Modulation of doxorubicin activity in cancer cells by conjugation with fatty acyl and terpenyl hydrazones

Article abstract of DOI:10.1016/j.ejmech.2010.01.037

Doxorubicin N-acylhydrazones derived from saturated, unsaturated and terpene-terminated fatty acids were tested for anticancer activity in cells of human HL-60 leukaemia, 518A2 melanoma, MCF-7/Topo breast and KB-V1/Vbl cervix carcinomas. In the latter, the N-heptadecanoyl hydrazone was more cytotoxic than its unsaturated C₁₈-fatty acyl analogues and even three times more than doxorubicin. The (menthoxycarbonyl)undecanoyl hydrazone was twice as active as doxorubicin in these multidrug resistant KB-V1/Vbl and in the 518A2 cells and also more efficacious in KB-V1 and MCF-7 cells that had been desensitised for doxorubicin. All hydrazones induced apoptosis albeit by slightly different mechanisms. While apoptosis induction by the menthoxymalonyl hydrazone was characterized by an upfront increase in caspase-8 activity, all other hydrazones elicited a hike in caspase-9 activity. Treatment of HL-60 and 518A2 cells with doxorubicin or its heptadecanoyl, linolenoyl, (menthoxycarbonyl)undecanoyl or menthoxymalonyl hydrazones also led to diverging increases of the ratio of bax to bcl-2 mRNA expression, of reactive oxygen species and of mitochondrial membrane damage.

Full text of DOI:10.1016/j.ejmech.2010.01.037

European Journal of Medicinal Chemistry 45 (2010) 1947–1954
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Modulation of doxorubicin activity in cancer cells by conjugation
with fatty acyl and terpenyl hydrazones
*
K. Effenberger, S. Breyer, R. Schobert
Organic Chemistry Laboratory, University of Bayreuth, 95440 Bayreuth, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Doxorubicin N-acylhydrazones derived from saturated, unsaturated and terpene-terminated fatty acids
were tested for anticancer activity in cells of human HL-60 leukaemia, 518A2 melanoma, MCF-7/Topo
breast and KB-V1/Vbl cervix carcinomas. In the latter, the N-heptadecanoyl hydrazone was more cyto-
toxic than its unsaturated C18-fatty acyl analogues and even three times more than doxorubicin. The
(menthoxycarbonyl)undecanoyl hydrazone was twice as active as doxorubicin in these multidrug
resistant KB-V1/Vbl and in the 518A2 cells and also more efficacious in KB-V1 and MCF-7 cells that had
been desensitised for doxorubicin. All hydrazones induced apoptosis albeit by slightly different mech-
anisms. While apoptosis induction by the menthoxymalonyl hydrazone was characterized by an upfront
increase in caspase-8 activity, all other hydrazones elicited a hike in caspase-9 activity. Treatment of
HL-60 and 518A2 cells with doxorubicin or its heptadecanoyl, linolenoyl, (menthoxycarbonyl)undecanoyl
or menthoxymalonyl hydrazones also led to diverging increases of the ratio of bax to bcl-2 mRNA
expression, of reactive oxygen species and of mitochondrial membrane damage.
Received 5 November 2009
Accepted 16 January 2010
Available online 25 January 2010
Keywords:
Doxorubicin
Fatty acids
Menthol
Multidrug resistance
Apoptosis
Ó 2010 Elsevier Masson SAS. All rights reserved.
1. Introduction
showed antitumour efficacy in mice bearing B16 melanoma [20].
DHA and other
u-3 fatty acids can bind to cognate receptors on
The Streptomyces metabolite doxorubicin (1) is indispensable for
the treatment of hematological cancers and solid tumours,
lymphomas, and soft tissue sarcomas despite side effects which
include myelosuppression and cardiac toxicity [1]. Its efficacy is
thought to be multifactorial [2] originating in part from intercalation
and alkylation of DNA [3], inhibition of RNA and DNA polymerases
[4] and of topoisomerase II [5], from interference with membranes
and the cellular homoeostasis of Fe and Ca [6–8], as well as from
accumulation of ceramides [9–11]. The role of reactive oxygen
species (ROS) is particularly multi-faceted and still under debate
[12]. ROS generated in cancer cells by supraclinical concentrations of
1 were found to induce lipid peroxidation leading to reactive frag-
ments that form DNA adducts [13]. A host of analogues, conjugates
and pharmaceutical formulations of 1 have been devised in order to
circumvent the unwanted side effects while retaining the desired
therapeutic impact [14–16]. As part of a project aimed at improving
the efficacy of anticancer drugs in resistant cancer cells by attach-
ment of fatty acid or terpene carriers [17–19] we now investigated
such conjugates with doxorubicin. Wang et al. had already reported
a conjugate with cis-4,7,10,13,16,19-docosahexaenoic acid (DHA) that
cancer cells and so exert a targeting effect [21]. They also seem to
play an important role in the delay of the progression of cancers
caused by ROS, presumably by modulating hormone receptors, Akt
kinase and the nuclear factor kB [22,23]. There is also evidence that
they inhibit the formation of the tumour growth promotor
13-hydroxy-octadecadienoic acid (13-HODE) [24], and that they
have a cardioprotective effect which could ameliorate the car-
diotoxicity of 1 [25]. Mei et al. reported an a-linolenic amide of 1
with improved uptake into cells [26]. Other fatty acid derivatives of 1
have not been systematically studied, yet. Neither have mono-
terpenes, some of which had occasionally been reported to be
weakly antitumoural, purportedly by interacting with membranes or
proteins. A few monoterpenes including R-limonene [27] and
S-perillyl alcohol [28,29] were clinically evaluated in cancer patients.
Terpene residues had previously shown beneficial effects of accu-
mulation and apoptosis induction in cancer cells when linked to
cytotoxic platinum(II) complexes [17]. Herein we describe new
N-acylhydrazones attached at C13 of doxorubicin. The acyl residues
were derived either from C17/18-fatty acids with different numbers
and positions of C]C bonds or from menthyl or bornyl esters con-
nected via alkyl spacers of different length. These conjugates were
tested for cytotoxicity in four cell lines of human cancers that are
typically treated with doxorubicin, namely leukaemia, melanoma,
breast carcinoma and cervix carcinoma. They were also scrutinized
* Corresponding author. Tel.: þ49 921 552679; fax: þ49 921 552671.
E-mail address: Rainer.Schobert@uni-bayreuth.de (R. Schobert).
0223-5234/$ – see front matter Ó 2010 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2010.01.037

1948
K. Effenberger et al. / European Journal of Medicinal Chemistry 45 (2010) 1947–1954
for their DNA affinity, for their mechanisms of apoptosis induction,
and for their influence on the mitochondrial membrane potential
and the levels of ROS in cancer cells.
Derivative 2h was also about 1.6-fold more efficacious than 1 in
MCF-7/Dox and KB-V1/Dox cells that had been desensitised for
doxorubicin by repeated treatment with clinically relevant doses
(MCF-7/Dox: 55 nM, KB-V1/Dox: 200 nM). This finding has some
bearing on the issues of acquired resistance and relapse prevention.
