Synthesis of multivalent lactose derivatives by 1,3-dipolar cycloadditions: selective galectin-1 inhibition

Article abstract of DOI:10.1016/j.carres.2006.04.028

Acetylene derivatives of phenylalanine, phenethylamine and the multifunctional unnatural amino acids, phenyl-bis-alanine and phenyl-tris-alanine, were synthesized and functionalized with 2-azidoethyl β-d-galactopyranosyl-(1→4)-β-d-glucopyranoside via regioselective copper(I)-mediated 1,3-dipolar cycloaddition to give a panel of mono-, di- and trivalent lactoside derivatives. Evaluation of the compounds as inhibitors against the tumour- and inflammation-related galectin-1, -3, -4N, -4C, -4, -7, -8N and -9N revealed a divalent compound with a K_d value as low as 3.2 μM for galectin-1, which corresponded to a relative potency of 30 per lactose unit as compared to the natural disaccharide ligand lactose. This divalent compound had at least one order of magnitude higher affinity for galectin-1 than for any of the other galectins investigated.

Full text of DOI:10.1016/j.carres.2006.04.028

Carbohydrate Research 341 (2006) 1353–1362
Synthesis of multivalent lactose derivatives by 1,3-dipolar
cycloadditions: selective galectin-1 inhibition
Johan Tejler,a,b, Erik Tullberg,a, Torbjo¨rn Frejd,a Hakon Lefflerb and Ulf J. Nilssona,*
aOrganic Chemistry, Lund University, PO Box 124, SE-221 00 Lund, Sweden
bSection MIG, Department of Laboratory Medicine, Lund University, So¨lvegatan 23, SE-223 62 Lund, Sweden
Received 9 February 2006; received in revised form 10 April 2006; accepted 15 April 2006
Available online 15 May 2006
Abstract—Acetylene derivatives of phenylalanine, phenethylamine and the multifunctional unnatural amino acids, phenyl-bis-ala-
nine and phenyl-tris-alanine, were synthesized and functionalized with 2-azidoethyl b-D-galactopyranosyl-(1!4)-b-D-glucopyrano-
side via regioselective copper(I)-mediated 1,3-dipolar cycloaddition to give a panel of mono-, di- and trivalent lactoside derivatives.
Evaluation of the compounds as inhibitors against the tumour- and inflammation-related galectin-1, -3, -4N, -4C, -4, -7, -8N and
-9N revealed a divalent compound with a Kd value as low as 3.2 lM for galectin-1, which corresponded to a relative potency of 30 per
lactose unit as compared to the natural disaccharide ligand lactose. This divalent compound had at least one order of magnitude
higher affinity for galectin-1 than for any of the other galectins investigated.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Keywords: Galectin; Lectin; Triazole; Multivalence; Cluster effect; Inhibition
1. Introduction
the intermolecular aggregative process and statistical ef-
fects. The intramolecular chelate effect appears when
multivalent ligands occupy multiple binding sites within
a protein. In contrast, when multivalent ligands bind to
multivalent receptors, large cross-linked aggregates may
form.15 The statistical effect comes from slower off rates
due to higher local concentrations of the binding moiety
(in our case lactose) around a single binding site.
Several research groups have exploited the idea of
using multivalent compounds to find high affinity galec-
tin ligands/inhibitors. Multivalent enhancement has
been seen for galectin-1,16,17 galectin-3,14,16–18 galectin-4
(N-terminal)19 and galectin-5.17 In these studies, a large
array of scaffolds have been examined for creating mul-
tivalent carbohydrate ligands including glycopolymers
and glycodendrimers.20–22
To optimize and study the glycoside cluster effect
between multivalent ligands and galectins, we have
developed new scaffolds between the linked carbo-
hydrate units based on polyfunctional unnatural amino
acids like phenyl-bis-alanine (PBA) and phenyl-tris-
alanine (PTA). These scaffolds have previously been
used for the preparation of amino-alcohol ligands, cage
Galectins are defined by two criteria established in 1994:1
‘affinity for b-galactosides and significant sequence
similarity in the carbohydrate-binding site . . .’, and now
include about 14 proteins in mammals and many in other
species.2 Galectins are implicated in different biological
events such as cancer3–7 and the regulation of immunity
and inflammation.8–11 In several of these biological
events, glycoconjugate binding is crucial, which makes
the development of galectin inhibitors important.
The recognition of glycoconjugates by lectins is typi-
cally characterized by low affinity and the biological sys-
tems use multivalent interactions to circumvent these
low affinities, a phenomenon referred to as ‘the glycoside
clustering effect’12–14 Several mechanisms may operate
to enhance the relative potency of multivalent ligands.
These processes include the intramolecular chelate effect,
*
Corresponding author. Fax: +46 46 2228209; e-mail: ulf.nilsson@
organic.lu.se
Contributed equally to this article.
0008-6215/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2006.04.028

1354
J. Tejler et al. / Carbohydrate Research 341 (2006) 1353–1362
molecules and dendrimers.23–25 Typically, the unnatural
PBA and PTA amino acids have been further function-
alized by peptide couplings to furnish more complex
structures. An important advantage with the choice of
PBA and PTA as scaffolds is that the two orthogonal
functionalities (amine and ester) enable derivatization
with two different molecular entities (e.g., a galectin
ligand and a fluorescent tag). A drawback of using stan-
dard peptide coupling reactions for PBA and PTA
functionalizations is that efficiencies and yields may vary
significantly with the choice of coupling partner. In
particular, carbohydrate carboxylic acid or amino deriv-
atives that carry unprotected hydroxyl groups are often
difficult to couple directly via amide bond formations.
To avoid these problems, we turned our attention to
other methods of attachment of unprotected galectin
ligands to PBA and PTA.
Inspired by a multitude of successful copper catalyzed
1,3-dipolar cycloaddition reactions between terminal
alkynes and azides, under mild conditions, for the synthe-
sis of intricate carbohydrate structures and the ready
availability of 2-azidoethyl b-D-galactopyranosyl-(1!4)-
b-D-glucopyranoside, we set out to synthesize suitable
alkyne precursors derived from the above PBA and
PTA. Acetylene derivatives of phenylalanine and of phen-
ethylamine were also prepared for comparative purposes.
In the presence of copper(I), the acetylene derivatives
were coupled with 1,4-regioselectivity20,26,27 with 2-azido-
ethyl b-D-galactopyranosyl-(1!4)-b-D-glucopyranoside.
The ligands were evaluated as inhibitors in a competitive
fluorescence polarization assay against galectin-1, -3 and
-7. They were also evaluated against intact galectin-4, the
C-terminal domain of galectin-4, as well as the N-terminal
domains of galectin-4, -8 and -9.28,29 In this paper, we
report the synthesis of these ligands and the relative
potency, as galectin inhibitors, of the multivalent ligands
as compared with their monovalent homologues.
romethane (method B), to give carbamates 2, 5 and 8.
