Carbohydrate triazoles and isoxazoles as inhibitors of galectins-1 and -3

Article abstract of DOI:10.1039/b517529a

Galactosides and lactosides bearing triazoles or isoxazoles, regiospecifically prepared by [1,3]-dipolar cycloadditions between alkynes, azides or nitrile oxides, provided specific galectin-1 and -3 inhibitors with potencies as low as 20 μM. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b517529a

View Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Carbohydrate triazoles and isoxazoles as inhibitors of galectins-1
and -3{
Denis Gigue`re,a Ramesh Patnam,a Marc-Andre´ Bellefleur,a Christian St-Pierre,b Sachiko Satob and
Rene´ Roy*a
Received (in Bloomington, IN, USA) 9th December 2005, Accepted 28th February 2006
First published as an Advance Article on the web 16th March 2006
DOI: 10.1039/b517529a
due to their stability.17 Isoxazoles are also useful from the point of
Galactosides and lactosides bearing triazoles or isoxazoles,
view of their stability under physiological pH and are easy to
make. 3,5-Disubstituted isoxazoles are more difficult to synthesise
but new methods have recently been discovered that facilitate their
synthesis (Scheme 1).16,18
regiospecifically prepared by [1,3]-dipolar cycloadditions
between alkynes, azides or nitrile oxides, provided specific
galectin-1 and -3 inhibitors with potencies as low as 20 mM.
Naturally occurring carbohydrate ligands for galectins19 have
low affinities, are too polar to be used as oral drugs, and possess
low physiological stabilities due to their acid sensitive glycosidic
bonds. A rational design approach for the development of new
classes of glycomimetic inhibitors with high affinity, stability, and
specificity is thus needed. Nilsson et al. have explored the
39-position of lactoside derivatives toward the synthesis of high
affinity inhibitors of galectin-3.20,21 Some N-39-triazole analogs
provided high affinity enhancement. However, the lengthy
synthetic scheme stimulated the impetus for a shorter synthesis.
We thus report herein the straightforward synthesis and evaluation
of O-39 triazole and isoxazole analogs of both galactosides and
lactosides. This strategy was also applied to the anomeric position.
The first alkyne adduct was synthesized from commercially
available galactosyl bromide 1 shown in Scheme 2. Phase transfer
catalyzed nucleophilic displacement22 and de-O-acetylation using
methanolic sodium methoxide afforded only phenyl 1-thio-b-D-
galactoside 2. Dibutylstannylene acetal formation with dibutyltin
oxide23 and in situ reaction with propargyl bromide allowed the
regioselective formation of a 3-propynyl ether. Finally, protection
under standard conditions provided intermediate 4.
Galectins are a family of cytosolic b-D-galactoside binding
proteins of which fourteen members have been identified in
mammals.1,2 Galectin-1 (Gal-1) is a homodimer composed of
subunits of approximately 130 amino acids and each subunit
folds as one compact globular domain.1 Galectin-3 (Gal-3) is
quite unique and has one carbohydrate recognition domain
(CRD) ending with a collagen-like repeat of peptides rich in
proline and glycine capable of self association.3,4 The roles of the
galectin family are not yet clear, but a striking common feature of
all galectins is the strong modulation of their expression during
development, differentiation stages, and under different physio-
logical or pathological conditions.2 Recent studies have demon-
strated that Gal-3 is involved in colon cancer metastasis,5 brain
tumor progression,6 inhibition of metastasis-associated cancer
cell adhesion,7 and may play a key role in innate immunity.8
Other reports suggest that Gal-39 and Gal-110 can regulate
apoptosis processes.11 It has also been reported that Gal-1 acts
as an insoluble host factor that promotes HIV-1 infectivity
through stabilization of virus attachment to host cells.12
Recent developments have been reported in the synthesis of
carbohydrate-based 1,2,3-triazoles.13,14 Meldal15 and Sharpless16
have solved the problem of 1,4-regioselectivity by using copper(I)
catalysts (Scheme 1). This non-concerted cycloaddition is powerful
for the synthesis of non-natural heterocycles which are attractive
In order to synthesize more hydrolytically stable analogs,
b-C-propynyl galactoside 6 was synthesised by ozonolysis of the
known b-C-allyl derivative 524 followed by the Ohira25 modifica-
tion of the Seyfert–Gilbert homologation reaction under mildly
basic conditions (Scheme 3).
