Synthesis and cytotoxic properties of new fluorodeoxyglucose-coupled chlorambucil derivatives

Article abstract of DOI:10.1016/j.bmc.2008.03.038

Frequently used in the treatment of malignant cells, alkylating agents, like most anticancer substances, produce adverse side effects caused by the toxicity of the agents toward normal tissues and lose efficiency through poor distribution to target sites.

Full text of DOI:10.1016/j.bmc.2008.03.038

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 5004–5020
Synthesis and cytotoxic properties of new
fluorodeoxyglucose-coupled chlorambucil derivatives
*
´
`
Bastien Reux, Valerie Weber, Marie-Josephe Galmier, Michele Borel,
Michel Madesclaire, Jean-Claude Madelmont, Eric Debiton and Pascal Coudert
Univ Clermont 1, UFR Pharmacie/Me´decine, Clermont-Ferrand F-63001, France
INSERM, U484, rue Montalembert, Clermont-Ferrand, F-63005, France
Centre Jean Perrin, Clermont-Ferrand, F-63005, France
Received 3 October 2007; revised 4 March 2008; accepted 14 March 2008
Available online 17 March 2008
Abstract—Frequently used in the treatment of malignant cells, alkylating agents, like most anticancer substances, produce adverse
side effects caused by the toxicity of the agents toward normal tissues and lose efficiency through poor distribution to target sites.
Our approach to developing more selective drugs with low systemic toxicity is based on the premise that the body distribution and
cell uptake of a drug can be altered by attaching a neoplastic cell-specific uptake enhancer, such as 2-fluoro-2-deoxyglucose (FDG),
the radiotracer most frequently used in PET for tumor imaging. Two properties of deoxyglucose, namely preferential accumulation
in neoplastic cells and inhibition of glycolysis, underpin this targeting approach. Here, we report the synthesis of 19 new chloram-
bucil glycoconjugates in which the alkylating drug is attached to the C-1 position of FDG, directly or via different linkages. This set
of compounds was evaluated for in vitro cytotoxicity against different human normal and tumor cell lines. There was a significant
improvement in the in vitro cytotoxicity of peracetylated glucoconjugates compared with the free substance. Four compounds were
finally selected for further in vivo studies owing to their lack of oxidative stress-inducing properties.
ꢀ 2008 Elsevier Ltd. All rights reserved.
1. Introduction
and intestinal epithelial cells. Non-selective cytotoxicity
is the main effect that limits the use of optimal doses
in most conventional chemotherapeutic drug regimens.
Consequently, protocols with sub-optimal doses are fre-
quently used, minimizing toxicity and patient suffering
but often resulting in inefficient treatment and unsatis-
factory prognosis (residual disease, tumor regrowth,
metastasis, drug resistance, etc.). This major issue dic-
tates research in selective chemotherapy by drug target-
ing—if a drug can be selectively targeted to a tumor to
provide a high concentration of drug selectively in the
target zone, with less drug reaching critical normal tis-
sues, then many problems associated with the systemic
toxicity of chemotherapy will be resolved. Cancer ther-
apy and patients’ quality of life can thereby be substan-
tially improved.
Selectively directing a drug into the vicinity of its in-
tended site of action has clear therapeutic advantages,
including reduced toxicity, smaller dose levels and espe-
cially enhanced efficacy. This is particularly true in can-
cer chemotherapy, where most of the pharmaceuticals
currently in use do not accumulate selectively in
malignant tissues or cells. The main limitations in con-
ventional cancer chemotherapy arise from the biodistri-
bution of cytotoxins in the body; lack of drug-specific
affinity toward tumor cells, need for a large total dose
of a drug to achieve high local concentration and sys-
temic toxicity leading to many negative, sometimes
life-threatening, side effects. Acute toxicity affects rap-
idly dividing normal cells such as bone marrow cells
Several reviews have examined new drug delivery sys-
tems (DDSs) to achieve drug targeting.1–3 DDS are
based on two main principles: (i) passive targeting,
using a vehicle, and (ii) active targeting, using a specific
vector and exploiting an increased affinity of the deliv-
ery system toward malignant cell or tissue components.
The first approach involves the development of
Keywords: Chlorambucil; Antitumor agents; 2-Fluoro-2-deoxy-D-glu-
cose derivatives; Drug carrier.
*
Corresponding author. Address: UFR Pharmacie, Laboratoire de
´
Chimie Therapeutique, 28, place Henri Dunant, 63001, Clermont-
Ferrand Cedex, France. Tel.: +33 4 7317 7991; fax: +33 4 7317
7090; e-mail: valerie.weber@u-clermont1.fr
0968-0896/$ - see front matter ꢀ 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.03.038

B. Reux et al. / Bioorg. Med. Chem. 16 (2008) 5004–5020
5005
reservoir-type systems to encapsulate a drug,4,5 such as
polymers, microcapsules, nanoparticles, and micellar or
liposomal drug preparations, which cause spontaneous
drug accumulation due to enhanced permeability and
retention effects.6 The second approach uses a tumor
characteristic: a drug is conjugated with a suitable vec-
tor to impart specific affinity for the target to a non-
specific drug.7 Active DDSs offer a wide scope using
various kinds of carriers, such as antibodies, peptides,
proteins (lectins,8 albumins), hormones, charged enti-
ties, mono-, oligo- and polysaccharides,9 and some
low molecular-weight ligands, such as folate or lac-
tose,10 for example. The two approaches can be com-
bined using DDS nanoparticles bearing active
carriers.11
tard derivative exploiting transmembrane glucose
transport.25,26 Glufosfamide shows lower medullary
toxicity and greater antitumor activity than ifosfamide.
These improvements are due to increased uptake of the
agent by the tumoral cells due to the glucose moi-
ety.27,28 Work on the affinity of some new antineoplas-
tic compounds for the GLUT1 carriers have also been
published; these new compounds, such as hexose keto-
C-glycoside conjugates,29 glucose–chlorambucil deriva-
tives30 and O-methylsulfonyl derivatives of glucose23
were designed to enhance selectivity for brain tumors.
However, none of these studies has produced a thera-
peutic drug.
To use the strongly increased uptake of sugars by tu-
mors, and also to potentiate the cytotoxic effect of
the chemotherapeutic agent carried, we chose a partic-
ular glucose derivative to act as selective drug carrier,
2-deoxy-2-fluoro-D-glucose (FDG). FDG is a structural
analogue of glucose, differing at the second carbon
atom by the substitution of a fluoro for a hydroxyl
group. Its radiolabeled fluorine-18 derivative31 is the
main tumor-specific radiopharmaceutical agent used
in tumor imaging by positron emission tomography
(PET).32 Widespread clinical use of 18F-FDG PET
has unequivocally demonstrated that most primary
and metastatic human tumors avidly trap FDG to
meet the increased glucose. Hence the use of an FDG
moiety coupled to a chemotherapeutic agent should
in principle make it possible to trap the glycoconjugate
inside the tumor cells.
Active targeting is largely concerned with immunocon-
jugates12 especially in antibody-directed enzyme pro-
drug therapy (ADEPT).13,14 The application of
monoclonal antibodies for tumor targetability of che-
motherapeutics appears promising. However, it will be
restricted to tumors expressing high levels of related
antigen. Our work is part of another strategy using
low molecular-weight ligands.
Previous work by our team has shown that drugs can
be targeted toward particular types of tumors, using
biochemical characteristics specific to these cells, such
as high melanin content in melanoma15 or the presence
of proteoglycans in chondrosarcoma.16,17 However,
more ubiquitous drug delivery systems aiming at solid
tumors could extend this type of approach to a much
greater number of cancers, given that solid tumors rep-
resent 89% of malignancies. To achieve this ubiquitous
targeting, we have chosen to exploit a well-docu-
mented, typical biochemical phenotype of all invasive
solid tumors that confers a selective growth advantage:
the upregulation of glycolysis, resulting in increased
rates of glucose consumption compared with normal,
non-transformed cells.18 It has long been recognized
that malignant cells most often exhibit increased glu-
cose metabolism19 and increased glucose transport.20
This upregulation of glycolysis observed especially in
neoplastic tissues is nearly ubiquitous in cancers and
increases with tumor aggressiveness. Even under aero-
bic conditions, neoplastic cells need a 19-fold increase
in glucose consumption per mole of ATP produced be-
cause they lose efficient production of ATP by the aer-
obic oxidative phosphorylation in mitochondria, and
therefore rely solely on glycolysis to provide the energy
requirements of rapidly replicating tissue. This leads to
progressive increases in glucose uptake with malignant
evolution.
In addition, nutrient deprivation has been shown to
cause cancer cell death, and some glucose analogues,
such as deoxyglucose (DG), can mimic glucose depriva-
tion by profoundly inhibiting glucose metabolism.
Some data suggest that neoplastic cells treated with
DG accelerate their own demise33–35 and that
DG36–38 and its tetraacetate ester39 enhance the cyto-
toxic effects of anticancer agents. Normal and trans-
formed cells respond to glucose antagonists in
opposing manners. FDG bearing a chemotherapeutic
may thus interfere with glucose metabolism like DG,
and sensitize tumoral cells to the cytostatic action of
the FDG conjugates.
In this respect, there are two main mechanisms whereby
FDG-cytotoxic conjugates may prove advantageous: (i)
increased glycolysis to support the metabolic require-
ments of malignant cells results in high glucose con-
sumption and may lead to enhanced intracellular
uptake of fluoroglycosylated drugs in tumor cells com-
pared with healthy ones and (ii) DG interferes with en-
ergy metabolism, resulting in cancer cell death by
apoptosis.
Thus, glycolysis-mediated targeting using a glucose
derivative as an active drug carrier could be useful
for delivering chemotherapy selectively to cancer cells.
Studies aimed at increasing tumoral selectivity by con-
jugation of therapeutics to sugars have already been
undertaken.21–23 One of the first anticancer agents de-
signed to exploit the upregulation of glycolysis and suc-
cessfully tested in clinical trials24 is the new drug
glufosfamide, a sugar-linked isophosphoramide mus-
These properties of FDG recently prompted us to initi-
ate a new tumor-targeting approach, using FDG as a
drug carrier.
Chlorambucil (CLB) is a DNA-alkylation-inducing
nitrogen mustard currently used in clinical treatment
especially against chronic lymphatic leukemia, lympho-

5006
B. Reux et al. / Bioorg. Med. Chem. 16 (2008) 5004–5020
mas, and advanced ovarian and breast carcinomas. Che-
motherapy protocols using CLB are limited by toxicity
with nausea, myelotoxicity and neurotoxicity. Like all
alkylating agents, its toxicity arises from its non-specific-
ity for tumor tissue when administered systemically and
from its oxidative stress-inducing properties in normal
cells. In the present study, chlorambucil was used as a
model drug to develop FDG-glycolysis-mediated target-
ing. Here, we describe the synthesis, characterization
and in vitro cytotoxicity of new glycoconjugates in
which the dialkylating antitumor compound chlorambu-
cil is attached to the C-1 position of the FDG skeleton
using a variety of linkages.
spacer arm or to the FDG residue by the formation of
an ester or an amide bond, also we choose in the present
work preferentially ester and amide functions, two of
the most-used bonds between a drug and the labile res-
idue in targeted drug strategies.30,42–47 Enzymatic and
non-enzymatic hydrolysis of ester and amide functional-
ity has been most often reported. For many drugs, pas-
sive hydrolysis may occur in aqueous media; the rate of
release will decrease in the order ester > amide.40
The structure of all the targeted CLB–FDG conjugates
varying by the nature of the spacer is given in Figure 1.
The introduction of a spacer arm between the carrier
and the drug is a strategy widely used to separate the ac-
tive moiety from the carrier. Ideally, as a conjugate is
meant to reach the malignant site, the linkage should
be sufficiently stable during circulation in blood stream
to maintain its chemical integrity until it reaches its tar-
get. However, after uptake into the cells, they should
have an intrinsic activity or release the active compound.
2. Results and discussion
2.1. Chemistry
Schemes 1–6, describe the syntheses of compounds with
linker modifications. 1,3,4,6-Tetra-O-acetyl-2-deoxy-2-
fluoro-b-D-glucopyranose 5 and its derivatives 6–13
(Scheme 1) are the key intermediates for all the syntheses
in which chlorambucil bonds directly or via a spacer to C-
1 of the sugar moiety. Compound 5 was prepared accord-
ing to the method described by Kovac.48 The first step of
the reaction involved the conversion of D-mannose 3 into
1,3,4,6-tetra-O-acetyl-b-D-mannopyranose 4 and is a
modification of the procedure described by Deferrari
et al.49 The second step gave compound 5 by introducing
a fluorine atom at C-2 of the D-mannose derivative, using
diethylaminosulfur trifluoride.48 The introduction of a
spacer via a glycolytic or amide function under Koe-
nigs–Knorr conditions needs the synthesis of compound
6, which was achieved with hydrobromic acid by the pro-
cedure described by Baer et al.50 The NMR data for this
compound agree with those already published.51
There are many ways to tether a drug to the carrier;
however, the carrier must always be anchored in a man-
ner that assures successful targetability. Considering
that many times, if the drug molecule is positioned too
closed to the carrier, the overall affinity of the conjugate
for the targeted receptor or the accessibility of the cyto-
toxic drug to its final destination could be reduced or
even ablated. Admittedly, steric interference is more
pronounced for small ligands (pteroates, amino acids,
sugar, etc.) than it is for larger ligands (protein, MAb,
polymer, nanoparticules, etc.). Thereby, initially in or-
der to evaluate the necessity of spacing between CLB
and the sugar carrier, we synthesized compounds 14,
15, and 17 where CLB was directly linked to the FDG
moiety through an ester, amide, or urea function. Then,
following the widely applied strategy, we also prepared
compounds with a spacer arm between the fluorosugar
moiety and CLB; the composition of the linker can vary
widely and the linker can be designed to contain stable
or cleavable functional groups, and at the same time res-
idues that may influence the physical properties such as
the hydrophilic-lipophilic balance. Indeed, depending on
the nature of the drug activity, it may be crucial to have
it dissociated from the carrier following endocytosis. In
our case, owing to CLB cancer cell-killing mechanism
through DNA alkylation, the overall carrier-drug conju-
gate or chlorambucil derivatives with ‘chemical baggage’
could be even cytotoxic. Thus, we decided to use linkers
possessing cleavable bond juxtaposed to the drug to al-
low the release of the original active chemical form
(CLB) at the opposite extremity of the linker and/or at
the level of the spacer-backbone bond, so that spacer-
CLB moieties can be similarly released and also have
intrinsic activity as nitrogen mustard.
The coupling of a moiety at C-1 of a sugar via an ester
bond required the preparation of compound 9, while an
amide bond needed the synthesis of the amino derivative
8, which was obtained from the a-bromo compound 6
through the b-anomeric azide 7. The a,b-glucopyranose
9 was obtained in quantitative yield from chemoselective
deacetylation of the peracetylated 5 with ammonium
carbonate in DMF.52
For the preparation of compounds possessing an ester
linkage between the active drug and the anchor point
of the carrier, the use of hydroxyl protecting groups
other than acetates proved necessary. Benzyl ethers were
chosen. These benzyl-protected sugars were obtained
starting from bromo-sugar 6. This compound was trea-
ted with sodium ethanethiolate to give the ethyl b-thio-
glycoside 1053 in 87 % yield. Replacement of the acetate
protecting group with benzyl ethers was achieved in two
54
steps: removal of acetyl groups by Zemplen transeste-
`
rification followed by treatment with benzyl bromide in
DMF in the presence of sodium hydride. The addition
of bromide to thioglyoside 11 gave the newly benzyl-
protected bromoglycoside 12.53 The 3,4,6-tri-benzyl-2-
deoxy-2-fluoroglucose 13 was synthesized from 3,4,
Many types of functions can be employed to assure the
in vivo cleavage which can be due to passive hydrolysis
or caused by a more specific mode: enzymatic, reductive
or pH-controlled release.40,41 CLB providing a carboxyl
group is appropriately functionalized to bind to the
6-tri-benzyl-2-deoxy-2-fluoroglucopyranosyl
12 using silver carbonate and water.
bromide