In another experiment the most efficacious compounds in KB-V1/
Vbl and MCF-7/Topo cells were also tested in these cells in the
presence of selective inhibitors of the respective ABC-transporters,
2. Chemistry
The hydrazones 2a–d were prepared from doxorubicin hydro-
chloride (1) and heptadecanoic acid (3a), oleic acid (3b),
a-linoleic
i.e., with 24
mM verapamil hydrochloride added to the P-gp over-
acid (3c), or -linolenic acid (3d) by a known procedure [30]
a
expressing KB-V1/Vbl cervix carcinoma cells and with 1.2
m
M
(Scheme 1). For the synthesis of hydrazones 2e–i, (ꢀ)-menthol and
(ꢀ)-borneol were first mono-esterified with either malonic acid,
hexane-1,6-dicarboxylic acid, nonane-1,9-dicarboxylic acid or
dodecane-1,12-dicarboxylic acid to give the corresponding
(menthoxycarbonyl)alkanoic acids 3e–h or (bornyloxycarbonyl)
undecanoic acid 3i. The acids 3 were then coupled with BocNHNH2
in the presence of 1-(3-dimethylaminopropyl)-3-ethylcarbo-
diimide hydrochloride (EDCI) in dry dimethylformamide (DMF) to
give the corresponding N-protected hydrazides RCONHNHBoc.
These were deprotected with trifluoroacetic acid (TFA) and the
resulting hydrazides 4 were condensed with 1 in the presence of
TFA in methanol.
fumitremorgin C added to the BCRP-rich MCF-7/Topo breast cancer
cells. While 1 was a good substrate for both transporters, apparent
from a 95% drop of its IC50 values upon addition of the inhibitors,
the derivatives 2 were more specific. The heptadecanoyl hydrazone
2a was a good substrate for the BCRP of the MCF-7/Topo cells, yet it
was less well pumped by the P-gp of the KB-V1/Vbl cells. Ideally,
a drug should not be affected by ABC-transporters and thus should
be characterized by a ratio of its IC50 values with and without
addition of specific inhibitors close to 1. This was the case for
derivative 2a in the KB-V1/Vbl cells (ratio ¼ 0.74).
Hydrazones are prone to hydrolytic cleavage at low pH [31,32]. Ina
control experiment
a mixture of 1 and (menthoxy-carbon-
yl)undecanoic acid 3h, two conceivable hydrolysis products of
hydrazone 2h, had a much lesser effect on the viability of the mela-
noma cells than intact 2h which is proof of its stability under the
applied conditions. Acid 3h alone had no effect at all. We also
quantified the cellular uptake of the hydrazones 2 by measuring the
decrease of the fluorescence intensity of the residual test compounds
in the medium after incubation of the cells. As the fluorescence of 1 is
much more intensive than that of any of its derivatives 2, a rapid
hydrolysis of the latter would have led to an increase of the fluores-
cence intensity in the medium. This was not observed. A direct
determination of the cellular fluorescence is not meaningful as its
intensity changes upon attachment of 1 and its derivatives to cell
membranes and upon their intercalation into DNA [33,34].
3. Biological evaluation
3.1. Inhibition of tumour cell growth
The hydrazones 2 were tested in vitro by the MTT assay against
four human cancer cell lines: HL-60 leukaemia, 518A2 melanoma,
and the multidrug resistant MCF-7/Topo breast and KB-V1/Vbl
cervix carcinomas (Table 1). They showed distinctly different
activity profiles (Fig. 1). In all cell lines the heptadecanoyl derivative
2a exhibited the greatest and the linoleyl derivative 2c the lowest
cytotoxic effect of the C17/18-acylhydrazones 2a–d after 72 h incu-
bation. Compound 2a was even three times more efficacious than 1
in the multidrug resistant cervix carcinoma cells. The terpenyl
derivatives 2e–i also showed a structure-dependent activity profile.
The menthyl derivatives with the shortest (2e) and the longest (2h)
tethers displayed an activity above average in all four cell lines. The
most potent (menthoxycarbonyl)undecanoyl hydrazone 2h was
twice as active as 1 in the melanoma and the cervix carcinoma cells,
while the activity profile of bornyl analogue 2i resembled that of 1.
3.2. Apoptosis induction
The apoptosis caused by hydrazones 2 was qualitatively visual-
ised through morphological changes and DNA fragmentation (cf.
Supporting information). 1 and its hydrazones 2a, 2d, 2e and 2h
O
R
NH
3
4
RCO2H
RCONHNH2
(ii)
O
OH
OH
O
N
O
OH
OH
(i)
OH
OH
OH
OH
O
O
OMe O
OMe O
O
O
NH2 × HCl
NH2 × HCl
1
2
OH
OH
R =
R =
CH2
CH2
CH2
CH2
O
O
O
5
5
5
5
2a
2b
2c
CH2
CH2
CH2
CH2
O
O
O
3
2f
6
2e
2g
O
O
CH2
O
O
9
9
2d
2h
2i
Scheme 1. Syntheses of acyl hydrazones 2 of doxorubicin hydrochloride (1). Reagents and conditions: (i) a) BocNHNH2, 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide (EDCI),
dimethylformamide; b) trifluoroacetic acid (TFA); (ii) TFA, methanol, room temperature; yields of 2: 80–99%.

K. Effenberger et al. / European Journal of Medicinal Chemistry 45 (2010) 1947–1954
1949
Table 1
Inhibitory concentrations IC50 in
a
mM of doxorubicin hydrochloride (1) and its
acylhydrazones 2 when applied to 518A2, HL-60, KB-V1/Vbl and MCF-7/Topo cells.
518A2
IC50, 24 h
IC50, 48 h
IC50, 72 h
1
0.34 ꢁ 0.11
0.17 ꢁ 0.09
0.35 ꢁ 0.09
0.39 ꢁ 0.11
1.8 ꢁ 0.5
0.12 ꢁ 0.04
0.11 ꢁ 0.05
0.42 ꢁ 0.17
1.6 ꢁ 0.6
2a
2b
2c
2d
2e
2f
2g
2h
2i
0.76 ꢁ 0.05
0.67 ꢁ 0.35
2.9 ꢁ 1.6
7.4 ꢁ 0.6
0.32 ꢁ 0.02
0.54 ꢁ 0.02
0.42 ꢁ 0.08
0.51 ꢁ 0.17
0.16 ꢁ 0.05
0.23 ꢁ 0.06
0.14 ꢁ 0.03
0.14 ꢁ 0.01
0.32 ꢁ 0.12
0.46 ꢁ 0.16
0.06 ꢁ 0.01
0.19 ꢁ 0.05
0.71 ꢁ 0.52
1.2 ꢁ 0.3
0.82 ꢁ 0.20
0.23 ꢁ 0.06
0.57 ꢁ 0.19
HL-60
1
2a
2b
2c
2d
2e
2f
2g
2h
2i
IC50, 24 h
IC50, 48 h
IC50, 72 h
Fig. 1. Cell line specificities of doxorubicin (1) and its acylhydrazones 2 as deviation of
the log (IC50/72 h) of individual derivatives from the mean over all compounds 1 and 2
log(IC50/72 h) values for a given cell line. Negative values indicate higher, positive
values lower than average activities. Mean log (IC50/72 h): ꢀ6.7 (518A2), ꢀ6.5 (HL-60),
ꢀ4.8 (KB-V1/Vbl) and ꢀ5.5 (MCF-7/Topo).