All terminal alkynes 4 and 6–9 reacted with a panel of azides
(10, 11, and 13) or nitrile oxide 12 to give product containing only
one regioisomer, summarized in Table 1. Alkyne 4 was treated
with two different azides (10 and 11) for the formation of triazoles
14 and 15, respectively, designed to maximize binding interactions
with arginine 144.20 Anomeric C-propynyl galactoside 6 reacted
Scheme 1
aDepartment of Chemistry, Universite´ du Que´bec a` Montre´al, P.O. Box
8888, Succ. Centre-Ville Montreal, Que., Canada H3C 3P8.
E-mail: roy.rene@uqam.ca; Fax: +1 514 987 4054;
Tel: +514 987 3000x2546
bResearch Center for Infectious Diseases, Faculty of Medicine,
Universite´ Laval, 2705 boul. Laurier, RC-9700 Sainte-Foy, Que.,
Canada G1V 4G2
{ Electronic supplementary information (ESI) available: Experimental
procedure and compound characterization data. See DOI: 10.1039/
b517529a
Scheme 2 Reagents and conditions: a) HSPh, TBAHS, 1 M Na2CO3,
AcOEt, 75%; b) NaOMe, MeOH, quant.; c) Bu2SnO, MeOH, then
Bu4NI, propargyl bromide, benzene, 78%; d) Ac2O, pyridine, 95%.
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 2379–2381 | 2379

View Online
amine 8 to afford triazole 19 in good yield. Finally, C3-symmetric
tris-lactoside 20 was prepared from the cycloaddition of 13 with
N,N9,N0-tripropargyl-1,3,5-carboxamidobenzene 9 (obtained in
82% yield by treatment of 1,3,5-benzenetricarboxylic acid with
oxalyl chloride then propargyl amine added dropwise).
All new compounds and references 21 (galactose) and 22
(lactose) were tested by inhibition of hemagglutination assay at a
concentration of 1 mM for both galectins. Assays were performed
using red blood cells, type O, fixed with 3% glutaraldehyde–
0.0025% NaN3 in PBS.12,27 Table 2 shows inhibitory properties
and relative affinity of our derivatives toward Gal-1 and -3. The
first overall observation was that none of our compounds bound
to human Gal-4, indicating that triazole and isoxazole derivatives
have better affinities and selectivities for Gal-1 and -3.28 Triazoles
prepared from a 3-O-propynyl spacer showed the most promising
family of specific Gal-3 inhibitors (3 and 14) among the tested
Scheme 3 Reagents and conditions: a) O3, Me2S, MeOH; b)
(MeO)2P(O)CHN2C(O)CH3, Na2CO3, MeOH, then Ac2O, pyridine,
86% over 3 steps.
with azide 10 to form stable triazole 16 while O-propynyl
galactoside 7 reacted with acetone nitrile oxide generated in situ
from acetone and ceric ammonium nitrate (CAN)18 and
benzonitrile oxide26 12 (prepared from benzhydroximoyl chloride
and pyridine) to provide the corresponding isoxazole heterocycles
17 and 18, respectively. To synthesize and evaluate anomeric
triazoles, lactosyl azide 13 reacted with N-Boc protected propargyl
Table 1 Synthesis of triazoles and isoxazoles from various alkynes, azides, and nitrile oxides
Entry
1a
Alkynes
Azides or nitrile oxides
Productsd
Yields (%)d
92
2a
4
97
94
3a
10
4b
78e
5c
7
61e
98
6a
7a
13
83
a
b
CuI, DIPEA, THF. CAN, acetone, molecular sieves, DCM. NCS, pyridine, CHCl3. Yields and products are for cycloaddition and
c
d
e
deprotection steps (NaOMe, MeOH, except for entry 1: NaOH/MeOH/H2O). Based on recovered starting material.
2380 | Chem. Commun., 2006, 2379–2381
This journal is ß The Royal Society of Chemistry 2006

View Online
Canadian Research Chair in Therapeutic Chemistry (RR) and
from the Canadian Institutes for Health Research (SS).
Table 2 Inhibitory properties and relative activity for Gal-1 and -3
Inhibitory properties
(mM)
Relative activitya
Compound
no.
Notes and references
Galectin-1 Galectin-3 Galectin-1
Galectin-3
1 S. H. Barondes, D. N. Cooper, M. A. Gitt and H. Leffler, J. Biol.
Chem., 1994, 269, 20807.
