Direct formation of β-glycosides of N-acetyl glycosamines mediated by rare earth metal triflates

Article abstract of DOI:10.1039/b807064d

A direct, mild and efficient protocol for the preparation of β-glycosides of N-acetyl glucosamine (GlcNAc) and N-acetyl galactosamine (GalNAc) has been developed using peracetylated β-GlcNAc and β-GalNAc as donors. All rare Earth metal triflate promoters

Full text of DOI:10.1039/b807064d

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Direct formation of b-glycosides of N-acetyl glycosamines mediated by rare
earth metal triflates†
Helle Christensen, Mira Steinicke Christiansen, Jette Petersen and Henrik Helligsø Jensen*
Received 25th April 2008, Accepted 19th June 2008
First published as an Advance Article on the web 28th July 2008
DOI: 10.1039/b807064d
A direct, mild and efficient protocol for the preparation of b-glycosides of N-acetyl glucosamine
(GlcNAc) and N-acetyl galactosamine (GalNAc) has been developed using peracetylated b-GlcNAc
and b-GalNAc as donors. All rare Earth metal triflate promoters screened were found to promote
glycosylation with Sc(OTf)3 being superior in terms of reaction rate. Simple alcohol glycosylation was
found to proceed smoothly in refluxing dichloromethane, whereas higher temperatures under
microwave conditions were needed to attain acceptable yields with less reactive, carbohydrate based
glycosyl acceptors. The protocol developed was applied to provide the first example of direct chemical
formation of a disaccharide using both GlcNAc as a glycosyl donor and acceptor. The a-acetate donor
was found to be significantly less reactive than the corresponding b-anomer necessitating higher
reaction temperatures under which glycoside anomerisation was found to occur. It was established, that
the anomerisation only took place in the presence of both Sc(OTf)3 and acetic acid.
useful catalysts for the synthesis of simple glycosides of GlcNAc
Introduction
(Scheme 1a). Bundle and co-workers have furthermore found 5,6-
The N-acetyl glucosamine (GlcNAc) unit is frequently used by
O-isopropylidene glucofuranosyl oxazoline (2, Scheme 1b) to be
Nature as a monosaccharide building block in a range of impor-
a useful donor with p-TsOH as the activating agent giving the
2-acetamido glucopyranosides as products.19
tant oligosaccharides like the core pentasaccharide, nodulation
factors, and carbohydrate epitopes.1 Accordingly, methodology
for the efficient preparation of GlcNAc glycosides remains an
important component of modern carbohydrate chemistry.2 Tradi-
tionally, the chemical synthesis of the b-GlcNAc glycosidic linkage
has most frequently been carried out by using glucosamine-
derived donors possessing a temporary protecting group of the 2-
amino functionality such as N-Troc (N-trichloroethoxycarbonyl),3
N-Phth (N-phthalimido),4 N,N-diacetyl,5 or azide.6 Among
the less used nitrogen camouflaging groups are N-Alloc (N-
allyloxycarbonyl),7 N-Cbz (N-benzyloxycarbonyl),7 N-TCA (N-
trichloroacetyl),8 N-TCP (tetrachlorophthalimido),9 N-Dts
(N-dithiasuccinoyl),10 N-DCPhth (N-4,5-dichlorophthalimido),11
N-DMM (N-dimethylmaleoyl),12 N,N-dibenzyl,13 and N-TDG
(N-thiodiglycoloyl).14 Although glycosylation generally functions
well with donors bearing these protecting groups their drawback
is that separate synthetic steps are required for introduction
and later interconversion to the biologically relevant 2-acetamido
substituent.
Scheme
1 Synthesis of GlcNAc b-glycosides. a) from pyranosyl
To minimise the number of synthetic steps, sugar oxazolines
have been explored as potential donors, which directly yield the
acetyl function without further manipulations. Traditionally, a
range of Lewis acids such as FeCl3,15 AlCl3,16 and TMSOTf2 have
been used with some success for glycopyranosyl oxazoline (1)
activation in chemical GlcNAc-ylation. Recently, stoichiometric
cupric salts17 and Yb(OTf)3 (30 mol%)18 were established as being
oxazoline,17,18 b) from furanosyl oxazoline,19 and c) direct glycosylation
from acetate donor, present study.
The use of oxazolines as donors for chemical glycosylation
is troubled by their preparation and low shelf stability. We
accordingly undertook a study inspired by the above mentioned
promising results18 to search for reaction conditions under which
GlcNAc-ylation could be conducted using a stable, storable
glycosyl donor.
Rare earth metal triflates have proven to be a highly valuable tool
in a wide range of Lewis acid mediated reactions. The advantage
of using these metals as catalysts is their high level of acidity in
combination with their stability towards moisture.20
Department of Chemistry, University of Aarhus, Langelandsgade 140, DK-
8000, Aarhus C, Denmark. E-mail: hhj@chem.au.dk; Fax: +45 86196199;
Tel: +45 89423963
† Electronic supplementary information (ESI) available: Further experi-
mental details. See DOI: 10.1039/b807064d
3276 | Org. Biomol. Chem., 2008, 6, 3276–3283
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
This work describes our recent results in using shelf stable and
commercially21 available GlcNAc-acetates (3) as glycosyl donors
and rare earth metal triflates as Lewis acid promoters, without
the need for purification of the oxazoline intermediate. The use of
catalytic amounts of rare earth metal triflates is believed to both
catalyse oxazoline formation and activation.
also showed the development of this more polar compound that
eventually disappeared after reaction completion. This strongly
indicates that the rare earth metal triflates catalyse both anomeric
activation leading to oxazoline formation and later breakdown of
this as described by Crasto and Jones.18
Due to the superior reactivity of Sc(OTf)3 in spite of it being the
most highly priced of the tested catalysts, we continued our screen
for the best reaction medium among the standard laboratory
solvents (entry 9–12, Table 1). These were found to result in
diminished yields compared to when CH2Cl2 was used at the same
temperature with dichloroethane (DCE) giving the best result
(6 hours, 69%, entry 9, Table 1). Both reaction times and yields
were the order CH2Cl2 > DCE > CH3CN > PhCH3 > THF.