0.22 ꢁ 0.10
0.10 ꢁ 0.03
0.86 ꢁ 0.00
0.56 ꢁ 0.05
2.8 ꢁ 0.2
0.08 ꢁ 0.01
0.28 ꢁ 0.07
0.56 ꢁ 0.41
3.8 ꢁ 1.2
1.4 ꢁ 0.2
1.2 ꢁ 0.6
5.6 ꢁ 0.5
5.1 ꢁ 0.4
1.2 ꢁ 0.07
0.26 ꢁ 0.11
0.33 ꢁ 0.11
0.49 ꢁ 0.17
0.25 ꢁ 0.19
0.19 ꢁ 0.09
0.81 ꢁ 0.19
0.23 ꢁ 0.10
0.33 ꢁ 0.05
0.37 ꢁ 0.25
0.11 ꢁ 0.05
0.13 ꢁ 0.05
0.57 ꢁ 0.18
0.39 ꢁ 0.18
0.40 ꢁ 0.22
0.30 ꢁ 0.10
0.23 ꢁ 0.07
cells. The resulting changes in caspase activities were ascertained at
regular intervals between 1 h and 24 h by a luminescent assay
employing caspase-specific luminogenic oligopeptide substrates
(Fig. 2). The compounds 1 and 2h caused similar caspase kinetics in
either cell line. In HL-60 cells they led to a strong increase of effector
caspase-3 commencing after 4 h and peaking after 12 h, while the
initiator caspase-8 rose continuously between 4 and 24 h. Caspase-9
started to increase after 8 h incubation with 2h but only after 12 h
when treated with 1. In the melanoma cells compounds 1 and 2h
effected relatively low levels of caspases-8 and -3 but a steep and
pronounced rise in the caspase-9 activity after 8 h with a summit
after 12 h. In contrast, compound 2e initiated a very strong
maximum of caspase-8 activity after 12 h in HL-60 cells followed by
a strong maximum of effector caspase-3 activity after 18 h. The
caspase-9 activity ascended only from 12 h on and reached merely
a tenth of the caspase-8 level after 24 h. In 518A2 cells hydrazone 2e
elicited an early rise in caspase-8 activity after 6 h and a down-
stream maximum of caspase-3 activity after 8 h at which point in
time the activity of the caspase-9 only began to ascend. The hep-
tadecanoyl hydrazone 2a caused a simultaneous activity increase of
caspases-3, -8 and -9 between 1 and 24 h in both cell lines with the
absolute levels being highest of caspase-8 in the leukaemia and of
caspase-9 in the melanoma cells. In HL-60 cells the linolenoyl
hydrazone 2d led to a simultaneous and steep rise of caspase-8 and
-3 starting after 8 h and peaking between 12 and 18 h. The activity of
caspase-9 remained comparably low. In 518A2 cells the hydrazone
2d gave a profile of caspase activities reminiscent of that of 2a:
a continuous and strong rise of caspase-9 activity from 2 h on,
concurrent with a flat and less marked increase of the activities of
caspases-3 and -8. The general attenuation of caspases-3 and -8
activity levels in 518A2 cells can be explained with their known
overexpression of anti-apoptotic Bcl-2 protein [37].
KB-V1/Vbl
IC50, 24 h
IC50, 48 h
35.6 ꢁ 8.5
–
IC50, 72 h
18.3 ꢁ 7.6
1.02 ꢁ 0.04
5.1 ꢁ 0.9
1
59.9 ꢁ 13.0
1b
–
–
2a
2ab
2b
2c
2d
2e
2f
2g
2h
2hb
2i
42.9 ꢁ 10.8
10.7 ꢁ 4.3
–
3.7 ꢁ 0.5
32.5 ꢁ 1.8
29.6 ꢁ 5.9
40.8 ꢁ 3.9
17.8 ꢁ 3.8
17.7 ꢁ 3.2
18.7 ꢁ 3.1
20.5 ꢁ 2.1
18.4 ꢁ 3.0
34.9 ꢁ 15.7
21.5 ꢁ 3.7
17.1 ꢁ 2.5
–
12.7 ꢁ 1.8
19.7 ꢁ 7.0
14.0 ꢁ 5.0
10.3 ꢁ 2.1
23.3 ꢁ 12.5
21.8 ꢁ 3.8
8.8 ꢁ 0.7
>100
79.6 ꢁ 6.8
30.5 ꢁ 4.6
–
0.67 ꢁ 0.09
24.2 ꢁ 7.2
68.8 ꢁ 18.8
20.7 ꢁ 6.7
MCF-7/Topo
IC50, 24 h
IC50, 48 h
1.3 ꢁ 0.5
–
IC50, 72 h
1.0 ꢁ 0.4
0.08 ꢁ 0.01
3.0 ꢁ 0.9
0.17 ꢁ 0.05
6.6 ꢁ 1.5
9.9 ꢁ 2.8
3.6 ꢁ 1.0
2.7 ꢁ 1.0
0.52 ꢁ 0.20
2.4 ꢁ 1.0
4.3 ꢁ 1.9
2.6 ꢁ 1.1
0.09 ꢁ 0.01
1.8 ꢁ 0.6
1
6.2 ꢁ 1.5
1c
–
–
2a
2ac
2b
2c
2d
2e
2ec
2f
2g
2h
2hc
2i
5.9 ꢁ 1.9
3.2 ꢁ 0.6
–
14.6 ꢁ 0.3
40.5 ꢁ 7.3
5.7 ꢁ 1.4
8.4 ꢁ 2.6
6.2 ꢁ 1.9
20.3 ꢁ 6.5
4.3 ꢁ 2.3
3.8 ꢁ 1.0
–
4.7 ꢁ 1.2
7.2 ꢁ 1.8
2.6 ꢁ 1.1
–
–
–
6.7 ꢁ 0.9
10.2 ꢁ 2.8
7.1 ꢁ 1.8
6.3 ꢁ 1.7
3.8 ꢁ 1.3
a
Values are derived from concentration-response curves obtained by measuring
the percentual absorbance of vital cells relative to untreated controls (100%) after
24, 48, and 72 h exposure of 518A2, HL-60, KB-V1/Vbl and MCF-7/Topo cells to test
compounds in the MTT assay. Values represent means of four independent
experiments ꢁ standard deviation.
b
With 24
With 1.2
mM verapamil hydrochloride added.
