Rapid access to glucopyranosyl-1,2,3-triazoles via Cu(I)-catalyzed reactions in water

Article abstract of DOI:10.3987/COM-04-10218

1-Azido-1-deoxy-2,3,4,6-tetra-O-acetyl-β-D-glucopyranose reacts with various terminal alkynes in the presence of CUSO₄/ascorbic acid in water to give the corresponding 1,4-disubstituted 1,2,3-triazoles, which are isolated in high yield and puri

Full text of DOI:10.3987/COM-04-10218

HETEROCYCLES, Vol. 63, No. 12, 2004, pp. 2719 - 2725
Received, 2nd September, 2004, Accepted, 21st October, 2004, Published online, 26th October, 2004
RAPID ACCESS TO GLUCOPYRANOSYL-1,2,3-TRIAZOLES VIA
Cu(I)-CATALYZED REACTIONS IN WATER
Rakesh A. Akula, David P. Temelkoff, Nicole D. Artis, and Peter Norris*
Department of Chemistry, Youngstown State University, 1 University Plaza,
Youngstown, OH 44555, U.S.A.
Abstract – 1-Azido-1-deoxy-2,3,4,6-tetra-O-acetyl-β-D-glucopyranose reacts
with various terminal alkynes in the presence of CuSO /ascorbic acid in water to
4
give the corresponding 1,4-disubstituted 1,2,3-triazoles, which are isolated in
high yield and purity by simply filtering the precipitate from the reaction mixture.
Several sugar-derived acetylenes react similarly to yield triazole-linked
disaccharide analogs.
With the growing appreciation of the roles that sugars play in numerous biological events there is a demand
for new compounds that may serve as tools with which to study such processes. Impressive advances have
been made in the synthesis of O-glycosides, as well as C-glycosides and N-glycosides, which serve as
1
glycomimetic analogs of parent O-glycosides. Numerous C-glycosidic structures are found in nature and
2
the N-glycoside motif appears in the nucleic acids and in N-linked glycopeptides.
The control of stereochemistry during O-glycosylation remains one of the ultimate challenges in synthetic
carbohydrate chemistry, and similar difficulties are encountered in C-glycoside formation. There are few
truly general approaches to each of these types of compound. The situation is somewhat different in the
N-glycoside field due to the fact that the most common precursors, i.e. glycosyl azides, are usually prepared
by stereospecific bimolecular displacement of a glycosyl halide, many of which are available in
3
diastereomerically pure form.
The formation of 1,2,3-triazoles at the anomeric position of carbohydrates has traditionally involved the
4
5
Huisgen 1,3-dipolar cycloaddition between a glycosyl azide and an alkyne partner, however the utility of
the process has been limited due to slow reactions with unactivated alkynes and the formation of isomeric

mixtures of 1,4- and 1,5-disubstituted triazole products. Since glycosyl-1,2,3-triazoles (2, Figure 1) could
potentially serve as mimics of glycosyl amides (1, Figure 1) we were interested in applying recently
developed methods for 1,2,3-triazole synthesis based on Cu(I) catalysis.
Sharpless and colleagues have suggested the 1,2,3-triazole ring as a useful way of conjugating biologically
6
active groups. Meldal and coworkers reported the use of Cu(I) halides to form peptido-1,4-substituted
7
1
,2,3-triazoles regiospecifically, and very shortly afterwards Sharpless and Fokin published their work on
the use of CuSO
4
in azide-acetylene “ligations” in which the Cu(I) catalyst is generated by reaction of
8
CuSO
4
with sodium ascorbate. Aside from the affordability of this reagent mixture, the fact that the
reactions were performed in aqueous medium made this system extremely attractive for rapid
glycosyl-1,2,3-triazole synthesis.
HO
HO
HO
HO
HO
HO
H
N
N
O
O
N
N
R
R
OH
OH
H
O
1
2
Figure 1
There have been several recent reports dealing with copper(I)-catalyzed 1,2,3-triazole synthesis in
carbohydrate systems. Santoyo-González and colleagues have used an organic-soluble catalyst to produce a
9
wide variety of triazole-linked neoglycoconjugates, and Gin and coworkers have used similar chemistry to
1
0
build cyclodextrin analogs. We now communicate our work on the rapid, regiospecific synthesis and
isolation of various glucosyl-1,4-disubstituted 1,2,3-triazoles using the Sharpless CuSO
4
/ascorbic acid
system using water as the solvent.
1
1
1
-Azido-1-deoxy-2,3,4,6-tetra-O-acetyl-β-D-glucopyranose (3, Eq. 1) is available on large scale and is
insoluble in water at room temperature. In initial experiments to determine if co-solvents where needed for
reactions with various acetylenes, it was found that uncatalyzed reactions were possible but were very slow
(
up to four days with 1-ethynyl-3-fluorobenzene) and resulted in the isolation of mixtures of 1,4- and
,5-disubstituted triazoles (4, Eq. 1). No co-solvent was necessary and the products readily precipitated and
1
could be isolated by filtration.
AcO
AcO
AcO
AcO
N
O
H
R
O
N
N
AcO
AcO
N3
H O, 70 o
C
R
AcO
AcO
2
3
4
Equation 1
With CuSO /ascorbic acid the reactions of azide (3) with various terminal acetylenes were complete within
4
1
2
eight hours and the products, isolated by filtration, were found to be a single triazole (5, Eq. 2, Table 1).