B. Reux et al. / Bioorg. Med. Chem. 16 (2008) 5004–5020
5007
Cl
N
OH
O
O
Cl
HO
HO
OH
HO
F
1, FDG
2, Chlorambucil
Cl
Cl
N
N
OR
O
OR
O
Cl
O
Cl
RO
RO
RO
RO
N
[X]
[X]
H
N
F
F
H
O
R = OAc (a), OH (b)
X = OC(=O) (14), NHC(=O) (15), NHC(=O)NH (17)
R = OAc (a), OH (b)
X = C6H4 (20), CH2 (21), (CH2)2 (31), (CH2)3 (22)
Cl
OR
OR
O
O
O
O
RO
RO
RO
RO
[X]
N
F
H
N
O
Cl
F
N
H
[Y]
O
Cl
N
R = OAc (a), OH (b)
X = NH, Y =O (34) ; Y = O, Y = NH (37)
R = OAc (40a), OH (40b)
Cl
Figure 1. Structures of FDG, chlorambucil and all target compounds.
To evaluate the importance of the type of linkage bind-
ing the carrier and the drug, we synthesized compounds
14, 15, and 17, where the active molecule was directly
linked to the FDG moiety through an ester, amide, or
urea function (Scheme 2).
5.85 ppm and resonates as a doublet of doublets
(J1,2 = 8.1 Hz, J1,F = 3.1 Hz).
For the preparation of compounds 20b, 21b and 22b for
which 4-aminobenzamide, 2-aminoethanamide or 4-
aminobutanamide were used, respectively, as spacer
arms between the sugar residue and chlorambucil, the
synthetic methods in Scheme 3 were adopted. The N-
acylation of the b-aminosugar 8 with 4-nitrobenzoyl
chloride, N-Cbz-glycine or 4-azidobutanoic acid 25
led, respectively, to derivatives 18a, 18b, and 18c in good
yields. The requisite azido acid 25 was readily accessible
in almost quantitative yield from the commercially
available 4-bromobutanoic acid ethyl ester 23 through
a two-step sequence: the formation of the corresponding
azido ester 24 by the substitution of bromide using so-
dium azide and hydrolysis of the ethyl ester with aque-
ous potassium hydroxide. Either the reduction of the
nitro or the hydrogenation of azido function or the
cleavage of the benzyloxycarbonyl (Cbz) protecting
group in compounds 18a–c by hydrogen provided the
corresponding amino derivatives 19a–c. These interme-
diates were then directly N-acylated with chlorambucil
activated by DCC and HOBt or by ethylchloroformate
and triethylamine, leading to enantiomerically pure
compounds 20a, 21a, and 22a. Final deacetylation by
The reaction of 8, 9, and 13 with chlorambucil, acti-
vated by the N,N0-dicyclohexylcarbodiimide (DCC)
procedure, using either 4-(N,N-dimethylamino)pyridine
(DMAP) or hydroxybenzotriazole (HOBt), gave the
corresponding esters 14 and amide 15. Using an ester
linkage between the sugar and chlorambucil, the hy-
droxyl compound 14c cannot be obtained from the
peracetylated one 14a, the linkage being cleaved during
`
the Zemplen deacetylation. Accordingly, 14c was pre-
pared from the benzyl derivative 14b after hydrogenol-
ysis (H2, Pd/C, quantitative). The acetylated compound
17a was obtained from chlorambucil in two steps
including the formation of the isocyanate 16 and its
reaction with b-amino-sugar 8. To synthesize isocya-
nate 16, sodium azide, ethylchloroformate and triethyl-
amine were first added to chlorambucil and gave the
corresponding acyl azide. Subsequent heating pro-
duced, after Curtius rearrangement, isocyanate 16 in
81% yield. The removal of the acetyl groups of glyco-
side 17a gave the expected compound 17b in only
45% yield. Derivatives 15 and 17 were enantiomerically
pure and only obtained as the b form. Compounds
14a, 14b, and 14c were isolated as an inseparable mix-
ture of the two anomers with a/b ratios of, respectively,
25:75, 85:15, and 55:45. The ratio was calculated from
`
the Zemplen method using sodium methoxide in metha-
nol afforded the fully deprotected species 20b, 21b, and
22b (enantiomerically pure as shown by NMR spectra)
in good yields. NMR and MS data agree with the pro-
posed structures.
1
their H NMR spectra: for example, H-1 of the a-ano-
mer 14c resonates as
(J1,2 = 3.9 Hz), whereas H-1 of the b-anomer shows
an extra coupling with the C-2 fluorine atom at
a
doublet at 6.41 ppm
Compounds 31 were to be obtained in a similar way as
for derivatives 20, 21, and 22, but the synthetic sequence
required the preparation of 3-azidopropionic acid. De-

5008
B. Reux et al. / Bioorg. Med. Chem. 16 (2008) 5004–5020
OH
OH
O
HO
HO
OH
3
(i), (ii), (iii)
23 %
OAc
OH
O
AcO
AcO
OAc
4
(iv)
79%
OAc
O
OAc
O
OAc
O
(ix), (x)
87 %
(v)
AcO
AcO
AcO
AcO
AcO
AcO
OAc
SEt
78 %
F
5
F
F
10
Br
6
(viii)
(xi), (xii)
(vi)
84 %
61 %
100 %
OAc
O
OBn
O
OAc
AcO
AcO
O
AcO
AcO
BnO
BnO
OH
N3
SEt
F
F
F
9
7
11
(vii)
98 %
(xiii)
72 %
OBn
O
OAc
O
BnO
BnO
AcO
AcO
NH2
F
F
Br
8
12
(xiv)
61 %
OBn
O
BnO
BnO
OH
F
13
Scheme 1. Reagents and conditions: (i) Ac2O, HClO4; (ii) PBr3, H2O; (iii) AcONa, H2O; (iv) DAST, dioxane; (v) HBr/AcOH, CHCl3; (vi) NaN3,
acetone/H2O; (vii) H2, Pd/C, THF/MeOH; (viii) (NH4)2CO3, DMF; (ix) EtSH, NaOMe; (x) Ac2O, AcONa; (xi) NaOMe, MeOH; (xii) BnBr, NaH,
KI; (xiii) Br2, Et2O; (xiv) Ag2CO3, H2O/acetone.
spite repeated attempts, this compound could not be ob-
tained pure and so another reaction pathway (Scheme 4)
was used, with preliminary coupling of chlorambucil
with the spacer before the N-glycosylation, in contrast
to the other pathway.
chlorambucil anchored to the spacer through an ester
bond, or to an O-glycoside 37 with chlorambucil an-
chored through an amide bond. Reaction of chlorambu-
cil with 4-nitrophenol, using DCC and DMAP as
condensing reagents, gave the ester 32 in good yield.
Hydrogenation of 32 catalyzed by Pd/C afforded the
reductive analogue 33. The protected b-glucopyranose
group was introduced under Koenigs–Knorr conditions
by N-glycosylation of the aniline 33 with 3,4,6-tri-ben-
zyl-2-deoxy-2-fluoroglucopyranose bromide 12 in the
presence of silver oxide as catalyst. The stereochemistry
of 34a was assigned based on the multiplicity of H-2 ob-
served in 1H NMR: H-2 resonates at 4.41 ppm as a dou-
Starting from 3-bromopropionic acid ethyl ester 26, the
substitution of bromide using sodium azide followed by
hydrogenolysis in the presence of Pd/C provided 3-ami-
nopropionic acid ethyl ester 28. N-Acylation of the lat-
ter with chlorambucil (C1COOEt, Et3N, CH2C12) gave
the amide 29. After hydrolysis to the carboxylic acid 30,
the N-glycoside 31a was prepared from the amino-sugar
8 under N-acylation conditions. Final deacetylation
gave the target compound 31b in 90% yield.
blet of triplets (J2,1 ꢀ J2,3 = Jax–ax = 7.5–8.5 Hz, J2,F
=
51.0 Hz). Hydrogenolysis of 34a (H2, Pd/C, EtOH) then
gave 34b.
We also prepared compounds 34 and 37 in which 4-ami-
nophenol was used as a spacer between the sugar moiety
and chlorambucil (Schemes 5 and 6). The introduction
of this moiety can lead to an N-glycoside 34 with
The glycosylation to 35 was conducted under Koenigs–
Knorr conditions by the reaction of bromide 6 with 4-
nitrophenol in the presence of silver oxide in acetoni-

B. Reux et al. / Bioorg. Med. Chem. 16 (2008) 5004–5020
5009
OR
O
OR
(i)
O
RO
RO
RO
OH
O
RO
F
F
O
Cl
9
: R = OAc
N
13 : R = OBn
14a : R = OAc ; 57 %
14b : R = OBn ; 98 %
14c : R = OH ; 35 %
(ii)
Cl
OAc
OAc
OH
O
(iii)
(iv)
58 %
O
O
AcO
AcO
AcO
HO
HO
N
NH2
N
AcO
51 %
H
H
F
F
F
O
Cl
O
Cl
8
N
N
15a
15b
Cl
Cl
Cl
Cl
N
N
Cl
O
Cl
(v), (vi)
81 %
O C
Cl
N
HO
2
16
Cl
N
N
OAc
OH
O
(viii)
45 %
(vii)
50 %
Cl
Cl
O
AcO
HO
N
N
N
N
AcO
HO
H
H
H
H
F
F
O
O
17a
17b
Scheme 2. Reagents and conditions: (i) chlorambucil, DCC, DMAP, CH2Cl2; (ii) HCOONH4, Pd/C, THF/EtOH; (iii) chlorambucil, DCC, HOBt,
DMF; (iv) NaOMe, MeOH; (v) ClCOOEt, Et3N, NaN3, acetone; (vi) toluene, reflux; (vii) 8, CH2Cl2/THF; (viii) NaOMe, MeOH.
OAc
O
OAc
O
OAc
O
(i)
(ii)
AcO
AcO
AcO
AcO
AcO
AcO
NH2
N
N
R
[X]
H
H
NH2
F
F
F
O
O
8
18a : R = 4-NO2C6H4 ; 67%
18b : R = CH2NHCbz ; 69 %
18c : R = (CH2)3N3 ; 86 %
19a : X = C6H4 ; 91 %
19b : X = CH2 ; 92 %
19c : X = (CH2)3 ; 96 %
Cl
N
OR
O
Cl
O
(iii)
RO
N
[X]
RO
H
N
H
F
O
R = OAc
R = OH
20a : X = C6H4 ; 80 %
21a : X = CH2 ; 64 %
22a : X = (CH2)3 ; 24 %
20b : X = C6H4 ; 74 %
21b : X = CH2 ; 98 %
22b : X = (CH2)3 ; 72 %
(iv)
O
O
O
(v)
(vi)
N3
Br
N3
EtO
EtO
HO
90 %
99 %
25
23
24
Scheme 3. Reagents and conditions: (i) 18a: 4-nitrobenzoyl chloride, Et3N, CH2Cl2; 18b: N-Cbz-glycine, DCC, HOBt, DMF; 18c, 25: ClCOOEt,
Et3N, CH2Cl2; (ii) H2, Pd/C; (iii) 20a, 22a: chlorambucil, ClCOOEt, Et3N, CH2Cl2; 21a: chlorambucil, DCC, HOBt, DMF; (iv) NaOMe, MeOH; (v)
NaN3, acetone/H2O, reflux; (vi) KOH, MeOH/H2O.
trile. After stirring for one day at room temperature, the
silver salts were filtered off and chromatography on sil-
ica gel afforded b-anomer 35 exclusively, in good yield.
The stereochemistry of 35, similar to that of 34a, was as-
signed based on its H NMR signature and particularly
on the multiplicity at and near the anomeric center: H-1
1

