Functionalized glycomers as growth inhibitors and inducers of apoptosis in human glioblastoma cells

Article abstract of DOI:10.1021/jm0205853

The effects of functionalized aryl β-D-glycopyranosides (glycomers) on the proliferation, survival, and apoptosis of human glioblastoma cells in culture were evaluated as a way to control tumor progression. The results showed that inhibition of growth and/or induction of apoptosis can be achieved by these molecules in human glioblastoma cells. Inhibition of DNA synthesis precedes induction of apoptosis and growth inhibition. The substituents at C-1, C-2, C-3,C-4, and C-6 on the pyranosidic scaffold are important to modulate the action and the efficacy of these molecules. Human fibroblasts and brain-derived endothelial cells were less sensitive to glycomers than tumor cells. Thus, functionalized aryl β-D-glycopyranosides represent a new class of molecules potentially able to control the progression of brain tumors.

Full text of DOI:10.1021/jm0205853

3600
J . Med. Chem. 2003, 46, 3600-3611
F u n ction a lized Glycom er s a s Gr ow th In h ibitor s a n d In d u cer s of Ap op tosis in
Hu m a n Gliobla stom a Cells
Stephen Hanessian,*,† Lijie Zhan,† Raymonde Bovey,‡ Oscar M. Saavedra,† and Lucienne J uillerat-J eanneret*,‡
Department of Chemistry, Universite´ de Montre´al, C. P. 6128, Succursale Centre-Ville, Montre´al, P. Q., Canada, H3C 3J 7, and
University Institute of Pathology, CHUV, Bugnon 25, CH1011 Lausanne, Switzerland
Received December 27, 2002
The effects of functionalized aryl â-D-glycopyranosides (glycomers) on the proliferation, survival,
and apoptosis of human glioblastoma cells in culture were evaluated as a way to control tumor
progression. The results showed that inhibition of growth and/or induction of apoptosis can be
achieved by these molecules in human glioblastoma cells. Inhibition of DNA synthesis precedes
induction of apoptosis and growth inhibition. The substituents at C-1, C-2, C-3,C-4, and C-6
on the pyranosidic scaffold are important to modulate the action and the efficacy of these
molecules. Human fibroblasts and brain-derived endothelial cells were less sensitive to
glycomers than tumor cells. Thus, functionalized aryl â-D-glycopyranosides represent a new
class of molecules potentially able to control the progression of brain tumors.
In tr od u ction
human epidermoid and colon carcinoma cells lines.10
Here we report on the synthesis of a new series of
glycomers by systematic functional group modifications
of an aryl glycopyranosidic scaffold, and their effects on
inhibition of proliferation and induction of apoptosis of
human glioblastoma cells. The effect of these compounds
toward human cerebral endothelial cells and normal
primary human fibroblasts was also studied.
Uncontrolled cell proliferation is involved in several
human diseases, including cancers of any origin. The
control of cancer growth by inhibition of cell division
and induction of tumor cell death (apoptosis), in con-
junction with other adjuvant anticancer treatments,
could provide a therapeutic strategy with improved
efficacy.
Glioblastomas are brain tumors with a high prolifera-
tive potential and a very poor prognosis, mainly due to
two reasons. First, glioblastoma tumor cells as well as
the cerebral vasculature forming the blood-brain bar-
rier express multiple resistance pathways to chemo-
therapeutic agents,1 precluding conventional treat-
ments.2 Second, while glioblastomas rarely metastasize
outside of the brain, they have a very high migratory
potential in the central nervous system,3-5 precluding
complete surgical resection. Thus, the development of
drugs able to control the growth of glioblastoma cells
without inducing drug resistance, would be of benefit
for the treatment of brain cancer.6 Disaccharide or
tetrasaccharide derivatives able to control glioblastoma
cell growth have been reported to be of potential
therapeutic interest.7,8 However, simpler functionalized
monosaccharide derivatives could be advantageous in
biological systems, due to their increased stability and
potential to be transported across physiological barriers
and cell membranes. Different binding capacities for
saccharide epitopes in brain tumors have been de-
scribed.9 Thus, the control of glioblastoma tumor cell
growth either by inhibition of cell division and/or
induction of cell apoptosis by carbohydrate-based mol-
ecules may be possible.
Syn th esis
Glycomers in the aryl D-glucopyranoside series were
synthesized from common precursors utilizing well-
established methods starting with the known11 3-O-
allyl-1,2,5,6-O-isopropylidene-D-glucofuranose 2 (Scheme
1). Acetolysis provided the tetraacetate 3, which was
converted to the â-iodophenyl glycoside 4 utilizing the
enhanced nucleophilicity of tributylstannyloxy 4-iodo-
benzene toward a glucopyranosyl bromide.12 Deacetyl-
ation and conversion to the 4,6-O-benzylidene deriva-
tive 6 allowed the preparation of a series of 2-esters 7a -
d . Cleavage of the acetal and selective sulfonylation
with 3-trifluoromethylbenzenesulfonyl chloride afforded
a set of glycomers 8a -e with different esters at C-2
(Scheme 1). Glycomer 8e was prepared from the corre-
sponding 3-O-methyl ether as described for 8b. The
sulfonamides 10a and 10b were prepared by displace-
ment of 8a and 8c with sodium azide to give the
corresponding 6-azido derivatives 9a and 9b, which
were reduced and N-sulfonylated (Scheme 2). Reductive
cleavage of the 4,6-O-benzylidene ring in 7a and 7c with
sodium cyanoborohydride in the presence of hydrogen
chloride13 afforded the 6-O-benzyl ethers 11a and 11b
(Scheme 3).
The glycosidation reaction shown in Scheme 1 also
afforded the p-iodophenyl R-D-glucopyranoside analogue
12 as a minor product (Scheme 4). This was transformed
through a series of steps similar to the â-anomer
affording glycomer 13.
The preparation of analogues in the D-galacto series
starting with â-D-galactopyranose pentaacetate was
achieved as described for the D-gluco series in Schemes
We have previously shown that functionalized aryl
â-D-glucopyranosides (glycomers) inhibit the growth of
* To whom correspondence should be addressed: S.H.: Fax: (514)
343-5728; phone: +1 514 343 6738; e-mail: stephen.hanessian@
umontreal.ca. L. J .-J .: or Fax: +41 21 314 7175; phone: 41 21 314
7173; e-mail: lucienne.juillerat@chuv.hospvd.ch.
† Universite´ de Montre´al.
‡ University Institute of Pathology.
10.1021/jm0205853 CCC: $25.00 © 2003 American Chemical Society
Published on Web 07/16/2003

Functionalized Glycomers as Growth Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3601
Sch em e 1a
Sch em e 4a
a
NaOMe,MeOH; (b) PhCH(OCH3)2, HBF4‚Et2O, DMF; (c)
RCOCl, Et3N, CH2Cl2; (d) TFA, H2O, THF; (e) 3-(trifluorometh-
yl)benzenesulfonyl chloride, Et3N, CH2Cl2.
Sch em e 5a
a
NaH, allyl bromide; (b) ion-exchange resin IR-20 (H+);
(c) Ac2O, pyridine; (d) HBr in AcOH, Ac2O, CH2Cl2, 0 °C; (e)
4-iodotributylstannyl phenoxide, SnCl4, benzene, reflux; (f)
NaOMe,MeOH; (g) PhCH(OCH3)2, HBF4.Et2O, DMF; (h) (RCO)2O,
pyridine (for 7a , 7b); RCOCl, Et3N, CH2Cl2 (for 7c, 7d ); (i) TFA,
H2O, THF; (j) 3-(trifluoromethyl)benzenesulfonyl chloride, Et3N,
CH2Cl2, (k) 3-(trifluoromethyl)benzenesulfonyl chloride, pyridine,
0 °C to rt.
Sch em e 2a
a
HBr in AcOH, Ac2O, CH2Cl2, 0 °C; (b) 4-iodotributylstannyl
phenoxide, SnCl4, benzene, reflux; (c) NaOMe, MeOH; (d) Bu2SnO,
allyl bromide, Bu4NBr, benzene, reflux; (e) PhCH(OCH3)2, HBF4‚
Et2O, DMF; (f) RCOCl, Et3N, CH2Cl2; (g) TFA, H2O, THF; (h)
3-(trifluoromethyl)benzenesulfonyl chloride, Et3N, CH2Cl2; (i)
NaN3, DMF; (j) PPh3, H2O, THF; (k) NaCNBH3, HCl, THF, 4A
MS.
a
NaN3, DMF; (b) PPh3, H2O, THF; (c) 3-(trifluoromethyl)ben-
zenesulfonyl chloride, Et3N, CH2Cl2.
Sch em e 3a
LNZ308 cell lines14 or human brain-derived endothelial
cells (HCEC),15 in cell cultures with and without expo-
sure to serum (fetal calf serum, FCS) were used. The
results showed that in the presence of FCS only low
inhibition of growth was observed. Removal of FCS
immediately prior to the addition of the glycomers
restored growth inhibition, and a preincubation period
of 24 h in the absence of FCS further increased the
growth inhibitory response. Glioblastoma cells were
found to be more sensitive to 8d than HCEC brain-
derived endothelial cells. The effect of cell density on
growth inhibition showed that subconfluent cells were
more sensitive to growth inhibition than confluent cells
under these experimental conditions. Thus, for screen-
ing and evaluation of glycomer efficacy, cells were
routinely split and left to adhere for 24 h in the presence
of FCS, then deprived of FCS for another 24 h period,
a
NaCNBH3, HCl, THF, 4A MS.
1-3. The 2-ester derivatives of the 6-sulfonates 16a ,b,
the sulfonamides 17a ,b , and the 6-O-benzyl ethers
18a ,b were easily prepared as shown in Scheme 5
(Scheme 5).
Eva lu a tion of th e Effects of F u n ction a lized Gly-
com er s in Hu m a n Gliobla stom a Cells. First, we
tested 8d as a prototype for the inhibition of cell growth
using the MTT (thiazole blue) assay, which determines
the number of adherent metabolically active cells. For
these experiments human glioblastoma LN18 and