Heating the reaction mixture to reflux in DCE gave an improved
yield of 82% in only 90 min (entry 13, Table 1). The same protocol
was investigated leaving out the catalyst showing no product
formation after 30 hours demonstrating the necessity of a Lewis
acid promoter (entry 14, Table 1).
Results and discussion
Various Lewis acid catalysts
Early in our study, inspired by the work of Crasto and Jones,18 we
were pleased to find that Yb(OTf)3 (15 mol%) with BnOH (4) in
refluxing CH2Cl2 efficiently catalysed the exclusive formation of
benzyl b-glycoside 5 in 75% yield from the stable, commercially
available b-peracetylated donor 3 (entry 1, Table 1). This promising
result led us to screen for optimal reaction conditions. All Lewis
acids tested were capable of promoting glycosylation with isolated
yields between 69–88% (entry 1–7, Table 1). Reaction times were
typically 20–22 hours but with Sc(OTf)3 and Cu(OTf)2 standing
out as significantly more active catalysts resulting in reaction
completion after only 4 and 8 hours, respectively (entry 2 and
entry 7, Table 1). This confirms the known fact that of this series
of catalysts, Sc(OTf)3 is known to have extraordinarily high activity
as a Lewis acid.20
Also CuBr2 (15 mol%), which had been found by Wittmann and
Lennartz17 to promote oxazoline glycosylation using a stoichio-
metric amount of Lewis acid gave the expected b-benzyl glycoside
(5) albeit in only 57% yield (entry 8, Table 1). Interestingly, in
contrast to the present results, Wittmann and Lennartz reported
that Cu(OTf)2 was not a promoter for the conversion of oxazoline
to glycoside.17
Various glycosylation reactions
After optimising reaction conditions, the use of various alcohols as
potential glycosyl acceptors were explored (Table 2). The electron
poor acceptor 2-bromoethanol (6,entry 1, Table 2) also yielded
the expected b-glycoside (7) with a significant prolonged reaction
time (27 hours) compared to BnOH (4). Glycosylation with
cyclohexanol (8, entry 2, Table 2) gave the expected b-cyclohexyl
glycoside (9) in 84% yield but the reaction time had gone from
4 hours with BnOH (4) to 48 hours with this more sterically
demanding acceptor. The protocol too offered the possibility
of preparing a glycoamino acid building block of serine (11,
50%, entry 3, Table 2) found in nuclear proteins.23 It is well-
established that carbamate protected serines (10) act as poor
glycosyl acceptors, a fact that has been ascribed to unfavourable
HO: → H–N hydrogen bond formation.24
To shed light on the reaction mechanism an experiment was
carried out in refluxing CH2Cl2 with Yb(OTf)3 (15 mol%) but
leaving out any acceptor alcohol. This led, according to TLC
analysis, to the development of a slightly more polar compound,
which was found to have the NMR spectral characteristics of
the expected oxazoline intermediate.22 Regardless of the choice of
catalyst, performing the reaction in the presence of e.g. BnOH (4)
Galactose (12) and glucose25 (14) based primary alcohols were
furthermore glycosylated in 78% and 85%, respectively (entry 4
and entry 5, Table 2).
Full consumption of acceptor with 4 equivalents of donor (3)
could not be reached in the case of 4-OH acceptor 1626 (entry 6,
Table 2). After heating for 72 hours only 21% of the disaccharide 17
could be isolated (64% based on recovered 16) giving an indication
of the scope of the present protocol. In this case adding more donor
or catalyst was found not to result in further conversion.
Table 1 Results of glycosylations according to Scheme 1c conducted with
BnOH (4) (3 equiv.)
Cat.
Price/
Entry T/◦Ca Solvent
(15 mol%) (€/5 g)b Time
Yieldc
N-Acetyl-galactosamine (GalNAc) has no less biological signif-
icance compared to GlcNAc so we were pleased to find Sc(OTf)3
also was an efficient catalyst to bring about galactosylation from
the corresponding GalNAc b-acetate donor (4-epi-3, entry 7,
Table 2).27 The reaction time with BnOH (4) as the acceptor was
considerably shorter than with the corresponding glucose config-
ured isomer (3) as expected, due to the smaller degree of transition
state destabilisation exerted by the C-4 axial substituent.28 It
was furthermore possible to galactosylate the hindered secondary
galactose based acceptor 19 in 65% yield (83% based on recovered
acceptor). The acceptor (19) was prepared by benzylidene ring-
opening of the known 24 with Et3SiH–TFA according to Scheme 2.
An efficient synthesis of this disaccharide (20) being a target
sequence for pulmonary pathogens was published recently using
Troc-protection of the galactosamine N-atom.29 We believe that
1
2
3
4
5
6
7
8
9
10
11
12
13
14
45
45
45
45
45
45
45
45
45
45
45
45
90
90
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
ClCH2CH2Cl Sc(OTf)3
THF
PhCH3
CH3CN
ClCH2CH2Cl Sc(OTf)3
ClCH2CH2Cl none
Yb(OTf)3
Sc(OTf)3
Sm(OTf)3 101.50
La(OTf)3
Dy(OTf)3
Nd(OTf)3
Cu(OTf)2
CuBr2
87.00
255.30
22 h 75%
4 h 81%
22 h 84%
22 h 81%
20 h 83%
20 h 88%
8 h 69%
>24 h 57%
6 h 69%
>28 h 48%
12 h 54%
8 h 62%
74.20
69.70
42.50
83.50
20.90
255.30
255.30
255.30
255.30
255.30
Sc(OTf)3
Sc(OTf)3
Sc(OTf)3
90 min 82%
>30 h
0%
a Oil bath temperature. b Prices listed on www.sigmaaldrich.com, March
2008. c Isolated yield after chromatography.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3276–3283 | 3277
©

View Article Online
Table 2 Sc(OTf)3 (15 mol%) promoted glycosylation in refluxing CH2Cl2
Entry
1
Da
Ab
Product
(D : A)c
1 : 3
Time
27 h
Yieldd
67%
3
2
3
3
3
1 : 3
48 h
72 h
84%
50%
1 : 1.5
4
3
2 : 1
24 h
78%
5
6
3
3
4 : 1
72 h
85%
4 : 1
72 h
21% (64%)
7
8
4-epi-3
4-epi-3
1 : 3
4 : 1
2 h
84%
72 h
65% (83%)
a Donor. b Acceptor. c Donor : acceptor ratio. d Isolated yield after chromatography (yield based on recovered acceptor).
