`
52. Ferrieres, V.; Bertho, J. N.; Plusquellec, D. Carbohydr.
27. Eserstam, R.; Rajaguru, T. P.; Jansson, P. E.; Weintraub,
A.; Albert, M. J. Eur. J. Biochem. 2002, 269, 3289–
3295.
Res. 1998, 311, 25–35.
53. Euzen, R.; Lopez, G.; Nugier-Chauvin, C.; Ferrieres, V.;
`
´
Plusquellec, D.; Remond, C.; O’Donohue, M. J. Eur. J.
28. Tzianabos, A.; Wang, J. Y.; Kasper, D. L. Carbohydr.
Res. 2003, 338, 2531–2538.
29. Jones, C.; Todeschini, A. R.; Agrellos, O. A.; Previato, J.
O.; Mendonc¸a-Previato, L. Biochemistry 2004, 43, 11889–
11897.
30. Besra, G. S.; Khoo, K. H.; McNeil, M. R.; Dell, A.;
Morris, H. R.; Brennan, P. J. Biochemistry 1995, 34, 4257–
4266.
Org. Chem. 2005, 4860–4869.
54. Hernandez, O.; Chaudhary, S. K.; Cox, R. H.; Poter, J.
Tetrahedron Lett. 1981, 22, 1491–1494.
55. Smoot, J. T.; Pornsuriyasak, P.; Demchenko, A. V.
Angew. Chem., Int. Ed. 2005, 44, 7123–7126.
56. Reitz, A. B.; Tuman, R. W.; Marchione, C. S.; Jordan, A.
D.; Bowden, C. R.; Maryanoff, B. E. J. Med. Chem. 1989,
32, 2110–2116.