3602 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17
Hanessian et al.
Ta ble 1. Glycomer Activity against Human Glioblastoma
LN18 and LNZ308 Cells
compound
R1
R2
LN18a
LNZ308
A. Glycomers with IC50 < 20 µM
8a
8b
8d
10a
10b
11a
11b
16b
17b
OSO2m-TMBb
OSO2m-TMB
OSO2m-TMB
NHSO2m-TMB
NHSO2m-TMB
OCH2Ph
Et
7.0 ( 1.5 11.5 ( 3.0
12.5 ( 2.0 13.0 ( 2.0
Me
i-Pr
Et
Pr
Et
Pr
i-Pr
i-Pr
12.5 ( 2.0
8.5 ( 2.0
12.5 ( 1.5 18.5 ( 2.5
6.5 ( 1.0 17.5 ( 2.0
11.5 ( 2.5 17.5 ( 4.5
7.5 ( 1.5 11.5 ( 2.5
18.5 ( 2.0 18.0 ( 2.0
OCH2Ph
OSO2m-TMB
NHSO2m-TMB
12.5 ( 2.0
7.5 ( 1.0
F igu r e 1. Growth inhibition of glioblastoma cells as a function
of systematic chemical modifications of the glucopyranoside
backbone. Cells were left to adhere for 24 h in the presence of
FCS and incubated for another 24 h in the absence of FCS.
Medium was changed and glycomers were added to cells at
increasing concentrations in the absence of FCS for 24 h. MTT
assay was performed during the last 2 h of incubation. Percent
of growth was calculated as the ratio of the MMT reduction
(absorbance at 540 nm) of treated versus control cells. Open
symbols: LN18 cells; closed symbols: LNZ308 cells A: modifica-
tion at position C-1; 0 9: 29; O b: 30; [ ]: 8b B: modification
at position C-2: 0 9: 8b; [ ]: 27 C: modification at position
C-3: 0 9: 8; [ ]: 28 D: modification at position C-4: 0 9:
8a ; [ ]: 32 E: modification at position C-6: [ ]: 8d ; 0 9:
11b; 2 4: 31; O b: 17b.
B. Glycomers with IC50 20-45 µM
8
8c
8e
13
16a
17a c
17d
18a
OSO2m-TFMBb
OSO2m-TFMB
OSO2m-TFMB
OSO2m-TFMB
OSO2m-TFMB
COMe
COPr
COPr
COPr
COPr
21 ( 2.5
43 ( 8.5
26 ( 3.5
25 ( 5.0
40 ( 4.0
25 ( 3.0
16 ( 2.5
20 ( 2.5
19 ( 2.5
41 ( 7.5
22 ( 2.0
42 ( 8.0
29 ( 3.5
26 ( 3.5
27 ( 4.5
37 ( 5.5
NHSO2m-TFMB COPr
NHSO2m-TFMB
OCH2Ph
H
COPr
a
Cells were grown for 24 h with FCS, then deprived of FCS
for 24 h. Glycomers were added to cells for 24 h and MTT assay
was performed. The % growth inhibition was calculated as the
ratio of MTT reduction (absorbance at 540 nm) of treated to control
cells and IC50 was calculated from dose-response curves as the
concentration inducing 50% of growth inhibition. Experiments
were done in triplicate and means ( standard deviations were
was preferred to hydrogen, bromo, and methoxy ana-
logues (Chart 1). No significant difference in efficacy was
observed between gluco- and galacto-isomers (Table 1A,
10a , 17b; 8a , 16b) or between the R/â-configuration of
the glycomers (Table 1B, 8c, 13).
b
c
calculated. TMB ) trifluoromethyl benzenesulfonyl. Phenyl-
glycoside analogue (I ) H).
before exposure to glycomers and evaluation of growth
by the MTT assay. Cell density seeding was adjusted
to have confluent cell layers in the control wells at the
end of the experiment period.
In position C-2, an aliphatic short-chain ester sub-
stituent improves efficacy, the propionate 8a or acetate
8b derivatives being the most favorable (compare 8b
and 27, Figure 1B), possibly representing the preference
of cellular esterases for short-chain alcohols, as we
observed using esters of 5-aminolevulinic acid.16 In
position C-3, an allyl ether is more favorable than an
methyl ether substitution (compare 8 and 28, Figure
1C). The C-4 position must be unsubstituted (compare
8a and 32, Figure 1D). In position C-6, substitution of
a sulfonate derivative with sulfonamide or ether deriva-
tives does not modify the effect, while an ester substitu-
tion is detrimental, which we interpreted as a result of
hydrolysis due to the presence of cellular esterases
(compare 8d , 11b, and 17b with 31, Figure 1E). Thus,
the structure of the most efficient molecule, as defined
Using the experimental conditions described above,
a systematic evaluation of glycomers with chemical
modifications in C-1, C-2, C-3, C-4, and C-6 positions
was performed (Table 1A,B) (Figure 1A-E). The results
showed that the glycomers could be grouped in three
classes: high efficacy with IC50 < 20 µM (Table 1A); low
efficacy with IC50 20-50 µM, Table 1B), and glycomers
with no efficacy (IC50 > 50 µM, Chart 1). Examination
of the functional features in these classes of aryl
glycopyranosides led to the identification of critical
groups involved in the observed inhibitory effects. Most
interesting was the nature of the p-phenyl substituent
at the anomeric position. The p-iodo group (Tables 1A,B)

Functionalized Glycomers as Growth Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3603
Ch a r t 1. Glycomers with IC50 > 50 µM against Human Glioblastoma LN18 and LNZ308 Cell Lines
Ch a r t 2 The Active Glycomers
evaluating the incorporation of tritiated thymidine to
quantitate DNA synthesis. The effects of 8c, a weak
inhibitor, and 11b, a potent inhibitor, on inhibition of
DNA synthesis were first determined in the two glio-
blastoma cell lines after 8 or 27 h of exposure (results
not shown). While 11b completely inhibited DNA syn-
thesis already after 8 h, 8c was much less efficient and
only partial inhibition was observed after 27 h exposure.
These results demonstrated that inhibition of DNA
synthesis is not due to a nonspecific effect of glycomers
in glioblastoma cells, but is related to the potency of the
drug. Exposure to 8a induced cell rounding and detach-
ment (as observed under the microscope), starting after
6-8 h. Inhibition of DNA synthesis was already ob-
served 2 h after addition of 8a and increased only
slightly with longer exposure time (Figure 2A). To
observe a decrease in the number of metabolically active
cells, using the MTT assay as a marker and the
quantification of protein content to ascertain cell num-
ber (Figure 2B) in the culture wells, it was necessary
to expose to 8a for periods longer than 8 h. A maximal
effect was observed after 24 h, which did not further
increase by incubating the cells for 48 h. Continuous
exposure to 8a was not necessary to observe an effect
on DNA synthesis, since diminution of DNA synthesis
was seen already after 5 min, reaching a maximum after
60 min exposure of glioblastoma cells (Figure 3A). Thus
the primary effect of these glycomers appears to be a
by MTT assay, has a p-iodophenyl substitution in
position C-1, a short-chain aliphatic ester in position
C-2, an allyl group in position C-3, no substituent at
position C-4, and an aromatic sulfonate, sulfonamide,
or ether, but not ester, substitution at position C-6
(Chart 2).
We then evaluated whether the inhibitory effects were
dependent on increased cell death or decreased prolif-
eration or both. Examination under the microscope of
cells exposed to glycomers in the presence of MTT, an
indicator of mitochondrial function, did not indicate
mitochondrial toxicity, since adherent and rounded cells
displayed similar violet coloration of mitochondria (not
shown). We first investigated growth inhibition by

3604 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17
Hanessian et al.
F igu r e 2. Inhibition of DNA synthesis precedes inhibition of
cell growth and loss of adherence. Cells were left to adhere
for 24 h in the presence of FCS, then incubated for another
24 h in the absence of FCS. Medium was changed, and 8a was
added to cells at increasing concentrations in the absence of
FCS for 2-24 h. Thymidine incorporation and MTT assay were
performed for the two last hours of incubation, and protein
content was determined at the end of the incubation period.
(A): Thymidine incorporation. Grey bars: no 8a ; white bars:
2 µM 8a ; black bars: 4 µM 8a . (B): MTT assay and protein
content. ]: 8 h exposure; 0: 24 h exposure; 2: 48 h exposure.
rapid initial inhibition (<5 min) of DNA synthesis in
glioblastoma cells, followed by death of cells (>8 h)
which is maximal at 24 h.
Endothelial cells derived from human brain (HCEC)
were less sensitive to inhibition of DNA synthesis in
response to short-time exposure to 8a than glioblastoma
cells (Figure 3B). The growth of normal primary human
fibroblasts, as determined by an MTT assay, was
inhibited at slightly higher concentrations of 8a (IC50
) 20-25 µM) than glioblastoma tumor cells (Figure 4A).
Inhibition of growth (MTT) and DNA synthesis (3HT
incorporation) (Figure 4B) in fibroblasts (Figure 4B,
upper panels) by another glycomer, 10a , was also
induced at higher glycomer concentration in fibroblasts
than in LN18 and LNZ308 glioblastoma cells (Figure
4B, lower panels). Thus, some concentration-dependent
selectivity toward tumor cells versus stromal cells can
also been obtained by some glycomers.
F igu r e 3. Continuous exposure to glycomers is not necessary
for effect in human glioblastoma cells and endothelial cells.
Endothelial cells are less sensitive than glioblastoma cells to
short time exposure to glycomers. Cells were left to adhere
for 24 h in the presence of FCS, then incubated for another
24 h in the absence of FCS. Medium was changed and 8a
was added to cells at 0, 4, 12, and 20 µM concentrations in
the absence of FCS for 5, 15 30, 60, or 180 min. Following
exposure, 8a was removed from the culture wells, medium
without FCS and without glycomer was added and incubation
was continued to reach 60 min, then thymidine incorporation
was performed for the two next hours. A: glioblastoma cells;
B: endothelial cells. Black bars: no 8a ; light gray bars: 4 µM
8a ; white bars: 12 µM 8a ; dark gray bars: 20 µM 8a .
We then studied whether growth inhibition of glio-
blastoma cells by the glycomers can also be achieved
through induction of apoptosis. To this end, we deter-
mined chromatin condensation in the nuclei of adherent
cells in culture using DAPI, a fluorescent dye labeling
double-stranded DNA (Figure 5A), or histological nuclear
and cytoplasmic staining (hematoxylin-eosin and
Giemsa histological stains) (Figure 5B), or quantitated
in cell extracts histone-associated DNA fragments (Table
2), which are end-point markers of apoptotic cell death.
Using quantification of nucleosome fragments, the apo-
ptosis index in the absence of glycomer was defined as
1.00. The results are shown in Table 2. Compounds 19
and 20 (Chart 1) did not induce apoptosis, based on the
observation that growth inhibition and loss of adhesivity
was not observed, while in the presence of the most