this more direct strategy offers an equally efficient approach to
this biologically important disaccharide.
served for some of the reactions listed in the above section, we
wondered whether microwave irradiation would make the present
approach for glycosylation more efficient by lowering the reaction
time. Given that microwave reactions are carried out in sealed
tubes it is furthermore possible to heat e.g. CH2Cl2 to temperatures
well above the boiling point at atmospheric pressure avoiding the
need of higher boiling solvents like DCE. The benefit of this is
both the easier later removal of a lower boiling solvent along with
Glycosylation under microwave conditions
Microwave irradiation as a mode of heating is becoming in-
creasingly popular in all aspects of organic chemistry including
carbohydrate chemistry.30–33 Due to the slow reaction rates ob-
3278 | Org. Biomol. Chem., 2008, 6, 3276–3283
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
We next decided to explore glycosylation of carbohydrate based
alcohols. Primary alcohol 14 could be glycosylated in as little
as 3 hours but a slightly lower yield was obtained (entry 3,
Table 3). Using microwave heating, a significantly higher level
of conversion was obtained with the troublesome secondary
alcohol of glucose acceptor 16 (entry 4, Table 3) after 16 hours
at 80 ◦C. Although this cannot be regarded as a high yield it
must be kept in mind that only one anomeric product is formed
and that no additional steps for protecting group interconver-
sion are required to attain the biologically relevant acetamido
function.
Some success in O-4 over O-3 regioselective galactosylation of
b-GlcNAc derivatives has in recent years been reported.34 Inspired
by these results it was decided to investigate whether the present
protocol also was able to selectively give a 1→4 linked disaccharide
and thereby a di-N-acetyl chitobiose derivative. Heating the novel
GlcNAc acceptor 21 (Scheme 2) with b-acetate 3 under microwave
irradiation in the presence of Sc(OTf)3 (15 mol%), however, was
found to selectively produce the 1→3 linked disaccharide in the
astonishing yield of 90% (entry 5, Table 3). The regiochemistry
of the product was established by the presence of H1ꢀ→C3 and
H3→C1ꢀ correlations in the HMBC spectrum of the disaccharide
(22) and confirmed by global deprotection and comparison to
known spectral data (Scheme 2c).35 This, to the best of our knowl-
edge, constitutes the first example of a chemical glycosylation
reaction where GlcNAc successfully has been used as both donor
and acceptor.
Scheme 2 Synthesis of galactose and N-acetyl glucosamine acceptors (a
and b), and regioselective glycosylation of 3,4-acceptor (21) followed by
deprotection to confirm regiochemistry of interglycosidic linkage (c).
the fact that DCE is less commonly available in comparison to
CH2Cl2 as a freshly distilled solvent in most chemical laboratories.
We first investigated the GlcNAc- and GalNac-ylation with
benzyl alcohol as the acceptor under conditions similar those
described above with conventional heating (Table 1, entry 2 and
Table 2, entry 7). When heating by microwave irradiation to 80 ◦C
it was yet again observed that galactosylation (5 min, entry 2,
Table 3) was faster than glucosylation (20 min, entry 1, Table 3).
The reaction time was hence brought down from several hours to
minutes while the yield essentially remained the same.
Table 3 Sc(OTf)3 (15 mol%) promoted glycosylation in CH2Cl2 under microwave irradiation (80 ◦C)
Entry
1
Da
Ab
Product
(D : A)c
1 : 3
Time
Yieldd
80%
3
20 min
2
3
4-epi-3
1 : 3
2 : 1
5 min
3 h
80%
75%
3
4
5
3
3
4 : 1
2 : 1
16 h
8 h
43% (68%)
90%
a Donor. b Acceptor. c Donor : acceptor ratio. d Isolated yield after chromatography (yield based on recovered acceptor).
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3276–3283 | 3279
©

View Article Online
Glycosylation with a-anomeric acetate
Heating benzyl glycoside 5 with a stoichiometric amount of
acetic acid leaving out the Sc(OTf)3 under otherwise identical
conditions also showed lack of anomerisation (entry 3, Table 4).
These observations could indicate that the glycoside (5) was
anomerically stable under the reaction conditions and that the
a-glycoside product (entry 1, Table 4) in some way originated
from using the a-acetate donor 3a. As a final control experiment
the benzyl glycoside was treated with both Sc(OTf)3 and acetic
acid and these conditions indeed did show anomerisation to
occur (entry 4, Table 4). This can only mean that a powerful
Lewis acid like triflic acid forms from the reaction of Sc(OTf)3
with acetic acid, which is capable of catalysing post-glycosylation
anomerisation. This, hence, only becomes relevant due to the
more forcing conditions required to activate the a-acetate donor
(3a).
Only b-acetates (3 and 4-epi-3) had to this point been explored
as the glycosyl donor (Scheme 1c). Despite the fact that this can
be easily synthesised21 on a large scale without the need for chro-
matographic purification of reaction intermediates, its preparation
involves several synthetic steps. We therefore wondered whether
an anomeric mixture of acetates being the product of a simple
acetylation reaction of D-N-acetylglucosamine, could be used with
equal success.36
Prolonged conventional heating of an a–b-mixture (3ab, a :
b = 4.6 : 1) in CH2Cl2 with BnOH (4)–Sc(OTf)3 according to the
standard procedure (entry 2, Table 1) was found only to yield the
benzyl glycoside (5) according to the proportion of b-donor (3).
Unreacted donor was accordingly shown to be exclusively the a-
anomeric acetate (3a) in accordance with the well-known fact that
axial donors are less reactive than the corresponding equatorial
donors.37 Changing reaction conditions from conventional reflux
to microwave heating at 80 ◦C in CH2Cl2 it was possible to
obtain some activation of the a-anomeric acetate but the reaction
was still rather sluggish. Full consumption of donor (3ab) was
achieved after heating to 110 ◦C for 5 hours in the presence of
benzyl alcohol (4) giving a the benzyl glycoside (5ab) in 80%
yield. The product, however, was obtained as a 1 : 4 anomeric
mixture of glycosides (entry 1, Table 4). This led us to investigate
whether glycosylation at this temperature with an a-donor led to
an a : b product mixture or rather glycosylation was accordingly
slow that anomerisation of the already formed product under-
went post-glycosylation anomerisation. The results are listed in
Table 4.