M fumitremorgin C added.
c
m
We then measured the mRNA expression of bcl-2 and its pro-
apoptotic counterpart bax in 518A2 and HL-60 cells following treat-
ment with 5
mM of compounds 1, 2a, 2d, 2e and 2h for up to 24 h (see
gave rise to distinct DNA ladder patterns upon incubation at 5
m
M
Supporting information). Fig. 3 depicts the more apoptosis-relevant
ratio of bax mRNA to bcl-2 mRNA expression in these two cell lines
after 6 h and 18 h. In the HL-60 cells this ratio fell initially for all tested
compounds yet reached elevated levels after 18 h exposure to
compounds 2d, 2e and 2h. In 518A2 cells, all test compounds led to
greatly increased ratios of bax mRNA to bcl-2 mRNA after 18 h. The
menthyl hydrazones 2e and 2h showed such an effect already after
6 h. In the case of the malonyl spacered hydrazone 2e an unusually
strong drop of the bcl-2 mRNA expression within the first 6 h was
responsible for the high bax/bcl-2 mRNA ratio.
concentrations with HL-60 cells for 24 h. A quantitative assessment
of apoptosis is difficult due to the intrinsic fluorescence of doxoru-
bicin derivatives which interferes with fluorescence assisted cell
sorting (FACS) and with Annexin-V-Fluos/PI or TUNEL assays. To
elucidate the mechanism of apoptosis we first checked the
involvement of caspases-3, -8, and -9 [35,36]. HL-60 leukaemia and
518A2 melanoma cells were treated with 5 mM of compounds 1, 2a,
2d, 2e and 2h which had all been cytotoxic above average in the
melanoma cells and with the exception of 2d also in the leukaemia

1950
K. Effenberger et al. / European Journal of Medicinal Chemistry 45 (2010) 1947–1954
Fig. 2. Caspases activation in HL-60 and 518A2 cells treated with 5
and ꢀ9 (triangle symbol) were quantified by a luminometric assay (Caspase-Glo, Promega) and are reported as relative luminescence intensities.
m
M of 1 or selected derivatives 2 for up to 24 h. The activities of caspases-3 (square symbol), ꢀ8 (lozenge symbol)
Only for incubation with doxorubicin (1) the extent of mRNA
expression corresponded to the levels of Bax and Bcl-2 proteins as
ascertained by Western blotting (cf. Supporting information). In
this case, in HL-60 cells the level of Bcl-2 protein rose continuously
over 24 h with only little bax being present, whereas in 518A2 cells
the Bcl-2 concentration decreased and that of Bax increased.
Similar Western blots for incubations with hydrazones 2e and 2h
were inconclusive.
3.3. Change of mitochondrial membrane potential and generation
of ROS
One of the earliest indicators of doxorubicin mediated apoptosis
is a change in the mitochondrial membrane potential DJm [38].
Hence, we analysed such changes in 518A2 and HL-60 cells treated
with 5
mM of 1 or hydrazones 2 for 6 h or 18 h using a kit from
Stratagene which is based upon the fluorescent cationic dye JC-1

K. Effenberger et al. / European Journal of Medicinal Chemistry 45 (2010) 1947–1954
1951
Table 3
ROS generation as NBT reduction (%)a in 518A2 and HL-60 cells upon exposure to
5
m
M of selected compounds for 24 and 72 h.
NBT reduction [%]a in 518A2
cells at 24 h/72 h
NBT reduction [%]a in HL-60
cells at 24 h/72 h
1
2.5 ꢁ 0.3/15.4 ꢁ 1.9
6.6 ꢁ 1.0/71.2 ꢁ 19.7
2.8 ꢁ 0.4/4.4 ꢁ 1.7
1.6 ꢁ 0.5/2.3 ꢁ 0.5
2.9 ꢁ 1.6/45.1 ꢁ 7.7
3.0 ꢁ 1.4/16.2 ꢁ 0.5
2.2 ꢁ 0.9/1.5 ꢁ 0.5
1.4 ꢁ 0.7/1.3 ꢁ 0.1
3.3 ꢁ 0.9/38.3 ꢁ 12.4
2.7 ꢁ 1.3/3.8 ꢁ 0.7
2.8 ꢁ 0.8/5.6 ꢁ 0.8
2.8 ꢁ 0.5/3.4 ꢁ 0.3
0.6 ꢁ 0.3/4.0 ꢁ 0.2
0.4 ꢁ 0.1/1.4 ꢁ 0.1
1.0 ꢁ 0.3/3.0 ꢁ 0.3
3.8 ꢁ 1.5/11.2 ꢁ 2.1
1.0 ꢁ 0.3/5.7 ꢁ 0.8
1.0 ꢁ 0.5/6.8 ꢁ 1.3
3.8 ꢁ 1.1/5.9 ꢁ 0.9
1.8 ꢁ 0.6/7.9 ꢁ 1.1
2a
2b
2c
2d
2e
2f
2g
2h
2i
Fig. 3. Bax/bcl-2 mRNA expression in 518A2 and HL-60 cells after exposure for 6 h or
18 h to 5
section.
mM of selected test compounds. Conditions as stated in the Experimental
a
Relative ROS generation (NBT reduction) as determined from the percentual
absorbance of diformazan relative to untreated controls (1%) after 24 and 72 h
exposure of 518A2 und HL-60 cells to 5 M of the test compounds. Values represent
m
means of four independent experiments ꢁ standard deviation.
[39]. The ratio of red (JC-1 aggregates in intact mitochondria) to
green fluorescence (JC-1 monomeres in the cytosol) is lowered in
apoptotic cells. We found that the reduction of intact mitochondria
was significant in HL-60 cells treated with 1, 2e or 2h (e.g., up to
77% after 18 h for 2h) whereas compounds 2a and 2d caused just
a small decrease of 30–40% (Table 2). In 518A2 cells the compounds
1, 2a and 2e–i also led to a decrease of the fluorescence ratio after
18 h, similar in magnitude to the decrease in HL-60 cells after 6 h.
Compounds 2b–d showed no notable effect.
doxorubicin as apparent from distinct anodic peak currents
(cf. Supporting information).
3.4. Interaction with plasmid DNA
To check whether DNA is a potential target of the hydrazones 2
they were incubated at different concentrations with pBR322
plasmid DNA and the effect on the electrophoretic mobilities of the
various DNA forms present was monitored (Fig. 4). In accordance
with its pronounced inhibitory effect on the tested cancer cells the
long-tethered menthyl derivative 2h strongly retarded the front-
most, covalently closed circular (cc) form due to an unwinding of its
supercoiled structure. This effect even exceeded that of doxorubicin
(1) at high concentrations. The retardation effects of the derivatives
2d and 2e were less pronounced. The saturated hydrazone 2a was
least effective. In essence, while there was no cogent relation
between DNA binding and reduction of the cell viability for each of
the tested hydrazones, DNA is arguably one of their targets and their
affinity for it strongly depends on the nature of the acyl residue.
The generation of ROS in HL-60 and 518A2 cells by the hydrazones
2 at a clinically relevant concentration of 5 mM was assessed with the
colorimetric nitroblue-tetrazolium (NBT) assay [40]. It is based on the
selective reduction of a yellow, water-soluble tetrazolium chloride to
an insoluble violet diformazan by superoxide O–2. The hydrazones 2a,
2d, 2e and 2h which exhibited the greatest viability reduction in the
melanoma cells after 72 h incubation also caused the most distinct
increase in ROS levels in these cells, most dramatically so the satu-
rated hydrazone 2a. For the leukaemia cells there was a similar
though less stringent relation between the cytotoxicity and the
degree of ROS generation by the derivatives 2 upon 72 h exposure
(Table 3). At grossly superclinical concentrations the ROS production
was no longer correlated to the cytotoxicity. The cyclic voltammo-
grams of compounds 2a, 2d, 2e and 2h recorded in air-saturated
dimethylformamide were reminiscent of that of doxorubicin yet
differed in the rate of the oxygen consumption via generation of ROS
by quinone–O2 redox cycling at potentials < ꢀ0.8 V vs. SCE. Deriva-
tives 2a, 2e, and 2h led to ROS formation with better reversibility and
a more complete regeneration of their 5,12-quinone core than
4. Discussion
A series of N-acylhydrazones of doxorubicin were tested in vitro
for eligibility as anticancer agents. They were derived from satu-
rated, unsaturated and menthyl- or bornyl-terminated fatty acids.