Functionalized Glycomers as Growth Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3605
F igu r e 4. Growth and DNA synthesis inhibition of primary human fibroblasts by 8a and 10a . Comparison with human
glioblastoma cells. Cells were grown in the presence of FCS, then incubated for 24 h in the absence of FCS. Medium was changed
and 8a or 10a was added to cells at increasing concentrations in the absence of FCS for 24 h. MTT assay or thymidine incorporation
was performed during the last 2 h of incubation. A: 8a inhibits growth in cultures of human fibroblasts from 11 different surgical
specimens and two glioblastoma cell lines. B: dose-dependent inhibition of growth (MTT, left) and DNA synthesis (3HT
incorporation, right) by 10a in four different cultures of human fibroblasts (upper panels) or in 2 glioblastoma cell lines (lower
panels).
efficient growth inhibitors 8a , 11a , and 11b (Table 1A),
nucleosome fragmentation was observed (Tables 2 and
3). However, 31 (Chart 1), which did not inhibit growth
and cell adherence after 24 h treatment, could induce
apoptosis. Glycomer 10a , which is a potent inhibitor of
DNA synthesis and cell growth at low concentration,
was a poorer inducer of apoptosis at these concentra-
tions. Thus, the concentration able to inhibit cell growth
or induce apoptosis could be dissociated for some func-
tionalized glycomers. When studied for a shorter time-
course (4-8 h of exposure), before cell detachment was
observed but at the time of cell rounding, glycomers 8a
and 11a were able to induce apoptosis as determined
by chromatin condensation and revealed by DAPI stain-
ing (Figure 5A) or nuclear staining by hematoxylin-eosin
or Giemsa histological stains (Figure 5B). These results
show that functionalized p-iodophenyl glycosides are
able to inhibit DNA synthesis and induce apoptosis in
human glioblastoma cells. Therefore they represent a
new class of molecules with a potential to control brain
tumor progression.
To exclude the possible alkylating effect of sulfonated
glycomers such as 8a , the efficacy of inhibition of DNA
synthesis or induction of apoptosis in glioblastoma cells
was directly compared with the sulfonamide 10a and
the ether 11a analogues (Table 3). We calculated the
efficacy of these glycomers by reporting the ratio of
thymidine incorporation or of apoptosis index to the
level of MTT reduction. These experiments indicate that
the efficacy indexes, at concentrations above the IC50
are similar for 8a , 10a , and 11a , and that an alkylating
of sulfonate glycomers is not necessary for effects.
Discu ssion
Glioblastoma tumors express binding sites for sac-
charide epitopes.9 A few studies reporting the evaluation
of carbohydrate-based antiproliferative agents for glio-
blastoma tumor cells have been published.7,8,17 We show
here that substituted aryl glycopyranosides (glycomers)
cause selective inhibition of DNA synthesis, thus con-
trolling the replicative/proliferative potential of glio-
blastoma cells, as well as induction of apoptosis. These

3606 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17
Hanessian et al.
Ta ble 2. Induction of Apoptosis by Selected Glycomers.
apoptosis index
total exposure,a
time [h]
compd
concn [µM]
LN18
LNZ308
8a
8c
8d
10a
11a
11bb
19
16
25
25
11
11
25
25
25
10
24
24
15
15
24
24
24
3.82
0.44
1.10
0.98
3.67
1.60
1.19
1.29
8.97
0.41
1.09
1.45
0.91
2.27
ND
20
ND
a
Cells were grown for 24 h with FCS, deprived of FCS for 24
h, then glycomers were added to cells. Quantification of nucleo-
some fragments was performed in the still adherent cells at the
end of the incubation. Apoptosis index was defined as 1.00 for
untreated cell culture wells under identical culture conditions than
b
the treated cells. Initial concentration 25 µM, precipitation of the
compound in the culture medium starting after 15 min of incuba-
c
tion. ND: not determined.
equally active as antiproliferative agents. Thus, these
compounds are not active by virtue of a covalent
alkylation mechanism. In view of the fact that glioblas-
tomas are polyclonal and heterogeneous tumors, both
cell lines displayed slightly different responses to apo-
ptosis induction. Using bosentan and Fas Ligand18 as
an alternative approach to aptoptosis, we also observed
a graded response of these cells. Interestingly, apoptosis
was induced in a shorter time and at lower concentra-
tion by 8a compare to other glycomers, for both cell
lines. It appears that subtle changes in the nature of
the ester group have significant effects on the induction
of apoptosis (compare 8a , 8c, 8d ) (Table 2). This trend
was also observed in the 6-O-benzyl series (compare 11a
to 11b). We anticipate that this may be possibly due to
the efficiency of cleavage by esterases.
Glioblastomas present both compact tumor areas and
brain tissue diffusely infiltrated by tumor cells. Depend-
ing on tumor cell density, the volume of tumor mass,
the presence of tumor-derived factors, and dysfunction
of the blood-brain barrier, extravasation of serum in
some tumor areas may be facilitated. The inhibitory role
of serum and the increased sensitivity to glycomers in
nonconfluent cell cultures indicate that glycomers may
be more effective on scattered tumor cells at the inva-
sion front in brain areas where there is an intact blood-
brain barrier compared to the primary tumor site, where
tumor cell density and serum leakage are high. The
passage of molecules through the cerebral vascular
system forming the blood-brain is a difficult task to
achieve. However, the fact that endothelial cells in
nonangiogenic areas are not actively dividing, thus
would be less sensitive to inhibition of DNA synthesis,
and that brain-derived endothelial cells are more resis-
tant to apoptosis induction than tumor cells, augurs well
for molecules that have a selective effect. The prelimi-
nary results presented here confirm that selectivity
between glioblastoma and brain-derived endothelial
cells or primary human fibroblasts used as a model for
normal untransformed cells may be achieved with these
functionalized glycomers.
F igu r e 5. 8a and 11a glycomers block mitosis and induce
nuclear condensation and apoptosis in human glioblastoma
cells. Cells were left to adhere for 24 h on histological slides
in the presence of FCS, then incubated for another 24 h in
the absence of FCS. Medium was changed, and 8a (A) or 11a
(B) was added for 7 h to cells in the absence of FCS. Then
cells were fixed, exposed to the DNA labeling agent DAPI (A-
D), and photographed under a fluorescent microscope. Alter-
natively following treatment, cells were fixed and stained using
either Giemsa (E,F) or hematoxylin-eosin (G,H) histological
stainings. (Part A) DAPI staining: LN18 (A,B) or LNZ308 (C,D)
cells were exposed to control medium (A,C) or 20 µM 8a (B,D),
then fixed in Carnoy and exposed to the DNA labeling agent
DAPI and photographed under a fluorescence microscope
(white arrowheads: apoptotic cells). (Part B) hematoxylin-eosin
and Giemsa stainings: LNZ308 cells were exposed to control
medium (E,G) or to 22.5 µM 11a (F,H), then fixed in 4%
buffered paraformaldehyde and stained using Giemsa staining
(E,F), which mainly stains nuclei or hematoxylin-eosin staining
(G,H), which stains both cytoplasm and nucleus (G: white
arrowheads: mitotic cells; H: black arrowheads: preapoptotic
nucleus displaying nuclear blebbing). Identical information
was obtained using LN18 cells (not shown).
combined effects would result in the control of glioblas-
toma growth.
The most active glycomers where p-iodophenyl â-D-
glucopyranosides or â-D-galactopyranosides with short
ester groups at C-2, an allyl ether at C-3, a free hydroxyl
at C-4, and sulfonate, sulfonamide or benzyl ether at
C-6. In the original series10 the active glycomers con-
tained a C-6 sulfonate, which is a potential alkylating
molecule for proteins or nucleic acids. We therefore
explored the chemically and biologically more stable
sulfonamide or benzyl ether derivatives which were
Con clu sion
While no efficient treatment exists for high-grade
glioblastoma, some drugs belonging to diverse classes,
including microtubule destabilizing agents, topoisomerase