3
14
15
16
17
18
19
20
. 5
1.25
. 5
5
2.5
1.25
not tested
0.02
1.25
5
. 5
. 5
. 5
. 5
. 10
40
. 10
10
20
40
10
. 10
. 10
. 10
. 10
2 L. Chiariotti, P. Salvatore, R. Frunzio and C. B. Bruni, Glycoconjugate
J., 2004, 19, 441; H. Horrie, Curr. Drug Targets, 2005, 6, 373.
3 D. N. W. Cooper and S. H. Barondes, Glycobiology, 1999, 9, 979.
4 J. Hirabayashi and K. Kasai, Glycobiology, 1993, 3, 297.
5 R. S. Bresalier, N. Mazurek, L. R. Stenberg, J. C. Byrd, C. K. Yunker,
P. Nangia-Makker and A. Raz, Gastroenterology, 1998, 115, 287.
6 B. N. Stillman, P. S. Mischel and L. G. Baum, Brain Pathol., 2005, 15,
124.
7 J. Zou, V. V. Glinsky, L. A. Landon, L. Matthews and S. L. Deutscher,
Carcinogenesis, 2005, 26, 309.
8 S. Sato and J. Nieminem, Glycoconjugate J., 2004, 19, 441; F. T. Liu,
Int. Arch. Allergy Immunol., 2005, 136, 385.
9 S. Califice, V. Castronovo and F. Van Den Brule, Int. J. Oncol., 2004,
25, 983; S. Nakahara, N. Oka and A. Raz, Apoptosis, 2005, 10, 267.
10 N. L. Perllo, K. E. Pace, J. J. Seilhamer and L. G. Baum, Nature, 1995,
378, 736; T. Szoke, K. Kayser, J. D. Baumhakel, I. Trojan, J. Furak,
L. Tislaviez, A. Horvath, K. Szluha, H.-J. Gabius and S. Andre´,
Oncology, 2005, 69, 167.
11 D. K. Hsu and F. T. Liu, Glycoconjugate J., 2004, 19, 507;
J. D. Hermandes and L. G. Baum, Glycobiology, 2002, 12, 127R.
12 M. Ouellet, S. Mercier, I. Pelletier, S. Bounou, J. Roy, J. Hirabayashi,
S. Sato and M. J. Tremblay, J. Immunol., 2005, 174, 4120.
13 J. A. F. Joosten, N. T. H. Tholen, F. A. E. Maate, A. J. Brouwer,
G. W. van Esse, D. T. S. Rijkers, R. M. J. Liskamp and R. J. Pieters,
Eur. J. Org. Chem., 2005, 3182; S. Hotha and S. Kashyap, J. Org.
Chem., 2006, 71, 364.
40
0.25
50
0.8
40 (13.3)c
1
1
3.2 (1.1)b
1
1
21 Gal
22 Lacc
a
50
0.8
Compounds 3 and 14–18 were compared to reference galactose 21 and
b
compound 20 was compared to lactose 22. Number in paren-
theses expresses the relative potency of each lactose unit in the triva-
lent derivative compared to lactose. c Lactose is y506 better than Gal.
compounds, while 15 did not have any activity, probably due to
the large size of the substituent on the triazole. The more stable
C-galactoside derivative 16 had inhibitory properties of 5 mM
against Gal-1 but no inhibition toward Gal-3. Isoxazoles carrying
two different substituents and aromatic 18 showed the best results
(1250 mM) having 40 times better affinity than the natural analog
21. No inhibition was observed against Gal-3 for 15–18, indicating
that no anomeric triazoles or isoxazoles had higher inhibitory
potency against Gal-3.
14 Y. Gao, A. Eguchi, K. Kakehi and Y. C. Lee, Bioorg. Med. Chem.,
2005, 13, 6151.
15 C. M. Tornoe, C. Christensen and M. Meldal, J. Org. Chem., 2002, 67,
3057.
16 F. Himo, T. Lovell, R. Hilgraf, V. V. Rostovtsev, L. Noodleman,
K. B. Sharpless and V. V. Fokin, J. Am. Chem. Soc., 2005, 127, 210.
17 H. Wamho, in Comprehensive Heterocyclic Chemistry; A. R. Katritzky,
C. W. Rees and K. T. Potts, eds., Pergamon: Oxford, 1984, Vol. 5,
p. 669.