5010
B. Reux et al. / Bioorg. Med. Chem. 16 (2008) 5004–5020
O
O
O
(i)
(ii)
EtO
Br
EtO
N3
EtO
NH2
47 %
81 %
26
27
28
Cl
Cl
N
N
O
O
Cl
O
O
Cl
(iv)
100 %
(iii)
HO
N
EtO
N
H
47 %
H
30
29
OAc
O
(v)
H
AcO
AcO
N
N
88 %
H
F
O
O
Cl
N
31a
Cl
OH
O
(vi)
H
HO
HO
N
H
N
83 %
F
O
O
Cl
N
31b
Cl
Scheme 4. Reagents and conditions: (i) NaN3, MeOH/H2O; (ii) H2, Pd/C, MeOH/THF; (iii) chlorambucil, ClCOOEt, Et3N, CH2Cl2; (iv) LiOH,
EtOH/H2O; (v) 8, ClCOOEt, Et3N, CH2Cl2; (vi) NaOMe, MeOH.
Cl
Cl
N
O2N
N
(i)
O
Cl
O
Cl
91 %
HO
O
2
32
Cl
Cl
OBn
O
BnO
BnO
H2N
N
N
N
(ii)
(iii)
O
Cl
O
Cl
H
F
79 %
35 %
O
O
33
34a
Cl
OH
O
(iv)
HO
HO
N
N
79 %
O
Cl
H
F
O
34b
Scheme 5. Reagents and conditions: (i) 4-nitrophenol, DCC, DMAP, CH2Cl2; (ii) H2, Pd/C, EtOH; (iii) 12, Ag2O, CH3CN; (iv) H2, Pd/C, EtOH/
THF.
resonates at 5.30 ppm as a doublet of doublets
(J1,2 = Jax–ax = 7.6 Hz, J1,F = 3.1 Hz) and H-2 resonates
at 4.63 ppm as a doublet of doublets of doublets
(J2,3 = Jax–ax = 9.2 Hz, J2,F = 50.4 Hz). Reduction of
the nitro group was achieved by hydrogenation to give
derivative 36. Thus, the compound was used as a reagent
for the two different syntheses leading to derivatives 37
and 40.
N-Acylation of amine 36 with activated chlorambucil
(DCC-HOBt/DMF) gave compound 37a. Deacetylation
then gave the expected compound 37b in quantitative
yield.
Finally, we describe the synthesis of glycoside 40 for
which a glycine moiety was added next to the aminophe-
nol to provide the spacer domain. Starting from aniline

B. Reux et al. / Bioorg. Med. Chem. 16 (2008) 5004–5020
5011
OAc
O
OAc
O
OAc
O
(i)
(ii)
AcO
AcO
AcO
AcO
AcO
AcO
O
O
77 %
100 %
F
F
F
Br
NO2
NH2
6
35
36
Cl
Cl
OAc
OH
O
O
AcO
AcO
HO
HO
O
N
O
N
(iv)
99 %
O
Cl
O
Cl
F
F
(iii)
N
N
H
H
79 %
37a
37b
36
74 %
(v)
OAc
O
OAc
O
AcO
AcO
AcO
AcO
(vi)
O
O
O
O
F
F
98 %
NHCbz
NH2
N
H
N
H
38
39
(vii)
O
82 %
OAc
O
AcO
AcO
O
F
H
N
N
H
O
Cl
N
40a
Cl
(viii)
O
99 %
OH
O
HO
HO
O
F
H
N
N
H
O
Cl
N
40b
Cl
Scheme 6. Reagents and conditions: (i) 4-nitrophenol, Ag2O, CH3CN; (ii) H2, Pd/C, EtOH/THF; (iii) chlorambucil, DCC, HOBt, DMF; (iv)
NaOMe, MeOH; (v) N-Cbz-glycine, DCC, HOBt, DMF; (vi) HCOONH4, Pd/C, EtOH/THF; (vii) chlorambucil, DCC, HOBt, DMF; (viii) NaOMe,
MeOH.
36, DCC/HOBt coupling to N-Cbz-glycine preceded
hydrogenolytic removal of the benzyl carbamate to give
39. N-Acylation with chlorambucil afforded the acetyl
compounds and chlorambucil are summarized in Table
1. IC50 values revealed that the peracetylated com-
pounds (15a, 17a, 20a, 21a, 22a, 31a, 37a, and 40a) were
more potent than chlorambucil whatever the cell line
considered. Most often the peracetylated compounds
showed a low IC50 value of 15 lM, whereas CLB per
se is only moderately cytotoxic in these conditions
(IC50 values above 40 lM except in PA1 cells, a very
sensitive cell line). Corresponding unprotected deriva-
tives showed a drop in activity (results not shown). This
observation had already been made on a series of hexose
keto-C-glycoside conjugates55 which prevent the uptake
of [14C]-glucose by human GLUT-1 glucose transporter,
and the authors connect this effect to lipophilicity. Con-
sidering compounds 15a and 17a varying only by one
connection, an amide or an urea function linking di-
rectly the FDG moiety and CLB, no difference on cyto-
`
derivative 40a and Zemplen deprotection gave the ex-
pected compound 40b. NMR and mass spectrometry
confirmed the structure and purity of the pure enantio-
mers 37 and 40.
2.2. Biological activity
2.2.1. In vitro cytotoxicity. The cytotoxic activities of the
compounds were evaluated in vitro against the six hu-
man tumor cell lines; breast adenocarcinoma MCF-7,
colon adenocarcinoma DLD-1, ovary adenocarcinoma
PA1, prostatic adenocarcinoma PC3, lung carcinoma
A549, melanoma M4Beu and a primary culture of hu-
man fibroblasts. The cytotoxicity results for some target

5012
B. Reux et al. / Bioorg. Med. Chem. 16 (2008) 5004–5020
Table 1. Cytotoxicity induced by the new synthesized compounds on human normal and tumor cells
Compound
IC50 (lM)
A549
Fibroblasts
PC3
MCF7
27
26.4 0.6
DLD1
M4Beu
PA1
1.3 0.2
15a
17a
20a
21a
22a
31a
37a
40a
CLB
11
13
4
4
2
2
4.4 0.9
7
5.4 0.4
7.1 0.9
27
27 11
9
4 1
1.4 0.3
8 2
2.7 0.6
7
3
3
2
3
2
4
5
2
0.6 0.2
0.3 0.1
1.0 0.1
5
8
2
2
1
5
7
6.8 0.9
3
4.1 0.1
4.1 0.1
10
3
2
1
5
13
2
5
4
12
13
9
9
15
3
3
0.27 0.04
0.24 0.08
0.76 0.06
1.6 0.5
7
>50
16
5
3.7 0.3
9.2 0.5
1.90 0.01
11
10
6
11
8
4
3
26 11
5
>50
14
5
2
14
>100
10
43 14
4
5
99
nd
47 17
>100
2.8 0.4
toxicity was observed. Substances possessing two amidic
functions correspond to the most cytotoxic compounds
namely 20a, 21a, 22a, 31a, and 40a with IC50 values
ranging from 0.24 to 16 lM. These data obtained for
compounds 20a, 21a, 22a, and 31a compared to com-
pounds 15 and 17 tend to show the necessity of a spacer
arm between CLB and the sugar carrier. In the set of
substances including compounds 20a, 21a, 22a, and
31a the presence of an aromatic spacer seems to favor
the global cytotoxicity slightly (20a compared with
21a, 22a, and 31a), whereas the length of the alkyl chain
of the spacer does not seem to be the deciding factor,
compounds possessing a linker with one, two, or three
carbons (respectively, 21a, 22a, and 31a) having compa-
rable cytotoxicities. Concerning the cell line, no real sen-
sitivity or resistance was evident for this set of
compounds which have almost the same cytotoxicity
on all cell lines used. Some differences in the pattern of
cytotoxicity were observed with compounds 15a, 17a,
and 37a, IC50 values are on the order of 27 lM on
MCF7 and M4Beu lines and generally smaller on the
other lines (except for compound 37a). This pattern is
comparable to CLB with IC50 values upper 100 lM on
MCF7 and M4Beu lines. Thereby, the use of amidic
bonds via the spacer linkage between the carrier and
CLB modified the cytotoxic pattern of the drug by eras-
ing the differences of sensitivity and improving activity
on all cell lines. For compound 37a, a relative resistance
of normal diploid human fibroblasts can be observed
(IC50 > 50 lM) and could be a useful property for a dif-
ferential effect between tumor and these healthy cells.
Such a resistance is also observed for compound 40a
but to a lesser extent. Unlike the other compounds,
the two last ones are O-glycosides. However, this resis-
tance of fibroblasts could not be considered as a discrim-
inating criterion or a proof of selectivity between
malignant and healthy cells since fibroblasts are not sur-
rogates of all normal cells and these cytotoxic activities
are obtained in vitro.
by the upregulation of glucose transporters, resulting in
accelerated influx in conjunction with higher intracellu-
lar levels and efficacy of hexokinase bound to tumor
mitochondria, leading to trapping and lower intracellu-
lar levels of phosphatase responsible for intracellular
retention.57,58 According to this pathway, the FDG–
CLB conjugate may be trapped inside the tumor cells
and so be more potent than CLB. Further developments
in vivo will, therefore, include the assessment of acute
toxicity and antitumor activity in tumor-bearing mice,
which is currently made and the demonstration of the
affinity of the conjugates for the neoplastic tissues. To
perform pharmacokinetic studies and biodistribution
in mice in order to prove the preferential targeting to
the tumors, the compound must be 14C radiolabeled,
and this radiolabeling is currently in progress.
2.2.2. Cellular stress. Most of the alkylating agents, and
nitrogen mustards in particular, are known to induce re-
duced-glutathione (GSH) depletion and ROS produc-
tion in cells, at least in vitro.59–62 This resulting
oxidative stress is implicated in toxic action on normal
cells, especially on respiratory cells61 or hepatocytes59
and could be partly responsible for adaptation and resis-
tance of tumor cells to DNA-crosslinking agents.62 We
therefore selected our drug candidates taking into con-
sideration their oxidative stress potency, which should
be minimal. To this end, we investigated both parame-
ters in the clonal murine L929 fibroblastic cell line after
a 4 h treatment with the cytotoxic compounds in the ser-
ies, including CLB (Table 2). In parallel, we carried out
the calcein-AM assay to ascertain a specific effect on
GSH and not an indirect drop due to cytolysis or plas-
mic membrane impairment. In accordance with previous
results reported in the literature,62 we observed a dra-
matic fall in GSH cellular content concomitant with
ROS production after CLB treatment from 6.25 lM, a
non-cytolytic concentration. In the set of new com-
pounds tested here, a similar decrease in GSH at low
concentration was observed only with 37a. In this last
case, in contrast to the CLB situation, no ROS produc-
tion was highlighted. 31a and 20a also showed a signif-
icant decrease, but at higher concentration (50 lM) and
concomitant to a cytolysis effect with the latter. Surpris-
ingly, 22a, and to a lesser extent 21a and 17a, induced a
seemingly direct production of ROS in cells. Investiga-
tion of the mechanism involved in these cellular re-
sponses lies outside the scope of this work, but a
All peracetylated compounds were shown to be much
more cytotoxic than CLB in vitro. This improvement
in potency could be correlated with a retention of the
conjugate in the tumor cells through a trapping mecha-
nism similar to that of FDG. The accumulation of FDG
in cancer cells56 is well exploited in PET technology with
18F-FDG and is assigned to metabolic trapping. This
preferential assimilation of FDG seems to be controlled