Functionalized Glycomers as Growth Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3607
Ta ble 3. Comparison of the Efficacy Indexes on Inhibition of DNA Synthesis and Induction of Apoptosis of Sulfonate (8a ),
Sulfonamide (10a ), and Ether (11a ) Substitution in Position C-6 of Glycomersa
8a
10a
apoptosis/MTT
11a
apoptosis/MTT
concn [µM]
3H-T /MTT
apoptosis/MTT
3H-T/MTT
3H-T/MTT
LN18
0
4.5
11.3
22.5
LN308
0
4.5
11.3
22.5
16.25
4.87
0.23
0.30
0.93
1.19
3.51
9.60
16.25
3.11
0.07
0.47
0.93
0.88
1.63
8.43
16.25
6.23
0.09
0.58
0.93
1.92
4.85
8.30
9.79
5.00
1.23
0.24
0.95
0.99
2.03
5.09
9.79
6.01
0.64
0.35
0.95
0.72
1.81
6.18
9.79
5.65
2.04
0.34
0.95
0.62
0.70
2.32
a
Cells were grown for 24 h with FCS, then deprived of FCS for 10 h. Glycomers were added to cells for 12 h, and MTT reduction
(absorbance at 540 nm) and incorporation of tritiated thymidine (cpm) or apoptosis (apoptosis index: amount of nucleosme fragments in
treated versus control cells) were determined in adherent cells. Efficacy indexes were obtained by calculating the ratio of either thymidine
incorporation or apoptosis index to MTT reduction.
Cell Cu ltu r e a n d Tr ea tm en ts. The cell culture reagents
were from Life Technologies or Gibco (both in Basel, Switzer-
land). The human cell lines were routinely maintained in
DMEM supplemented with 10% FCS and detached from the
plate using trypsin-EDTA. Cells were seeded in 48-well plates
(Costar) as monolayers for 24 h in DMEM-10% FCS to half
confluence and, unless otherwise stated, incubated 24 h in
DMEM without FCS, then exposed for the time indicated to
the glycomers under investigation. Then either thymidine
incorporation to evaluate DNA synthesis, or protein content
and MTT reduction to evaluate cell viability, or determination
of nucleosome-DNA fragments to quantitate apotosis was
performed. Experiments were done in triplicate and means and
standard deviations (SD) were calculated.
Th ym id in e In cor p or a tion . Thymidine incorporation was
used to ascertain DNA synthesis. Following treatment, cells
were exposed to 0.8 µCi/mL [3H]-thymidine (Amersham Phar-
macia, Du¨bendorf, Switzerland) for 2-3 h, and radioactivity
was quantitated in a beta-counter (LKB) after precipitation
with 10% trichloracetic acid and solubilization in 0.1% SDS-
0.1 N NaOH and 5 mL Optiphase scintillation cocktail (Wallac,
Fisher Chemicals, England).
Eva lu a tion of Cell P r olifer a tion by MTT. MTT ((3,4,5-
dimethylthiazol-yl)-2,5-diphenyl tetrazolium); Sigma, Buchs,
Switzerland) was used to quantify the number of metabolically
active cells. Following glycomer treatment, cells were exposed
to 0.2 mg/mL MTT in DMEM medium for 2 h, supernatant
was aspirated, and the precipitated formazan was dissolved
in 0.1 N HCl in 2-propanol and quantified at 540 nm in a
multiwell plate reader (iEMS Reader MF, Labsystems, Bio-
concepts, Switzerland).
Ap op tosis Deter m in a tion a n d Qu a n tifica tion . Apopto-
sis was determined in cells in culture by evaluation of nuclear
chromatin condensation using the fluorescent nuclear marker
4′,6′-diamidino-2-phenylindolylhydrochloride, (DAPI, Boehring-
er Roche). Cells were grown on histological sides and exposed
to glycomers according to the same protocols than cells grown
in culture wells. Following fixation in methanol/acetic acid (3:
1), slides were exposed to 1 µg/mL DAPI for 30 min at 37 °C,
then mounted in 20% glycerol in PBS and examined and
photographed under a fluorescent microscope (360/500 nm,
excitation/emission wavelength, respectively). Alternatively,
following buffered paraformaldehyde fixation, Giemsa and
hematoxylin-eosin staining to visualize nuclei were performed
according to standard procedures.
inhibitors, inhibitors of damage-repair enzymes, and
alkylating and intercalating agents, are under experi-
mental or clinical (re)evaluation.2,5,6,19
We have evaluated a series of new substituted aryl
â-D-glycopyranosides that exhibit specific effects on
growth inhibition and apoptosis induction in human
glioblastoma cells in culture. Studies of systematic
substitution at selected positions of the glycopyranosidic
backbone indicate that the most important groups for
biological activity are the p-iodophenyl in position C-1,
no substitution in position C-4, while substitutions at
positions C-2, C-3, and C-6 exert fine-tuning of the
effects (Chart 2). The principal event induced by expo-
sure of glioblastoma cells to functionalized glycomers
is an inhibition of DNA synthesis, ultimately resulting
in loss of adhesive properties and induction of apoptosis.
Thus, these molecules represent interesting new poten-
tial agents to control the progression of glioblastomas.
Exp er im en ta l Section
Solvents were distilled under positive pressure of dry
nitrogen before use and dried by standard methods; THF and
ether, from K/benzophenone; CH2Cl2 and toluene from CaCl2.
All commercially available reagents were used without further
purification. All reactions were performed under nitrogen
atmosphere. NMR (1H, 13C) spectra were recorded on AMX-
300 and ARX-400 spectrometers in CDCl3 or CD3OD with
tetramethylsilane as the internal standard. In some cases
carbon resonances were coincident. Low- and high-resolution
mass spectra were recorded on VG Micromass, AEI-MS 902,
or Kratos MS-50 spectrometers using fast atom bombardment
(FAB) or electrospray techniques. Optical rotations were
recorded on a Perkin-Elmer 241 polarimeter in a 1 dm cell at
ambient temperature. Analytical thin-layer chromatography
was performed on Merck 60F254 precoated silica gel plates.
Visualization was performed by UV or by development using
KMnO4 or FeCl3 solutions. Flash column chromatography were
performed using (40-60 µm) silica gel at increased pressure.
Melting points recorded were uncorrected.
Biology. Cells. The human LN18 and LNZ308 glioblastoma
cell lines were developed from surgical specimens of human
glioblastoma14 (a kind gift of AC. Diserens and N. de Tribolet,
Neurosurgery Division, CHUV, Lausanne). Human brain-
derived endothelial cells15 (HCEC cells) were a kind gift of Dr.
D. Stanimirovic, Ottawa, Canada. Human primary urogenital
fibroblasts were cultured from surgical biopsies performed
during delivery or reconstructive surgery in women, according
to a protocol approved by the Ethics Committee of the CHUV
in Lausanne, using the explant technique. All cells were grown
in DMEM medium (Gibco-BRL, Basel, Switzerland) containing
4.5 g/L glucose, 10% fetal calf serum (FCS), and antibiotics.
Apoptosis was quantified in the adherent cell population at
the end of the incubation using the Cell Death Detection
ELISAPLUS (Roche, Rotkreuz, Switzerland), a photometric
enzyme-linked immunoassay for quantitative in vitro deter-
mination of cytoplasmic histone-associated DNA fragments
(mono- and oligo-nucleosomes), a marker of DNA breakdown
and laddering, which is considered as a landmark of apoptosis.
Determinations were performed according to the supplier’s
instructions. Using this method, the increase in absorbance