1
Purity was >90% in all cases as judged by H NMR spectra. The 1,4-disubstituted triazole products
correspond to the major isomer formed in the uncatalyzed reactions, which is the accepted outcome for such
1
3,14
conventional Huisgen dipolar cycloadditions.
Additionally, nOe experiments on triazole 5f (R = Ph)
show clear interaction between H-1 of the pyranose ring and the proton attached to the triazole ring (Eq. 2),
which would be unlikely if the product where the 1,5-isomer.
AcO
AcO
AcO
AcO
AcO
AcO
N
O
H
R
O
N
N
R
N3
AcO
CuSO
4
AcO
H
H
ascorbic acid
H O, 70
3
5
o
C
2
nOe
Equation 2
The simplicity of this method suggests it would be useful for the synthesis of libraries of triazole-linked
15
oligosaccharide analogs. To illustrate this, reaction of azide (3) with D-glucuronic acid-derived alkyne (6)
Scheme 1) in water in the presence of CuSO /ascorbic acid gave a single triazole (7, Scheme 1) in 76%
(
4
1
6
yield, which again was isolated in very pure form simply by filtration and drying. A related
D-galactose-derived alkyne (8, Scheme 1) required the use of t-BuOH as co-solvent due to the low
solubility of isopropylidene derivatives versus acetate-protected sugars. However, the product triazole (9,
Scheme 1) could be isolated in 84% yield by evaporating most of the t-BuOH and allowing the product to
1
8
precipitate from the remaining aqueous mixture.
O
O
O
O
O
O
O
AcO
AcO
AcO
AcO
OAc
O
AcO
AcO
AcO
N
AcO
N
O
N
N
AcO
O
O
N
N
O
6
8
O
O
O
O
O
O
AcO
OAc
OAc
3
AcO
O
H
H
AcO
AcO
CuSO4
CuSO4
7
ascorbic acid
H O, 70
ascorbic acid
H O/t-BuOH, 70 oC
9
o
O
C
2
2
Scheme 1
ACKNOWLEDGEMENTS
The American Chemical Society’s Project SEED is thanked for a stipend for N.D.A. This work was
sponsored by the National Institutes of Health (R15-AI053112-01).
REFERENCES AND NOTES
1
. For recent reviews in O-glycoside, C-glycoside, and glycosyl amino acid synthesis see
Carbohydrates,’ ed. by H. M. I. Osborn, Elsevier, Oxford, UK, 2003.
. B. G. Davis, Chem. Rev., 2002, 102, 579.
‘
2