Conclusion
The results presented in this article demonstrate that scandium
triflate efficiently catalyses the activation of b-GlcNAc and b-
GalNAc tetraacetates in the presence of a variety of glycosyl
acceptors resulting in the exclusive formation of the b-glycoside.
The donor substrates are shelf stable and have the bene-
fit of directly producing 2-acetamido substituent circumvent-
ing the use of traditional phthalimido- or Troc protecting
groups.
For slowly reacting alcohols, microwave irradiation was found
to significantly diminish reaction time and in certain cases result
in higher acceptor turnover. a-Acetates were found to have poor
glycosyl donor properties and the reaction conditions needed to
bring about full conversion resulted in product anomerisation
catalysed by both Sc(OTf)3 and the acetic acid formed during
the reaction.
Since we expected that the rare earth metal triflate was
responsible for the anomerisation we were surprised that treatment
of benzyl glycoside 5 (entry 2, Table 4) with Sc(OTf)3 resulted in
complete isolation of starting material in its b-anomeric form.
Table 4 Reaction conducted in CH2Cl2 under microwave irradiation (110 ◦C) for 5 hours
Entry
1
Da
Ab
Sc(OTf)3
15 mol%
AcOH
none
a : b product ratioc
21 : 79
2
3
4
none
none
none
15 mol%
none
none
0 : 100
0 : 100
28 : 72
1 equiv.
1 equiv.
15 mol%
a Donor; b Acceptor. c a : b product ratio of benzyl glucoside (5ab) as determined by NMR analysis of the crude reaction mixture.
3280 | Org. Biomol. Chem., 2008, 6, 3276–3283
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
We believe that this protocol offers a convenient approach for
the synthesis of b-glycosides of GlcNAc and GalNAc that will be
of high value to carbohydrate chemistry.
Allyl O-(2-acetamido-3,4,6-tri-O-acetyl-2-deoxy-b-D-
galactopyranosyl)-(1→4)-2,3,6-tri-O-benzyl-a-D-
galactopyranoside (20)
Dry toluene was removed under reduced pressure from a
round bottomed flask containing a stirrer bar, b-D-N-acetyl-
galactosamine tetraacetate (4-epi-3) (133 mg, 0.34 mmol) and
galactose acceptor 19 (84 mg, 0.17 mmol). When dry, the
mixture was dissolved in CH2Cl2 (1.0 mL) and Sc(OTf)3 (13 mg,
0.026 mmol) was added before the mixture was heated to reflux
under an atmosphere of nitrogen. After 24 h another portion of
donor (67 mg, 0.17 mmol) and Sc(OTf)3 (13 mg, 0.026 mmol)
were added. After a total of 67 h no further reaction was observed
by TLC analysis (EtOAc–pentane 2 : 1). The reaction mixture
was diluted with CH2Cl2 (20 mL) and washed with NaHCO3
(sat. aq.), dried (MgSO4) and purified by column chromatography
(EtOAc–pentane 1 : 4 → 2 : 1). This gave un-reacted acceptor
(18 mg) and the desired disaccharide (91 mg, 65%, 83% rcv.)
as a colourless oil. Rf(EtOAc–pentane 1 : 1) 0.12. 1H-NMR
(400 MHz, CDCl3) d 7.43–7.25 (m, 15H, ArH), 5.97–5.87 (m,
Experimental section
General methods
All reagents except otherwise stated were used as purchased
without further purification. Solvents were dried according to
standard procedures. Columns for flash chromatography were
˚
packed with silica gel (60 A). TLC plates (Kieselgel 60 F254
)
were visualised by use of a “Ce-Mol” solution (Ce(IV)sulfate
(10 g) and ammonium molybdate (15 g) dissolved in 10% H2SO4
(1 L)) and heated until coloured spots appeared. 1H and 13C
NMR experiments were recorded on a Varian Mercury 400
NMR instrument and assigned using DEPT-135, gHMQC, and
gCOSY-experiments. Mass spectral analyses were carried out as
electrospray experiments on a Micromass LC-TOF instrument.
Optical rotation was measured on a PE-314 polarimeter. [a]D
values are given in 10−1 deg cm2 g−1. Melting points were measured
on a Bu¨chi B-540.
1H, CH=CH2), 5.53 (d, 1H, JNH,2 8.4 Hz, NH), 5.31 (b d, Jvic
ꢀ
=
17.2 Hz, CH CHH), 5.26 (d, 1H, J 3.2 Hz, H4 ), 5.20 (b d, 1H,
=
Jvic 10.4 Hz, CH CHH), 4.91 (d, 1H, J1,2 3.6 Hz, H1), 4.87 (d,
1H, Jgem 10.4 Hz, PhCHH), 4.70–4.55 (m, 7H, PhCHH, PhCH2,
Microwave experiments were carried out on in a Biotage
Initiator (Biotage, Sweden). Reaction times listed refer to ‘hold
time’ at the specified temperature.
ꢀ
ꢀ
ꢀ
ꢀ
=
PhCH2, H1 , H3 ), 4.23–3.96 (m, 8H, CH2CH CH2, H2 , H6 a,
H6ꢀb, H3, H4, H5), 3.81 (dd, 1H, J2,3 10.0 Hz, H2), 3.77 (t, 1H,
J 7.6 Hz, H5ꢀ), 3.71 (dd, 1H, J5,6a 6.0 Hz, J6a,6b 10.0 Hz), 3.64
(dd, 1H, J5,6b 6.4 Hz, H6b), 2.15, 1.99, 1.95, 1.55 (C(O)CH3). 13C-
NMR (100 MHz, CDCl3) d 170.7, 170.6, 170.2 (CO), 138.6, 138.4,
General procedure for the synthesis of benzyl 2-acetamido-3,4,6-
tri-O-acetyl-2-deoxy-b-D-glucopyranoside (5) with various
catalysts and solvents
138.0, 129.2, 128.9, 128.7, 128.6, 128.1, 128.0, 127.8, 127.7 (Ar),
ꢀ
=
134.0 (CH CH2), 118.4 (CH=CH2), 103.4 (C1 ), 95.7 (C1), 78.1,
77.9, 77.7 (C2, C3, C4), 74.9, 73.6, 73.0 (PhCH2), 72.3 (C3ꢀ), 71.1
Donor 3 (200 mg, 0.51 mmol) and catalyst (0.081 mmol) were
dissolved in dry solvent (2.5 mL) and benzyl alcohol (4) (0.17 mL,
1.6 mmol) was added. The mixture was heated to reflux with
condenser under an atmosphere of nitrogen until TLC analysis
indicated that the reaction gave no further conversion. The reac-
tion mixture was worked up by diluting with CH2Cl2 and washing
the organic layer with water, followed by extraction with CH2Cl2.