3608 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17
Hanessian et al.
at 405 nm is proportional to apoptosis. The enrichment in
apoptotic cell proportion (apoptosis index) was calculated as
the ratio of absorbance of treated cells/absorbance of untreated
cells as previously described.20
P r otein Con cen tr a tion . Protein content was determined
with the BCA protein assay kit (Pierce, Socochim, Switzerland)
and bovine serum albumin as standard.
was stirred overnight at room temperature, MeOH (0.2 mL)
was added, and the mixture was stirred for 1 h, then
concentrated. The residue was dissolved in CH2Cl2 and pro-
cessed in the usual way. Chromatography (hexanes-ethyl
acetate, 3/1) gave 7a (87 mg, 96%) as a white foam; [R]D -8.7(c
0.67, CHCl3); IR (KBr) 1746 cm-1; 1H NMR (300 MHz, CDCl3)
δ 7.55 (d, 2H), 7.47-7.44 (m, 2H), 7.38-7.34 (m, 3H), 6.74 (d,
2H), 5.87-5.74 (m, 1H, vinyl-H), 5.55 (s, 1H, CHPh), 5.27-
5.23 (m, 2H, 2-H and vinyl-H), 5.12 (dd, 1H, vinyl-H), 5.02 (d,
1H, J 1,2 ) 7.8 Hz, 1-H), 4.38-4.32 (m, 2H, 6-H and allyl-H),
4.17-4.08 (dd, 1H, allyl-H), 3.84-3.68 (m, 3H, 6-H, 4-H and
3-H), 3.59-3.51 (m, 1H, 5-H), 2.33 (q, 2H), 1.14 (t, 3H); 13C
NMR (75 MHz, CDCl3): δ 173.22(0), 157.26(0), 138.92, 137.42(0),
134.96, 129.54, 128.74, 126.42, 119.59, 117.40(-), 101.73,
100.17, 86.38(0), 81.44, 78.86, 73.74(-), 72.74, 68.97(-), 66.95,
28.04(-), 9.69; MS (FAB,) m/z 566.1 [M]+.
Ch em istr y. p-Iod op h en yl 3-O-Allyl-â-D-glu cop yr a n o-
sid e (5). To a solution of 311 (2.38 g, 6.13 mmol) in CH2Cl2 (20
mL) at 0 °C was added dropwise HBr (30 wt % in acetic acid,
14 mL). After being stirred for 20 min at 0 °C, the reaction
mixture was diluted with CH2Cl2, washed with cold water
twice, cold saturated aqueous NaHCO3 solution, and cold water
again, dried over Na2SO4, and concentrated. The syrupy
residue was dissolved in dry CH2Cl2 (20 mL). A mixture of
4-iodophenol (1.35 g, 6.13 mmol) and bis[tri-n-butyltin] oxide
(1.56 mL) was refluxed (100 °C) in benzene (250 mL) with a
Dean-Stark apparatus. After 1 h, the solvent was removed
in vacuo, and the crude tributylstannyl 4-iodophenoxide was
added to the glycosyl bromide in CH2Cl2 (30 mL), after which
SnCl4 (6.7 mL, 1 M in CH2Cl2) was added dropwise at 0 °C.
After being stirred 30 min, the reaction mixture was allowed
to warm to room temperature, stirred for 4 h, and quenched
with dilute NaHCO3 and ether, and the organic layer was
processed as usual. Purification by flash column chromatog-
raphy (hexanes-ethyl acetate 2/1) gave 4 (2.84 g, 75%) as a
white foam; [R]D -22.0 (c 1.065, CH3Cl); IR (KBr) 1758, 1727
cm-1; 1H NMR (400 MHz, CDCl3) δ 7.59-7.55 (m, 2H), 6.77-
6.74 (m, 2H), 5.82-5.75 (m 1H, vinyl-H), 5.26-5.10 (m, 4H,
2-H, 4-H and vinyl-Hs), 4.94 (d, 1H, J 1,2 ) 7.8 Hz, 1-H), 4.22-
(dd, J 5,6 ) 5.8 Hz and J 6′,6 ) 12.3 Hz, 6-H), 4.14 (dd, 1H, J 5,6′
) 2.7 Hz, 6′-H), 4.11-4.09 (fm, 2H, allyl-Hs), 3.77-3.73 (m,
1H, 5-H), 3.68(t, J 3,4 ) J 2,3 ) 9.3 Hz, 3-H), 2.10 (s, 6H), 2.07
(s, 3H); 13C NMR (75 MHz, CDCl3) δ 170.99(0), 169.68(0),
169.46(0), 157.23(0), 138.77, 134.46, 119.55, 117.63(-), 99.49,
86.29(0), 79.90, 73.36(-), 72.74, 72.52, 69.77, 62.65(-), 21.46,
21.26, 21.13; MS (FAB) m/z 548.0 [M]+.
To a solution of 4 (548 mg, 1.0 mmol) in dried MeOH (7
mL) was added a catalytic amount of NaOMe (pH ∼ 9). After
2 h, the solution was neutralized with Amberlite IR-200 (H+)
and filtered, and the filtrate was evaporated to a residue that
was used in the next step without further purification. 1H
NMR (400 MHz, CDCl3) δ 7.59 (dt, 2H), 6.91 (dt, 2H), 6.08-
5.98 (m 1H, vinyl-H), 5.30 (ddt, 1H, vinyl-H), 5.14 (ddt, 1H,
vinyl-H), 4.88 (d, 1H, J 1,2 ) 7.8 Hz, 1-H), 4.44-4.34 (m, 2H,
ally-Hs), 3.88 (dd, J 5,6 ) 2.0 Hz and J 6′,6 ) 12.1 Hz, 6-H), 3.69
(dd, 1H, J 5,6′ ) 5.3 Hz, 6′-H), 3.52 (dd, J 2,3 ) 9.1 Hz, 2-H), 3.47-
3.44 (m, 2H), 3.36-3.31 (m, 2H); 13C NMR (100 MHz, CDCl3)
δ 159.210), 139.57, 137.11, 120.31, 116.91(-), 102.29, 85.84,
85.51(0), 78.20, 75.33(-), 75.02, 71.13, 62.53(-).
p-Iod op h en yl 2-O-Acetyl-3-O-a llyl-4,6-O-ben zylid en e-
â-D-glu cop yr a n osid e (7b). Prepared as described above to
give 7b (120 mg, 98%) as a white foam; [R]D -10.0 (c 0.51,
1
CHCl3); IR 1747 cm-1; H NMR (300 MHz, CDCl3) δ 2.08 (s,
3H); MS (FAB), m/z 553.1 [MH+].
p-Iod op h en yl 3-O-Allyl-4,6-O-ben zylid en e-2-O-bu tyr yl-
â-D-glu cop yr a n osid e (7c). To a solution of 6 (50 mg, 0.1
mmol) in CH2Cl2 were added Et3N (15 µL, 0.11 mmol) and
butyryl chloride (11 µL, 0.11 mmol), and the mixture was
stirred at room temperature. After 2 h, another portion of Et3N
(15 µL, 0.11 mmol) and butyryl chloride (11 µL, 0.11 mmol)
was added. After TLC indicated the completion of the reaction,
MeOH (50 µL) and Et3N (20µL) were added, and the solution
was stirred for 1 h. The solvent was removed in vacuo and
the residue dissolved in CH2Cl2, processed as usual, and
purified by chromatography (hexanes-ethyl acetate, 3/1) to
give 7c (56 mg, 99%) as a white foam; [R]D -11.6 (c 0.58,
CHCl3); IR (KBr) 1746 cm-1; HRMS calcd for C26H30O7I:
581.10364; found: 581.10140.
p-Iod op h en yl 3-O-Allyl-4,6-O-ben zylid en e-2-O-isobu -
tyr yl-â-D-glu copyr an oside (7d). Prepared as described above
to give 7d (60 mg, 94%) as a white foam; [R]D -11.4 (c 0.175,
CHCl3); MS (FAB) m/z 581.0 [MH]+.
p-Iod op h en yl 3-O-Allyl-6-O-(m -tr iflu or om eth yl)ben ze-
n esu lfon yl-â-D-glu cop yr a n osid e (8). To a solution of 5 (42
mg, 0.1 mmol) in dry pyridine (1.5 mL) was added 3-(trifluo-
romethyl)benzenesulfonyl chloride (19 mL, 0.12 mmol) drop-
wise at 0 °C. The reaction mixture was stirred for 1 h at 0 °C
and then allowed to warm to room temperature. After TLC
indicated completion of the reaction, the solvent was removed
in vacuo and the residue dissolved in CH2Cl2, processed as
usual, and purified by chromatography (hexanes-ethyl ac-
etate, 5/3) to yield 8 (53 mg, 85%) as a white foam; [R]D -33.3
(c1.5, CH3Cl); IR (neat/NaCl) 1485, 1327, 1239, 1187 cm-1; 1H
NMR (400 MHz, CDCl3) δ 8.14 (s, 1H), 8.01 (d, 1H), 7.81 (d,
1H), 7.57-7.52 (m, 3H), 6.73-6.69 (m, 2H), 6.00-5.91 (m, 1H,
vinyl-H), 5.32 (dd, 1H, vinyl-H), 5.23 (dd, 1H, vinyl-H), 4.77
(d, 1H, J 1,2 ) 7.7 Hz, 1-H), 4.49-4.25 (m, 4H, 6-H, 6′-H and
p-Iod op h en yl 3-O-Allyl-4,6-O-ben zylid en e-â-D-glu cop y-
r a n osid e (6). To a solution of 5 (∼1.0 mmol) in DMF (18 mL)
were added benzaldehyde dimethyl acetal (225 µL, 1.5 mmol)
and tetrafluoroboric acid (54% in ether, 2.1 mL, 1.5 mmol).
The reaction mixture was stirred overnight at room temper-
ature, the solvent removed in vacuo, and the residue dissolved
in CHCl3, filtered through a short layer of silica gel, concen-
trated, and purified by chromatography (hexanes-ethyl ac-
etate 5/1) to yield 6 (434 mg, 85%, over two steps) as a white
allyl-Hs), 3.67-3.62 (m, 2H, 2-H and 5-H), 3.54 (dt, 1H, J 3,4
)
J 4,5 ) 8.9 Hz, J 4,
) 2.8 Hz, 4-H), 3.78(t, 1H, J 2,3 ) 8.9 Hz,
OH
3-H), 2.65 (d, 1H, 2-OH), 2.52 (d, 1H, 4-OH); 13C NMR (75 MHz,
CDCl3) δ 156.46(0), 138.27, 136.76(0), 134.50, 130.96, 130.36-
(0), 130.33, 124.88(0), 124.84, 118.79, 117.68(-), 100.36, 85.67-
(0), 82.84, 73.81(-), 73.64, 73.16, 69.00(-), 68.72; HRMS calcd
for C22H22F3IO8S: 630.00322; found: 630.00387.
1
foam; [R]D -11.7 (c 0.75, CHCl3); H NMR (400 MHz, CDCl3)
δ 7.60 (dd, 2H), 7.50-7.48 (m, 2H), 7.40-7.26 (m, 3H), 6.83
(dd, 2H), 6.02-5.92 (m, 1H, vinyl-H), 5.57 (s, 1H, CHPh), 5.32
p-Iod op h en yl 3-O-Allyl-2-O-p r op ion yl-6-O-(m -tr iflu o-
r om eth yl)ben zen esu lfon yl-â-D-glu cop yr a n osid e (8a ). The
reaction mixture of 7a (140 mg, 0.24 mmol) and TFA (2.4 mL)
in water/THF (5/20 mL) was refluxed at 85 °C for 2 h, and the
reaction was monitored by TLC (hexanes-ethyl acetate 1/1).
The crude was transferred to a separatory funnel, diluted with
CH2Cl2, and washed with saturated NaHCO3 solution, and the
organic phase was processed as usual. The residue was
purified by chromatography (hexanes-ethyl acetate 2/1) to
yield 4-iodophenyl 3-O-allyl-2-O-propionyl-â-D-glucopyranoside
(100 mg, 98%) as a white foam. To a solution of this compound
(67 mg, 0.14 mmol) in pyridine (2.0 mL) containing a catalytic
amount of DMAP was added 3-(trifluoromethyl)benzenesulfo-
(dq, 1H, vinyl-H), 5.22 (dq, 1H, vinyl-H), 5.00 (d, 1H, J 1,2
)
7.7 Hz, 1-H), 4.48 (dq, 1H, allyl-H), 4.36 (dd, 1H, J 5,6 ) 5.0
Hz, J 6,6′ ) 10.5 Hz, 6-H), 4.29 (dq, 1H, allyl-H), 3.84-3.64 (m,
4H, 2-H, 3-H, 4-H and 6′-H), 3.59-3.55 (m, 1H, 5-H), 2.68 (d,
1H, 2-OH); 13C NMR (75 MHz, CDCl3) δ 157.10(0), 138.88,
137.44(0), 135.03, 129.52, 128.73, 126.39, 119.70, 118.13(-),
101.72, 101.43, 86.40(0), 81.46, 80.32, 74.16(-), 74.10, 68.95-
(-), 67.00; HRMS calcd for C22H24O6I: 511.06177; found:
511.05980.
p-Iod op h en yl 3-O-Allyl-4,6-O-ben zylid en e-2-O-p r op io-
n yl-â-D-glu cop yr a n osid e (7a ). A solution of 6 (83 mg, 0.16
mmol) and propionic anhydride (0.07 mL) in pyridine (2 mL)