Table 1. Alkynes used in this work and yields for resultant glucosyl-1,2,3-triazoles.
%Yielda
92
Entry
a
Alkyne
Glucosyl triazole product
AcO
N
O
N
N
AcO
AcO
CO Et
CO Et
2
2
AcO
H
AcO
N
N
O
N
N
b
c
(CH ) Me
AcO
(CH ) Me
89
81
2
4
2 4
AcO
AcO
H
H
AcO
O
N
N
AcO
(CH ) Me
2 6
(
(
(
CH ) Me
2
6
AcO
AcO
AcO
N
N
N
N
O
N
N
d
e
f
AcO
(CH ) Me
2 7
78
94
85
CH ) Me
2
7
AcO
AcO
H
H
H
H
AcO
O
N
N
AcO
(CH ) Me
CH ) Me
2 9
2
9
AcO
AcO
AcO
O
N
N
AcO
AcO
AcO
AcO
O
N
N
g
AcO
8
2
AcO
AcO
F
F
AcO
N
O
N
N
h
i
AcO
61
79
80
AcO
AcO
H
H
AcO
N
O
N
N
AcO
OMe
OMe
AcO
AcO
AcO
AcO
AcO
N
O
N
N
j
AcO
H
a
1
13
Satisfactory H NMR, C NMR, and HRMS spectral data were obtained for all compounds.

3
4
5
. R. M. Cicchillo and P. Norris, Carbohydr. Res., 2000, 328, 431 and references cited therein.
. R. Huisgen, R. Knorr, L. Moebius, and G. Szeimies, Chem. Ber., 1965, 98, 4014.
. A. San-Felix, R. Alvarez, S. Velazquez, E. De Clerq, J. Balzarini, and M. J. Camarasa, Nucleosides
Nucleotides, 1995, 14, 595.
6. H. C. Kolb, M. G. Finn, and K. B. Sharpless, Angew. Chem., Int. Ed., 2001, 40, 2004.
7. C. W. Tornøe, C. Christensen, and M. Meldal, J. Org. Chem., 2002, 67, 3057.
8. V. V. Rostovtsev, L. G. Green, V. V. Fokin, and K. B. Sharpless, Angew. Chem., Int. Ed., 2002, 41,
2
596.
9
. F. Pérez-Balderas, M. Ortega-Muñoz, J. Morales-Sanfrutos, F. Hernández-Mateo, F. G. Calvo-Flores,
J. A. Calvo-Asín, J. Isac-García, and F. Santoyo-González, Org. Lett., 2003, 5, 1951.
1
1
1
0. K. D. Bodine, D. Y. Gin, and M. S. Gin, J. Am. Chem. Soc., 2004, 126, 1638.
1. F. M. Ibatullin and K. A. Shabalin, Synth. Commun., 2000, 30, 2819 and references cited therein.
2. Typical procedure for synthesis of 1-glucopyranosyl-4-substituted 1,2,3-triazoles (5): Azide (3) (1.0 g,
2.68 mmol), CuSO
4
(0.01 g, 0.04 mmol), ascorbic acid (0.1 g, 0.56 mmol) and phenylacetylene (0.3
o
mL, 2.70 mmol) were heated together in water (15 mL) at 70 C for 8 h. The mixture was cooled to rt
and then in an ice bath. The solid was filtered and washed with water (10 mL) and methanol (10 mL)
1
to leave a pale yellow powder (1.08 g, 85%), which was >90% pure by H NMR spectrum and
homogenous by TLC. Recrystallization from 95% ethanol gave 5f as a fluffy white solid: mp 198-202
o
o
1
C; [α]
D 2 2 3
-85.8 (c 1.0, CH Cl ); H NMR (400 MHz, CDCl ) δ 1.88, 2.03, 2.07, 2.08 (4 x s, 3H each,
4
x COCH ), 4.03 (m, 1H, H-5), 4.15 (dd, 1H, J = 1.8, 12.5 Hz, H-6), 4.32 (dd, 1H, J = 5.1, 12.6 Hz,
3
H-6’), 5.27 (t, 1H, J = 9.5 Hz, H-2), 5.44 (t, 1H, J = 9.3 Hz, H-3), 5.52 (t, 1H, J = 9.5 Hz, H-4), 5.93
1
3
(
d, 1H, J = 9.2 Hz, H-1), 7.25-7.84 (m, 5H, Ar-H), 8.00 (s, 1H, triazole-H); C NMR (100 MHz,
CDCl
3
) δ 21.5, 21.8 (double intensity), 22.0, 62.7, 68.9, 71.3, 73.9, 76.3, 86.9, 118.8, 126.9, 129.6,
(+Na): 498.1488, found
1
4
29.9, 130.9, 149.5, 169.8, 169.9, 170.3, 170.8; HRMS calcd for C22
H
25
N
3
O
9
1
4
98.1451.
1
1
3. P. Norris, D. Horton, and B. R. Levine, Heterocycles, 1996, 43, 2643 and references cited therein.
4. Triazole (5f) has been reported previously: M. T. Garcia-Lopez, G. Garcia-Munoz, J. Iglesias, and R.
Madronero, J. Heterocycl. Chem., 1969, 6, 639.
1
1
5. S. Sabesan, ‘Preparation of Triazole-linked Oligosaccharides via Cycloaddition of Alkyne with Azide
Sugars.’ U.S. Patent 6,664,399, December 16, 2003 (Chem. Abstr., 2004, 140, 667).
6. Alkyne (6) was prepared in 95% yield by treating 1,2,3,4-tetra-O-acetyl-β-D-glucopyranuronosyl
1
8
o
o
1
chloride with propargyl alcohol in pyridine: mp 138-141 C; [α]
400 MHz, CDCl ) δ 2.04, 2.05, 2.07, 2.13 (4 x s, 3H each, 4 x COCH
alkyne-H), 4.24 (d, 1H, J = 9.5 Hz, H-5), 4.71 (m, 2H, CH ), 5.15 (dd, 1H, J = 7.7, 9.0 Hz, H-2), 5.25
D
+10.8 (c 1.0, CH
2 2
Cl ); H NMR
(
3
3
), 2.53 (t, 1H, J = 2.5 Hz,
2