Unfortunately, anomeric triazole 19 wasn’t soluble enough for
testing even with 5% DMSO added. The C3-symmetrical lactoside
20 was designed for the reason described below. First, studies have
demonstrated that some galectins are dimeric and create a soluble
network in the presence of a multivalent ligand.29 Thus,
glycoclusters may increase affinity enhancement due to multivalent
effects and formation of soluble cross-linked lattices. Glycoclusters
with a valency of three were synthesized because it was previously
demonstrated that C3-symmetrical saccharide had good affinity
with galectins30 and symmetrical analogs provided simpler analysis
due to their intrinsic symmetry. As expected, trivalent lactoside 20
provided inhibitory properties of 20 mM against Gal-1 for relative
affinity of 40 that are 13 times better for each lactose unit.
Surprisingly, the multivalent effect did not exist for Gal-3 with
inhibitory properties of 250 mM and relative affinity of 3.2 which is
almost one lactose unit by galectins.
18 K.-I. Itoh, S. Takahashi, T. Ueki, T. Sugiyama, T. Takahashi and
C. A. Horiuchi, Tetrahedron Lett., 2002, 43, 7035.
19 H. Leffler and S. H. Barondes, J. Biol. Chem., 1986, 22, 10119.
20 P. So¨rme, Y. Qian, P.-G. Nyholm, H. Leffler and U. J. Nilsson,
ChemBioChem, 2002, 3, 183; P. So¨rme, P. Arnoux, B. Kahl-Knutsson,
H. Leffler, J. M. Rini and U. J. Nilsson, J. Am. Chem. Soc., 2005, 127,
1737; I. Cumpstey, A. Sundin, H. Leffler and U. J. Nilsson, Angew.
Chem., Int. Ed., 2005, 44, 5110.
21 B. A. Salameh, H. Leffler and U. J. Nilsson, Bioorg. Med. Chem. Lett.,
2005, 15, 3344.
22 R. Roy, F. D. Tropper, S. Cao and J. M. Kim, Phase Transfer Catalyst
Mechanism and Syntheses, ACS Symp. Ser., 1997, 659, 163.
23 T. B. Grindley, Synthetic oligosaccharides, ACS Symp. Ser., 1994, 560,
51.
24 T. Uchiyama, V. P. Vassilev, T. Kajimoto, W. Wong, H. Huang,
C.-C. Lin and C.-H. Wong, J. Am. Chem. Soc., 1995, 117, 5395.
25 S. Ohira, Synth. Commun., 1989, 19, 561.
26 F. M. Hausser and X. Hu, Org. Lett., 2002, 4, 977.
27 W. T. Butler, J. Immunol., 1963, 90, 663.
28 Concentration at 0.5 mM for galectin-4.
29 T. K. Dam, H.-J. Gabius, S. Andre´, H. Kaltner, M. Lensch and C. F.
Brewer, Biochemistry, 2005, 44, 12564; N. Ahmad, H.-J. Gabius,
S. Sabesan, S. Oscarson and C. F. Brewer, Glycobiology, 2004, 14, 817.
30 S. Andre´, B. Liu, H.-J. Gabius and R. Roy, Org. Biomol. Chem., 2003,
1, 3909.
In conclusion, isoxazoles and triazoles have potential as Gal-1
selective inhibitors over other galectins and compared well with
known inhibitors.20,21,31–33 The best inhibitors among the tested
series were triazole 14 and anomeric isoxazole 18 with inhibitory
properties of 1250 mM for both inhibitors. Simple 3-propynyl
galactoside 3 was a good candidate against Gal-3 and is a potential
lead structure for the further development of novel inhibitors.
Finally, we developed a potent trivalent inhibitor (20) of galectins
with inhibitory properties of 20 mM. It is probable that formation
of C3-symmetric analogs of 15 or 18 would provide even better
results. Although the above compounds are notably less efficient
than those described by Nilsson et al.,20,21 we used inhibition of
hemagglutination assays known to require higher concentrations.
This work received support from the Natural Science and
Engineering Research Council of Canada (NSERC) for a
31 R. T. Lee, Y. Ichikawa, H. J. Allen and Y. C. Lee, J. Biol. Chem., 1990,
265, 7874.
32 H. Ahmed, H. J. Allen, A. Sharma and K. L. Matta, Biochemistry,
1990, 29, 5315.
33 D. Gigue`re, S. Sato, C. St-Pierre, S. Sirois and R. Roy, Bioorg. Med.
Chem. Lett., 2006, 16, 1668.
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 2379–2381 | 2381