Functionalized Glycomers as Growth Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3609
nyl chloride (27 µL, 1.2 equiv) at 0 °C and the solution stirred
for 30 min, then allowed to warm to room temperature. The
solvent was removed in vacuo, the residue dissolved in CH2-
Cl2 and washed with water, 1 N HCl, and water, and the
organic layer separated and concentrated. The residue was
purified by chromatography (hexanes-ethyl acetate 2/1) to
yield 8a (91 mg, 95%) as a white foam; [R]D -24.9 (c 0.515,
CHCl3); IR (film) 1745, 1611 cm-1; 1H NMR (400 MHz, CDCl3)
δ 8.14 (s, 1H), 8.02 (d, 1H), 7.81 (d, 1H), 7.57-7.52 (m, 3H),
6.64 (d, 2H), 5.91-5.81 (m, 1H, vinyl-H), 5.30-5.18 (m, 2H,
vinyls-Hs), 5.10 (dd, 1H, J 2,3 ) 9.0 Hz, 2-H), 4.87 (d, 1H, J 1,2
) 7.9 Hz, 1-H), 4.48 (dd, 1H, J 5,6 ) 1.2 Hz, J 6,6′ ) 11.3 Hz,
6-H), 4.36 (dd, 1H, J 5,6′ ) 4.9 Hz, 6′-H), 4.22-4.17 (m, 2H, allyl-
Hs), 3.74-3.64 (m, 2H, 4-H and 5-H), 3.49 (t, J ) 9.3 Hz, 3-H),
2.99 (s, 1H), 2.34 (q, 2H), 1.15(t, 3H); 13C NMR (75 MHz,
CDCl3) δ 173.27(0), 157.15(0), 138.84 137.20(0), 134.66, 131.59,
131.04, 131.0, 130.54, 125.44, 125.39, 119.26, 118.23(-), 99.37,
86.18(0), 82.25, 74.05(-), 73.84, 72.54, 69.57(-), 69.52,
28.09(-), 9.61; HRMS calcd for C25H26F3IO9S: 686.02944;
found 686.02488.
p -Iod op h en yl 3-O-Allyl-2-O-a cet yl-6-O-(m -t r iflu or o-
m eth yl)ben zen esu lfon yl-â-D-glu cop yr a n osid e (8b). Pre-
pared as described above to yield 8b (90%) as a white foam;
[R]D -20.4 (c 0.69, CHCl3); MS (FAB), m/z, 672.0; HRMS: calcd
for C24H24F3IO9S: 672.01379; found 672.01332.
p -Iod op h en yl 3-O-Allyl-2-O-b u t yr yl-6-O-(m -t r iflu or o-
m eth yl)ben zen esu lfon yl-â-D-glu cop yr a n osid e (8c). Pre-
pared as described above to yield 8c (94%) as a white foam;
[R]D -28.8 (c 0.7, CHCl3); HRMS calcd for C26H29O9ISF3:
701.05292; found: 701.05570.
MeOH (50 µL) and Et3N (9 µL) were added, and the mixture
was stirred for 1 h, diluted with CH2Cl2, and processed as
usual. The residue was chromatographed (hexanes-ethyl
acetate 2/1) to yield 10a (96%) as a white foam; [R]D -33.0 (c
0.385, CHCl3); IR (KBr) 1743, 1484, 1327, 1236, 1164 cm-1
;
1H NMR (400 MHz, CDCl3) δ 8.10 (s, 1H), 8.01 (d, 1H), 7.80
(d, 1H), 7.61 (t, 1H), 7.52 (dd, 2H), 6.66 (dd, 2H), 5.90-5.81-
(m, 1H, vinyl-H), 5.54 (t, 1H, NH), 5.28-5.09 (m, 3H, vinyls-
Hs and 2-H), 4.92 (d, 1H, J 1,2 ) 8.0 Hz, 1-H), 4.26-4.16 (m,
2H, allyl-Hs), 3.72 (dt, 1H, 4-H), 3.60-3.47 (m, 3H, 3-H, 5-H,
and OH), 3.38-3.35 (m, 2H, 6-H and 6′-H), 2.33 (q, 2H), 1.13
(t, 3H); 13C NMR (75 MHz, CDCl3) δ 173.57(0), 157.16(0),
141.53(0), 138.94, 134.91, 130.65, 130.53, 129.90, 129.86,
127.91, 119.25, 117.92(-), 99.51, 86.24(0), 81.86, 74.72, 74.09-
(-), 72.67, 70.65, 44.06(-), 28.10(-), 9.63; MS (FAB) m/z 708.0
[MNa]+; HRMS calcd for C25H26F3INO8SNa: 708.03519; found
708.03877.
p -Iod op h en yl 3-O-Allyl-2-O-b u t yr yl-6-d eoxy-6-N-(m -
t r iflu or om et h yl)ben zen esu lfon yl)-â-D-glu cop yr a n osid e
(10b). Prepared as described above to yield 10b (25 mg, 83%)
as a white foam; [R]D -21.8 (c 1.30, CHCl3); MS (FAB) m/z
722.0 [MNa]+; HRMS calcd for C26H29O8NSF3INa: 722.05084;
found: 722.04890.
p-Iod op h en yl 3-O-Allyl-6-O-ben zyl-2-O-p r op ion yl-â-D-
glu cop yr a n osid e (11a ). 1N HCl was added at room temper-
ature to a solution of 7a (28 mg, 0.05 mmol) and sodium
cyanoborohydride (39 mg, 0.6 mmol) in THF (10 mL) contain-
ing molecular sieves until the evolution of gas ceased. The
mixture was diluted with CH2Cl2 and water and filtered over
Celite, the filtrate was extracted with CH2Cl2 and processed
as usual, and the product was purified with flash column
chromatography (hexanes-ethyl acetate, 2/1) to give 11a (25
mg, 89%) as a white foam; [R]D -11.1 (c 0.54, CHCl3); IR (neat/
p-Iod op h en yl 2-O-Isobu tyr yl-3-O-a llyl-6-O-(m -tr iflu o-
r om eth yl)ben zen esu lfon yl-â-D-glu copyr an oside (8d). Pre-
pared as described for 8a to yield 8d (90%) as a white foam;
[R]D -17.4 (c 0.5, CHCl3); HRMS calcd for C26H29O9ISF3:
701.05292; found: 701.05765.
1
NaCl) 1745 cm-1; H NMR (400 MHz, CDCl3) δ 7.48 (d, 2H),
7.29-7.21 (m, 5H), 6.70 (d, 2H), 5.87-5.78 (m, 1H, vinyl-H),
5.24-5.11 (m, 3H, 2-H and vinyl-Hs), 4.86 (d, 1H, J 1,2 ) 7.9
Hz, 1-H), 4.51 (AB, 2H, J ) 11.9 Hz and J ) 1 6.4 Hz,
CH2Ph), 4.17-4.15 (fm, 2H, allyl-Hs), 3.80-3.66 (m, 3H, 6,
6′-H and 4-H), 3.60-3.54 (m, 1H, 5-H), 3.45 (t, 1H, J ) 9.2
Hz, 3-H), 2.76 (d, 1H, 4-OH), 2.30 (q, 2H), 1.12 (t, 3H); 13C
NMR (75 MHz, CDCl3) δ 173.27(0), 157.44(0), 138.78, 138.02-
(0), 134.93, 128.90, 128.30, 128.14, 119.52, 117.79(-), 99.66,
85.98(0), 82.50, 74.96, 74.16(-), 73.83(-),72.69, 71.83,
70.44(-), 53.86, 28.11(-), 9.64; HRMS: calcd for C25H29IO7:
568.09580,: found 568.10654.
p-Iod op h en yl 3-O-a llyl-6-a zid o-6-d eoxy-2-O-p r op ion yl-
â-D-glu cop yr a n osid e (9a ). A mixture of 8c (600 mg, 0.9
mmol) and NaN3 (340 mg, 5.4 mmol) in DMF (9 mL) was
stirred 1 h at 60-65 °C, and the solvent was removed in vacuo.
The residue was partitioned between CH2Cl2 and water, the
aqueous phase was extracted with excess of CH2Cl2, and the
combined organic layer was dried over Na2SO4 and concen-
trated. The crude product was purified by flash column
chromatography (hexanes-ethyl acetate 5/2) to yield 9a (375
mg, 86%) as a white foam; [R]D -52.3 (c 0.325, CHCl3); IR
1
(KBr) 2102, 1744 cm-1; H NMR (400 MHz, CDCl3) δ 7.61-
p-Iod op h en yl 3-O-Allyl-6-O-ben zyl-2-O-bu tyr yl-â-D-glu -
cop yr a n osid e (11b). Prepared as described above to yield 11b
(95%) as a white foam; [R]D ) -26.2 (c 0.45, CHCl3); IR (KBr)
1744 cm-1; HRMS calcd for C25H29IO7: 582.11145: found
582.11234.
p-Iod op h en yl 2,4,6-Tr i-O-a cetyl-3-O-a llyl-R-D-glu cop y-
r a n osid e (12). Obtained as a byproduct of 4, in <10%, white
foam; [R]D +119 (c 0.45, CHCl3); 1H NMR (400 MHz, CDCl3) δ
7.59-7.57(m, 2H), 6.87-6.84(m, 2H), 5.85-5.81 (m 1H, vinyl-
H), 5.68(d, 1H, J 1,2 ) 3.7 Hz, 1-H), 530-5.08(m, 3H, 4-H and
vinyl-Hs), 4.94(dd, 1-H, J 2,3 ) 10.0 Hz, 2-H), 4.27-4.11 (m,
4H, 6-H, 6′-H and allyl-Hs), 4.03 (t, 1H, 3-H), 3.98-3.94 (m,
1H, 5-H), 2.10(s, 6H), 2.07(s, 3H); 13C NMR (75 MHz, CDCl3)
δ 171.04(0), 170.48(0), 169.79(0), 156.42(0), 138.88, 134.80,
119.37, 117.25(-), 94.93, 86.06(0), 77.09, 74.28(-), 73.05,
69.82, 68.98, 62.25(-), 21.24, 21.09; MS (FAB), m/z 452.2 [M]+.
7.58 (fm, 2H), 6.79-6.77 (fm, 2H), 5.94-5.84 (m, 1H, vinyl-
H), 5.32-5.21 (m, 3H, 2-H, and vinyl-Hs), 4.94 (d, 1H, J 1,2
)
7.8 Hz, 1-H), 4.27-4.13 (m, 2H, allyl-Hs), 3.70-3.62 (m, 3H,
4-H, 6-H and 6′-H), 3.53-3.49 (m, 2H, 3-H and 5-H), 2.65 (d,
1H, OH), 2.36 (q, 2H), 1.19 (t, 3H); 13C NMR (75 MHz, CDCl3)
δ 173.24(0), 157.31(0), 138.90, 134.65, 119.72, 118.31(-), 99.91,
86.47(0), 82.52, 75.49, 73.91(-), 72.83, 70.78, 51.91(-), 28.14-
(-), 9.65; HRMS: calcd for C18H22O6IN3: 503.05534; found:
503.05765.
4-Iod op h en yl 3-O-Allyl-6-a zid o-6-d eoxy-2-O-bu tyr yl-â-
D-glu cop yr a n osid e (9b). Prepared as described above to yield
9b (125 mg, 83%) as a white foam; [R]D -81.8 (c 0.5, CHCl3);
IR (KBr) 2102, 1745 cm-1; MS (FAB) m/z 518.1[MH]+, 540.2
[MNa]+; HRMS calcd for C19H25O6IN3: 518.07880; found:
518.07960.
p-Iod op h en yl 3-O-Allyl-2-O-p r op ion yl-6-d eoxy-6-N-(m -
t r iflu or om et h yl)b en zen esu lfon yl-â-D-glu cop yr a n osid e
(10a ). A mixture of 9a (352 mg, 0.7 mmol) and triphenylphos-
phine (220 mg, 0.84 mmol) in THF (3.5 mL) containing water
(15 µL) was stirred at room temperature for 20 h, then
concentrated in vacuo. The residue was purified by flash
column chromatography (CH2Cl2-MeOH-NH4OH 9:1:0.25,
lower layer used as eluant) to yield 4-iodophenyl 3-O-allyl-6-
amino-2-O-butyryl-6-deoxy-â-D-glucopyranoside (242 mg, 96%)
as a white foam. To a solution of the amine (48 mg, 0.1 mmol)
and Et3N (15 µL, 0.11 mmol) in CH2Cl2 (2.0 mL) was added
3-(trifluoromethyl)benzenesulfonyl chloride (18 µL 0.11 mmol)
at room temperature. After the mixture was stirred overnight,
p -Iod op h en yl 3-O-Allyl-2-O-b u t yr yl-6-O-(m -t r iflu or o-
m eth yl)ben zen esu lfon yl-â-D-glu cop yr a n osid e (13). Pre-
pared as described for 8a to yield 13 (50% overall) as a colorless
syrup; [R]D +68.3 (c 0.12, CDCl3); 1H NMR (400 MHz, CDCl3)
δ 8.18(s, 1H), 8.07 (d, 1H), 7.93 (d, 1H), 7.72 (t, 1H), 7.54 (dd,
2H), 6.73 (dd, 2H), 5.98-5.87 (m, 1H, vinyl-H), 5.52(d, 1H, J 1,2
) 3.6 Hz, 1-HR), 5.31 (dd, 1H, vinyl-H), 5.22 (dd, 1H, vinyl-
H), 4.75 (dd, 1H, J 2,3 ) 9.9 Hz, 2-H), 4.42 (dd, J 5,6 ) 3.3 Hz,
J 5,6′ ) 11.2 Hz, 6-H), 4.37-4.30 (m, 2H), 4.23 (dd, 1H, J ) 5.9
Hz, J ) 12.7 Hz), 3.91-3.83 (m, 2H), 3.65 (t, 1H, J ) 9.1 Hz,
3-H), 2.70-2.59 (broad, 1H, OH), 2.32 (t, 2H), 1.62 (tq, 2H),
0.92 (t, 3H); 13C NMR (100 MHz, CDCl3) δ 173.17, 156.51,
138.91, 138.00, 134.87, 131.54, 130.96, 130.49, 125.43, 119.23,