When MeCN or THF were used as solvent, the combined organic
extracts were washed with brine. The mixture was then dried
over anhydrous MgSO4 and purified by column chromatography
(pentane–EtOAc 1 : 1 → EtOAc) to give the benzyl glycoside
5 as white crystals in varying yields as listed in Table 1. Rf
(EtOAc): 0.58. Mp (uncorr.): 165–167 ◦C. LRMS(ES+): calcd.
for C21H27NO9Na: 460.2, found: 460.2. Spectral data was in
accordance with previously reported results.38
ꢀ
ꢀ
=
(C5 ), 69.8 (C6), 69.4 (C5), 68.6 (CH2CH CH2), 66.7 (C4 ), 61.5
(C6ꢀ), 50.7 (C2ꢀ), 23.2, 21.0, 20.9, 20.9 (C(O)CH3). [a]D21 24.8 (c 1.0,
CHCl3), HRMS(ES+): calcd. for C44H53NO14Na: 842.3364; found:
842.3370.
Benzyl 2-acetamido-3,4,6-tri-O-acetyl-2-deoxy-b-D-
galactopyranoside (18)
Benzyl alcohol (4) (88 lL, 0.85 mmol) and Sc(OTf)3 (21 mg,
0.042 mmol) were added to b-D-N-acetyl-galactosamine tetraac-
etate (4-epi-3) (110 mg, 0.28 mmol) in CH2Cl2 (0.84 mL). The
reaction mixture was heated to reflux on an oil bath (2 h) or 80 ◦C
under microwave conditions (5 min) before the reaction mixture
was diluted with CH2Cl2 (20 mL) and washed with NaHCO3
(sat. aq.), dried (MgSO4), concentrated, and purified by column
chromatography (EtOAc–pentane 1 : 1 → EtOAc). This gave the
benzyl galactoside 18 as a white powder (104 mg, 84%, oil bath)
(99 mg, 80%, microwave). Rf(EtOAc) 0.37. 1H-NMR (400 MHz,
CDCl3) d 7.33–7.25 (m, 5H, ArH), 5.73 (d, 1H, J2,NH 8.4 Hz, NH),
5.32 (d, 1H, J3,4 3.2 Hz, H4), 5.20 (dd, 1H, J2,3 11.2 Hz, H3),
4.87 (d, 1H, Jgem 12.0 Hz, PhCHHO), 4.65 (d, 1H, J1,2 8.4 Hz,
H1), 4.58 (d, 1H, PhCHHO), 4.19–4.00 (m, 3H, H5, H6a, H6b),
3.90–3.87 (m, 1H, H2), 2.11, 2.03, 1.94, 1.85 (s, 12H, C(O)CH3).
13C-NMR (100 MHz, CDCl3) d 170.7, 170.6, 170.5 (CO), 137.3,
128.7, 128.4, 128.2 (Ar), 100.0 (C1), 70.9 (double intensity), 70.2,
67.1, 61.8 (C3, C4, C5, C6, OCH2Ph), 51.5 (C2), 23.6, 20.9 (triple
intensity, C(O)CH3). [a]2D1 −50.2 (c 1.0, CHCl3) (lit.: [a]2D3 −55
General procedure for the coupling of 2-acetamido-1,3,4,6-tetra-O-
acetyl-2-deoxy-b-D-glucopyranose to various alcohols
Glycosyl acceptor was added to a dry solution of donor 3 and
catalyst (0.15 eq.). The mixture was heated to reflux with condenser
under nitrogen, until TLC analysis indicated that the reaction
gave no further conversion. The reaction mixture was worked
up by diluting with CH2Cl2 and washing with water, followed
by extraction with CH2Cl2. The combined organic extracts were
then dried over anhydrous MgSO4, filtered, concentrated in vacuo,
and purified by flash column chromatography on silica using an
appropriate eluent.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3276–3283 | 3281
©

View Article Online
(c 1.0, CHCl3).39 HRMS(ES+): calcd. for C21H27NO9Na: 460.1584;
found: 460.1580.
J5,6b 5.2 Hz, H6b), 3.66 (t, 1H, H4), 6.59–3.55 (m, 1H, H5), 2.00,
1.83 (s, 6H, C(O)CH3). 13C-NMR (CDCl3, 100 MHz) d 172.2,
170.7 (CO), 138.1, 137.6, 128.7, 128.6, 128.1, 128.0, 127.9 (Ar),
100.1 (C1), 75.9, 74.8, 73.9, 70.6, 70.5, 70.4 (C3, C4, C5, C6,
PhCH2O1, PhCH2O6), 54.2 (C2), 23.4, 21.2 (C(O)CH3).
Benzyl 2-acetamido-6-O-benzyl-2-deoxy-b-D-glucopyranoside (21)
Donor 3 (4.92 g, 12.6 mmol) and benzyl alcohol (4) (4.0 mL,
0.0378 mol) were coupled according to the general procedure
with Sc(OTf)3 (0.933 g, 0.188 mmol) as catalyst in dry CH2Cl2
(50 mL). After 4 h TLC analysis indicated no further reaction.
The reaction mixture was worked up as described in the general
procedure and purified by column chromatography (pentane–
EtOAc 1 : 1 → EtOAc) to give benzyl glycoside 5 (4.31 g, 78%).