Their stability under in vitro test conditions with various cancer
cells was evident from their bioactivities which differed from those
of conceivable cleavage products, and from the decay of
Table 2
Percentage of intact mitochondriaa upon treatment of 518A2 and HL-60 cells with
5
m
M of selected compounds.
Intact mitochondria [%]a in 518A2
cells after 6 h/18 h
Intact mitochondria [%]a in HL-60
cells after 6 h/18 h
1
98 ꢁ 10/74 ꢁ 19
113 ꢁ 11/80 ꢁ 16
96 ꢁ 4/99 ꢁ 4
83 ꢁ 8/28 ꢁ 4
94 ꢁ 8/59 ꢁ 4
103 ꢁ 1/104 ꢁ 4
88 ꢁ 11/87 ꢁ 3
103 ꢁ 11/71 ꢁ 11
76 ꢁ 6/36 ꢁ 4
93 ꢁ 8/66 ꢁ 6
84 ꢁ 13/68 ꢁ 4
90 ꢁ 5/23 ꢁ 5
78 ꢁ 4/65 ꢁ 7
2a
2b
2c
2d
2e
2f
2g
2h
2i
80 ꢁ 11/97 ꢁ 2
120 ꢁ 7/96 ꢁ 34
93 ꢁ 13/61 ꢁ 18
82 ꢁ 3/79 ꢁ 11
83 ꢁ 2/72 ꢁ 11
108 ꢁ 2/71 ꢁ 17
66 ꢁ 7/73 ꢁ 4
a
Ratio of red to green fluorescence relative to untreated controls (100%) after 6 h
M of the test compounds.
and 18 h exposure of 518A2 and HL-60 cells to 5
m
Fig. 4. Modification of gel electrophoretic mobility of pBR322 plasmid DNA when
incubated with different concentrations of 1 and selected conjugates 2 (oc ¼ open
Determined with the Mitochondrial Membrane Detection Kit (Stratagene, La Jolla,
CA, USA). Values represent means of four independent experiments ꢁ standard
deviation.
circular, cc ¼ covalently closed circular DNA form). Concentrations (in
mM): lane
1(leftmost): 0, lane 2: 5, lane 3: 10, lane 4: 25.

1952
K. Effenberger et al. / European Journal of Medicinal Chemistry 45 (2010) 1947–1954
fluorescence of residual test compounds in the media over the time
of incubation which discounted the liberation of the strongly
fluorescent doxorubicin. Of the four C17/18-N-fatty acylhydrazones
2a–d, the saturated heptadecanoyl derivative 2a reduced the
metabolic activity of all cells more than the unsaturated analogues.
In multidrug resistant KB-V1/Vbl cervix carcinoma cells it was even
three times more efficacious than doxorubicin. This is partly owed
to the fact that it is much less good a substrate for the P-gp efflux
pumps of these cells. This was demonstrated by addition of the P-gp
inhibitor verapamil. Of the five terpenylacylhydrazones 2e–i the
(menthoxycarbonyl)undecanoyl derivative 2h performed much
better than average in all tested cell lines and exceeded doxorubicin
in the melanoma and the cervix carcinoma cells. 2h was also much
more active in KB-V1 and MCF-7 cells that had been desensitised
for doxorubicin by repeated exposure to clinically relevant doses.
Derivatives with a bornyl residue (2i) or shorter tethers (2e–g) had
different activity profiles. Hydrazone 2i resembled doxorubicin
with a loss of activity mainly in the multidrug resistant MCF-7/Topo
and KB-V1/Vbl cells, while 2e was still efficacious above average in
all cells albeit less than 2h. The menthyl hydrazones 2f and 2g with
medium-length tethers were least active.
The mode of cytotoxic action of the hydrazones 2 was largely
apoptotic. 518A2 melanoma cells are reluctant to enter into the
mitochondrial pathway of apoptosis due to a high level of anti-
apoptotic bcl-2 protein. While we found by real time PCR that all
hydrazones that reduce the viability of 518A2 cells above average,
i.e., 2a, 2d, 2e, 2h as well as 1, led to a distinct long-term decrease
of bcl-2 mRNA expression, the precise apoptotic mechanism and
involvement of caspases varied for the individual cell lines and
test compounds. For instance, apoptosis of 518A2 melanoma cells
treated with 2a, 2d or 2h was characterized by an early onset of
initiator caspase-9 activity. In contrast, apoptosis elicited in
518A2 or in HL-60 cells by the menthoxymalonyl hydrazone 2e
was accompanied by high initiator caspase-8 activity. The
observed quick and steep hike of the ratio of bax mRNA to bcl-2
mRNA in 518A2 cells treated with compound 2e originates from
a genuine slump of the bcl-2 mRNA expression. Apoptosis
induced by doxorubicin (1) and its derivatives 2e and 2h in HL-60
and 518A2 cells also proceeds with a swift and distinct loss of
mitochondrial membrane potential regardless of the divergent
caspase kinetics.
6. Experimental protocols
Melting points were recorded on a GALLENKAMP apparatus and
are uncorrected. IR spectra were recorded on a PERKIN-ELMER
Spectrum One FT-IR spectrophotometer equipped with an ATR
sampling unit. Nuclear magnetic resonance (NMR) spectra were
recorded under conditions as indicated on a BRUKER Avance 300
spectrometer. Chemical shifts are given in parts per million (
d)
down-field from tetramethylsilane as internal standard for 1H and
13C. Mass spectra were recorded using a VARIAN MAT 311A (EI).
Microanalyses indicated by the symbols of the elements were within
ꢁ0.2% of the theoretical values for all new doxorubicin hydrazones.
For chromatography MERCK silica gel 60 (230–400 mesh) was used.
Fluorescence intensity was measured with an SFM 25 spectrofluo-
rometer (KONTRON Instruments, Zurich, Switzerland), excitation
490 nm, emission 594 nm. Fluorescence spectra were recordered on
an FP-6500 fluorescence spectrophotometer (JASCO, Tokyo, Japan)
between 500 and 700 nm. Uptake of the derivatives and mito-
chondrial membrane potential was measured using a CM Infinite
F200 fluorescence plate reader (TECAN, Crailsheim, Germany).
Absorbance for the MTT (570 nm, 630 nm) and the NBT assay
(630 nm, 405 nm) was measured using a Power Wave ꢂ 340 ELISA
plate reader (BIO-TEK Instruments, Winooski, VT, USA). All starting
compounds other than doxorubicin were purchased from ALDRICH
and used without further purification.