3610 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17
Hanessian et al.
117.89, 95.25, 85.06, 79.13, 74.51, 72.95, 70.25, 69.43, 69.25,
36.48, 18.73, 13.97; HRMS calcd for C26H28F3IO9S: 700.04509;
found: 700.05943.
721.9 [MNa]+; HRMS calcd for C26H29F3SIO8NaN: 722.05084;
found: 722.03310.
p-Iod op h en yl 3-O-Allyl-6-O-ben zyl-2-O-bu tyr yl-â-D-ga -
la ctop yr a n osid e (18a ). Prepared as described for 11a to yield
18a (91%) as a white foam; [R]D +5.0 (c 1.105, CH3Cl); 1H NMR
(400 MHz, CDCl3) δ 7.53 (d, 2H),7.36-7.31 (m, 5H), 6.79 (d,
2H), 5.93-5.80 (m, 1H, vinyl-H), 5.44 (dd, 1H, J 2,3 ) 9.7 Hz,
2-H), 5.32-5.20 (m, 2H, vinyl-Hs), 4.90 (d, 1H, J 1,2 ) 8.0 Hz,
1-H), 4.58 (s, 2H, CH2PH), 4.21-4.15 (m, 1H, allyl-H), 4.11
(d, 1H, J 4,5 ) 0, 4-H), 4.07-4.01 (m, 1H, allyl-H), 3.89-3.74
(m, 3H, 5-H, 6-H and 6′-H), 3.55 (dd, 1H, J 3,4 ) 3.4 Hz, 3-H),
2.32 (t, 2H), 1.70-1.63 (m, 2H), 0.95 (t, 3H); 13C NMR (75 MHz,
CDCl3) δ 172.54(0), 157.43(0), 138.71, 138.24, 134.25, 128.87,
128.25, 128.20, 119.56, 118.45(-), 99.70, 85.81(0), 78.93, 74.45,
74.23(-), 71.25, 70.32, 69.45(-), 66.60, 36.62(-), 18.92(-),
14.01; HRMS calcd for C26H31IO7: 582.11145; found 582.11432.
p-Iod op h en yl 3-O-Allyl-6-O-ben zyl-2-O-isobu tyr yl-â-D-
ga la ctop yr a n osid e (18b). Prepared as described for 11a to
yield 18b (95%) as a white foam; [R]D +1.1 (c 1.135, CH3Cl);
HRMS calcd for C26H31IO7: 582.11145; found 582.12876.
p-Iod op h en yl 3-O-Allyl-4,6-O-ben zylid en e-2-O-bu tyr yl-
â-D-ga la ctop yr a n osid e (15a ). Prepared as described for 7a
to give 15a (76%) as a white foam; [R]D -10.4 (c 1.0, CHCl3);
IR (neat/NaCl) 1747, 1485 cm-1; 1H NMR (400 MHz, CDCl3) δ
7.56-7.53 (m, 4H), 7.37-7.34 (m, 3H), 6.78 (dd, 2H) 5.94-
5.83(m, 1H, vinyl-H), 5.63-5.57 (m, 2H, 2-H and CHPh), 5.32-
5.19 (m, 2H, vinyl-Hs), 4.99 (d, 1H, J 1,2 ) 8.0 Hz 1-H), 4.36-
4.34 (m, 2H, 4-H and 6-H), 4.22-4.10 (m, 3H, H6′ and allyl-
Hs), 3.67 (dd, 1H, J 2,3 )10.1 Hz and J 3,4 ) 3.4 Hz, 3-H), 3.57-
3.55 (m, 1H, 5-H), 2.31 (t, 2H), 1.69-1.63 (m, 2H), 0.94 (t, 3H);
13C NMR (100 MHz, CDCl3) δ 172.36(0), 157.54(0), 138.65,
137.80(0), 134.97, 129.46, 128.57, 126.87, 119.79, 117.92(-),
101.62, 100.10, 85.90(0), 77.37, 73.54, 70.99(-), 69.85,
69.38(-), 67.26, 36.63(-), 18.96(-), 14.04; MS (FAB), m/z 581.0
[MH]+.
p-Iod op h en yl 3-O-Allyl-4,6-O-ben zylid en e-2-O-isobu -
tyr yl-â-D-ga la ctop yr a n osid e (15b). Prepared as described
7a to give 15b (95%) as a white foam; [R]D -9.1 (c 0.515,
Ack n ow led gm en t. The authors thank the Swiss
League and Research against Cancer (grant KFS 947-
09-1999), the Swiss National Science Foundation for
Scientific Research (grant 3152-059219.99), and the
Swiss Society for Multiple Sclerosis for financial sup-
port. We are very grateful to Drs. Sylvain Meyer,
Lausanne, for providing the human surgical specimens
for fibroblast cultures and R. C. J anzer for critical
reading of the manuscript. We also thank NSERC for
generous financial assistance through the Medicinal
Chemistry Chair Program (S.H., L.Z.).
1
CHCl3); IR (neat/NaCl) 1746, 1485 cm-1; H NMR (400 MHz,
CDCl3) δ 2.60-2.56 (m,1H), 1.21-1.17(m, 6H); MS(FAB), m/z
579.1 [M - H]+, 603.1 [MNa]+. See also Supporting Informa-
tion.
p -Iod op h en yl 3-O-Allyl-2-O-b u t yr yl-6-O-(m -t r iflu or o-
m eth yl)ben zen esu lfon yl-â-D-galactopyr an oside (16a). Pre-
pared as described for 8a to give 16a (82%) as a white foam;
[R]D -33.3 (c 1.0, CHCl3); IR (neat/NaCl) 1744, 1485, 1369,
1
1327, 1185, 1075 cm-1; H NMR (400 MHz, CDCl3) δ 8.15 (s,
1H), 8.04 (d, 1H, J ) 7.9 Hz), 7.88 (d, 1H, J ) 7.9 Hz), 7.61(t,
1H, J ) 7.9 Hz), 7.54 (dd, 2H, J ) 2.1 Hz and 6.9 Hz), 6.70
(dd, 2H, J ) 2.1 Hz and 6.9 Hz), 5.87-5.80 (m, 1H, vinyl-H),
5.35 (dd, 1H, J 1,2 ) 8.1 Hz and J 2,3 ) 9.8 Hz, 2-H), 5.31-5.22
(m, 2H, vinyl-H), 4.88(d, 1H, 1-H), 4.42-4.32 (m, 2H, 6-H and
6′-H), 4.19-4.01 (m, 3H, allyl-H and 4-H), 3.95-3.92 (m, 1H,
5-H), 3.57 (dd, 1H, J 3,4 ) 3.4 Hz, 3-H), 2.52 (s, 1H, OH), 2.30
(t, 2H, J ) 7.3), 1.68-1.61 (m, 2H), 0.93 (t, 3H); 13C NMR (100
MHz, CDCl3) δ 171.93(0), 156.57(0), 138.23, 136.56(0), 133.37,
130.98, 130.56, 130.03, 124.84, 118.74, 118.33(-), 99.78,
85.53(0), 77.72, 72.15, 71.02(-), 69.25, 68.86(-), 65.53,
Su p p or tin g In for m a tion Ava ila ble: Copies of LC/MS
data for selected compounds. This material is available free
of charge via the Internet at http://pubs.acs.org.
Refer en ces
(1) Bredel, M. Anticancer drug resistance in primary human brain
tumors. Brain Res. Rev. 2001, 35, 161-204.
(2) Brandes, A. A.; Basso, U.; Pasetto, L. M.; Ermani, M. New
strategy development in brain tumor therapy. Curr. Pharm. Des.
2001, 7, 1553-1580.
35.99(-), 18.33(-), 13.43; HRMS calcd for
701.05292; found: 701.05889.
C26H29O9ISF3:
(3) Berens, M. E.; Giese, A. Biology and oncology of invasive glioma
cells. Neoplasia 1999, 1, 208-219.
p-Iod op h en yl 3-O-Allyl-2-O-isobu tyr yl-6-O-(m -tr iflu o-
r om eth yl)ben zen esu lfon yl-â-D-ga la ctop yr a n osid e (16b).
Prepared as described for 8a to yield 16b (95%) as a white
foam; [R]D -31.5 (c 0.355, CHCl3); IR (neat/NaCl) 1743, 1485,
1328, 1186, 1074 cm-1; HRMS calcd for C26H28F3SIO9Na:
723.03485; found: 723.03310.
(4) Prados, M. D.; Levin, V. Biology and treatment of malignant
glioma. Sem. Oncol. 2000, 27, 1-10.
(5) Hofer, S.; Herrmann, R. Chemotherapy for malignant brain
tumors of astrocytic and oligodendroglial lineage. J . Cancer Res.
Clin. Oncol. 2001, 127, 91-95.
(6) Newton, H. B. Novel chemotherapeutic agents for the treatment
of brain cancer. Exp. Opinion Invest. Drugs 2000, 9, 2815-2829.
(7) Nieto-Sampedro, M.; Bailon, C.; Fernandez-Mayoralas, A.; Martin-
Lomas, M.; Mellstro¨m, B.; Naranjo, J . R. Experimental brain
glioma : growth arrest and destruction by a blood-group-related
tetrasaccharide. J . Neuropath. Exp. Neurol. 1996, 55, 169-177.
(8) Aguilera, B.; Romera-Ramirez, L.; Abad-Rodriguez, J .; Corrales,
G.; Nieto-Sampiedro, M. Novel disaccharide inhibitors of human
glioblastoma cell division. J . Med. Chem. 1998, 41, 4599-4606.
(9) Camby, I.; Decaestecker, C.; Gordower, L.; DeDecker, R.; Kacem,
Y.; Lemmers, A.; Siebert, H. C.; Bovin, N. V.; Wesseling, P.;
Danguy, A.; Salmon, I.; Gabius, H. J .; Kiss, R. Distinct differ-
ences in binding capacity to saccharide epitopes in supratentorial
pilocytic astrocytomas, astrocytomas, anaplastic astrocytomas
and glioblastomas. J . Neuropath. Exp. Neurol. 2001, 60, 75-
84.
p -Iod op h en yl 3-O-Allyl-2-O-b u t yr yl-6-d eoxy-6-N-(m -
tr iflu or om eth yl)ben zen e su lfon yl-â-D-ga la ctop yr a n osid e
(17a ). Prepared as described for 10a to yield 17a (82%) as a
white foam; [R]D -19.0 (c 1.225, CHCl3); IR (neat/NaCl) 1742,
1485, 1327, 1165, 1072 cm-1 1H NMR (400 MHz, CDCl3) δ
;
8.12 (s, 1H), 8.02 (d, 1H,), 7.80 (d, 1H), 7.63-7.57 (m, 3H),
6.71 (d, 2H), 5.90-5.80 (m, 1H, vinyl-H), 5.36 (dd, 1H, J 2,3
)
9.7 Hz, 2-H), 5.31-5.18 (m, 3H, vinyl-Hs and 6-NH), 4.92 (d,
1H, J 1,2 ) 8.1 Hz 1-H), 4.20-4.04 (m, 3H, allyl-Hs and 4-H),
3.84-3.82 (m, 1H, 5-H), 3.57 (dd, 1H, J 3,4 ) 3.4 Hz, 3-H), 3.40-
3.37 (m, 2H, 6-H and 6′-H), 2.68 (s, 1H, 4-OH), 2.30 (t, 2H),
1.69-1.62 (m, 2H), 0.94 (t, 3H); 13C NMR (100 MHz, CDCl3) δ
172.64(0), 157.04(0), 141.44(0) 138.91, 133.97, 132.41(0), 131.96,
130.55, 130.42, 129.84, 124.45, 124.40, 118.99, 118.86(-),
99.17, 85.96(0), 78.42, 73.68, 71.52(-), 69.87, 66.84, 43.89(-)
36.56(-), 18.89(-), 13.98; MS (FAB) m/z 722.0 [MNa]+; HRMS
calcd for C26H29F3SIO8NNa: 722.05084; found: 722.04820. See
also Supporting Information.
(10) Hanessian, S.; Saavedra, O. M.; Xie, F.; Amboldi, N.; Battistini,
C. Design and synthesis of functionalized glycomers as nonpep-
tidic ligands for SH2 binding and as inhibitors of A-431 human
epidermoid and HT-29 colon carcinoma cell lines. Bioorg. Med.
Chem. Lett. 2000, 10, 439-442.
(11) Bessel, E. M.; Westwood, J . M. (R,S)-2,3-Ethoxypropyl ethers
and glycosides of D-glucopyranose. Carbohydr. Res. 1992, 25, 11-
21.
(12) Ogawa, T.; Matsui, M. An approach to the synthesis of glyco-
sides: enhancement of nucleophilicity of hydroxyl gorups by
trialkylstannylation. Carbohydr. Res. 1976, 51, C13-C18.
(13) Garegg, P.; Hultberg, H.; Wallin, S. A novel reductive ring
opening of carbohydrate benzylidene acetals. Carbohdr. Res.
1982, 108, 97-101.
p-Iod op h en yl 3-O-Allyl-6-d eoxy-2-O-isobu tyr yl-6-N-(m -
t r iflu or om e t h yl)b e n ze n e su lfon yl-â-D-ga la ct op yr a n o-
sid e (17b). Prepared as described for 10a to yield 17b (95%)
as a white foam; [R]D -22.6 (c 0.97, CHCl3); IR (neat/NaCl)
1
1741, 1484, 1072 cm-1; H NMR (300 MHz, CDCl3) δ 2.55-
2.43(m, 1H), 1.12-1.08 (m, 6H); MS (FAB) m/z 697.9 [M]+,