The divalent acetylene derivatives 7 and 8 were synthe-
sized from the corresponding diamine 6 using method A
and method B, respectively. Diamine 6 was synthesized
in two steps from the corresponding protected phenyl-
bis-didehydro-alanine derivative by a modification of
the literature procedure,23 which uses the chiral Rh(I)-
Et-DuPhos catalyst to synthesize diastereomerically
enriched (S,S)-PBA.31 However, due to unreliably low
stability of Rh(I)-Et-DuPhos and hence unpredictable
reactivity, together with a predicted limited importance
of stereoconfiguration of PBA and PTA for galectin bind-
ing, we opted for a cheaper and more robust unselective
catalyst. The styrene double bonds were thus saturated
by hydrogenation using Wilkinson’s catalyst at 60 psi
for 5 days followed by hydrogenolytic removal of the
carboxybenzyloxy protective group to afford the amine.
This methodology was analogously applied to the prepa-
ration of the triamine precursor for the synthesis of tris
acetylene derivative 10 using method A.25
2.2. Synthesis of triazoles
The 1,3-dipolar cycloaddition between the acetylene
derivatives 1–5, 7, 8 and 10 and 2-azidoethyl b-D-galac-
topyranosyl-(1!4)-b-D-glucopyranoside 1132 was per-
formed using N,N-diisopropylethylamine (DIPEA) as
base and copper (I) iodide as mediator (Scheme 1).
The 1,3-dipolar cycloaddition under these conditions
has the advantage of forming triazoles at room temper-
ature. Previous results have shown higher yields using
toluene as compared to acetonitrile as solvent,20 but
due to solubility problems, acetonitrile was used for all
reactions in this work except for the formation of 14.
The reaction times were between 2 and 7 days and yields
varied between 53% and 80% for the formation of
mono- and divalent structures 12–16 and 18–19, whereas
the yield for the trivalent 17 was 33%, which corre-
sponds to 69% per cycloaddition.
2. Results and discussion
2.3. Galectin binding
2.1. Synthesis of acetylenes
Dissociation constants for ligands 12–19 with galectin-1,
-3, -4N, -4C, -4, -7, -8N and -9N were determined by
competitive fluorescence polarization28,29 in solution
and compared to the methyl b-glycoside of lactose 20
as a reference compound. While galectin-3 could be
evaluated at room temperature, the other galectins were
studied at 4 ꢁC to achieve interaction strengths high
enough to assure accurate measurements. Importantly,
the fluorescence polarization method measures in solution
the affinity of ligands to the galectins, which typically
function as soluble proteins in their biological environ-
ments. Other methods include solid phase assays and a
direct comparison between these methods may be diffi-
cult due to matrix interference.33 This is apparent in pre-
Amine precursors phenethyl amine, L-phenylalanine and
D,L-phenylalanine methyl esters, and PBA and PTA
methyl esters 6 and 7 were coupled with either of the
two types of acetylenic precursors (Table 1). In the first
case, propiolic amides 1,30 3, 4 and 10 were formed by cou-
pling propiolic acid with the corresponding amine using 2-
ethoxy-1-ethoxycarbonyl-1,2-dihydroquinoline (EEDQ)
in dichloromethane (method A), as this activating agent
has in our hands been found to give particularly satisfac-
tory results in the synthesis of propargylic amides. In the
second case, the amine was treated with propargyl chloro-
formate in the presence of triethylamine and a catalytic
amount of 4-dimethylaminopyridine (DMAP) in dichlo-

J. Tejler et al. / Carbohydrate Research 341 (2006) 1353–1362
1355
Table 1. Synthesis of acetylene derivatives
Amine
Methoda (yield)
A (78%)
Product
H
N
NH
2
Ph
Ph
Ph
1
2
O
O
H
N
O
NH
2
B (32%)
A (61%)
Ph
Ph
O
O
Ph
Ph
Ph
OMe
OMe
OMe
HN
HN
HN
NH Cl
3
3
O
O
O
A (51%)
B (41%)
Ph
Ph
OMe
NH Cl
3
4
O
O
O
OMe
O
OMe
NH Cl
3
O
5
MeO2C
O
CO2Me
MeO2C
CbzHN
CO2Me
NHCbz
iH2, Pd/C
iiA (33%)
NH
HN
HN
O
6
7
MeO2C
CO2Me
O
O
NH
O
O
6
iH2, Pd/C
iiB (38%)
8
O
CbzHN
MeO2C
CO2Me
HN
CO2Me
CbzHN
MeO2C
MeO2C
O
iH2, Pd/C
iiA (20%)
NH
NHCbz
H
N
O
9
MeO2C
10
a Method A: amine precursor, EEDQ and propiolic acid in CH2Cl2. Method B: amine precursor, propargyl chloroformate, Et3N and DMAP in
CH2Cl2.
vious results where both a solid phase assay and a fluo-
rescence titration on the same multivalent ligands were
compared.17 In this example for a divalent ligand, the
solid phase assay gave a relative potency of 750 per lac-
tose moiety, while a fluorescence titration on the same
ligand gave a relative potency of 10.
It is apparent that the aglycon fragment has an effect
on the affinity of the ligands for some of the galectins
(Table 2). All monovalent ligands showed better affinity
to galectin-1, -3 and -4N than lactoside 20. In particular,
13 (Kd = 24 lM) showed a pronounced affinity for
galectin-1 with a relative potency of 8 as compared to

1356
J. Tejler et al. / Carbohydrate Research 341 (2006) 1353–1362
HO
HO
OH
O
H
N
OH
O
O
HO
O(CH2)2N3
O
OH
OH
1
11
CuI, DIPEA, CH3CN
HO
HO
OH
O
OH
O
O
HO
O
O
OH
N
N
OH
HN
12
N
H
N
H
R
R
R
O
O
N
12
13
66%
80%
O
O
H
N
H
N
R
14
60%
15
60%
O
O
OMe
O
OMe
O
R
HN
CO2Me
MeO2C
CO2Me
17
33%
16
53%
MeO2C
NH
HN
R
O
R
O
NH
R
O
H
N
O
MeO2C
R
H
N
MeO2C
CO2Me
R
O
18
56%
19
73%
R
O
NH
HN
O
R
O
O
OMe
O
O
HO
HO
OH
O
OH
O
R =
O
HO
O
OH
N
N
OH
N
Scheme 1. Synthesis of 1,4-substituted 1,2,3-triazoles (12–19) via 1,3-dipolar cycloaddition reactions exemplified with 12.
lactose (Kd = 190 lM). The longer linker in phenethyl-
amine derived carbamate 13 results in a higher relative
potency towards galectin-1 and -7 compared to the
shorter linker in the phenethylamine derived amide 12.
This effect is not seen in galectin-3, -4, -4N, -4C, -8N
or -9N. Comparing the affinity of the amide derivatives
14, 15 and 12, the presence or absence of the methyl es-
ter side chain does not have any effect. However, com-
paring carbamate derivatives 13 and 18, the derivative
without the ester side chain gives better affinities in
galectins-1, -3, -7 and -9N, whereas no effect was observed
for other galectins. In general, the carbamate linker
showed stronger affinity for galectin-1, which is not ob-
served for other galectins. The diastereomerically en-
riched derivative 15 shows better affinity for galectins-1,
-4N, -4C and -8N, than the diastereomeric mixture 14
does.
relative to lactose, which we believe may sometimes be
misleading as effects from direct interactions between
the galectins and lactose aglycon structures are not
taken into account. A significant glycosidic cluster effect
relative to methyl b-lactoside 20 can be seen with all mul-
tivalent ligands 16, 17 and 19 for galectin-1 and to a lesser
extent for galectin-3 and -4N (Table 3). However, when
the cluster effects are calculated with respect to the corres-
ponding monomers 14 or 18, the cluster effects are much
lower for galectin-1, almost unaffected for galectin-4N,
and basically not observable for galectin-3. This clearly
illustrates that direct interactions between lactose agly-
cons can be important and that cluster effects should be
calculated with respect to a reference monomer instead
of lactose when analyzing multivalent galectin inhibitors.