This (4.31 g, 9.86 mmol) was dissolved in dry CH3OH (25 mL)
under a N2-atmosphere. With a syringe, a catalytic amount of a
freshly prepared NaOCH3 solution was transferred to the reaction
mixture. When no further reaction was observed by TLC analysis
(EtOAc–CH3OH 9 : 1), Amberlite (IR-120H+) was added. The
reaction mixture was stirred for 10 min before Amberlite was
filtered off and the resulting mixture concentrated under vacuum
to give the crude triol. This was then dissolved in dry CH3CN
(80 mL) before benzaldehyde dimethyl acetal (1.6 mL, 10.6 mmol)
and PTSA (68 mg, 0.36 mmol) were added. The heterogeneous
reaction mixture was stirred at ambient temperature for 1 h and
then heated to 50 ◦C for 30 min. The solid was filtered off, washed
with Et2O and dissolved in dry pyridine–acetic anhydride (1 : 1,
40 mL) with 4-DMAP (25 mg, 0.21 mmol) and stirred at ambient
temperature overnight. The reaction mixture was quenched with
H2O (ice bath), the aqueous phase extracted 4 times with CH2Cl2,
dried over MgSO4, and concentrated under vacuum to give
crude benzyl 2-acetamido-2-O-acetyl-4,6-O-benzylidene-2-deoxy-
b-D-glucopyranoside, which was sufficiently pure for further
Purified 3-O-acetyl monosaccharide (980 mg, 2.21 mmol) was
then dissolved in dry CH3OH under an N2-atmosphere. With a
syringe a catalytic amount of a freshly prepared NaOCH3 solution
was transferred to the reaction mixture. After 10 minutes of
stirring, the reaction mixture consisted of a white solid. CH3OH
(enough to dissolve the solid) and Amberlite (IR-120H+) were
added, and after stirring the Amberlite was filtered of and the
resulting mixture was concentrated under reduced pressure to give
the desired product in quantitative yield as a white powder (23%
over◦5 steps). Rf (EtOAc–CH3OH 6 : 1) 0.44. Mp (uncorr.) 175–
178 C (EtOAc–CH3OH).40 1H-NMR (DMSO-d6, 400 MHz): d
7.72 (d, 1H, JNH,2 9.2 Hz, NH), 7.34–7.23 (m, 10 H, ArH), 5.11
(b s, 1H, OH), 4.97 (b s, 1H, OH), 4.73 (d, 1H, Jgem 12.6 Hz,
PhCHHO1), 4.53 (s, 2H, PhCH2O6), 4.49 (d, 1H, PhCHHO1),
4.38 (d, 1H, J1,2 8.4 Hz, H1), 3.76 (d, 1H, J6a,6b 10.8 Hz, H6a), 3.52
(dd, 1H, J5,6b 6.0 Hz, H6b), 3.50 (q, 1H, H2), 3.32–3.27 (under
H2O, H3, H5), 3.10 (t, 1H, J3,4 = J4,5 9.0 Hz, H4), 1.79 (s, 3H,
COCH3); 1H-NMR (DMSO-d6 + drop of D2O, 400 MHz) d 7.34–
7.23 (m, 10H, ArH), 4.70 (d, 1H, Jgem 12.0 Hz, PhCHHO1), 4.50 (s,
2H, PhCH2O6), 4.45 (d, 1H, PhCHHO1), 4.36 (d, 1H, J1,2 8.4 Hz,
H1), 3.80–3.68 (under DHO signal, 1H, H6a), 3.53 (dd, 1H, J5,6b
6.0 Hz, J6a,6b 10.8 Hz, H6b), 3.49 (t, 1H, H2), 3.31–3.24 (m, 2H, H3,
H5), 3.10 (t, 1H, J 9.2 Hz, H4), 1.79 (s, 3H, C(O)CH3). 13C-NMR
(DMSO-d6, 100 MHz): d 170.6 (CO), 139.2, 138.5, 128.9, 128.1,
127.9 (Ar), 101.3 (C1), 76.1, 74.5 (C3, C5), 73.0 (PhCH2O6), 71.1
(C4), 70.4 (PhCH2O1), 70.2 (C6), 55.9 (C2), 23.5 (C(O)CH3). [a]D22
−33.6 (c 1.0, CH3OH). HRMS(ES+): calcd. for C22H27NO6Na,
427.1736; found, 427.1727.
1
reaction (1.843 g, 42%, 3 steps). H-NMR (CDCl3, 400 MHz)
d 7.43–7.25 (m, 10H, Ar-H), 5.52 (s, 1H, PhCHO2), 5.32 (b s, 1H,
NH), 5.16 (t, J3,4 = J2,3 9.8 Hz, H3), 4.93 (d, 1H, Jgem 12.2 Hz,
PhCHH), 4.60 (d, 1H, PhCHH), 4.53 (d, 1H, J1,2 8.8 Hz, H1),
4.39 (dd, 1H, J5,6a 5.4 Hz, J6a,6b 10.2 Hz, H6a), 4.15 (m, 1H, H2),
3.84 (t, 1H, H6b), 3.73 (t, 1H, H4), 3.48 (ddd, 1H, H5), 2.06, 1.82
(s, 6H, C(O)CH3). 13C-NMR (CDCl3, 100 MHz) d 171.5, 170.3
(CO), 137.2, 129.4, 128.7, 128.5, 128.2, 128.1, 126.4 (Ar), 101.6,
100.9 (PhCHO2, C1), 78.8, 72.1, 71.0, 68.9, 66.8 (C3, C4, C5, C6,
PhCH2O1), 54.7 (C2), 23.5, 21.1 (C(O)CH3).