B. Reux et al. / Bioorg. Med. Chem. 16 (2008) 5004–5020
5013
Table 2. Cellular modifications induced by active new synthesized compounds on L929 murine cells
Compound
Growth inhibition at 48 h
ROS production1
GSH depletion2
Cytolysis3
17a
20a
21a
22a
31a
37a
40a
CLB
40%
50%
70%
50%
50%
ND
40%
30%
+++
—
—
—
+
+
—
++
—
—
—
—
—
—
++++
—
—
—
+
++++
—
++++
—
++++
Each compound was tested at four concentrations (6.25, 12.5, 25, 50 lM) for 4 h and cellular response was then assayed as described in Materials and
1
2
3
Methods. Positive response is defined as dichlorofluorescein production P 150% and GS-bimane or calcein production 6 80%. These results are
representative of two independent experiments. ++++:positive response at 6.25 lM, +++ at 12.5 lM, ++ at 25 lM, + at 50 lM.
specific mechanism seems to be involved, since 21a, 22a,
and 31a, which differ only in the length of the carbon
spacer, had different patterns. Finally, according to their
cytotoxicity on normal or tumor cells and to their mod-
erate ability or inability to induce oxidative stress, we
chose 20a, 21a, 31a, and 40a to investigate their pharma-
codynamic properties in an animal context.
with vanillin in sulfuric acid. Mass spectra were re-
corded on a Bruker Esquire-LC spectrometer. Electro-
spray ionization mass spectrometry (ESI-MS) was used
in positive mode. Microanalysis was performed by the
Central Analysis Service (CNRS, Vernaison, France)
for C, H, N, and the results were within 0.4% of the
theoretical values, unless otherwise stated.
To summarize, the synthesis of new carbohydrate-cou-
pled CLB derivatives with FDG as the sugar-moiety is
3.1.2.
3,4,6-Tri-O-acetyl-1-[4-{4-[bis(2-chloroeth-
yl)amino]phenyl}butyrate]-2-deoxy-2-fluoro-a,b-D-gluco-
pyranose (14a). To a solution of 3,4,6-tri-O-acetyl-2-
reported. These studies demonstrate
a significant
improvement in the in vitro cytotoxicity of peracetylated
glucoconjugates compared with the free substance. On
the basis of these promising data, the next step will be
to evaluate the acute toxicity and in vivo antitumor
activity of the selected compounds on animals bearing
tumors and to investigate in vivo the targeting of these
new agents to tumor tissue.
deoxy-2-fluoro-a,b-D-glucopyranose
9
(330 mg,
1.07 mmol) in CH2Cl2 (50 mL), chlorambucil (490 mg,
1.61 mmol), DCC (376 mg, 1.78 mmol), and DMAP
(8.7 mg, 0.071 mmol) were added and stirred at room
temperature for 22 h. The solid was removed by filtra-
tion, and the filtrate was washed with a 1 N aqueous
solution of acetic acid and water, dried over MgSO4,
and concentrated under vacuum. The resulting residue
was purified by silica gel chromatography, using cyclo-
hexane/ethyl acetate (7:3) as eluent, to yield 14a
(360 mg, 57%) as a syrup (a/b ratio = 25:75): IR (CCl4
solution) m 1763, 1366, 1240, 1223, 1075, 1038; 1H
NMR (200 MHz, CDCl3) d 7.12 (d, 2H, Jo = 8.7 Hz,
HAra), 7.10 (d, 2H, Jo = 8.7 Hz, HArb), 6.69 (d, 2H,
HAra), 6.67 (d, 2H, HArb), 6.50 (d, 1H, J1,2 = 3.9 Hz,
H-1a), 5.85 (dd, 1H, J1,2 = 8.1 Hz, J1,F = 3.1 Hz, H-
1b), 5.60 (td, 1H, J2,3 = J3,4 = 9.6 Hz, J3,F = 12.1 Hz,
H-3a), 5.43 (td, 1H, J2,3 = J3,4 = 9.2 Hz, J3,F = 14.2 Hz,
H-3b), 5.14 (t, 1H, J4,5 = 9.8 Hz, H-4a), 5.12 (t, 1H,
J4,5 = 9.7 Hz, H-4b), 4.70 (ddd, 1H, J2,F = 48.6 Hz, H-
2a), 4.49 (td, 1H, J2,F = 51.1 Hz, H-2b), 4.40–4.30 (m,
2H, H-6aa, H-6ab), 4.16–4.04 (m, 3H, H-6ba, H-6bb,
H-5a), 3.90 (ddd, 1H, H-5b), 3.75–3.64 (m, 16H,
N(CH2CH2Cl)2ab), 2.62 (m, 4H, CH2Ph ab), 2.48 (m,
4H, COCH2ab), 2.10–1.94 (m, 22H, OAc,
CH2CH2CH2ab); 13C NMR (50 MHz, CDCl3)d
171.55, 171.38, 170.59, 170.18, 169.94, 169.61, 169.54
(CO ab); 144.63, 144.59 (CArNab), 130.31, 130.16
(CArCH2ab), 129.91, 129.87, 112.35 (CHArab), 91.30
(C-1b, J1,F = 24.1 Hz), 88.39 (C-2b, J2,F = 190.9 Hz),
3. Experimental
3.1. Chemistry
3.1.1. General. All solvents and reagents obtained from
commercial sources were used without further purifica-
tion. Nuclear magnetic resonance spectra (1H NMR
and 13C NMR) were performed on a Bruker AM 200
(4.5 T) spectrometer. Chemical shifts are reported in
parts per million relative to the internal tetramethylsil-
ane standard for 1H NMR and the solvent for 13C
NMR (acetone-d, d 29.8 ppm; DMSO-d6, d 39.5 ppm;
CDCl3, d = 77.2 ppm) The abbreviations used for signal
patterns are: br, broad; s, singlet; d, doublet; dd, doublet
of doublets; ddd, doublet of doublet of doublets; t, trip-
let; dt, doublet of triplets; q, quadruplet; qt, quintuplet;
m, multiplet. Coupling constants (J values) are in Hertz.
Infrared (IR) spectra were recorded on an FTIR-Nicolet
Impact 410 spectrophotometer. Melting points were
determined on an electrothermal digital apparatus
(Reichert) and are uncorrected. Medium-pressure col-
umn chromatography was carried out on silica gel 60
(Chromagel, 35–70 lm, SDS) using the indicated solvent
mixture expressed as volume/volume ratios. Analytical
thin-layer chromatography (TLC) was conducted on
precoated silica gel plates (SDS, 60 F254, 0.2 mm thick)
with detection by ultraviolet light and (or) visualization
88.31 (C-1a, J1,F = 22.2 Hz), 86.40 (C-2a, J2,F
=
193.7 Hz), 72.90 (C-5b), 72.87 (C-3b, J3,F = 19.4 Hz),
70.73 (C-3a, J3,F = 19.4 Hz), 69.71 (C-5a), 67.76
(C-4a/b, J4,F = 7.2 Hz), 65.47 (C-4a/b, J4,F = 7.6 Hz),
61.46 (C-6ab), 53.75 (CH2Nab), 40.57 (CH2Clab),
33.82, 33.31 (CH2Phab, COCH2ab), 26.54, 26.46

5014
B. Reux et al. / Bioorg. Med. Chem. 16 (2008) 5004–5020
(CH2CH2CH2ab), 20.78, 20.72, 20.64 (CH3ab); MS (ESI)
m/z 594.24 [M+1]+; Anal. (C26H34Cl2FNO9) C, H, N.
(CH2Cl), 35.68 (COCH2), 33.89 (CH2Ph), 26.77
(CH2CH2CH2), 20.82, 20.76, 20.70 (CH3); MS (ESI)
m/z 593.30 [M+1]+; Anal. (C26H35Cl2FN2O8) C, H, N.
3.1.3. 1-[4-{4-[Bis(2-chloroethyl)amino]phenyl}butyrate]-
2-deoxy-2-fluoro-a,b-D-glucopyranose (14c). To a sus-
pension of 10% palladium on charcoal (1.5 g) in THF/
EtOH (12:24 mL) was added the benzylated glycoside
14 b (1.78 g, 2.41 mmol), followed by ammonium for-
mate (850 mg, 13.5 mmol). The mixture was refluxed
for 3.5 h and then allowed to cool to room temperature,
filtered on Celite, and concentrated under vacuum. The
crude product was purified by silica gel chromatogra-
phy, eluting with cyclohexane/ethyl acetate (1:9), to
yield the debenzylated product 14c (400 mg, 35%) as a
brown oil (a/b ratio = 55:45): IR(NaCl) m 3418, 1738,
3.1.5. 4-{4-[Bis(2-chloroethyl)amino]phenyl}-N-(2-deoxy-
2-fluoro-b-D-glucopyranosyl) butanamide (15b). To a stir-
red solution of acetylated compound 15a (140 mg,
0.24 mmol) in methanol (7 mL) was added sodium
methoxide (1 mg). After 4 h at room temperature, the
mixture was evaporated to dryness and purified on silica
gel (10:90 to 0:100 cyclohexane/ethyl acetate gradient) to
afford 15b (63 mg, 57%) as a solid: mp 168 ꢁC; IR (KBr)
1
m 3310, 1655, 1519, 1092, 1057; H NMR (200 MHz,
acetone-d6) d 7.81 (d, 1H, J = 9.1 Hz, NH), 7.06 (d,
2H, Jo = 8.8 Hz, HAr), 6.70 (d, 2H, HAr), 5.19 (td, 1H,
J1,2 = J1,NH = 9.2 Hz, J1,F = 2.3 Hz, H-1), 4.77 (d, 1H,
J = 4.4 Hz, OH), 4.39 (br s, 1H, OH), 4.06 (td, 1H,
J2,3 = 9.0 Hz, J2,F = 50.9 Hz, H-2), 3.80–3.58 (m, 12H,
N(CH2CH2Cl)2, 2H-6, H-3, H-4), 3.63–3.32 (m, 2H,
OH, H-5), 2.51 (t, 2H, J = 7.6 Hz, CH2Ph), 2.21 (t,
2H, J = 7.9 Hz, COCH2), 1.84 (qt, 2H, CH2CH2CH2);
13C NMR (50 MHz, acetone-d6) d 173.47 (NHCO),
145.55 (CArN), 131.43 (CArCH2), 130.36, 113.07 (CHAr),
92.42 (C-2, J2,F = 183.6 Hz), 79.10 (C-5), 78.06 (C-1,
J1,F = 23.2 Hz), 76.76 (C-3, J3,F = 16.8 Hz), 71.51 (C-4,
J4,F = 7.7 Hz), 62.47 (C-6), 53.89 (CH2N), 41.62
(CH2Cl), 36.02 (COCH2), 34.70 (CH2Ph), 28.07
(CH2CH2CH2); MS (ESI) m/z 467.15 [M+1]+; Anal.
(C20H29Cl2FN2O5) C, H, N.
1
1078 cmꢁ1; H NMR (200 MHz, acetone-d6)d 7.18 (d,
4H, Jo = 8.7 Hz, HArab), 6.81 (d, 4H, HArab), 6.41 (d,
1H, J1,2 = 3.9 Hz, H-1a), 5.85 (dd, 1H, J1,2 = 8.1 Hz,
J1,F = 3.1 Hz, H-1b), 5.02 (br s, 1H, OH), 4.90 (br s,
1H, OH), 4.71 (br s, 1H, OH), 4.53 (ddd, 1H,
J2,3 = 9.4 Hz, J2,F = 48.7 Hz, H-2a), 4.25 (td, 1H,
J2,3 = 8.5 Hz, J2,F = 52.0 Hz, H-2b), 4.19–3.49 (m,
26H, N(CH2CH2Cl)2ab, H-3ab, H-4ab, H-5 ab, 2H-
6ab), 2.65 (t, 4H, J = 7.6 Hz, CH2Phab), 2.55–2.41 (m,
4H, COCH2ab), 2.00–1.93 (m, 4H, CH2CH2CH2ab);
13C NMR (50 MHz, acetone-d6)d 171.86, 171.81
(COab), 145.25 (CArNab), 130.57 (CArCH2ab), 129.96,
112.70 (CHArab), 92.02 (C-2b, J2,F = 184.2 Hz), 92.00
(C-1b, J1,F = 21.7 Hz), 89.88 (C-2a, J2,F = 184.5 Hz),
89.47 (C-1a, J1,F = 22.6 Hz), 78.00 (C-5b), 75.43 (C-3b,
J3,F = 16.6 Hz),
75.18
(C-5a),
72.51
(C-3a,
3.1.6. N-(3,4,6-Tri-O-acetyl-2-deoxy-2-fluoro-b-D-gluco-
pyranosyl)-N0-{3-(4-[bis(2-chloroethyl)amino]phenyl)propyl}-
urea (17a). To aminosugar 8 (200 mg, 0.65 mmol) dis-
solved in CH2Cl2/THF (1.5:2 mL) was added isocyanate
16 (215 mg, 0.72 mmol) in CH2Cl2 (0.5 mL), and the
mixture was stirred overnight at room temperature.
After evaporation of the solvent, the concentrate was
purified by column chromatography on silica gel with
ethyl acetate/petroleum benzine (6:4 to 7:3 gradient) as
developing solvent to give compound 17a (200 mg,
50%) as a red powder: mp 91 ꢁC; IR (NaCl) m 1746,
J3,F = 16.8 Hz), 70.44 (C-4a/b, J4,F = 8.0 Hz), 70.16 (C-
4a/b, J4,F = 8.1 Hz), 61.62, 61.57 (C-6ab), 53.47
(CH2Nab), 41.20 (CH2Cl ab), 33.99, 33.92, 33.58,
33.37
(CH2Phab,
COCH2ab),
27.20,
27.05
(CH2CH2CH2ab); MS (ESI) m/z 468.54 [M+1]+; Anal.
(C20H28Cl2FNO6) C, H, N.
3.1.4. N-(3,4,6-Tri-O-acetyl-2-deoxy-2-fluoro-b-D-gluco-
pyranosyl)-4-{4-[bis(2-chloroethyl)amino]phenyl}butanamide
(15a). Aminosugar 8 (189 mg, 0.61 mmol) was dissolved
in anhydrous DMF (10 mL) and chlorambucil (187 mg,
0.61 mmol), DCC (127 mg, 0.61 mmol), and HOBt
(92 mg, 0.67 mmol) were added. The solution was stirred
for 20 h at room temperature, filtered, concentrated in
vacuo, and purified by silica gel chromatography (cyclo-
hexane/ethyl acetate 6:4) to give 15a (184 mg, 51%) as a
white solid: IR (KBr) m 3327, 1750, 1676, 1520, 1244,
1069, 1032; 1H NMR (200 MHz, CDCl3) d 7.08 (d,
2H, Jo = 8.6 Hz, HAr), 6.69 (d, 2H, HAr), 6.04 (d, 1H,
J = 9.3 Hz, NH), 5.44–5.28 (m, 2H, H-3, H-1), 5.03 (t,
1H, J3,4 = J4,5 = 9.7 Hz, H-4), 4.31 (dd, 1H,
J5,6a = 4.4 Hz, J6a,6b = 12.5 Hz, H-6a), 4.26 (td, 1H,
J2,F = 50.8 Hz, J1,2 = J2,3 = 9.2 Hz, H-2), 4.06 (dd, 1H,
J5,6b = 2.0 Hz, H-6b), 3.86 (ddd, 1H, H5), 3.76–3.58
(m, 8H, N(CH2CH2Cl)2), 2.58 (t, 2H, J = 7.3 Hz,
CH2Ph), 2.25 (t, 2H, J = 7.3 Hz, COCH2), 2.08, 2.07,
2.04 (each s, 3 · 3H, OAc), 1.94 (m, 2H, CH2CH2CH2);
13C NMR (50 MHz, CDCl3) d 173.24, 170.68, 169.92,
169.85 (CO), 144.13 (CArN), 131.00 (CArCH2), 129.88,
112.70 (CHAr), 88.50 (C-2, J2,F = 191.2 Hz), 77.62 (C-
1), 73.78 (C-5), 73.50 (C-3, J3,F = 19.3 Hz), 67.93 (C-4,
J4,F = 7.0 Hz), 61.65 (C-6), 53.89 (CH2N), 40.45
1
1364, 1230, 1029; H NMR (200 MHz, CDCl3) d 7.06
(d, 2H, Jo = 8.6 Hz, HAr), 6.63 (d, 2H, HAr), 5.46–5.23
(m, 2H, H-1, H-3), 5.01 (t, 1H, J3,4 = J4,5 = 9.7 Hz, H-
4), 4.84 (br t, 1H, J = 5.1 Hz, NHCH2), 4.33 (dd, 1H,
J6a,6b = 12.5 Hz, J5,6a = 4.3 Hz, H-6a), 4.20 (td, 1H,
J2,F = 50.5 Hz, J1,2 = J2,3 = 9.2 Hz, H-2), 4.06 (dd, 1H,
J5,6b = 2.1 Hz, H-6b), 3.82 (ddd, 1H, H-5), 3.65 (m,
8H, N(CH2CH2Cl)2), 3.23 (m, 2H, NHCH2), 2.55 (t,
2H, J = 7.4 Hz, CH2Ph), 2.09, 2.07, 2.05 (each s,
3 · 3H, OAc), 1.78 (qt, 2H, CH2CH2CH2); 13C NMR
(50 MHz, CDCl3) d 170.76, 170.03, 169.91 (COCH3),
156.84 (NHCO), 144.43 (CArN), 130.53 (CArCH2),
129.63, 112.29 (CHAr), 88.54 (C-2, J2,F = 190.6 Hz),
79.59 (C-1, J1,F = 22.2 Hz), 73.67 (C-3, J3,F = 19.5 Hz),
73.20 (C-5), 68.15 (C-4, J4,F = 7.1 Hz), 61.72 (C-6),
53.63 (CH2N), 40.61 (CH2Cl), 40.03 (NHCH2), 31.99,
31.87 (CH2Ph, CH2CH2CH2), 20.73 (CH3); MS (ESI)
m/z 608.19 [M+1]+; Anal. (C26H36Cl2FN3O8) C, H, N.
3.1.7. N-{3-(4-[Bis(2-chloroethyl)amino]phenyl)propyl}-
N0-(2-deoxy-2-fluoro-b-D-glucopyranosyl)urea (17b). To
a
stirred solution of acetylated compound 17a