6.1. Chemistry
6.1.1. Doxorubicin heptadecanoyl hydrazone (2a) – typical
procedure
6.1.1.1. N-Boc-Heptadecanoyl hydrazine. A mixture of heptadeca-
noic acid 3a (270 mg, 1 mmol), DMF (15 mL), H2NNHBoc (159 mg,
1.2 mmol) and EDCl (575 mg, 3 mmol) was stirred at room
temperature for 12 h. Water was added and the mixture was
extracted with ethyl acetate. The organic layers were washed with
brine, dried with Na2SO4 and the volatiles were evaporated. The
crude product was purified by column chromatography (Rf 0.46;
ethyl acetate/cyclohexane 1:1; silica gel 60). Yield: 270 mg
(0.7 mmol, 70%), white solid of mp 47 ꢃC; nmax (ATR)/cmꢀ1 3275,
2923, 2853, 1725, 1670, 1467, 1368, 1246, 1159, 907, 729; 1H NMR
(300 MHz, CDCl3)
d
0.81 (t, J ¼ 6.9 Hz, 3H),1.18 (br, 26H),1.39 (s, 9H),
The hydrazones 2a, 2d, 2e and 2h which are cytotoxic above
average in the melanoma cells also caused a significant increase of
1.57 (m, 2H), 2.16 (t, J ¼ 7.8 Hz, 2H), 7.07 (br, 1H), 8.51 (br, 1H); 13C
NMR (75.5 MHz, CDCl3) d 13.9, 22.6, 25.2, 28.0, 29.2, 29.3, 29.4, 29.5,
reactive oxygen species in these cells at clinically relevant 5
mM
31.8, 33.8, 81.4, 155.9, 172.9; m/z (EI) 311 (10) [Mþ–C4H9O], 253 (33),
concentrations, especially the saturated hydrazone 2a. Whether
this is just coincidental or causative for the efficacy of these
compounds remains unclear. For the leukaemia cells there was
a less stringent relation between the viability depression and the
degree of ROS generation by the derivatives 2. On the other hand,
DNA seemed to be a potential target for all tested hydrazones with
their binding affinity depending on the nature of the acyl residue. In
accordance with its impact on cancer cell viability the menthyl
derivative 2h interacted most strongly with plasmid DNA.
118 (6), 98 (4), 85 (4), 71 (6), 57 (100).
6.1.1.2. Heptadecanoylhydrazine (4a). N-Boc-heptadecanoylhy-
drazine (270 mg, 0.7 mmol) was dissolved in dichloromethane
(15 mL) and treated with trifluoroacetic acid (TFA; 5 mL). The
resulting mixture was stirred at room temperature for 15 min,
the solvent was evaporated, and the crude product was purified
by column chromatography (Rf 0.29; ethyl acetate/cyclohexane
1:1; silica gel 60). Yield: 114 mg (0.39 mmol, 57%); colourless
oil; nmax (ATR)/cmꢀ1 3319, 3200, 2919, 2849, 1668, 1629, 1537,
5. Conclusions
1462, 1400, 1376, 719; 1H NMR (300 MHz, CDCl3)
d 0.84 (t,
J ¼ 6.9 Hz, 3H), 1.21 (br, 26H), 1.61 (m, 2H), 2.59 (t, J ¼ 7.7 Hz,
In summary, N-acylhydrazones of doxorubicin can complement
or even surpass the properties of the parent drug in terms of
cytotoxic efficacy, cell line specificity, breach of multidrug resis-
tance, mechanism of apoptosis and combination of molecular
targets if the end group and the length and degree of saturation of
the acyl residue are aptly chosen. Minute alterations in any of these
structural features may change the biological profile considerably.
This is proof that these acylhydrazone appendages are more than
just lipophilic shuttle groups.
2H), 5.04 (br, 2H), 8.49 (br, 1H); 13C NMR (75.5 MHz, CDCl3)
d
14.2, 22.6, 24.7, 29.3, 29.4, 29.5, 29.6, 31.9, 32.0, 32.7, 176.2; m/
z (EI) 253 (8) [MþeN2H3], 211 (3), 169 (4), 138 (3), 127 (9), 114
(33), 72 (100), 57 (23).
6.1.1.3. Doxorubicin heptadecanoyl hydrazone (2a). 1 (78 mg,
0.13 mmol) and heptadecanoylhydrazine 4a (50 mg, 0.18 mmol)
were dissolved in anhydrous methanol (15 mL) and the mixture
was treated with TFA (7.8
mL). After stirring overnight at room

K. Effenberger et al. / European Journal of Medicinal Chemistry 45 (2010) 1947–1954
1953
temperature and in the dark all volatiles were removed in vacuum,
the residue was redissolved in methanol (5 mL) and diethyl ether
(35 mL) was added to precipitate the product. The red solid was
collected by centrifugation and dried in vacuum. Yield: 91 mg
(80%); Rf 0.86 (CH2Cl2/MeOH 1:1); nmax (ATR)/cmꢀ1 3373, 3214,
2921, 2851, 1720, 1662, 1615, 1578, 1524, 1446, 1403, 1280, 1206,
1112, 1081, 1010, 983, 866, 817, 762; 1H NMR (300 MHz, DMSO)
using the homogeneous luminescent assay. After cleavage of the
luminogenic substrate the luminescence was quantified with
a Tecan Genios Plus plate reader and assumed to be proportional to
the caspase activities [41].
6.2.3. Real-time PCR for detection of bcl-2 and bax mRNA
expression
d
0.84 (t, J ¼ 6.9 Hz, 3H),1.1–1.3 (m, 28H),1.15 (d, J ¼ 6.7 Hz, 3H),1.72
HL-60 and 518A2 cells were treated with 5 mM of the test
(d, J ¼ 12.7 Hz, 1H), 1.89 (d, J ¼ 12.7 Hz, 1H), 2.15 (m, 2H), 2,43 (m,
2H), 2.82 (d, J ¼ 18.5 Hz, 1H), 2.98 (d, J ¼ 18.5 Hz, 1H), 3.24 (br, 1H),
3.61 (m, 1H), 3.96 (s, 3H), 4.13 (q, J ¼ 6.7 Hz), 4.59 (s, 2H), 4.89 (br,
1H), 5.28 (br, 1H), 7.62 (d, J ¼ 7.9 Hz, 1H), 7.86 (dd, J ¼ 7.9, 8.4 Hz,
1H), 7.99 (d, J ¼ 8.4 Hz, 1H), 10.32 (s, 1H), 13.33 (br, 1H), 14.11 (br,
compounds for up to 24 h. The total RNA was extracted using
peqGOLD RNAPure (peqLab, Erlangen, Germany) according to the
manufacturer’s instructions. The synthesis of cDNA was conducted in
a 21 mL reaction mixture starting with 80 ng total RNA using the
reverse transcription system (Promega, Madison, WI, USA). Quanti-
tative PCR was performed using the SYBR GREEN PCR-Kit LightCycler
fast start DNA Master for a LightCycler 2.0 System (Roche Diagnostic,
Mannheim, Germany) and threshold numbers were determined
using the LightCycler Software, version 3.5. All genes examined were
normalised to a house-keeping gene encoding GAPDH. The primers
were obtained from Qiagen: GAPDH (Hs_Gapdh_2_SG QuantiTect
Primer Assay, Cat.No.: QT01192646), bax (Hs_Bax_2_SG QuantiTect
Primer Assay, Cat.No.: QT00031192), bcl-2 (Hs_Bcl2_1_SG QuantiTect
Primer Assay, Cat.No.: QT00025011).