Nevertheless, the most pronounced relative potency
effects per monomer unit are 7.7 for 19 and 7.2 for 16
(30 and 11 relative to methyl b-lactoside 20), possibly
due to 19 and 16 cross-linking two separate galectin-1
The literature on cluster effects between galectins and
multivalent ligands invariably reports the cluster effects

J. Tejler et al. / Carbohydrate Research 341 (2006) 1353–1362
1357
Table 2. Dissociation constants (lM) of 12–19 and relative potencies compared to methyl b-lactoside 20 as determined in a fluorescence polarization
assaya
Compd
Valency
Galectin-1
Galectin-3
Rel
Galectin-4N
Galectin-4C
Kd
190
80
24
120
80
Rel
Kd
Kd
Rel
Kd
Rel
20
12
13
14
15
16
17
18
19
1
1
1
1
1
2
3
1
2
1
220
66
66
75
70
30
17
86
27
1
540
230
220
460
200
110
22
1
1200
2200
1600
2300
1100
710
360
1700
570
1
2.4
7.9
1.6
2.4
23
26
3.9
59
3.3
3.3
2.9
3.1
7.3
13
2.6
8.1
2.3
2.5
1.2
2.7
4.9
25
0.5
0.8
0.5
1.1
1.7
3.3
0.7
2.1
8.3
7.4
49
3.2
220
110
2.5
4.9
Galectin-4
Galectin-7
Galectin-8N
Galectin-9N
Kd
Rel
Kd
Rel
Kd
Rel
Kd
Rel
20
12
13
14
15
16
17
18
19
1
1
1
1
1
2
3
1
2
1400
1300
1100
1700
1400
750
360
1300
630
1
91
380
100
210
270
42
32
240
96
1
52
63
61
94
62
29
19
58
35
1
23
51
48
97
85
27
18
110
31
1
1.1
1.3
0.8
1
1.9
3.9
1.1
2.2
0.2
0.9
0.4
0.3
2.2
2.8
0.4
0.9
0.8
0.9
0.6
0.8
1.8
2.7
0.9
1.5
0.5
0.5
0.2
0.3
0.9
1.3
0.2
0.7
a At 4 ꢁC except for galectin-3 and 8N, which was evaluated at room temperature.
Table 3. Cluster effects of the di- and trivalent compounds 16, 17 and 19 calculated relative to methyl b-lactoside 20 and to the corresponding
monomers: Mon = compounds 14 or 18a
Compd
Mon
Val
Galectin-1
Cluster effect
Galectin-3
Cluster effect
Galectin-4N
Cluster effect
Galectin-4C
Cluster effect
20
11
8.6
30
Mon
20
Mon
20
Mon
20
Mon
16
17
19
14
14
18
2
3
2
7.2
5.4
7.7
3.7
4.3
4.1
1.3
1.5
1.6
2.5
8.2
2.5
2.1
7.0
1.0
0.8
1.1
1.1
1.6
2.1
1.5
Galectin-4
Cluster effect
Galectin-7
Cluster effect
Galectin-8N
Cluster effect
Galectin-9N
Cluster effect
20
Mon
20 Mon
20
Mon
20
Mon
16
17
19
14
14
18
2
3
2
0.9
1.3
1.1
1.1
1.6
1.0
1.1
0.9
0.5
2.5
2.2
1.3
0.9
0.9
0.7
1.6
1.6
0.8
0.4
0.4
0.4
1.8
1.8
1.8
a At 4 ꢁC except for galectin-3 and 8N, which was evaluated at room temperature.
molecules leading to aggregation.15 Furthermore, the
An unexpected cluster effect of 7 (8.2 relative to
carbamate derivative 19 displays higher affinity towards
galectin-1 compared to the corresponding amide deriva-
tive 16, which is consistent with the observations for the
monovalent ligands 18 and 14. Interestingly, trimer 17
has a less pronounced relative potency per lactose
moiety of 8.6 for galectin-1. This suggests that trivalent
ligands form aggregates less efficiently, which is in anal-
ogy with previous studies.17 Unfortunately, the limited
number of multivalent ligands evaluated, together with
the relatively high flexibility of their aglycons, does not
permit useful structure–activity relationship to be estab-
lished with the aid of galectin crystal structures; it is
clear that both the aglycon structure and level of multi-
valency of 12–19 influence the binding by galectins.
methyl b-lactoside 20) was observed with trivalent 17
in galectin-4N, while no such effect was observed for
galectin-4C or intact galectin-4. This may suggest that
galectin-4N either can interact with more than one
lactose unit of 17 or that galectin-4N form aggregates.
It should be noted that the probe used in the fluores-
cence polarization competitive inhibitions involving in-
tact galectin-4 has a strong preference for galectin-4C
over galectin-4N. Hence, it is not surprising that the re-
sults for intact galectin-4 are similar to those with galec-
tin-4C, as an inhibitor binding to the N-terminal of
intact galectin-4 will not inhibit the assay reporting fluo-
rescent probe that selectively binds the C-terminal of
intact galectin-4. Ideally, intact galectin-4 should be

1358
J. Tejler et al. / Carbohydrate Research 341 (2006) 1353–1362
evaluated with an unfortunately not yet discovered fluo-
rescent probe that binds efficiently to both N- and C-ter-
minal domain.
were recorded on a Shimadzu FTIR-8300 instrument.
Fluorescence polarization experiments and calculations
were performed in general as described28 and performed
at 4 ꢁC except for galectin-3 and -8N, which were done
at ambient temperature. Each experiment was per-
formed at least in duplicate. Concentrations and probes
used for galectins-1, -3, -7, -8N, and 9N were as
described.28,29 For galectin-4 a probe with an A-tetra
saccharide was used, which was synthesized in analogy
3. Conclusions
The copper-mediated 1,3-dipolar cycloaddition proved
to be a convenient mode of attachment for the synthesis
of unprotected carbohydrate ligands to PBA and PTA.