B. Reux et al. / Bioorg. Med. Chem. 16 (2008) 5004–5020
5015
(200 mg, 0.33 mmol) in methanol (12 mL) was added so-
dium methoxide (2 mg), and the mixture was stirred at
room temperature for 3 h. After neutralization with
IRC 50 Amberlite ion-exchange resin (H+), filtration
and evaporation to dryness, the residue was purified
on silica gel (10% methanol in ethyl acetate) to yield
compound 17b (75 mg, 45%) as an oil: IR (KBr) m
decomposition at 230 ꢁC; IR (KBr) m 3301, 1658, 1519,
1
1089, 1045; H NMR (200 MHz, DSMO) d 10.12 (s,
1H, PhNHCO), 9.04 (d, 1H, J = 8.9 Hz, C1NH), 7.83
(d, 2H, J = 8.8 Hz, HAr), 7.67 (d, 2H, Jo, HAr), 7.02
(d, 2H, J0o ¼ 8:6 Hz, HAr), 6.64 (d, 2H, J0o, HAr), 5.54
(d, 1H, J = 5.4 Hz, OH), 5.19 (m, 2H, OH, H-1), 4.59
(t,
1H,
J = 5.6 Hz,
OH),
4.29
(td,
1H,
1
3348, 1750, 1078, 1028; H NMR (200 MHz, acetone-
J1,2 = J2,3 = 8.9 Hz, J2,F = 50.6 Hz, H-2), 3.70–3.05 (m,
13H, N(CH2CH2Cl)2, H-3, H-4, H-5, 2H-6), 2.48 (m,
2H, CH2Ph), 2.31 (t, 2H, J = 7.3 Hz, COCH2), 1.81
(qt, 2H, CH2CH2CH2); 13C NMR (50 MHz, DMSO-
d6) d 171.54, 166.02 (CO), 144.44 (CArN), 142.47
(CArNH), 129.72 (CArCH2), 129.32, 128.40 (CHAr),
127.56 (CArCO), 118.15, 111.91 (CHAr), 91.10 (C-2,
d6) d 7.09 (d, 2H, Jo = 8.8 Hz, HAr), 6.72 (d, 2H, HAr),
6.34 (d, 1H, J = 9.5 Hz, C1NHCO), 5.80 (t, 1H,
J = 5.8 Hz, NHCH2), 5.09 (td, 1H, J1,2 = 9.3 Hz,
J1,F = 2.0 Hz, H-1), 4.80 (d, 1H, J = 4.4 Hz, OH), 4.43
(d, 1H, J = 4.4 Hz, OH), 4.02 (td, 1H, J2,F = 50.9 Hz,
J2,3 = 8.9 Hz, H-2), 3.84–3.67 (m, 12H, N(CH2CH2Cl)2,
H-3, H-4, 2H-6), 3.40–3.36 (m, 2H, H-5, OH), 3.16 (t,
2H, J = 6.4 Hz, NHCH2), 2.53 (t, 2H, J = 7.7 Hz,
CH2Ph), 1.74 (qt, 2H, CH2CH2CH2); 13C NMR
(50 MHz, acetone-d6) d 158.49 (NHCO), 145.38 (CArN),
131.37 (CArCH2), 130.23, 113.01 (CHAr), 92.30 (C-2,
J2,F = 195.0 Hz), 79.97 (C-1, J1,F = 25.0 Hz), 78.50 (C-
5), 76.73 (C-3, J3,F = 16.9 Hz), 71.39 (C-4,
J4,F = 7.2 Hz), 62.36 (C-6), 53.84 (CH2N), 41.62
(CH2Cl), 40.27 (NHCH2), 32.84, 32.55 (CH2Ph,
CH2CH2CH2); MS (ESI) m/z 482.16 [M+1]+; Anal.
(C20H30Cl2FN3O5) C, H, N.
J2,F = 182.2 Hz),
78.73
(C-5),
77.42
(C-1,
J1,F = 22.9 Hz), 75.18 (C-3, J3,F = 16.2 Hz), 69.85 (C-4,
J4,F = 7.6 Hz), 60.55 (C-6), 52.22 (CH2N), 41.14
(CH2Cl), 35.85 (COCH2), 33.54 (CH2Ph), 26.85
(CH2CH2CH2); MS (ESI) m/z 586.14 [M+1]+; Anal.
(C27H34Cl2FN3O6) C, H, N.
3.1.10.
N-[2-(3,4,6-Tri-O-acetyl-2-deoxy-2-fluoro-b-D-
glucopyranosylamino)-2-oxoethyl]-4-{4-[bis(2-chloroethyl)-
amino]phenyl} butanamide (21a). Compound 21a was
prepared according to the procedure described for com-
pound 15a, starting from amino derivative 19b (1.02 g,
2.80 mmol) and chlorambucil (939 mg, 3.09 mmol).
Purification on silica gel (4:6 to 2:8 petroleum benzine/
ethyl acetate gradient) gave compound 21a (945 mg,
64%) as a white solid: mp 106 ꢁC; IR (KBr) m 3368,
1750, 1653, 1519, 1230, 1069, 1034; 1H NMR
(200 MHz, CDCl3) d 8.08 (d, 1H, J = 9.1 Hz, C1NH),
7.06 (d, 2H, Jo = 8.5 Hz, HAr), 6.63 (d, 2H, HAr), 6.46
(br t, 1H, J = 4.6 Hz, CH2NH), 5.41–5.29 (m, 2H, H-
1, H-3), 5.03 (t, 1H, J3,4 = J4,5 = 9.7 Hz, H-4), 4.40 (td,
1H, J1,2 = J2,3 = 9.0 Hz, J2,F = 50.3 Hz, H-2), 4.26 (dd,
1H, J5,6a = 4.2 Hz, J6a,6b = 12.7 Hz, H-6a), 4.09–4.01
(m, 3H, H-6b, COCH2NH), 3.84 (ddd, 1H,
J5,6b = 1.7 Hz, H-5), 3.75–3.57 (m, 8H, N(CH2CH2Cl)2),
2.56 (t, 2H, J = 7.3 Hz, CH2Ph), 2.26 (t, 2H, J = 7.3 Hz,
COCH2), 2.08, 2.04, 1.99 (each s, 3 · 3H, OAc), 1.92 (qt,
2H, CH2CH2CH2); 13C NMR (50 MHz, CDCl3) d
173.98, 170.71, 170.01, 169.84, 169.74 (CO), 145.58
(CArN), 130.36 (CArCH2), 129.78, 112.36 (CHAr),
88.37 (C-2, J2,F = 191.0 Hz), 77.38 (C-1), 73.76–73.29
(C-3, C-5), 68.02 (C-4, J4,F = 6.7 Hz), 61.77 (C-6),
53.70 (CH2N), 43.68 (COCH2NH), 40.67 (CH2Cl),
3.1.8. N-(3,4,6-Tri-O-acetyl-2-deoxy-2-fluoro-b-D-gluco-
pyranosyl)-4-(4-{4-[bis(2-chloroethyl)amino]phenyl}butan-
amido)benzamide (20a). Compound 20a was prepared
according to the procedure described for compound
18c, starting from chlorambucil (217 mg, 0.72 mmol)
and amine 19a (277 mg, 0.65 mmol). The crude product
was purified by column chromatography using petro-
leum benzine/ethyl acetate (5:5 to 3:7) as the eluent, to
give compound 20a (370 mg, 80%) as a beige powder:
mp 165 ꢁC; IR (KBr) m 3325, 1749, 1670, 1520, 1365,
1
1244, 1066, 1036; H NMR (200 MHz, CDCl3) d 7.80
(d, 2H, Jo = 8.6 Hz, HAr), 7.66 (s, 1H, PhNHCO), 7.59
(d, 2H, Jo, HAr), 7.39 (d, 1H, J = 9.1 Hz, C1NH), 7.07
(d, 2H, J0o ¼ 8:6 Hz, HAr), 6.63 (d, 2H, J0o, HAr), 5.58
(br t, 1H, J1,2 = 9.0 Hz, H-1), 5.40 (td, 1H,
J2,3 = J3,4 = 9.2 Hz, J3,F = 13.8 Hz, H-3), 5.06 (t, 1H,
J4,5 = 9.8 Hz, H-4), 4.55 (td, 1H, J2,F = 50.3 Hz, H-2),
4.32 (dd, 1H, J5,6a = 4.0 Hz, J6a,6b = 12.4 Hz, H-6a),
4.06 (dd, 1H, J5,6b = 1.5 Hz, H-6b), 3.86 (ddd, 1H, H-
5), 3.73–3.55 (m, 8H, N(CH2CH2Cl)2), 2.61 (t, 2H,
J = 7.3 Hz, CH2Ph), 2.38 (t, 2H, J = 7.0 Hz, COCH2),
2.07–1.73 (m, 11H, CH2CH2CH2, 3OAc); 13C NMR
(50 MHz, MeOD) d 174.73, 172.27, 171.47, 171.34,
169.88 (CO), 145.99 (CArN), 143.91 (CArNH), 131.60
(CArCH2), 130.66, 129.66 (CHAr), 129.30 (CArCO),
120.23, 113.54 (CHAr), 89.46 (C-2, J2,F = 187.7 Hz),
79.24 (C-1, J1,F = 23.2 Hz), 74.95 (C-3, J3,F = 19.4 Hz),
74.60 (C-5), 69.55 (C-4, J4,F = 7.3 Hz), 63.03 (C-6),
54.53 (CH2N), 41.67 (CH2Cl), 37.41 (COCH2), 35.20
(CH2Ph), 28.50 (CH2CH2CH2), 20.58 (CH3); MS (ESI)
m/z 712.33 [M+1]+; Anal. (C33H40Cl2FN3O9) C, H, N.
35.50
(CH2CH2CH2), 20.79, 20.74 (CH3); MS (ESI) m/z
(COCH2CH2),
34.02
(CH2Ph),
27.19
650.26 [M+1]+; Anal. (C28H38Cl2FN3O9) C, H, N.
3.1.11.
4-{4-[Bis(2-chloroethyl)amino]phenyl}-N-[2-(2-
deoxy-2-fluoro-b-D-glucopyranosylamino)-2-oxoethyl]-
`
butanamide (21b). Zemplen deacetylation as for 17b and
chromatography on silica gel (10% methanol in ethyl
acetate) gave compound 21b (100 mg, 98%) as a white
powder: mp 110 ꢁC; IR (KBr) m 3315, 1647, 1519,
1
1249, 1081, 1039; H NMR (200 MHz, acetone–d6) d
3.1.9. 4-(4-{4-[Bis(2-chloroethyl)amino]phenyl}butanami-
do)-N-(2-deoxy-2-fluoro-b-D-glucopyranosyl)benzamide
`
(20b). Zemplen deacetylation as for 17b and chromatog-
raphy on silica gel (5% methanol in ethyl acetate) affor-
8.19 (d, 1H, J = 9.3 Hz, C1NH), 7.51 (t, 1H,
J = 5.5 Hz, CH2NH), 7.08 (d, 2H, Jo = 8.6 Hz, HAr),
6.70 (d, 2H, HAr), 5.21 (td, 1H, J1,2 = 9.1 Hz,
J1,F = 1.8 Hz, H-1), 4.99 (br d, 1H, J = 3.4 Hz, OH),
4.62 (br s, 1H, OH), 4.15 (td, 1H, J2,3 = 8.8 Hz,
ded compound 20b (105 mg, 74%) as a white powder:

5016
B. Reux et al. / Bioorg. Med. Chem. 16 (2008) 5004–5020
J2,F = 50.8 Hz, H-2), 4.09 (br s, 1H, OH), 3.95 (d, 2H,
COCH2NH), 3.79–3.66 (m, 11H, N(CH2CH2Cl)2, H-3,
2H-6), 3.42 (m, 2H, H-4, H-5), 2.53 (t, 2H, J = 7.5 Hz,
CH2Ph), 2.27 (t, 2H, J = 7.3 Hz, COCH2CH2), 1.86
(qt, 2H, CH2CH2CH2); 13C NMR (50 MHz, acetone-
d6) d 174.39, 171.08 (CO), 145.45 (CArN), 131.49
2CH2CH2CH2); 13C NMR (50 MHz, MeOD) d 176.30,
176.18 (NHCO), 145.98 (CArN), 131.73 (CArCH2),
130.60, 113.54 (CHAr), 92.27 (C-2, J2,F = 184.4 Hz),
79.75 (C-5), 78.59 (C-1, J1,F = 23.3 Hz), 76.93 (C-3,
J3,F = 16.9 Hz), 71.11 (C-4, J4,F = 7.6 Hz), 62.35 (C-6),
54.55 (CH2N), 41.71 (CH2Cl), 39.75 (CH2CH2CH2NH),
36.55 (CH2CH2CH2Ph), 35.23 (CH2CH2CH2Ph), 34.34
(CH2CH2CH2NH), 28.94 (CH2CH2CH2Ph), 26.34
(CH2CH2CH2NH); MS (ESI) m/z 552.14 [M+1]+; Anal.
(C24H36Cl2FN3O6) C, H, N.
(CArCH2), 130.36, 113.04 (CHAr), 92.13 (C-2, J2,F
=
183.9 Hz), 79.08 (C-5), 78.13 (C-1, J1,F = 23.3 Hz),
76.55 (C-3, J3,F = 16.7 Hz), 70.94 (C-4, J4,F = 4.0 Hz),
62.02 (C-6), 53.89 (CH2N), 43.31 (COCH2NH), 41.65
(CH2Cl), 35.87 (COCH2CH2), 34.76 (CH2Ph), 28.27
(CH2CH2CH2); MS (ESI) m/z 524.12 [M+1]+; Anal.
(C22H32Cl2FN3O6) C, H, N.
3.1.14.
N-[3-(3,4,6-Tri-O-acetyl-2-deoxy-2-fluoro-b-D-
glucopyranosylamino)-3-oxopropyl]-4-{4-[bis(2-chloroeth-
yl)amino]phenyl}butanamide (31a). Compound 31a was
prepared according to the procedure described for com-
pound 18c, starting from acid 30 (33 mg, 0.09 mmol)
and aminosugar 8 (24.5 mg, 0.08 mmol). The crude
product was purified by silica gel chromatography (2:8
to 1:9 petroleum benzine/ethyl acetate gradient), to give
expected compound 31a (47 mg, 88%) as a beige solid:
mp 103 ꢁC; IR (KBr) m 3309, 1751, 1676, 1647, 1519,
3.1.12.
N-[4-(3,4,6-Tri-O-acetyl-2-deoxy-2-fluoro-b-D-
glucopyranosylamino)-4-oxobutyl]-4-{4-[bis(2-chloroethyl)-
amino]phenyl}butanamide (22a). Compound 22a was
prepared according to the procedure described for com-
pound 18c, starting from chlorambucil (130 mg,
0.43 mmol) and amine 19c (168 mg, 0.43 mmol). The
crude product was purified by column chromatography
using ethyl acetate as developing solvent, to give com-
pound 22a (70 mg, 24%) as a beige powder: mp 95 ꢁC;
IR (NaCl) m 3295, 1749, 1647, 1542, 1366, 1230, 1069,
1034; 1H NMR (200 MHz, CDCl3) d 8.26 (d, 1H,
J = 9.0 Hz, C1NH), 7.07 (d, 2H, Jo = 8.6 Hz, HAr),
6.63 (d, 2H, HAr), 5.86 (br t, 1H, J = 6.0 Hz,
CH2NHCO), 5.45–5.34 (m, 2H, H-1, H-3), 5.06 (t, 1H,
1
1367, 1228, 1066, 1034; H NMR (200 MHz, CDCl3) d
7.90 (d, 1 H, J = 8.9 Hz, C1NH), 7.04 (d, 2H,
Jo = 8.5 Hz, HAr), 6.61 (d, 2H, HAr), 6.37 (br t, 1H,
J = 5.5 Hz, CH2NHCO), 5.46–5.30 (m, 2H, H-1, H-3),
5.03 (t, 1H, J3,4 = J4,5 = 9.8 Hz, H-4), 4.40 (td, 1H,
J1,2 = J2,3 = 9.1 Hz, J2,F = 50.4 Hz, H-2), 4.28 (dd, 1H,
J6a,6b = 12.5 Hz, J5,6a = 4.3 Hz, H-6a), 4.04 (dd, 1H,
J5,6b = 1.4 Hz, H-6b), 3.84 (m, 1H, H-5), 3.72–3.47 (m,
10H, N(CH2CH2Cl)2, CH2NH), 2.52 (m, 4H, CH2Ph,
COCH2CH2NH), 2.19–2.04 (m, 11H, COCH2CH2CH2,
3OAc), 1.87 (m, 2H, CH2CH2CH2); 13C NMR (50
MHz, MeOD) d 176.25, 174.45 (CONH), 172.19,
171.43, 171.27 (COCH3), 145.99 (CArN), 131.81
J3,4 = J4,5 = 9.8 Hz,
H-4),
4.47
(td,
1H,
J1,2 = J2,3 = 9.1 Hz, J2,F = 50.3 Hz, H-2), 4.29 (dd, 1H,
J6a,6b = 12.5 Hz, J5,6a = 4.3 Hz, H-6a), 4.08 (dd, 1H,
J5,6b = 2.1 Hz, H-6b), 3.83 (ddd, 1H, H-5), 3.76–3.58
(m, 8H, N(CH2CH2Cl)2), 3.43 (m, 1H, CHNH), 3.17
(m, 1H, CH’NH), 2.56 (t, 2H, J = 7.3 Hz, CH2Ph),
2.30–2.17 (m, 4H, 2CH2CO), 2.07, 2.05, 2.04 (each s,
3 · 3H, OAc), 2.00–1.76 (m, 4H, 2CH2CH2CH2); 13C
NMR (50 MHz, CDCl3) d 174.45, 173.62 (NHCO),
170.76, 170.04, 169.75 (COCH3), 144.52 (CArN),
130.48 (CArCH2), 129.78, 112.32 (CHAr), 88.42 (C-2, J
2,F = 189.8 Hz), 77.63 (C-1, J1,F = 23.2 Hz), 73.65 (C-
(CArCH2), 130.63, 113.57 (CHAr), 89.57 (C-2, J2,F
=
188.6 Hz), 78.51 (C-1, J1,F = 22.9 Hz), 74.81 (C-3,
J3,F = 19.5 Hz), 74.53 (C-5), 69.50 (C-4, J4,F = 7.3 Hz),
62.98 (C-6), 54.58 (CH2N), 41.72 (CH2Cl), 36.44,
35.15 (COCH2CH2NH, COCH2CH2CH2Ph), 28.89
(CH2CH2CH2), 20.58 (CH3); MS (ESI) m/z 664.33
[M+1]+; Anal. (C29H40Cl2FN3O9) C, H, N.
5), 73.64 (C-3, J3,F = 22.6 Hz), 68.06 (C-4, J4,F
=
6.7 Hz), 61.82 (C-6), 53.69 (CH2N), 40.65 (CH2Cl),
38.42 (CH2CH2CH2NH), 36.10 (CH2CH2CH2Ph), 34.19
(CH2CH2CH2Ph), 33.59 (CH2CH2CH2NH), 27.56
(CH2CH2CH2Ph), 26.42 (CH2CH2CH2NH), 20.86,
20.79, 20,71 (CH3); MS (ESI) m/z 678.29 [M+1]+; Anal.
(C30H42Cl2FN3O9) C, H, N.
3.1.15.
4-{4-[Bis(2-chloroethyl)amino]phenyl}-N-[3-(2-
deoxy-2-fluoro-b-D-glucopyranosylamino)-3-oxopropyl]-
butanamide (31b). Zemplen deacetylation as for 17b and
`
chromatography on silica gel (10% methanol in ethyl
acetate) gave compound 31b (80 mg, 83%) as a white so-
lid: mp 179 ꢁC; IR (KBr) m 3525, 3305, 1647, 1544, 1091,
1
3.1.13.
4-{4-[Bis(2-chloroethyl)amino]phenyl}-N-[4-(2-
1043; H NMR (200 MHz, acetone-d6) d 8.05 (d, 1H,
deoxy-2-fluoro-b-D-glucopyranosylamino)-4-oxobutyl]butan-
J = 9.1 Hz, C1NH), 7.06 (m, 3H, HAr, CH2NHCO),
6.72 (d, 2H, Jo = 8.8 Hz, HAr), 5.19 (td, 1H,
J1,2 = 9.1 Hz, J1,F = 2.3 Hz, H-1), 4.84 (d, 1H,
J = 4.5 Hz, OH), 4.48 (d, 1H, J = 4.5 Hz, OH), 4.09
(td, 1H, J2,3 = 9.0 Hz, J2,F = 50.9 Hz, H-2), 3.80–3.68
(m, 11H, N(CH2CH2Cl)2, H-3, 2H-6), 3.45–3.38 (m,
5H, H-4, H-5, CH2NH, OH), 2.47 (m, 4H, CH2Ph,
COCH2CH2NH), 2.15 (m, 2H, COCH2CH2CH2), 1.83
(m, 2H, CH2CH2CH2); 13C NMR (50 MHz, MeOD) d
176.19, 174.58 (NHCO), 145.92 (CArN), 131.73
`
amide (22b). Zemplen deacetylation as for 17b and chro-
matography on silica gel (10% methanol in ethyl acetate)
gave compound 22b (69 mg, 72%) as a solid: mp 161 ꢁC;
IR (KBr) m 3525, 3309, 1641, 1550, 1519, 1355, 1093,
1
1045; H NMR (200 MHz, acetone-d6) d 8.13 (d, 1H,
J = 9.2 Hz, C1NH), 7.18–7.05 (m, 3H, HAr, CH2NH),
6.72 (d, 2H, Jo = 8.7 Hz, HAr), 5.19 (td, 1H,
J1,2 = 9.0 Hz, J1,F = 2.2 Hz, H-1), 4.78 (br s, 1H, OH),
4.45 (br s, 1H, OH), 4.12 (td, 1H, J2,3 = 9.0 Hz, J2,F
=
50.9 Hz, H-2), 3.90–3.51 (m, 12H, N(CH2CH2Cl)2, H-
3, H-4, 2H-6), 3.40 (m, 2H, OH, H-5), 3.23 (q, 2H,
J = 6.4 Hz, CH2NH), 2.52 (t, 2H, J = 7.4 Hz, CH2Ph),
2.28–2.13 (m, 4H, 2CH2CO), 1.96–1.73 (m, 4H,
(CArCH2), 130.60, 113.49 (CHAr), 92.20 (C-2, J2,F =
184.8 Hz), 79.71 (C-5), 78.50 (C-1, J1,F = 23.1 Hz),
76.90 (C-3, J3,F = 16.8 Hz), 71.13 (C-4, J4,F = 7.5 Hz),
62.38 (C-6), 54.53 (CH2N), 41.69 (CH2Cl), 36.52,