Functionalized Glycomers as Growth Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3611
(14) Diserens, A. C.; de Tribolet, N.; Martin-Achard, G.; Gaide, A.
C.; Schnegg J . F.; Carrel, S. Characterization of an established
human malignant glioma cell line : LN18. Acta Neuropathol.
1981, 53, 21-28.
tenyl cytosine (NSC 375575) in human glioblastoma cells. Oncol.
Res. 1997, 11-118, 1997.
(18) Egidy, G.; Peduto Eberl, L.; Valdenaire, O.; Irmler, M.; Majdi,
R.; Diserens, A. C.; Fontana, A.; J anzer, R. C.; Pinet, F.;
J uillerat-J eanneret, L. The endothelin system in human glio-
blastoma. Lab. Invest. 2000, 80, 1681-1689.
(19) Uckun, F. M. Rationally designed anti-mitotic agents with pro-
mitotic activity. Curr. Pharm. Des. 2001, 7, 1627-1639.
(20) Peduto Eberl, L.; Valdenaire, O.; Saintgiorgio, V.; J eannin, J .
F.; J uillerat-J eanneret, L. Endothelin receptor blockade potenti-
ates FasL-induced apoptosis in rat colon carcinoma cells. Int. J .
Cancer 2000, 86, 182-187.
(15) Muruganandan A.; Herz, L. M.; Monette, R.; Durkin, J . P.;
Stanimirovic, D. Development of immortalized human cere-
brovascular endothelial cell line as an in vitro model of the
human blood-brain barrier. FASEB J . 1997, 11, 1187-1197.
(16) Uehlinger, P.; Zellweger, M.; Wagnie`res, G.; J uillerat-J eanneret,
L.; van den Bergh, H.; Lange, N. 5-aminolevulinic acid and its
derivatives : physical chemical properties and protoporphyrin
IX formation in cultured cells. J . Photochem. Photobiol. B: Biol.
2000, 54, 72-80.
(17) Agbaria, R.; Kelley, J . A.; J ackman, J .; Viola, J . J .; Ram, Z.;
Oldfield, E.; J ohns, D. G. Antiproliferative effects of cyclopen-
J M0205853