Benzyl O-(2-acetamido-3,4,6-tri-O-acetyl-2-deoxy-b-D-
glucopyranosyl)-(1→3)-2-acetamido-6-O-benzyl-2-deoxy-b-D-
glucopyranoside (22)
Donor 3 (310 mg, 0.80 mmol) and 3,4-acceptor 21 (160 mg,
0.40 mmol) were dissolved in CH2Cl2 (2 mL) before Sc(OTf)3
(59 mg, 0.12 mmol) was added. The reaction mixture was heated
to 80 ◦C under microwave irradiation for 8 h before it was cooled
and poured into a separation funnel containing AcOEt (20 mL)
and NaHCO3 (sat. aq. 20 mL). The aqueous layer further was
extracted with AcOEt (2 × 20 mL) before the combined organic
extracts were washed with H2O (10 mL) and reduced to half
the volume under reduced pressure. The remaining solution was
cooled in a refrigerator for 2 h before the precipitate was isolated by
vacuum filtration. This gave disaccharide 22 as a colourless powder
To a stirred solution of crude benzyl 2-acetamido-2-O-
acetyl-4,6-O-benzylidene-2-deoxy-b-D-glucopyranoside (1.840 g,
4.17 mmol) in dry CH2Cl2 (5 mL) and triethylsilane (3.3 mL,
◦
20.9 mmol) at 0 C under an atmosphere of N2 was added TFA
(1.6 mL, 20.9 mmol) over 5 min. The reaction mixture was stirred
on an ice bath for 15 min. and then at ambient temperature
for 30 min. The mixture was transferred to a separation funnel
containing NaHCO3 (25 mL, sat. aq. Caution: development of
CO2). The aqueous phase was extracted three times with CH2Cl2
(25 mL) and the combined organic layers dried over MgSO4 and
concentrated under vacuum. The product was purified by column
chromatography (EtOAc–pentane 3 : 1 → EtOAc–CH3OH 10 : 1)
to give the desired dibenzyl glucoside (0.991 g 54%) as a colourless
1
(275 mg, 90%). H-NMR (DMSO-d6, 400 MHz): d 7.75 (d, 1H,
JNH,2 8.8 Hz, H2/H2ꢀ), 7.70 (d, 1H, JNH,2 8.8 Hz, H2ꢀ/H2), 7.34–
7.25 (m, 10H, ArH), 5.26 (t, 1H, J2 ,3 = J3 ,4 9.8 Hz, H3ꢀ), 4.82 (d,
ꢀ
ꢀ
ꢀ ꢀ
1H, J1 ,2 8.8 Hz, H1ꢀ), 4.81 (t, 1H, H4ꢀ), 4.74 (d, 1H, Jgem 12.4 Hz,
PhCHHO1), 4.53 (s, 2H, PhCH2O6), 4.50 (d, 1H, PhCHHO1),
4.44 (d, 1H, J1,2 8.0 Hz, H1), 4.41 (b s, 1H, OH), 4.16 (dd, 1H,
ꢀ
ꢀ
1
powder. Rf (EtOAc) 0.26. H-NMR (CDCl3, 400 MHz) d 7.32–
J5 ,6 a 5.4 Hz, J6 a,6 b 12.0 Hz, H6ꢀa), 4.03 (dd, 1H, J5 ,6 b 2.6 Hz,
H6ꢀb), 3.83–3.78 (m, 1H, H5ꢀ), 3.75 (b d, 1H, J6a,6b 9.6 Hz, H6a),
3.65 (t, 1H, 9.2 Hz, H3), 3.60–3.52 (m, 3H, H2, H6b, H2ꢀ), 3.37–
3.33 (m, 1H, H5), 3.24 (t, 1H, H4), 1.97, 1.95, 1.90, 1.83, 1.73 (s,
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ ꢀ
7.27 (m, 10H, ArH), 6.46 (d, 1H, JNH,2 9.2 Hz, NHAc), 5.08 (t,
1H, J2,3 = J3,4 9.2 Hz, H3), 4.83 (d, 1H, J 12.4 Hz, PhCHHO1),
4.60–4.54 (m, 4H, H-1, PhCHHO1, PhCH2O6), 3.99 (q, 1H, H2),
3.86 (b s, 1H, OH), 3.81 (m, 1H, H6a), 3.72 (dd, 1H, J6a,6b 11.2 Hz,
3282 | Org. Biomol. Chem., 2008, 6, 3276–3283
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
15H, C(O)CH3). 13C-NMR (DMSO-d6, 100 MHz): d 170.7, 170.3,
170.1, 170.0, 170.0 (CO), 139.3, 138.6, 128.9, 128.8, 128.1, 128.0,
128.0, 127.9 (Ar), 101.1 (C1), 100.8 (C1ꢀ), 83.5 (C3), 75.8 (C4), 73.0
(PhCH2O6), 72.7 (C3ꢀ), 71.2 (C5ꢀ), 70.4 (PhCH2O1), 70.1 (C6),
69.5, 69.5 (C2ꢀ, C4ꢀ), 62.7 (C6ꢀ), 54.5, 54.4 (C2, C2ꢀ), 23.9, 23.4
(NHC(O)CH3), 21.1, 21.1, 21.1 (OC(O)CH3). [a]2D1 −8.5 (c 1.0,
CH3CO2H). HRMS(ES+): calcd. for C36H46N2O14Na: 753.2847;
found: 753.2853.
17 V. Wittmann and D. Lennartz, Eur. J. Org. Chem., 2002, 1363–1367.
18 C. F. Crasto and G. B. Jones, Tetrahedron Lett., 2004, 45, 4891–4894.
19 Y. Cai, C.-C. Ling and D. R. Bundle, Org. Lett., 2005, 7, 4021–4024.
20 S. Kobayashi, M. Sugiura, H. Kitagawa and W. W.-L. Lam, Chem. Rev.,
2002, 102, 2227–2302.
21 b-D-GlcNAc tetraacetate (3) could alternatively be prepared on a large
scale by N-acetylation using Ac2O in CH2Cl2–Et3N from the b-D-
glucosammonium chloride tetraacetate: H. Myszka, D. Bednarczyk,
M. Najdar and W. Kaca, Carbohydr. Res., 2003, 338, 133–141.
22 D. J. Chambers, G. R. Evans and A. J. Fairbanks, Tetrahedron, 2004,
60, 8411–8419.
23 F. I. Comer and G. W. Hart, J. Biol. Chem., 2000, 275, 29179–29182.
24 L. Szabo`, Y. Li and R. Polt, Tetrahedron Lett., 1991, 5, 585–588.
25 C.-R. Shie, Z.-H. Tzeng, S. S. Kulkarni, B.-J. Uang, C.-Y. Hsu and S.-C.
Hung, Angew. Chem., Int. Ed., 2005, 44, 1665–1668.