The inhibitor multivalency proved to have a significant
influence on the affinity of the ligands against galectin-
1 and -4N. Galectin-1 showed a preference for ligands
with a carbamate linker and a moderate, but significant,
glycoside cluster effect with all multivalent ligands 16, 17
and 19 suggesting the formation of cross-linked aggre-
gates. The strongest glycoside cluster effect of 7.7 (30 rel-
ative to methyl b-lactoside 20) for galectin-1 was found
for the divalent carbamate 19, which compares favour-
ably with the best multivalent lactoside inhibitor of
galectin-1 reported.17 Carbamate 19 had a Kd as low
as 3.2 lM against galectin-1, which reflects a promising
selectivity as this Kd is one order of magnitude or more
lower than for any other galectin investigated. A glyco-
sidic cluster effect of 7 (8.2 relative to methyl b-lactoside
20) was also seen for 17 in galectin-4N. The results here-
in hold promise for the development of galectin-1 and
4N inhibitors and particularly useful research tools as
an efficient inhibitor (e.g., 19 against galectin-1) would,
via its ester functionality, allow for conjugation of differ-
ent reporter structures such as fluorescent labels,
enzymes, biotin or solid surfaces.
with the methods used in Oberg et al.34 (to be described
¨
elsewhere). The probe was used at 0.1 lM as for the
other probes, and galectin-4N at 5 lM, galectin-4C at
0.5 lM, and intact galectin-4 at 0.5 lM. DL-Phenyl ala-
nine methyl ester hydrochloride, 2-benzyloxycarbonyla-
mino-3-[3-(2-benzyloxycarbonylamino-2-methoxycarbo-
nyl-vinyl)-phenyl]-acrylic acid methyl ester and 2-ben-
zyloxycarbonylamino-3-[3,5-bis-(2-benzyloxycarbonyla-
mino-2-methoxycarbonyl-vinyl)-phenyl]-acrylic
acid
methyl ester were synthesized according to literature
procedures.35–37
4.2. Propynoic acid phenethyl-amide (1)
Propiolic acid (0.1 mL, 1.6 mmol) and EEDQ (442 mg,
1.8 mmol) were added to a stirred solution of phenethyl-
amine (0.2 mL, 1.6 mmol) in CH2Cl2 (10 mL) at rt. The
mixture was stirred overnight and then washed with
2.5 M HCl (2 · 10 mL), and dried (Na2SO4). After filtra-
tion, 600 mg of silica was added to the filtrate and the
solvent was removed under reduced pressure. The resi-
due thus absorbed on silica was purified by column
chromatography (5:1, petroleum ether/EtOAc) to give
the title compound (219 mg, 78%), which was recrystal-
lized from CH2Cl2/pentane to give 1 as thin white
needles the melting point of which was 56–58 ꢁC (lit.
4. Experimental
1
53–54 ꢁC) and the H NMR data were in accordance
4.1. General methods
with the literature data.30
All commercial chemicals were used without further
purification. Thin layer chromatography (TLC) was car-
ried out on 60F254 silica (Merck) and visualization was
made by UV light followed by heating with aqueous sul-
furic acid. Column chromatography was performed on
4.3. Phenethyl-carbamic acid prop-2-ynyl ester (2)
Propargylchloroformate (0.5 mL, 5.1 mmol), Et3N
(0.6 mL, 4.6 mmol) and DMAP (57 mg, 0.4 mmol) were
added to a stirred solution of phenethylamine (0.58 mL,
4.6 mmol) in CH2Cl2 (20 mL) at rt. The mixture was
stirred for 2 days or until TLC analysis of the reaction
mixture did not indicate any phenethylamine (ninhydrin
staining). The mixture was then washed with HCl 1 M
(2 · 20 mL) and dried (Na2SO4). After filtration, 2 g of
silica was added to the filtrate and the solvent was
removed under reduced pressure. The residue thus
absorbed on silica was purified by column
chromatography (5:1, petroleum ether/EtOAc) to give
2 as a clear oil (300 mg, 32%): 1H NMR (300 MHz,
CDCl3): d 7.34–7.18 (m, 5H, ArH), 4.87 (br, 1H, NH
partially exchanged), 4.68 (s, 2H, –OCH2–), 3.47 (q,
˚
silica gel (Amicon Matrex 35–70 lm, 60 A). Reversed
phase chromatography was performed on Waters Sep-
Pack Vac 35cc C18 columns. NMR experiments were
recorded with Bruker ARX 300 MHz or Bruker DRX
400 MHz spectrometers at ambient temperature. 1H
NMR assignments were derived from COSY experi-
ments. Optical rotations were measured with a Perkin–
Elmer 341 or a 241 polarimeter. High-resolution
fast bombardment mass spectra HRMS (FAB) were
recorded with a JEOL SX-120 instrument or HRMS
(ESI) Micromass Q-TOF micro spectrometer. Melting
points were measured on a Gallenkamp Melting Point
Apparatus MFB-595 and are uncorrected. IR spectra

J. Tejler et al. / Carbohydrate Research 341 (2006) 1353–1362
1359
2H, J = 13.2 Hz, J = 6.5 Hz, N–CH2–), 2.83 (t, 2H,
J = 6.9 Hz, Ar–CH2–), 2.47 (t, 1H, J = 2 Hz, CCH);
13C NMR (75 MHz, CDCl3): d 155.5, 138.7, 128.9,
128.8, 126.7, 78.4, 74.7, 52.6, 42.4, 36.1; IR (NaCl film):
3375, 2937, 2142, 1715, 1534 cmꢀ1; FABMS m/z calcd
for [C12H14NO2+H]+: 204.1025. Found: 204.1026.
to remove the catalyst and then the solvent was removed
under reduced pressure to give 6 as a brownish oil in
quantitative yield. 1H NMR and MS data were in accor-
dance with the literature data.23
4.7. 3-[3-(2-Methoxycarbonyl-2-propionylamino-ethyl)-
phenyl]-2-propionylamino-propionic acid methyl ester (7)
4.4. 3-Phenyl-2-propynoylamino-propionic acid methyl
ester (4)
10% Pd/C (100 mg) was added to a stirred solution of 6
(0.8 g, 1.5 mmol) in CH3OH (50 mL) at rt. The slurry
was hydrogenated at 1 atm overnight or until no starting
material could be seen on TLC. The slurry was then
filtered through a plug of Celite and the solvent was
removed under reduced pressure to give the diamine as
a brownish oil. The diamine was suspended in CH2Cl2
(50 mL). Propiolic acid (130 lL, 2.1 mmol) and EEDQ
(0.6 g, 2.31 mmol) were then added to the solution and
the resulting mixture was stirred for 2 days at rt and
then washed with HCl 2 M (3 · 50 mL). The combined
aqueous phases were extracted once with CH2Cl2
(50 mL) and the combined organic phases were dried
(Na2SO4). After filtration, 600 mg of silica was added
to the filtrate and the solvent was removed under re-
duced pressure. The residue thus absorbed on silica
was purified by column chromatography (5:1, petroleum
ether/EtOAc) to give 7 as a yellow syrup (88 mg, 33%):
1H NMR (400 MHz, CDCl3): d 7.25–7.20 (m, 1H, ArH),
7.01–6.98 (m, 2H, Ar–H), 6.92 (d, 1H, J = 11 Hz, ArH),
6.65 (d, 1H, J = 8 Hz, NH partially exchanged), 6.53 (d,
1H, J = 8 Hz, NH, partially exchanged) 4.95–4.86 (m,
2H, N–CH–C@O), 3.74 (s, 6H, –OCH3), 3.19–3.02 (m,
4H, Ar–CH2–), 2.85 (d, 2H, J = 6.8 Hz, –CCH); 13C
NMR (100 MHz, CDCl3): d 171.2, 151.7, 136.0, 130.5,
129.2, 128.5, 79.9, 74.6, 53.7, 53.0, 37.9; IR (NaCl film):
3255, 3019, 2109, 1736, 1638 cmꢀ1; FABMS m/z calcd
for [C20H21N2O6+H]+: 385.1400. Found: 385.1403.
The S-enantiomer of this compound has been prepared
previously.38 Compound 4 was therefore prepared anal-
ogously in racemic form (51%) mp 73.9–75.7 ꢁC. The IR
data of racemic 4 was in agreement with those of the
enantiomerically enriched compound.