1
3
(
t, 1H, J = 9.5 Hz, H-4), 5.33 (t, 1H, J = 9.0 Hz, H-3), 5.78 (d, 1H, J = 7.7 Hz, H-1); C NMR (100
MHz, CDCl ) δ 21.8 (triple intensity), 22.0, 54.8, 69.9, 71.2, 72.8, 73.8, 77.2, 77.5, 92.3, 166.7, 169.7,
70.1, 170.4, 170.8; HRMS calcd for C17 11 (+Na): 423.0903, found 423.0907. Triazole (7) was
prepared by heating azide (3) (746 mg, 2.0 mmol) and alkyne (6) (800 mg, 2.0 mmol) with CuSO
3
1
20
H O
4
(
(
0.025 mL of a 1M aqueous solution) and ascorbic acid (0.25 mL of a 1M aqueous solution) in water
o
15 mL) at 70 C for 12 h. After cooling to rt the precipitate was collected on a glass frit and washed
o
o
D
with cold methanol. Triazole (7) (1.2 g, 76%) was isolated as a tan solid: mp 204-208 C; [α] -5.8 (c
1
1
1
1
5
.2, CH
2
Cl
2
); H NMR (400 MHz, CDCl
3
, glu = glucopyranose ring, gluA = glucuronic acid ring) δ
), 3.99 (m, 1H, H-5glu), 4.14 (dd,
H, J = 2.2, 12.9 Hz, H-6glu), 4.20 (d, 1H, J = 9.7 Hz, H-5gluA), 4.29 (dd, 1H, J = 4.9, 12.6 Hz, H-6’glu),
.11-5.48 (m, 8H, H-2glu, H-2gluA, H-3glu, H-3gluA, H-4glu, H-4gluA, -OCH -triazole), 5.76 (d, 1H, J = 7.7
.89, 1.99, 2.01, 2.04, 2.08, 2.10, 2.12 (7 x s, total 24 H, 8 x COCH
3
2
1
3
Hz, H-1gluA), 5.88 (d, 1H, J = 9.0 Hz, H-1glu), 7.89 (s, 1H, triazole-H); C NMR (100 MHz, CDCl
3
) δ
1.4, 21.7, 21.8 (quadruple intensity), 21.9, 22.0, 59.9, 62.6, 68.7, 69.8, 71.2, 71.3, 72.9, 73.7, 73.8,
6.2, 86.8, 92.3, 124.0, 143.0, 167.2, 169.7, 169.8, 170.1, 170.2, 170.5, 170.7, 170.8, 171.4; HRMS
20 (+Na): 796.2025, found 796.2031.
7. Alkyne (8) was prepared in 80% yield by treating 1,2:3,4-di-O-isopropylidene-D-galactopyranose
2
7
39 3
calcd for C31H N O
1
o
o
1
with propargyl bromide and KOH in MeCN: mp 50-54 C; [α]
D
-3.2 (c 1.0, CH
), 2.44 (t, 1H, J = 2.0 Hz,
alkyne-H), 3.67 (dd, 1H, J = 7.1, 10.1 Hz, H-6), 3.78 (dd, 1H, J = 5.2, 10.2 Hz, H-6’), 4.00 (m, 1H,