B. Reux et al. / Bioorg. Med. Chem. 16 (2008) 5004–5020
5017
35.16 (COCH2CH2NH, COCH2CH2CH2Ph), 28.87
(CH2CH2CH2); MS (ESI) m/z 538.06 [M+1]+; Anal.
(C23H34Cl2FN3O6) C, H, N.
3, J3,F = 19.9 Hz), 72.10 (C-5), 68.14 (C-4,
J4,F = 7.3 Hz), 61.91 (C-6), 53.66 (CH2N), 40.63
(CH2Cl), 36.79 (COCH2), 34.03 (CH2Ph), 27.21
(CH2CH2CH2), 20.81, 20.79, 20.68 (CH3); MS (ESI)
m/z 685.30 [M+1]+; Anal. (C32H39Cl2FN2O9) C, H, N.
3.1.16. N-[4-(4-{4-[Bis(2-chloroethyl)amino]phenyl}butan-
oyloxy)phenyl]-2-deoxy-2-fluoro-b-D-glucopyranosylamine
(34b). A solution of the benzylated derivative 34a
(189 mg, 0.23 mmol) in EtOH/THF (5:2.5 mL) was stir-
red under hydrogen atmosphere in the presence of 10%
palladium on charcoal (47 mg) for 18 h at room temper-
ature. After filtration of the catalyst on Celite and evap-
oration of the solvent, the crude product was purified by
silica gel chromatography (cyclohexane/ethyl acetate
1:9) to yield the deprotected product 34b (100 mg,
78%) as a white powder (a/b ratio =20:80): IR (KBr) m
3349, 1750, 1519, 1201, 1179, 1135, 1078, 1012; 1H
NMR (200 MHz, acetone-d6) d 7.10 (d, 2H, Jo = 8.5 Hz,
HAr), 6.86 (d, 2H, Jo0 ¼ 9:1 Hz, HAr), 6.75 (2d, 4H, HAr),
6.00 (d, 1H, J1,NH = 8.4 Hz, NHb), 5.76 (d, 1H,
J1,NH = 4.8 Hz, NHa), 5.27 (br t, 1H, J1,2 = 5.2 Hz, H-
1a), 4.83 (br t, 1H, J1,2 = 8.5 Hz, H-1b), 4.75 (d, 1H,
J = 4.5 Hz, OHb), 4.61 (d, 1H, J = 4.4 Hz, OHa), 4.41
(d, 1H, J = 4.0 Hz, OHb), 4.38 (d, 1H, J = 5.0 Hz,
OHa), 4.13 (td, 1H, J2,3 = 8.7 Hz, J2,F = 51.8 Hz, H-
2b), 4.01 (m, 1H, H-3b), 3.82–3.38 (m, 13H,
N(CH2CH2Cl)2, H-4, H-5, 2H-6, OH), 2.60 (t, 2H,
J = 7.5 Hz, CH2Ph), 2.50 (t, 2H, J = 7.3 Hz, COCH2),
1.94 (qt, 2H, J = 7.4 Hz, CH2CH2CH2); 13C NMR (50
MHz, acetone-d6) 172.74 (CO), 145.65, 145.00 (2CArN),
143.91 (CArO), 130.98 (CArCH2), 130.41, 122.94, 114.84,
113.06 (CHAr), 92.84 (C-2, J2,F = 184.7 Hz), 83.67 (C-1,
J1,F = 20.2 Hz), 77.92 (C-5), 77.01 (C-3, J3,F = 16.9 Hz),
71.73 (C-4, J4,F = 7.7 Hz), 62.53 (C-6), 53.86 (NCH2),
41.59 (CH2Cl), 34.54, 33.93 (CH2Ph, COCH2), 27.71
(CH2CH2CH2); MS (ESI) m/z 559.10 [M+1]+; Anal.
(C26H33Cl2FN2O6) C, H, N.
3.1.18.
4-{4-[Bis(2-chloroethyl)amino]phenyl}-N-[4-(2-
deoxy-2-fluoro-b-D-glucopyranosyloxy)phenyl]butanamide
(37b). Zemplen deacetylation as for 17b and precipita-
tion in diisopropylic ether gave compound 37b (1.1 g,
98%) as a white solid: mp 170 ꢁC; IR (KBr) m 3293,
1667, 1521, 1221, 1073, 1049; H NMR (200 MHz, ace-
`
1
tone-d6) d 9.07 (s, 1H, NH), 7.61 (d, 2H, Jo = 9.0 Hz,
HAr), 7.10 (d, 2H, Jo0 ¼ 8:7 Hz, HAr), 7.03 (d, 2H, Jo,
HAr), 6.73 (d, 2H, J0o, HAr), 5.19 (dd, 1H,
J1,2 = 7.7 Hz, J1,F = 3.8 Hz, H-1), 4.88 (d, 1H,
J = 4.6 Hz, OH), 4.58 (d, 1H, J = 4.6 Hz, OH), 4.24
(td, 1H, J2,3 = 8.3 Hz, J2,F = 51.5 Hz, H-2), 3.89–3.56
(m, 14H, N(CH2CH2Cl)2, H-3, H-4, H-5, 2H-6, OH),
2.59 (t, 2H, J = 7.5 Hz, CH2Ph), 2.36 (t, 2H,
J = 7.4 Hz, COCH2), 1.96 (qt, 2H, CH2CH2CH2); 13C
NMR (50 MHz, acetone-d6) d 171.45 (NHCO), 153.93
(CArO), 145.53 (CArN), 135.49 (CArNH), 131.43
(CArCH2), 130.37, 121.26, 117.68, 113.07 (CHAr), 99.59
(C-1, J1,F = 23.4 Hz), 93.13 (C-2, J2,F = 184.9 Hz),
77.79 (C-5), 76.11 (C-3, J3,F = 17.1 Hz), 71.21 (C-4,
J4,F = 7.9 Hz), 62.31 (C-6), 53.95 (CH2N), 41.56
(CH2Cl), 36.93 (COCH2), 34.86 (CH2Ph), 28.19
(CH2CH2CH2); MS (ESI) m/z 559.30 [M+1]+; Anal.
(C26H33Cl2FN2O6) C, H, N.
3.1.19. N-{2-[4-(3,4,6-Tri-O-acetyl-2-deoxy-2-fluoro-b-D-
glucopyranosyloxy)phenyl]amino-2-oxoethyl}-4-{4-[bis(2-
chloroethyl)amino]phenyl}butanamide (40a). Compound
40a was prepared according to the procedure described
for compound 15a, starting from amino derivative 39
(1.5 g,
3.28 mmol)
and
chlorambucil
(1.10 g,
3.1.17.
N-[4-(3,4,6-Tri-O-acetyl-2-deoxy-2-fluoro-b-D-
3.62 mmol). Purification on silica gel (5:5 to 1:9 cyclo-
hexane/ethyl acetate gradient) gave compound 40a
(2 g, 82%) as a white solid: mp 130 ꢁC; IR(KBr) m
3307, 1755, 1650, 1509, 1367, 1247, 1068, 1044; 1H
NMR (200 MHz, CDCl3) d 9.29 (s, 1H, PhNH), 7.49
(d, 2H, Jo = 9.0 Hz, HAr), 7.05–6.96 (m, 4H, HAr), 6.86
(t, 1H, J = 4.6 Hz, CH2NH), 6.59 (d, 2H, J0o ¼ 8:7 Hz,
HAr), 5.41 (td, 1H, J2,3 = J3,4 = 9.2 Hz, J3,F = 14.5 Hz,
H-3), 5.14–5.05 (m, 2H, H-4, H-1), 4.56 (ddd, 1H,
J1,2 = 7.8 Hz, J2,F = 50.4 Hz, H-2), 4.29 (dd, 1H,
J6a,6b = 12.3 Hz, J5,6a = 5.3 Hz, H-6a), 4.16–4.10 (m,
3H, CH2NH, H-6b), 3.84 (ddd, 1H, J5,6b = 2.3 Hz,
J4,5 = 9.9 Hz, H-5), 3.73–3.56 (m, 8H, N(CH2CH2Cl)2),
2.55 (t, 2H, J = 7.4 Hz, CH2Ph), 2.31 (t, 2H,
J = 7.3 Hz, COCH2CH2), 2.11, 2.06, 2.05 (each s,
3 · 3H, OAc), 1.93 (qt, 2H, CH2CH2CH2); 13C NMR
(50 MHz, CDCl3) d 174.18, 170.64, 170.15, 169.65,
167.16 (CO), 153.33 (CArO), 144.55 (CArN), 133.75
(CArNH), 130.36 (CArCH2), 129.78, 121.40, 118.12,
112.32 (CHAr), 99.36 (C-1, J1,F = 23.4 Hz), 89.15 (C-2,
J2,F = 190.8 Hz), 72.80 (C-3, J3,F = 19.8 Hz), 72.14 (C-
5), 68.18 (C-4, J4,F = 7.3 Hz), 61.93 (C-6), 53.68
(NCH2), 44.55 (COCH2NH), 40.64 (CH2Cl), 35.61
(COCH2CH2), 34.10 (CH2Ph), 27.36 (CH2CH2CH2),
20.81, 20.69 (CH3); MS (ESI) m/z 742.35 [M+1]+; Anal.
(C34H42Cl2FN3O10) C, H, N.
glucopyranosyloxy)phenyl]-4-{4-[bis(2-chloroethyl)amino]-
phenyl}butanamide (37a). Compound 37a was prepared
according to the procedure described for compound
15a, starting from amino derivative 36 (318 mg,
0.80 mmol) and chlorambucil (243 g, 0.80 mmol). Purifi-
cation on silica gel (cyclohexane/ethyl acetate 6:4) gave
compound 37a (431 mg, 79%) as a white solid: mp
110–113 ꢁC; IR (KBr) m 3326, 1757, 1671, 1519, 1221,
1069, 1044; 1H NMR (200 MHz, CDCl3) d 7.44 (d,
2H, Jo = 9.0 Hz, HAr), 7.23 (br s, 1H, NH), 7.09 (d,
2H, J0o ¼ 8:6 Hz, HAr), 7.02 (d, 2H, Jo, HAr), 6.63 (d,
2H, J0o, HAr), 5.41 (td, 1H, J3,4 = J2,3 = 9.2 Hz,
J3,F = 14.5 Hz, H-3), 5.10 (t, 1H, J4,5 = 9.5 Hz, H-4),
5.07 (dd, 1H, J1,F = 2.7 Hz, J1,2 = 7.5 Hz, H-1), 4.56
(ddd, 1H, J2,F = 50.4 Hz, H-2), 4.30 (dd, 1H,
J5,6a = 5.3 Hz, J6a,6b = 12.3 Hz, H-6a), 4.15 (dd, 1H,
J5,6b = 2.1 Hz, H-6b), 3.79 (ddd, 1H, H-5), 3.73–3.58
(m, 8H, N(CH2CH2Cl)2), 2.62 (t, 2H, J = 7.3 Hz,
CH2Ph), 2.34 (t, 2H, J = 7.3 Hz, COCH2), 2.12, 2.08,
2.05 (each s, 3 · 3H, OAc), 2.11–1.98 (m, 2H,
CH2CH2CH2); 13C NMR (50 MHz, CDCl3) d 171.22,
170.66, 170.13, 169.65 (CO), 153.16 (CArO), 144.48
(CArN), 133.94 (CArNH), 130.52 (CArCH2), 129.83,
121.40, 118.12, 112.26 (CHAr), 99.39 (C-1,
J1,F = 23.2 Hz), 89.09 (C-2, J2,F = 190.6 Hz), 72.78 (C-

5018
B. Reux et al. / Bioorg. Med. Chem. 16 (2008) 5004–5020
3.1.20. 4-{4-[Bis(2-chloroethyl)amino]phenyl}butanamide-
N-{2-[4-(2-deoxy-2-fluoro-b-D-glucopyranosyloxy)phenyl]-
tration tested. This percentage did not modify cellular
growth. Fifty microliters of a 4· solution in MEM was
then added and a 48 h continuous drug exposure proto-
col was used. The cytotoxic effect of compounds on tu-
mor cells was then tested using the Resazurin reduction
test.
`
amino-2-oxoethyl} (40b). Zemplen deacetylation as for
17b and precipitation in EtOH gave compound 40b
(83 mg, 99%) as a white solid: mp 114 ꢁC; IR (KBr) m
3312, 1684, 1510, 1231, 1075; H NMR (200 MHz, ace-
1
tone-d6)d 9.22 (br s, 1 H, PhNHCO), 7.56 (d, 2 H,
Jo = 8.9 Hz, HAr), 7.43 (br t, 1H, CH2NHCO), 7.09 (d,
2H, J0 ¼ 8:6 Hz, HAr), 7.03 (d, 2H, Jo, HAr), 6.72 (d,
2H, J0oo, HAr), 5.19 (dd, 1H, J1,2 = 7.9 Hz, J1,F = 2.6 Hz,
H-1), 4.88 (br s, 1H, OH), 4.61 (br s, 1H, OH), 4.23
(td, 1H, J2,3 = 8.2 Hz, J2,F = 51.2 Hz, H-2), 4.01–3.45
(m, 16H, N(CH2CH2Cl)2, H-3, H-4, H-5, 2H-6,
CH2NH, OH), 2.55 (t, 2H, J = 7.6 Hz, CH2Ph), 2.29
(t, 2H, J = 7.4 Hz, CH2CH2CO), 1.89 (qt, 2H,
CH2CH2CH2); 13C NMR (50 MHz, acetone-d6) d
173.83, 168.37 (CO), 154.23 (CArO), 145.55 (CArN),
134.78 (CArNH), 131.56 (CArCH2), 130.40, 121.52,
117.73, 113.09 (CHAr), 99.47 (C-1, J1,F = 23.3 Hz),
93.16 (C-2, J2,F = 184.8 Hz), 77.82 (C-5), 76.09 (C-3,
J3,F = 16.8 Hz), 71.18 (C-4, J4,F = 7.6 Hz), 62.27 (C-6),
53.94 (NCH2), 44.16 (COCH2NH), 41.65 (CH2Cl),
3.2.2.1. Resazurin reduction test. The resazurin reduc-
tion test (RRT) was carried out according to the proto-
col described previously.63 Briefly, plates were rinsed
with 200 lL PBS (37 ꢁC., Gibco) using a multichannel
dispenser (Labsystems, Helsinki, Finland) and emptied
by overturning on absorbent toweling. Then, 150 lL
of a 25 lg/mL solution of resazurin in MEM without
SVF or phenol red was added to each well. The plates
were incubated for 1 h at 37 ꢁC in a humidified atmo-
sphere with 5% of CO2 for fluorescence development
by living cells. Fluorescence was then measured on the
automated 96-well plate reader Fluoroskan Ascent FLTM
(Labsystems) using an excitation wavelength of 530 nm
and an emission wavelength of 590 nm. The fluorescence
is proportional to the number of living cells in the well
and IC50 (drug concentration required to decrease final
cell population by 50%) was calculated from the curve
of concentration-dependent cell number decrease, de-
fined as the fluorescence in experimental wells as a per-
centage of that in control wells, with blank values
subtracted.
35.85
(COCH2CH2),
34.82
(CH2Ph),
28.40
(CH2CH2CH2); MS (ESI) m/z 616.31 [M+1]+; Anal.
(C28H36Cl2FN3O7) C, H, N.
3.2. Biological assays
3.2.1. Cell culture. Normal human fibroblasts were pur-
chased from Promocell (Heidelberg, Germany). This
frozen culture was obtained from foreskin waste from
a 6-year-old Caucasian male and the cells used in this
work were from the seventh to twelfth passage of the
culture. M4Beu, a human melanoma cell line, was estab-
3.2.3. Evaluation of cellular stress. L929 cells were plated
at a density of 5 · 103 cells per well in 96-well micro-
plates (Nunclon?, Nunc, Roskilde, Denmark) in 150 ll
of culture medium and were incubated for 48 h before
treatment with the compound tested to reach approxi-
mately 90% of confluence. Stock solutions of each com-
pound were prepared in DMSO and kept at ꢁ20 ꢁC
until use. The percentage of DMSO was kept at 0.5%
(v/v) whatever the concentration tested.
´
lished in the laboratory of Dr. J.F. Dore (INSERM,
Unit 218, Lyon, France) from metastatic biopsy speci-
mens and maintained in cell culture for almost 20 years
in our lab. Breast cancer adenocarcinoma MCF 7, pros-
tatic adenocarcinoma PC 3, colon adenocarcinoma
DLD-1, lung non-small cell carcinoma A 549, ovary
adenocarcinoma PA1 human cell lines and L 929 murine
cell line were purchased from the European Collection
of Cell Cultures (ECACC; Salisbury, United Kingdom).
3.2.3.1. Evaluation of cellular ROS production. Pro-
duction of reactive oxygen species (ROS) was evaluated
by 20,70-dichlofluorescin-diacetate (DCFH-DA; Sigma)
assay as previously described with some modifications.63
Briefly, L929 cells were washed once and incubated in
100 ll of HBSS/Hepes 15 mM, pH 7.4 (HBSS/H) con-
taining 20 lM DCFH-DA for 1 h at 37 ꢁC. After wash-
ing, treatment was performed in HBSS/H and the cells
were incubated again at 37 ꢁC. Positive control was
tert-butyl hydroperoxide at 100 lM. Fluorescence was
measured after 4 h at 485/530 nm. The results were ex-
pressed as the percentage of control. ROS production
induction was considered significant when the value
was above 150%.
Stock cell cultures were maintained as monolayers in
75 cm2 culture flasks in Glutamax Eagle’s minimum
essential medium with Eagle’s salts (MEM; Invitrogen,
Cergy-Pontoise, France) supplemented with 10% fetal
calf serum (Sigma, Saint-Quentin-Fallavier, France),
1 mM sodium pyruvate (Invitrogen), 1X vitamins solu-
tion (Invitrogen), 1X non essential amino acids solution
(Invitrogen) and 4 lg/ml of gentamicin (Invitrogen).
Cells were grown at 37 ꢁC in a humidified atmosphere
containing 5% CO2.
3.2.3.2. Evaluation of cellular reduced glutathione
content. GSH content was evaluated as previously de-
scribed with some modifications.63 After washing, treat-
ment was performed in HBSS/H and the cells were
incubated again at 37 ꢁC for 4 h. Positive control was
100 lM diethylmaleate (DEM, positive control). The
cells were then washed with 200 lL of HBSS/H and
incubated again for 15 min in 150 lL of HBSS/H con-
taining 7.5 lM monochlorobimane (Invitrogen). Fluo-
3.2.2. Cell growth inhibition assay. Cells were plated at a
density of 5 · 103 cells per well in 96-well microplates
(Nunc, Roskilde, Denmark) in 150 lL of culture med-
ium and were allowed to adhere for 16 h before treat-
ment with the compound tested. A stock solution of
each compound was prepared in dimethylsulfoxide
(DMSO) and kept at ꢁ20 ꢁC until use. The percentage
of DMSO was kept at 0.5% (v/v) whatever the concen-