4.5. 3-Phenyl-2-prop-2-ynyloxycarbonylamino-propionic
acid methyl ester (5)
Propargyl chloroformate (1 mL, 10 mmol), DMAP
(244 mg, 2 mmol) and Et3N (2.8 mL, 20 mmol) were
added to a stirred solution of DL-phenylalanine methyl
ester hydrochloride (2.2 g, 10 mmol) in CH2Cl2
(50 mL) at rt. The mixture was stirred for 2 days,
washed with HCl 1 M (2 · 50 mL) and brine (50 mL)
and dried (Na2SO4). After filtration, 6 g of silica was
added to the filtrate and the solvent was removed under
reduced pressure. The residue thus absorbed on silica
was purified by column chromatography (4:1, petroleum
1
ether/EtOAc) to give 5 as yellow oil (1.1 g, 41%): H
NMR (300 MHz, CDCl3): d 7.35–7.25 (m, 3H, ArH),
7.14 (d, 2H, Ar–H), 5.24 (br d, 1H, J = 7.2 Hz, NH par-
tially exchanged), 4.67–4.62 (m, 3H, –OCH2–, N–CH–
C@O), 3.72 (s, 3H, OCH3), 3.2–3.05 (m, 2H, ArCH2–),
2.47 (t, 1H, J = 2.4 Hz, CCH); 13C NMR (75 MHz,
CDCl3): d 171.9, 154.9, 135.7, 129.4, 128.8, 127.4,
78.1, 75.0, 55.0, 52.9, 52.5, 38.3; IR (NaCl film): 3337,
2959, 2115, 1726, 1518 cmꢀ1; FABMS m/z calcd for
[C14H16NO4+H]+: 262.1079. Found: 262.1076.
4.8. 3-[3-(2-Methoxycarbonyl-2-prop-2-ynyloxycarbon-
ylamino-ethyl)-phenyl]-2-prop-2-ynyloxycarbonylamino-
propionic acid methyl ester (8)
4.6. 2-Benzyloxycarbonylamino-3-[3-(2-benzyloxycarb-
onylamino-2-methoxycarbonyl-ethyl)-phenyl]-propionic
acid methyl ester (6)
Propargyl chloroformate (0.22 mL, 1.6 mmol), DMAP
(20 mg, 0.2 mmol) and Et3N (0.17 mL, 1.8 mmol) were
added to a stirred solution of diamine (0.23 g, 0.8 mmol)
in CH2Cl2 (50 mL). The mixture was stirred at rt for
2 days, washed with HCl 1 M (50 mL) and with brine
(50 mL) and dried (Na2SO4). After filtration, 600 mg
of silica was added to the filtrate and the solvent was
evaporated under reduced pressure. The residue thus
absorbed on silica was purified with column chromato-
graphy (10:1 ! 2:1, petroleum ether/EtOAc) to give 8
2-Benzyloxycarbonylamino-3-[3-(2-benzyloxycarbonyl-
amino-2-methoxycarbonyl-vinyl)-phenyl]-acrylic acid methyl
ester (0.8 g, 1.5 mmol) was added to absolute EtOH
(100 mL) and the obtained slurry was degassed on an
ultrasonic bath for 15 min and then purged with argon
for
a further 15 min. Then Wilkinson’s catalyst
(163 mg, 0.2 mmol) was added and the mixture was
hydrogenated at rt at 60 psi over 5 days. Silica (2.4 g)
was then added to the solution and the solvent was
removed under reduced pressure. The residue thus
absorbed on silica was passed through a short plug of
silica eluting with petroleum ether/EtOAc 1:1 (200 mL)
1
as a yellow syrup (0.14 g, 38%): H NMR (400 MHz,
CDCl3): d 7.22–7.17 (m, 1H, Ar–H), 6.98 (d, 2H,
J = 5.3 Hz, Ar–H), 6.92 (t, 1H, J = 22 Hz, Ar–H), 5.47
(d, 1H, J = 8 Hz, NH partially exchanged), 5.34 (d,
1H, J = 8 Hz), 5.47–4.61 (m, 6H, N–CH–CO, O–CH2–),

1360
J. Tejler et al. / Carbohydrate Research 341 (2006) 1353–1362
3.72 (d, 6H, J = 4.2 Hz, O–CH3), 3.14–2.99 (m, 4H, Ar–
CH2–), 2.47 (q, 2H, J = 4.8 Hz, J = 2.4 Hz, CCH); 13C
NMR (100 MHz, CDCl3): d 171.7, 154.9, 136.2, 130.5,
129.0, 128.3, 78.3, 75.0, 55.0, 53.0, 52.6, 38.3; IR (NaCl
film): 3315, 2942, 2120, 1731, 1523 cmꢀ1; FABMS m/z
calcd for [C22H25N2O8+H]+: 445.1611. Found:
445.1608.
CH2–), 2.86–2.82 (m, 3H, CCH); 13C NMR (75 MHz,
CDCl3): d 171.1, 151.7, 136.5, 129.6, 77.4, 74.6, 53.6,
53.0, 37.8; IR (NaCl): 3454, 2936, 2102, 1725,
1528 cmꢀ1; FABMS m/z calcd for [C27H28N3O9+H]+:
538.1826. Found: 538.1821.
4.11. Lactosyl amide (12)
4.9. 2-Benzyloxycarbonylamino-3-[3,5-bis-(2-benzyloxy-
carbonylamino-2-methoxycarbonyl-ethyl)-phenyl]-propi-
onic acid methyl ester (9)
2-Azidoethyl b-D-galactopyranosyl-(1!4)-b-D-gluco-
pyranoside 1132 (32 mg, 77 lmol), 1 (37 mg, 0.21 mmol)
and CuI (17 mg, 0.09 mmol) were dissolved in CH3CN
(2 mL). N,N-Diisopropylethylamine (14 lL, 77 lmol)
was added and the reaction mixture was stirred over
6 days followed by concentration under reduced pres-
sure. The residue was dissolved in H2O and applied on
to C-18 silica (5 g). Elution with a gradient of CH3OH
2-Benzyloxycarbonylamino-3-[3,5-bis-(2-benzyloxycar-
bonylamino-2-methoxycarbonyl-vinyl)-phenyl]-acrylic
acid methyl ester (0.8 g, 1 mmol) was added to a mixture
of CH3OH/EtOAc (1:1) and the obtained slurry was de-
gassed on an ultrasonic bath for 15 min and then purged
with argon for a further 15 min. Then Wilkinson’s cata-
lyst (100 mg, 0.1 mmol) was added and the mixture was
hydrogenated at rt at 60 psi over 5 days. Silica (2.4 g)
was then added to the solution and the solvent was
removed under reduced pressure. The residue thus
absorbed on silica was passed through a sort plug of sil-
ica eluting with petroleum ether/EtOAc (1:1, 200 mL) to
remove the catalyst and then the solvent was removed
under reduced pressure. The residue was coevaporated
with toluene (3 · 5 mL) to give 9 as a brownish paste
20
in H2O and lyophilization gave 12 (30 mg, 66%): ½aꢁD
+24; 1H NMR (400 MHz, CH3OD): d 8.46 (s, 1H, H-tri-
azole), 7.30–7.18 (m, 5H, Ar–H), 4.69 (t, 2H, J = 5 Hz,
–CH2–CH2–N), 4.34 (d, 2H, J = 7.8 Hz, H-10, H-1),
4.26–4.21 (m, 1H, O–CH2–CH2), 4.06–4.00 (m, 1H, O–
CH2–CH2), 3.90 (dd, 1H, J = 2.4 Hz, J = 12.1 Hz, H-6),
3.84–3.81 (m, 2H), 3.78, (dd, 1H, J = 7.4 Hz, J = 11.4
Hz, H-6), 3.69, (dd, 1H, J = 4.6 Hz, J = 11.4 Hz, H-6),
3.63–3.48 (m, 7H), 3.44–3.40 (m, 1H, H-5), 3.27 (t,
1H, J = 8.4 Hz, H-20), 2.91 (t, 2H, J = 7.6 Hz, J = 7.2
Hz, Ar–CH2); ESIMS m/z calcd for [C25H36N4O12+
H]+: 585.2408. Found: 585.2295.