1H); 13C NMR (75.5 MHz, DMSO)
d 14.6, 17.3, 22.7, 24.6, 28.4, 28.7,
29.1, 29.3, 29.4, 29.5, 31.6, 31.7, 34.3, 49.1, 56.6, 57.1, 66.6), 70.1, 75.3,
99.7, 110.9, 111.1, 119.4, 120.2, 120.4, 134.7, 135.7, 136.3, 136.7, 156.6,
157.1, 161.3, 174.5, 186.9. Anal C45H65N3O11 (C, H, N).
6.2. Biological studies
6.2.1. MTT assay. Cell lines and culture conditions
The HL-60 cells were obtained from the German Collection of
Biological Material (DSMZ), Braunschweig, the 518A2 cells from the
Department of Oncology and Hema of the Martin-Luther University
Halle, Germany, and the KB-V1/Vbl and the MCF-7/Topo cells from
the Institute of Pharmacy of the University Regensburg, Germany.
The HL-60 cells were grown in RPMI-1640, the 518A2 and the
KB-V1/Vbl cells in Dulbecco’s Modified Eagle Medium (D-MEM,
Gibco), supplemented in each case with 10% fetal calf serum (FCS),
6.2.4. Changes of the mitochondrial membrane potential
were determined with the mitochondrial membrane detection
kit (Stratagene, La Jolla, CA, USA). Following treatment with 5
mM of
the test compounds the cell samples were centrifuged at 400 g for
5 min. The pellets were resuspended in 500
mL of diluted JC-1
solution (0.2ꢂ), incubated at 37 ꢃC for 15 min (HL-60) or 35 min
100 IU/mL penicillin G, 100
amphotericin B and 250
mg/mL streptomycin sulfate, 0.25
mg/mL
(518A2) and then centrifuged again for 5 min at 400 g. After one
mg/mL gentamycine (all from Gibco,
washing step the pellets were resuspended in 100 mL PBS and
Egenstein, Germany). The MCF-7/Topo cells were grown in E-MEM
medium (Sigma) supplemented with 2.2 g/L NaHCO3, 110 mg/L
sodium pyruvate and 5% FCS. The cells were maintained (5% CO2,
95% humidity, 37 ꢃC) in 75-mL culture flasks (Nunc, Wiesbaden,
Germany) and serially passaged following trypsinisation with 0.05%
trypsin/0.02% EDTA (PAA laboratories, Co¨lbe, Germany).
transferred into a well of a black 96-well plate. The red fluorescence
(excitation 585 nm, emission 590 nm) and green fluorescence
(excitation 510 nm, emission 527 nm) were measured using
a fluorescence plate reader (Infinite F200, Tecan, Germany). The
ratio of red to green fluorescence is decreased in apoptotic cells
compared to vital cells [39].
Cells were seeded out in 96-well tissue culture plates and
cultured for 24 h. Inhibitors of P-pg or BCRP were optionally added.
Incubation of the cells following treatment with the test compounds
was continued for 24, 48 or 72 h. MTT [3-(4,5-dimethylthiazol-2-yl)-
2,5-diphenyltetrazolium bromide] in PBS (5 mg/mL) was added to
a final concentration of 0.05% (HL-60, 518A2) or 0.1% (KB-V1, MCF-7).
After 2 h the formazan precipitate was dissolved in 10% sodium
dodecylsulfate in DMSO containing 0.6% acetic acid in the case of the
HL-60 cells. For the adherent 518A2, KB-V1 and MCF-7 cells the
microplates were swiftly turned to discard the medium before
adding the solvent mixture. The microplates were gently shaken in
the dark for 30 min and absorbance at 570 nm and 630 nm (back-
ground) was measured with an ELISA plate reader. All experiments
were carried out in quadruplicate; the percentage of viable cells was
calculated as the mean ꢁ SD with controls set to 100%.
6.2.5. Generation of reactive oxygen species (NBT assay)
HL-60 cells (0.5 ꢂ 106/mL) were plated in 96-well culture plates
and test compounds were added after 24 h incubation at 37 ꢃC to
a final conc. of 5
and the remaining cells were resuspended in 100
and incubated for 1 h. The reduced NBT was solubilized with 100
of 2M KOH and 130 L DMSO for 30 min. The absorbance was
mM. The medium was removed by centrifugation
mL of 0.1% NBT
mL
m
measured at 630 nm and 405 nm (background) using an ELISA plate
reader. The adherent 518A2 cells (5 ꢂ 104/mL) were seeded out in
96-well tissue culture plates after trypsinisation and incubated for
24 h at 37 ꢃC to allow attachment, then treated similarly to the HL-
60 cells, only that the medium was removed prior to incubation
with NBT for 4 h. All experiments were run in double quadrupli-
cates [40,42].
6.2.2. Caspase activity assay
6.2.6. Interaction with pBR322 plasmid DNA
Caspase activities were determined by luminometric assays
(Caspase-Glo 3/7, 8 and 9, Promega, USA), according to the manu-
facturer’s instructions. Cellular proteins were extracted from HL-60
The interaction of doxorubicin derivatives with pBR322 plasmid
DNA was studied by agarose gel electrophoresis following the
general method by Huq et al. [43]. Briefly, aliquots (20 L) of a stock
m
and 518A2 cells following treatment with 5
mM of the test
TE buffer solution of pBR322 plasmid DNA (73.5 mg/mL) were
compounds in a lysis buffer containing 50 mM Tris–HCl, pH 7.4,
150 mM NaCl, 1% Triton-X-100 and 1 ꢂ EDTA-free protease inhib-
itor mix (Calbiochem). Cell lysates were incubated at 4 ꢃC for
15 min, centrifuged at 800 g for 10 min and the precipitates were
discarded. Protein concentrations were measured using the Brad-
ford Reagent (Sigma) and bovine serum albumin as a standard. Cell
incubated with different concentrations of the derivatives (0, 5, 10,
and 25
M). Samples were incubated in the dark at 37 ꢃC for 24 h,
then treated with 2 L loading buffer [6 mL glycerol; 20 mg xylene
L of
the resulting mixture was loaded onto a 1% agarose gel and elec-
trophoresis was run in a 0.5 ꢂ TBE buffer (45 mM Tris-base, pH 8.3;
45 mM boric acid; 1.25 mM EDTA) for 4 h at 66 V. The gel was
m
m
cyanole; 1 mL 5 ꢂ TBE (Trisborate-EDTA) buffer; 3 mL H2O]. 18
m
lysates (15 mg cellular protein) were analyzed for caspase activity

1954
K. Effenberger et al. / European Journal of Medicinal Chemistry 45 (2010) 1947–1954
[15] R. Graeser, N. Esser, H. Unger, I. Fichtner, A. Zhu, C. Unger, F. Kratz, Invest. New
Drugs 28 (2010) 14–19.
stained in 0.5 ꢂ TBE buffer containing ethidium bromide (0.5
mg/
mL, Calbiochem) and visualised under UV light (LTF Labortechnik,
Wasserburg/B, Germany).