26 M. P. DeNinno, J. B. Etienne and K. C. Kimberley, Tetrahedron Lett.,
1995, 36, 669–672.
27 This donor was prepared as the b-acetate only using the procedure given
by M. Dowlut, D. G. Hall and O. Hindsgaul, J. Org. Chem., 2005, 70,
9809–9813.
28 M. Bols, X. Liang and H. H. Jensen, J. Org. Chem., 2002, 67, 8970–
8974; H. H. Jensen and M. Bols, Acc. Chem. Res., 2006, 39, 259–265;
K. M. Koeller and C.-H. Wong, Chem. Rev., 2000, 100, 4465–4493.
29 R. Autar, R. M. J. Liskamp and R. J. Pieters, Carbohydr. Res., 2005,
340, 2436–2442.
30 A. Corsaro, U. Chiacchio, V. Pistara and G. Romeo, Curr. Org. Chem.,
2004, 8, 511–538.
31 Y. Yoshimura, H. Shimizu, H. Hinou and S.-I. Nishimura, Tetrahedron
Lett., 2005, 46, 4701–4705.
32 Microwaves in Organic Synthesis, ed. A. Loupy, Wiley-VCH, Weinheim,
2nd edn, 2006.
33 K. Larsen, K. Worm-Leonhard, P. Olsen, A. Hoel and K. J. Jensen,
Org. Biomol. Chem., 2005, 3, 3966–3970.
Acknowledgements
We thank OChem Graduate School, Department of Chemistry,
University of Aarhus for support and Sarah Allman for suggesting
the use of microwave irradiation.
References
1 R. A. Dwek, Chem. Rev., 1996, 96, 683–720.
2 A. F. G. Bongat and A. V. Demchenko, Carbohydr. Res., 2007, 324,
374–406.
3 T. B. Windholz and D. B. R. Johnston, Tetrahedron Lett., 1967, 27,
2555–2557; U. Ellervik and G. Magnusson, Carbohydr. Res., 1996, 280,
251–260.
4 J. Banoub, P. Boullanger and D. Lafont, Chem. Rev., 1992, 92, 1167–
1195; R. U. Lemieux, T. Takeda, B. Y. Chung, Synthetic Methods
for Carbohydrates, ed. H. S. El Khadem, ACS Symposium Series,
Washington DC, 1976, 39, pp. 90–115.
5 J. C. Castro-Palomino and R. R. Schmidt, Tetrahedron Lett., 1995, 36,
6871–6874.
34 M. L. Bohn, M. I. Colombo, P. L. Pisano, C. A. Stortz and E. A.
Ru´veda, Carbohydr. Res., 2007, 342, 2522–2536; S. I. v. Kasteren, H. B.
Kramer, H. H. Jensen, S. J. Campbell, J. Kirkpatrick, N. J. Oldham,
D. C. Anthony and B. G. Davis, Nature, 2007, 446, 1105–1109; J. Xue,
S. D. Khaja, R. D. Locke and K. L. Matta, Synlett, 2004, 861–865; U.
Ellervik and G. Magnusson, J. Org. Chem., 1998, 63, 9314–9322.
35 (a) For spectral data of 1→3 linked disaccharide, see: A. Rohlenova´, M.
6 For the highly stereoselective construction of the 1,2-trans-b-glucosidic
linkage using 2-azido-2-deoxy glucosyl donors, see T. Tsuda, S.
Nakamura and S. Hashimoto, Tetrahedron, 2004, 60, 10711–10737;
R. R. Schmidt, M. Behrendt and A. Toepfer, Synlett, 1990, 694–696.
7 P. Boullanger, M. Jouineau, B. Bouammali, D. Lafont and G. Descotes,
Carbohydr. Res., 1990, 202, 151–164.
8 G. Blatter, J. Beau and J. Jacquinet, Carbohydr. Res., 1994, 260, 189–202.
9 J. S. Debenham, R. Madsen, C. Roberts and B. Fraser-Reid, J. Am.
Chem. Soc., 1995, 117, 3302–3303.
10 E. Meinjohanns, M. Meldal, H. Paulsen and K. Bock, J. Chem. Soc.,
Perkin Trans. 1, 1995, 405–415; K. J. Jensen, P. R. Hansen, D. Venugopal
and G. Barany, J. Am. Chem. Soc., 1996, 118, 3148–3155.
11 H. Shimizu, Y. Ito, Y. Matsuzaki, H. Iijima and T. Ogawa, Biosci.,
Biotechnol., Biochem., 1996, 60, 73–76.
12 M. R. E. Aly, J. C. Castro-Palomino, E.-S. I. Ibrahim, E.-S. H. El-Ashry
and R. R. Schmidt, Eur. J. Org. Chem., 1998, 2305–2316.
13 H. Jiao and O. Hindsgaul, Angew. Chem., Int. Ed., 1999, 38, 346–348.
14 J. C. Castro-Palomino and R. R. Schmidt, Tetrahedron Lett., 2000, 41,
629–632.
ˇ
Ledvina, D. Saman and K. Bezousˇka, Collect. Czech. Chem. Commun.,
2004, 69, 1781–1804; (b) For spectral data of 1→4 linked disaccharide,
see: S. Singh, J. Packwood, C. J. Samuel, P. Critchley and D. H. G.
Crout, Carbohydr. Res., 1995, 279, 293–305.
36 K. P. Ravindranathan and R. A. Field, Tetrahedron, 1997, 53, 11753–
11766.
37 G.-J. Boons and K. J. Hale, Organic Synthesis with Carbohydrates,
Sheffield Academic Press, 2000.
38 B. Aguilera, A. Ferna´ndez-Mayoralas and C. Jaramillo, Tetrahedron,
1997, 53, 5863–5876.
39 K. L. Matta, E. A. Johnson and J. J. Barlow, Carbohydr. Res., 1973, 26,
215–218.
40 Melting point in accordance with that reported in: S. S. Rana and K. L.
Matta, Carbohydr. Res., 1983, 117, 101–111.
15 M. Kiso and L. Anderson, Carbohydr. Res., 1979, 72, C12–C14.
16 T. Osawa, Chem. Pharm. Bull., 1960, 8, 597–610.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3276–3283 | 3283
©