1
(0.77 g, 97%). H NMR and MS data were in accor-
dance with the literature data.37
4.12. Lactosyl carbamate (13)
4.10. 3-[3,5-Bis-(2-methoxycarbonyl-2-propynoylamino-
ethyl)-phenyl]-2-propynoylamino-propionic acid methyl
ester (10)
2-Azidoethyl b-D-galactopyranosyl-(1!4)-b-D-gluco-
pyranoside 1132 (32 mg, 77 lmol), 2 (49 mg, 0.24 mmol)
and CuI (34 mg, 0.19 mmol) were dissolved in CH3CN
(2 mL). N,N-Diisopropylethylamine (14 lL, 77 lmol)
was added and the reaction mixture was stirred over
5 days followed by concentration under reduced pres-
sure. The residue was dissolved in H2O and applied on
to C-18 silica (5 g). Elution with a gradient of CH3OH
10% Pd/C (150 mg) was added to a stirred solution of 9
(0.75 g, 0.95 mmol) in CH3OH (75 mL). The slurry was
hydrogenated overnight or until no starting material
could be seen on TLC (ninhydrin staining). The slurry
was then filtered through a plug of Celite and the sol-
vents were removed under reduced pressure to give the
crude triamine intermediate as a brownish paste, which
was suspended in CH2Cl2 (40 mL). Propiolic acid
(204 lL, 3.3 mmol) and EEDQ (0.9 g, 3.6 mmol) were
added to the solution and the resulting mixture was stir-
red for 2 days at rt and then washed with HCl 2 M
(3 · 40 mL). The combined aqueous phases were
extracted once with CH2Cl2 (50 mL) and the combined
organic phases were dried (Na2SO4). After filtration,
900 mg of silica was added to the filtrate and the solvent
was removed under reduced pressure. The residue thus
absorbed on silica was purified by column chromatogra-
phy (4:1 ! 1:4, petroleum ether/EtOAc) to give 10 as a
brown syrup (84 mg, 20%): 1H NMR (400 MHz,
CDCl3): d 6.81–6.70 (m, 4H, Ar–H), 6.58–6.49 (m, 2H,
NH partially exchanged), 4.95–4.84 (m, 3H, –N–CH–
CO–), 3.73 (s, 9H, OCH3), 3.17–2.96 (m, 6H, Ar–
20
in H2O and lyophilization gave 13 (25 mg, 80%): ½aꢁD
1
+29; H NMR (400 MHz, CH3OD): d 8.09 (s, 1H, tri-
azole), 7.28–7.16 (m, 5H, Ar–H), 5.12 (s, 2H, O–CH2–
C), 4.65 (t, 2H, J = 5 Hz, –CH2–CH2–N), 4.34 (d, 1H,
J = 7.8 Hz, H-1), 4.33 (d, 1H, J = 7.6 Hz, H-10), 4.25–
4.20 (m, 1H, O–CH2–CH2), 4.03–3.98 (m, 1H, O–CH2–
CH2), 3.90 (dd, 1H, J = 2.3 Hz, J = 12.1 Hz, H-6),
3.85–3.70 (m, 4H), 3.58–3.46 (m, 5H), 3.42–3.40 (m, 1H,
H-5), 3.35–3.30 (m, 2H, CH2–NH–CO, partly hidden in
solvent residual peak), 3.25 (t, 1H, J = 8.4 Hz, H-2),
2.77 (t, 2H, J = 7.4 Hz, Ar–CH2); ESIMS m/z calcd
for [C26H38N4O13+H]+: 615.2514. Found: 615.2526.
4.13. Lactosyl methylester amide (14)
2-Azidoethyl b-D-galactopyranosyl-(1!4)-b-D-gluco-
pyranoside 1132 (33 mg, 81 lmol), 4 (40 mg, 0.17 mmol)

J. Tejler et al. / Carbohydrate Research 341 (2006) 1353–1362
1361
and CuI (21 mg, 0.11 mmol) were dissolved in toluene
(3 mL). N,N-Diisopropylethylamine (15 lL, 81 lmol)
was added and the reaction mixture was stirred for 2 days
followed by concentration under reduced pressure. Puri-
fication by flash chromatography (4:1, CH2Cl2/CH3OH)
3.11 (dd, 1H, J = 8.2 Hz, J = 14.0 Hz, –NH–CH–CO,
partly hidden), 3.11 (dd, 1H, J = 8.3 Hz, J = 13.7 Hz,
–NH–CH–CO); ESIMS m/z calcd for [C48H70N8O28+
H]+: 1207.4378. Found: 1207.4337.
1
gave 14 (31 mg, 60%). H NMR (300 MHz, CH3OD): d
4.16. Trivalent lactosyl methylester carbamate (17)
8.47 (s, 1H, H-triazole), 7.30–7.20 (m, 5H, Ar–H), 4.69
(t, 2H, J = 4.8 Hz, –CH2–CH2–N), 4.60 (m, 1H, NH),
4.35 (br d, 2H, J = 7.2 Hz, H-1, H-10), 4.27–4.20 (m,
1H, O–CH2–CH2), 4.06–4.99 (m, 1H, O–CH2–CH2),
3.90–3.66 (m, 8H), 3.60–3.45 (m, 5H), 3.43–3.37 (m, 1H,
H-5), 3.28–3.21 (m, 3H, H-2, Ar–CH2), 3.19–3.12 (m,
1H, NH–CH–CO); ESIMS m/z calcd for [C27H38-
N4O14+ Na]+: 665.2282. Found: 665.2265.
2-Azidoethyl b-D-galactopyranosyl-(1!4)-b-D-gluco-
pyranoside 1132 (62 mg, 0.15 mmol), 10 (21 mg, 39 lmol)
and CuI (30 mg, 0.15 mmol) were dissolved in CH3CN
(3 mL). N,N-Diisopropylethylamine (26 lL, 150 lmol)
was added and the reaction mixture was stirred over
2 days followed by concentration under reduced pressure.
The residue was dissolved in H2O and applied on to C-18
silica (5 g). Elution with a gradient of CH3OH in H2O and
lyophilization gave 17 (23 mg, 33%): 1H NMR (400 MHz,
D2O): d 8.46–8.44 (m, 3H, triazole), 7.09–7.01 (m, 3H,
Ar–H), 4.77–4.73 (m, 6H, –CH2–CH2–N, partly hidden
in solvent residual peak), 4.47–4.43 (m, 6H, H-1, H-10),
4.32–4.29 (m, 3H, O–CH2–CH2), 4.16–4.13 (m, 3H,
O–CH2–CH2), 3.94–3.90 (m, 6H), 3.83–3.53 (m, 39H),
3.32–3.08 (m, 9H, H-2, Ar–CHa, NH–CH–CO); ESIMS
m/z calcd for [C69H102N12O42+Na]+: 1793.6112. Found:
1793.6088.