H-5), 4.17-4.33 (m, 4H, H-2, H-4, propargyl CH ), 4.61 (dd, 1H, J = 2.3, 8.0 Hz, H-3), 5.55 (d, 1H, J
) δ 25.7, 26.1, 27.2, 27.3, 59.6, 67.8, 69.8, 71.5, 71.7,
2.2, 75.8, 80.7, 97.4, 109.6, 110.3; HRMS calcd for C H O (+Na): 321.1314, found 321.1297.
2 2
Cl ); H NMR
(400 MHz, CDCl
3
) δ 1.33, 1.35, 1.46, 1.55 (4 x s, 3H each, 4 x CH
3
2
1
3
=
5.1 Hz, H-1); C NMR (100 MHz, CDCl
3
7
1
5
22
6
Triazole (9) was formed in a similar fashion to compound (7) with the addition of t-BuOH as a
co-solvent. Thus, azide (3) (2.61 g, 7.0 mmol), alkyne (8) (2.09 g, 7.0 mmol), CuSO (0.05 mL of a
M aqueous solution) and ascorbic acid (0.5 mL of a 1M aqueous solution) in water were heated in a
4
1
o
mixture of water (10 mL) and t-BuOH (10 mL) at 70 C for 90 min. After evaporating the t-BuOH the
precipitate was filtered and washed with cold water to give triazole (9) (3.97 g, 84%) as a colorless
o
o
1
powder: mp 160-163 C; [α]
D
-44.3 (c 1.1, CHCl
3
); H NMR (400 MHz, CDCl
3
, glu = glucopyranose
), 1.88, 2.04, 2.08,
), 3.68 (dd, 1H, J = 7.0, 10.1 Hz, H-6gal), 3.74 (dd, 1H, J = 5.6, 10.2
Hz, H-6’gal), 4.00 (m, 2H, H-5glu, H-5gal), 4.12 (dd, 1H, J = 1.5, 12.5, H-6glu), 4.25-4.33 (m, 3H,
H-6’glu, H-2gal, H-3gal), 4.61 (dd, 1H, J = 2.2, 7.9 Hz, H-4gal), 4.68 (m, 2H, -OCH -triazole), 5.24 (t,
H, J = 9.4 Hz, H-4glu), 5.40 (m, 2H, H-2glu, H-3glu), 5.56 (d, 1H, J = 4.9 Hz, H-1gal), 5.89 (d, 1H, J =
ring, gal = galactopyranose ring) δ 1.34, 1.35, 1.45, 1.54 (4 x s, 3H each, 4 x CH
.09 (4 x s, 3H each, 4 x COCH
3
2
3
2
1
8
1
3
.8 Hz, H-1glu), 7.84 (s, 1H, triazole-H); C NMR (100 MHz, CDCl ) δ 21.4, 21.7, 21.8, 21.9, 25.7,
3

2
1
6
6.1, 27.2, 27.3, 62.7, 65.8, 67.9, 68.8, 70.5, 71.4, 71.6, 71.7, 72.2, 73.8, 76.1, 86.7, 97.4, 109.6, 110.3,
41 3
22.0, 146.9, 169.8, 170.3, 170.8, 171.4; HRMS calcd for C29H N O15 (+Na): 694.2435, found
94.2433.
1
8. M. Tosin and P. V. Murphy, Org. Lett., 2002, 4, 3675.