B. Reux et al. / Bioorg. Med. Chem. 16 (2008) 5004–5020
5019
14. Bouvier, E.; Thirot, S.; Schmidt, F.; Monneret, C. Bioorg.
Med. Chem. 2004, 12, 969–977.
15. Moins, N.; Papon, J.; Seguin, H.; Gardette, D.; Moreau,
M. F.; Labarre, P.; Bayle, M.; Michelot, J.; Gramain, J.
C.; Madelmont, J. C.; Veyre, V. Nucl. Med. Biol. 2001, 28,
799–808.
rescence was measured at 393/460 nm. The results were
expressed as the percentage of control after subtraction
of blank values. GSH depletion was considered signifi-
cant when the value was below 80%.
3.2.3.3. Evaluation of cytolysis. To investigate poten-
tial cytolysis induced by the active compounds, we used
calcein-AM assay. Briefly, cells were washed once with
200 lL of Hank’s balanced salts solution buffered by
Hepes 15 mM, pH 7.4 (HBSS/H; Invitrogen). Then,
100 lL of each tested compound at specified concentra-
tion or the vehicle (DMSO 0.5% v/v final) or Tween 80
at 0.1 % m/v (positive control) in complete medium were
added to the well. Tests were done in triplicate. After 4 h
of incubation at 37 ꢁC, the absence of detached cells was
examined by inverse contrast-phase microscopy and
100 lL of working solution of calcein-AM (4 lM in
HBSS/H, Invitrogen) was added. After 1 h of incuba-
tion, the fluorescent signal of calcein was read on a fluo-
rometer (Fluoroscan Ascent FL, Labsystem) using an
excitation wavelength of 485 nm and an emission wave-
length of 530 nm. The percentage of membrane-undam-
aged cells was calculated from the formula:
[(RFUsample–RFU100%)/ (RFUvehicle ꢁ RFU100%] * 100
where RFU100% is the value of the background obtained
in the 100% cell lysis control (Tween 80 at 0.5% m/v).
Results are expressed as % of control. Cytolysis was
considered significant when this value was below 80%.
16. Rapp, M.; Giraud, I.; Maurizis, J. C.; Galmier, M. J.;
Madelmont, J. C. Bioconjugate Chem. 2003, 14, 500–506.
17. Giraud, I.; Rapp, M.; Maurizis, J. C.; Madelmont, J. C.
Bioorg. Med. Chem. 2003, 11, 5007–5012.
18. Gatenby, R. A.; Gillies, R. J. Nat. Rev. Cancer 2004, 4,
891–899.
19. Warburg, O. The metabolism of tumors; Richard Smith:
New York, 1931, 29–169.
20. Flier, J. S.; Mueckler, M. M.; Usher, P.; Lodish, H. F.
Science 1987, 235, 1492–1495.
21. Cantuaria, G.; Magalhaes, A.; Angioli, R.; Mendez, L.;
Mirhashemi, R.; Wang, J.; Wang, P.; Penalver, M.;
Averette, H.; Braunschweiger, P. Cancer 2000, 88, 381–388.
22. Iglesias-Guerra, F.; Candela, J. I.; Blanco, E.; Alcudia, F.;
´
´
Vega-Perez, J. M. Chirality 2002, 14, 199–203.
23. Sorg, B. L.; William, E. H.; Kliem, H. C.; Mier, W.;
Wiessler, M. Carbohydr. Res. 2005, 340, 181–189.
24. Briasoulis, E.; Pavlidis, N.; Terret, C.; Bauer, J.; Fiedler,
W.; Scho¨ffski, P.; Raoul, J. L.; Hess, D.; Selvais, R.;
Lacombe, D.; Bachmann, P.; Fumoleau, P. Eur. J. Cancer
2003, 39, 2334–2340.
25. Pohl, J.; Bertram, B.; Hilgard, P.; Nowrousian, M. R.;
Stuben, J.; Wiessler, M. Cancer Chemother. Pharmacol.
¨
1995, 35, 364–370.
26. Seker, H.; Bertram, B.; Burkle, A.; Kaina, B.; Pohl, J.;
¨
Koepsell, H.; Wiessler, M. Br. J. Cancer 2000, 82, 629–634.
27. Stuben, J.; Port, R.; Bertram, B.; Bollow, U.; Hull, W. E.;
¨
Schaper, M.; Pohl, J.; Wiessler, M. Cancer Chemother.
Pharmacol. 1996, 38, 355–365.
Supplementary data
28. Veyhl, M.; Wagner, K.; Volk, C.; Gorboulev, V.; Baum-
garten, K.; Weber, W. M.; Schaper, M.; Bertram, B.;
Wiessler, M.; Koepsell, H. Proc. Natl. Acad. Sci. U.S.A.
1998, 95, 2914–2919.
29. Uriel, C.; Egron, M. J.; Santarromana, M.; Scherman, D.;
Antonakis, K.; Herscovici, J. Bioorg. Med. Chem. 1996, 4,
2081–2090.
Chemical syntheses, characterization of all intermediates
and elemental analysis data of target compounds 14a,
14c, 15a–b, 17a–b, 20a–b, 21a–b, 22a–b, 31a–b, 34b,
37a–b, and 40a–b. Supplementary data associated with
this article can be found, in the online version, at
doi:10.1016/j.bmc.2008.03.038.
30. Halmos, T.; Santarromana, M.; Antonakis, K.; Scherman,
D. Eur. J. Pharmacol. 1996, 318, 477–484.
31. Beuthien-Baumann, B.; Hamacer, K.; Oberdorfer, F.;
Steinbach, J. Carbohydr. Res. 2000, 327, 107–118.
32. Gambhir, S. S. Nat. Rev. Cancer 2002, 2, 683–693.
33. Aft, R. L.; Zhang, F. W.; Gius, D. Br. J. Cancer 2002, 87,
805–812.
References and notes
1. Torchilin, V. P. Eur. J. Pharm. Sci. 2000, 11, S81–S91.
2. Van Zutphen, S.; Reedijk, J. Coord. Chem. Rev. 2005, 249,
2845–2853.
34. Lin, X.; Zhang, F.; Bradbury, C. M.; Kaushal, A.; Li, L.;
Spitz, D. R.; Aft, R. L.; Gius, D. Cancer Res. 2003, 63,
3413–3417.
35. Bell, S. E.; Quinn, D. M.; Kellett, G. L.; Warr, J. R. Br. J.
Cancer 1998, 78, 1464–1470.
3. Denny, W. A. Eur. J. Med. Chem. 2001, 36, 577–595.
4. Torchilin, V. P. Nat. Rev. 2005, 4, 145–160.
5. Hoste, K.; De Winne, K.; Schacht, E. Int. J. Pharm. 2004,
277, 119–131.
6. Maeda, H.; Sawa, T.; Konno, T. J. Control. Release 2001,
74, 47–61.
7. Beyer, U.; Roth, T.; Schumacher, P.; Maier, G.; Unold,
A.; Frahm, A. W.; Fiebig, H. H.; Unger, C.; Kratz, F. J.
Med. Chem. 1998, 41, 2701–2708.
8. Wirth, M.; Fuchs, A.; Wolf, M.; Ertl, B.; Gabor, F.
Pharma. Res. 1998, 15, 1031–1037.
9. Luo, Y.; Prestwich, G. D. Bioconjugate Chem. 1999, 10,
755–763.
10. Han, J. H.; Oh, Y. K.; Kim, D. S.; Kim, C. K. Int. J.
Pharm. 1999, 188, 39–47.
11. Yamazaki, N.; Kojima, S.; Yokoyama, H. Curr. Appl.
Phys. 2005, 5, 112–117.
12. Garnett, M. G. Adv. Drug Deliv. Rev. 2001, 53, 171–216.
13. Leu, Y. L.; Roffer, S. R.; Chern, J. W. J. Med. Chem.
1999, 42, 3623–3628.
36. Gupta, S.; Mathur, R.; Dwarakanath, B. S. Cancer Biol.
Ther. 2005, 4, 87–94.
37. Maschek, G.; Savaraj, N.; Priebe, W.; Braunschweiger, P.;
Hamilton, K.; Tidmarsh, G. F.; De Young, L. R.;
Lampidis, T. J. Cancer Res. 2004, 64, 31–34.
38. Dwarakanath, B. S.; Khaitan, D.; Rvindranath, T. Cancer
Biol. Ther. 2004, 3, 864–870.
39. Olivares, E.; Kadiata, M. M.; Malaisse, W. J. Oncol. Rep.
1999, 6, 1309–1311.
40. Soyez, H.; Schacht, E.; Vanderkerken, S. Adv. Drug Deliv.
Rev. 1996, 21, 81–106.
41. Leamon, C. P.; Reddy, J. A. Adv. Drug Deliv. Rev. 2004,
56, 1127–1141.
42. Mehta, G.; Sambaiah, T.; Maiya, B. G.; Sirish, M.;
Dattagupta, A. Tetrahedron Lett. 1994, 35, 4201–4204.

5020
B. Reux et al. / Bioorg. Med. Chem. 16 (2008) 5004–5020
43. Vega-Perez, J. M.; Candela, J. I.; Blanco, E.; Iglesias-
Guerra, F. Tetrahedron 1999, 55, 9641–9650.
44. Hou, Y.; Wang, J.; Andreana, P. R.; Cantauria, G.;
Tarasia, S.; Sharp, L.; Braunschweiger, P. G.; Wang, P. G.
Bioorg. Med. Chem. Lett. 1999, 9, 2255–2258.
54. (a) Zemplen, G. Ber. Dtsch. Chem. Ges. 1926, 59B, 1254–
1266; (b) Zemplen, G.; Kiss, D. Ber. Dtsch. Chem. Ges.
1927, 60B, 165–170.
55. Uriel, C.; Egron, M. J.; Santarromana, M.; Scherman, D.;
Antonakis, K.; Herscovici, J. Bioorg. Med. Chem. 1996, 4,
2081–2090.
45. Iglesias-Guerra, F.; Candela, J. I.; Bautista, J.; Alcu-
´
dia, F.; Vega-Perez, J. M. Carbohydr. Res. 1999, 316,
71–84.
56. Pauwels, E. K. J.; Ribeiro, M. J.; Stoot, J. H. M.;
`
McCready, V. R.; Bourguignon, M.; Maziere, B. Nucl.
Med. Biol. 1998, 25, 317–322.
46. Sugahara, S.; Kajiki, M.; Kuriyama, H.; Kobayashi, T.
Biol. Pharm. Bull. 2002, 25, 632–641.
47. Wook Lee, J.; Lu, J. Y.; Low, P. S.; Fuchs, P. L. Bioorg.
Med. Chem. 2002, 10, 2397–2414.
57. Smith, T. A. D. Nucl. Med. Biol. 2001, 28, 1–4.
58. Nelson, C. A.; Wang, J. Q.; Leav, I.; Crane, P. D. Nucl.
Med. Biol. 1996, 23, 533–541.
48. Kovac, P. Carbohydr. Res. 1986, 153, 168.
49. Deferrari, J. O.; Gros, E. G.; Mastronardi, I. O. Carbo-
hydr. Res. 1967, 4, 432–434.
50. Baer, H. H.; Hernandez Mateo, F.; Siemsen, L. Carbohy-
dr. Res. 1990, 195, 225–245.
51. Kovac, P.; Yeh, H. J. C.; Glaudemans, C. P. J. Carbohydr.
Res. 1987, 169, 23–34.
52. Tai, V. W.-F.; O’Reilly, M. K.; Imperiali, B. Bioorg. Med.
Chem. 2001, 9, 1133–1140.
59. Khan, S.; Ramwani, J. J.; O’Brien, P. J. Biochem.
Pharmacol. 1992, 43, 1963–1967.
60. Liu, H.; Lightfoot, R.; Stevens, J. L. J. Biol. Chem. 1996,
271, 4805–4812.
61. Rappeneau, S.; Baeza-Squiban, A.; Jeulin, C.; Marano, F.
Toxicol. Sci. 2000, 54, 212–221.
62. Das, G. C.; Bacsi, A.; Shrivastav, M.; Hazra, T. K.;
Boldogh, I. Mol. Carcinog. 2006, 45, 635–647.
63. Debiton, E.; Madelmont, J. C.; Legault, J.; Bart-
homeuf, C. Cancer Chemother. Pharmacol. 2003, 51,
474–482.
53. McCarter, J. D.; Yeung, W.; Chow, J.; Dolphin, D.;
Withers, S. G. J. Am. Chem. Soc. 1997, 119, 5792–5797.