4.14. Lactosyl methylester amide (15)
2-Azidoethyl b-D-galactopyranosyl-(1!4)-b-D-gluco-
pyranoside 1132 (30 mg, 73 lmol), 3 (24 mg, 0.11 mmol)
and CuI (34 mg, 0.18 mmol) were dissolved in CH3CN
(2 mL). N,N-Diisopropylethylamine (14 lL, 77 lmol)
was added and the reaction mixture was stirred over
3 days followed by concentration under reduced pressure.
Purification by flash chromatography (4:1, CH2Cl2/
20
CH3OH) gave 15 (28 mg, 60%): ½aꢁ ꢀ31; 1H NMR
D
(400 MHz, CH3OD): d 8.47 (s, 1H, H-triazole), 7.29–
7.18 (m, 5H, Ar–H), 4.68 (t, 2H, J = 5 Hz,
–CH2–CH2–N), 4.59 (br s, 0.5H, N–H, partly
exchanged), 4.35 (d, 1H, J = 7.4 Hz, H-10), 4.35 (d, 1H,
J = 7.8 Hz, H-1), 4.25–4.20 (m, 1H, O–CH2–CH2),
4.05–4.00 (m, 1H, O–CH2–CH2), 3.89 (dd, 1H, J =
2.4 Hz, J = 12.2 Hz, H-6), 3.83–3.67 (m, 7H), 3.59–3.46
(m, 5H), 3.43–3.40 (m, 1H, H-5), 3.30–3.21 (m, 3H, H-2,
Ar–CH2), 3.16 (dd, 1H, J = 8.5 Hz, J = 13.9 Hz, NH–
CH–CO); ESIMS m/z calcd for [C27H38N4O14+H]+:
643.2463. Found: 643.2435.
4.17. Lactosyl methylester carbamate (18)
2-Azidoethyl b-D-galactopyranosyl-(1!4)-b-D-gluco-
pyranoside 1132 (29 mg, 71 lmol), 5 (39 mg, 0.11mmol)
and CuI (14 mg, 0.073 mmol) were dissolved in CH3CN
(2 mL). N,N-Diisopropylethylamine (14 lL, 77 lmol)
was added and the reaction mixture was stirred over
2 days followed by concentration under reduced pressure.
Purification by flash chromatography (4:1, CH2Cl2/
1
CH3OH) gave 18 (27 mg, 56%): H NMR (400 MHz,
CH3OD): d 8.06, 8.05 (2s, 1H, triazole-H, from the two
diastereomers), 7.28–7.17 (m, 5H, Ar–H), 5.09 (s, 2H,
O–CH2–C), 4.63 (t, 2H, J = 5 Hz, –CH2–CH2–N), 4.60
(br s, 0.3H, NH, partly exchanged), 4.43 (dd, 1H, J =
5.6 Hz, J = 9.0 Hz, CH–CO), 4.35 (d, 1H, J = 7.5 Hz,
H-10), 4.34 (d, 1H, J = 7.9 Hz, H-1), 4.25–4.19 (m, 1H, O–
CH2–CH2), 4.03–3.97 (m, 1H, O–CH2–CH2), 3.90 (dd,
1H, J = 2.3 Hz, J = 12.1 Hz, H-6), 3.84–3.75 (m, 3H),
3.72–3.68 (m, 4H), 3.62–3.48 (m, 5H), 3.44–3.40 (m, 1H,
H-5), 3.25 (t, 1H, J = 8.2 Hz, J = 8.3 Hz, H-2), 3.12 (dd,
1H, J = 5.5 Hz, J = 13.8 Hz, Ar–CHa), 2.92 (dd, 1H,
J = 9.1 Hz, J = 13.8 Hz, Ar–CHb); ESIMS m/z calcd
for [C28H40N4O15+H]+: 673.2568. Found: 673.2535.
4.15. Divalent lactosyl methylester amide (16)
2-Azidoethyl b-D-galactopyranosyl-(1!4)-b-D-glucopyr-
anoside 1132 (46 mg, 0.11 mmol), 7 (15 mg, 40 lmol)
and CuI (26 mg, 0.14 mmol) were dissolved in CH3CN
(2 mL). N,N-Diisopropylethylamine (20 lL, 116 lmol)
was added and the reaction mixture was stirred over
6 days followed by concentration under reduced pres-
sure. The residue was dissolved in H2O and applied on
to C-18 silica (5 g). Elution with a gradient of CH3OH
1
in H2O and lyophilization gave 16 (26 mg, 53%): H
NMR (400 MHz, CH3OD): d 8.48 (2s, 1H, triazole-H,
from the diastereomers), 7.21–7.09 (m, 4H, Ar–H),
4.69 (t, 4H, J = 4.9 Hz, –CH2–CH2–N), 4.59 (s, 0.7H,
NH), 4.36 (br d, 4H, J = 7.6 Hz, H-10, H-1), 4.26–4.20
(m, 2H, O–CH2–CH2), 4.06–4.01 (m, 2H, O–CH2–
CH2), 3.89 (br dd, 2H), 3.84–3.67 (m, 14H), 3.61–3.47
(m, 10H), 3.43–3.40 (m, 2H, H-5), 3.33–3.18 (m, 6H,
H-2, Ar–CHa, partly hidden in solvent residual peak),
4.18. Divalent lactosyl methylester carbamate (19)
2-Azidoethyl b-D-galactopyranosyl-(1!4)-b-D-gluco-
pyranoside 1132 (74 mg, 0.18 mmol), 8 (28 mg, 64 lmol)
and CuI (38 mg, 0.20 mmol) were dissolved in CH3CN
(2 mL). N,N-Diisopropylethylamine (31 lL, 178 lmol)
was added and the reaction mixture was stirred over

1362
J. Tejler et al. / Carbohydrate Research 341 (2006) 1353–1362
7. Liu, F.-T.; Rabinovich, G. A. Nat. Rev. Cancer 2005, 5,
29–41.
8. Almkvist, J.; Karlsson, A. Glycoconjugate J. 2004, 19,
575–581.
9. Rabinovich, G. A.; Toscano, M. A.; Ilarregui, J. M.;
Rubinstein, N. Glycoconjugate J. 2004, 19, 565–573.
10. Sato, S.; Nieminen, J. Glycoconjugate J. 2004, 19, 583–591.
11. Hirashima, M.; Kashio, Y.; Nishi, N.; Yamauchi, A.;
Imaizumi, T.-A.; Kageshita, T.; Saita, N.; Nakamura, T.
Glycoconjugate J. 2004, 19, 593–600.
12. Lee, Y. C.; Lee, R. T. Acc. Chem. Res. 1995, 28, 321–
327.
13. Lundquist, J. J.; Toone, E. J. Chem. Rev. 2002, 102, 555–
578.
14. Pohl, N. L.; Kiessling, L. L. Synthesis 1999, 1515–1519.
15. Sacchettini, J. C.; Baum, L. G.; Brewer, C. F. Biochemistry
2001, 40, 3009–3015.
16. Andre, S.; Frisch, B.; Kaltner, H.; Desouza, D. L.;
Schuber, F.; Gabius, H.-J. Pharm. Res. 2000, 17, 985–990.
17. Vrasidas, I.; Andre, S.; Valentini, P.; Bock, C.; Lensch,
M.; Kaltner, H.; Liskamp, R. M. J.; Gabius, H.-J.; Pieters,
R. J. Org. Biomol. Chem. 2003, 1, 803–810.
2 days followed by concentration under reduced pres-
sure. The residue was dissolved in H2O and applied on
to C-18 silica (5 g). Elution with a gradient of CH3OH
in H2O and lyophilization gave 19 (59 mg, 73%): 1H
NMR (400 MHz, CH3OD): d 8.07 (br s, 2H, triazole),
7.19–7.04 (m, 4H, Ar–H), 5.09 (br s, 4H, O–CH2–C),
4.63 (br t, 4H, J = 4.2 Hz, –CH2–CH2–N), 4.60
(s, 1.5H, NH), 4.45–4.43 (m, 2H, CH–CO), 4.36 (d, 2H,
J = 7.4 Hz, H-10), 4.35 (d, 2H, J = 7.7 Hz, H-1),
4.24–4.19 (m, 2H, O–CH2–CH2), 4.03–3.98 (m, 2H,
O–CH2–CH2), 3.90 (br dd, 2H, J = 2.3 Hz, J = 12.1 Hz,
H-6), 3.85–3.68 (m, 14H), 3.63–3.47 (m, 10H), 3.43–
3.41 (m, 2H, H-5), 3.26 (br t, 2H, J = 7.9 Hz, H-2),
3.09 (br dd, 2H, Ar–CHa), 2.91 (br dd, 2H,
J = 9.0 Hz, J = 13.4 Hz, Ar–CHb); ESIMS m/z calcd
for [C50H74N8O30+H]+: 1267.4589. Found: 1267.4578.
Acknowledgements
18. Andre, S.; Liu, B.; Gabius, H.-J.; Roy, R. Org. Biomol.
Chem. 2003, 1, 3909–3916.
The authors would like to thank Barbro Kahl-Knutson
for excellent help with fluorescence polarization analysis
and galectin production, Susanne Carlsson for providing
19. Wu, A. M.; Wu, J. H.; Liu, J.-H.; Singh, T.; Andre, S.;
Kaltner, H.; Gabius, H.-J. Biochimie 2004, 86, 317–326.
20. Fazio, F.; Bryan, M. C.; Blixt, O.; Paulson, J. C.; Wong,
C.-H. J. Am. Chem. Soc. 2002, 124, 14397–14402.
21. Andre, S.; Pieters, R. J.; Vrasidas, I.; Kaltner, H.;
Kuwabara, I.; Liu, F.-T.; Liskamp, R. M. J.; Gabius,
H.-J. ChemBioChem 2001, 2, 822–830.
22. Lee, R. T.; Lee, Y. C. Glycoconjugate J. 2001, 17, 543–551.
23. Ionescu, R. D.; Blom, A.; Frejd, T. Tetrahedron: Asym-
metry 2003, 14, 2369–2380.
¨
galectins 8N and 9N, and Christopher Oberg for provid-
ing the probe used for galectin-4. Support from the Lund
University Research School of Medicinal Sciences, the
Swedish Research Council, the programs ‘Glycoconju-
gates in Biological Systems’ and ‘Chemistry for the Life
Sciences’ sponsored by the Swedish Strategic Research
Foundation, and the Royal Physiographic Society in
Lund is acknowledged.
24. Ionescu, R. D.; Frejd, T. Chem. Commun. 2001, 1088–
1089.
25. Ritzen, A.; Frejd, T. Eur. J. Org. Chem. 2000, 3771–3782.
26. Perez-Balderas, F.; Ortega-Munoz, M.; Morales-Sanfru-
tos, J.; Hernandez-Mateo, F.; Calvo-Flores, F. G.; Calvo-
Asin, J. A.; Isac-Garcia, J.; Santoyo-Gonzalez, F. Org.
Lett. 2003, 5, 1951–1954.
Supplementary data
1H NMR of 2, 5, 7–8, 10 and 12–19. Supplementary
data associated with this article can be found, in the
online version, at doi:10.1016/j.carres.2006.04.028.
27. Tornoe, C. W.; Christensen, C.; Meldal, M. J. Org. Chem.
2002, 67, 3057–3064.
28. Sorme, P.; Kahl-Knutsson, B.; Huflejt, M.; Nilsson Ulf, J.;
Leffler, H. Anal. Biochem. 2004, 334, 36–47.
29. Cumpstey, I.; Carlsson, S.; Leffler, H.; Nilsson, U. J. Org.
Biomol. Chem. 2005, 3, 1922–1932.
30. Kunishima, M.; Kawachi, C.; Morita, J.; Terao, K.;
Iwasaki, F.; Tani, S. Tetrahedron 1999, 55, 13159–13170.
31. Burk, M. J.; Feaster, J. E.; Nugent, W. A.; Harlow, R. L.
J. Am. Chem. Soc. 1993, 115, 10125–10138.
32. Chernyak, A.; Oscarsson, S.; Turek, D. Carbohydr. Res.
2000, 329, 309–316.
References
1. Barondes, S. H.; Castronovo, V.; Cooper, D. N.; Cum-
mings, R. D.; Drickamer, K.; Feizi, T.; Gitt, M. A.;
Hirabayashi, J.; Hughes, C.; Kasai, K. Cell 1994, 76, 597–
598.
33. Pieters, R. J. Trends. Glycosci. Glycotechnol. 2004, 16,
243–254.
34. Oberg, C. T.; Carlsson, S.; Fillion, E.; Leffler, H.; Nilsson,
U. J. Bioconjugate Chem. 2003, 14, 1289–1297.
35. Zielinski, T.; Achmatowicz, M.; Jurczak, J. Tetrahedron:
Asymmetry 2002, 13, 2053–2059.
2. Leffler, H.; Carlsson, S.; Hedlund, M.; Qian, Y.; Poirier,
F. Glycoconjugate J. 2004, 19, 433–440.
3. Grassadonia, A.; Tinari, N.; Iurisci, I.; Piccolo, E.;
Cumashi, A.; Innominato, P.; D’Egidio, M.; Natoli, C.;
Piantelli, M.; Iacobelli, S. Glycoconjugate J. 2004, 19, 551–
556.
¨
36. Schmidt, U.; Lieberknecht, A.; Wild, J. Synthesis 1984,
53–60.
4. Bidon-Wagner, N.; Le Pennec, J.-P. Glycoconjugate
J. 2004, 19, 557–563.
37. Ritzen, A.; Basu, B.; Wallberg, A.; Frejd, T. Tetrahedron:
Asymmetry 1998, 9, 3491–3496.
5. Takenaka, Y.; Fukumori, T.; Raz, A. Glycoconjugate
J. 2004, 19, 543–549.
38. Papanastassiou, Z. B.; Bruni, R. J.; White, E. J. Med.
Chem. 1967, 10, 701–706.
6. van den Brule, F.; Califice, S.; Castronovo, V. Glycocon-
jugate J. 2004, 19, 537–542.