[16] R. Zucchi, R. Danesi, Curr. Med. Chem. Anti-cancer Agents
151–171.
3 (2003)
[17] G. Bernhardt, B. Biersack, S. Bollwein, R. Schobert, M. Zoldakova, Chem. Biodiv.
5 (8) (2008) 1645–1659.
Acknowledgement
[18] R. Schobert, B. Biersack, A. Dietrich, A. Grotemeier, T. Mu¨ller, B. Kalinowski,
S. Knauer, W. Voigt, R. Paschke, J. Med. Chem. 50 (2007) 1288–1293.
[19] S. Breyer, K. Effenberger, R. Schobert, Chem. Med. Chem. 4 (2009) 761–768.
[20] Y. Wang, L. Li, W. Jiang, Z. Yang, Z. Zhang, Bioorg. Med. Chem. Lett. 16 (2006)
2974–2977.
[21] L.A. Sauer, R.T. Dauchy, D.E. Blask, Cancer Res. 60 (2000) 5289–5295.
[22] U.N. Das, Prostaglandins Leukot. Essent. Fatty Acids 70 (2004) 539–552.
[23] N.K. Narayanan, B.A. Narayanan, B.S. Reddy, Int. J. Oncol. 26 (2005) 785–792.
[24] L.A. Sauer, R.T. Dauchy, D.E. Blask, Biochem. Pharmacol. 61 (2001) 1455–1462.
[25] B. De Roos, Y. Mavrommatis, I.A. Brouwer, Brit. J. Pharmacol. 158 (2009)
413–428.
We thank the Deutsche Forschungsgemeinschaft for financial
support (grant Scho 402/8-2), Ribosepharm GmbH, Gra¨felfing
(Germany) for a free batch of doxorubicin, and Dr. Matthias Ocker
(University Hospital Erlangen) for kindly providing the lumino-
metric caspase kit and the facilities for measuring the Bcl-2 and Bax
mRNA expressions.
[26] M. Huan, S. Zhou, Z. Teng, B. Zhang, X. Liu, J. Wang, Q. Mei, Bioorg. Med. Chem.
Lett. 19 (2009) 2579–2584.
Appendix. Supporting information
[27] D.M. Vigushin, G.K. Poon, A. Boddy, J. English, G.W. Halbert, C. Pagonis,
M. Jarman, R.C. Coombes, Cancer Chemother. Pharmacol. 42 (1998) 111–117.
[28] M.J. Stark, Y.D. Burke, J.H. McKinzie, A.S. Ayoubi, P.L. Crowell, Cancer Lett. 96
(1995) 15–21.
Supplementary data associated with this article can be found in
the online version at doi:10.1016/j.ejmech.2010.01.037.
[29] H. Bailey, G. Ripple, M. Gould, R. Arzoomanian, D. Alberti, C. Feierabend,
K. Simon, K. Morgan, K. Tutsch, M. Pomplun, A. Wahamaki, R. Marnocha,
G. Wilding, Ann. Oncol. 9 (1998) 107.
References
[30] T. Kaneko, D. Willner, I. Monkovı´c, J.O. Knipe, G.R. Braslawsky, R.S. Greenfield,
D.M. Vyas, Bioconjug. Chem. 2 (1991) 133–141.
[31] C.C. Lee, A.T. Cramer, F.C. Szoka Jr., J.M. Fre´chet, Bioconjug. Chem. 17 (2006)
1364–1368.
[32] R.S. Greenfield, T. Kaneko, A. Daues, M.A. Edson, K.A. Fitzgerald, L.J. Olech,
J.A. Grattan, G.L. Spitalny, G.R. Braslawsky, Cancer Res. 50 (1990) 6600–6607.
[33] H.P. Xie, X. Chu, J.H. Jiang, H. Cui, G.L. Shen, R.Q. Yu, Spectrochim. Acta A 59
(2003) 743–749.
[34] K. Barabas, J.A. Sizensky, W.P. Faulk, J. Biol. Chem. 267 (1992) 9437–9442.
[35] S. Wang, E.A. Konorev, S. Kotamraju, J. Joseph, S. Kalivendi, B. Kalyanaraman,
J. Biol. Chem. 279 (2004) 25535–25543.
[36] T. Yan, S. Deng, A. Metzger, U. Go¨dtel-Armbrust, A. Porter, L. Wojnowski, Mol.
Cancer Ther. 8 (2009) 1075–1085.
[37] L. Benimetskaya, K. Ayyanar, N. Kornblum, D. Castanotto, J. Rossi, S. Wu, J. Lai,
B.D. Brown, N. Popova, P. Miller, H. McMicken, Y. Chen, C.A. Stein, Clin. Cancer
Res. 12 (2006) 4940–4948.
[38] E. Bossy-Wetzel, D.D. Newmeyer, D.R. Green, EMBO J. 17 (1998) 37–49.
[39] S. Desagher, A. Osen-Sand, A. Nicholas, R. Eskes, S. Montessuit, J. Cell Biol. 144
(1999) 891–901.
[1] J. Robert, Anthracyclines. in: L.B. Grochow, M.M. Ames (Eds.), A Clinician’s
Guide to Chemotherapy, Pharmacokinetics and Pharmacodynamics. Williams
& Wilkins, Baltimore, 1998, pp. 93–173.
[2] D.A. Gewirtz, Biochem. Pharmacol. 57 (1999) 727–741.
[3] R.C. Young, R.F. Ozols, C.E. Myers, N. Engl. J. Med. 305 (1981) 139–153.
[4] F. Zunino, R. Gambetta, A. Di Marco, Biochem. Pharmacol. 24 (1975) 309–311.
[5] M. Binaschi, M. Bigioni, A. Cipollone, C. Rossi, C. Goso, C.A. Maggi, G. Capranico,
F. Animati, Curr. Med. Chem. Anti-cancer Agents 1 (2001) 113–130.
[6] I.N. Pessah, E.L. Durie, M.J. Schiedt, I. Zimanyi, Mol. Pharmacol. 37 (1990)
503–514.
[7] S.G. Oakes, J.J. Schlager, K.S. Santone, R.T. Abraham, G.J. Powis, J. Pharmacol.
Exp. Ther. 252 (1990) 979–983.
[8] S.S. Bielack, R. Erttmann, B. Kempf-Bielack, K. Winkler, Eur. J. Cancer 32 (1996)
1652–1660.
[9] E. Delpy, S.N. Hatem, N. Andrieu, C. de Vaumas, M. Henaff, C. Ru¨ cker-Martin,
J.P. Jaffre´zou, G. Laurent, T. Levade, J.J. Mercadier, Cardiovasc. Res. 43 (1999)
398–407.
[10] Y.A. Haimovitz-Friedman, R.N. Kolesnick, Z. Fuks, Br. Med. Bull. 53 (1997)
539–553.
[40] G.A.W. Rook, J. Steele, S. Umar, H.M.J. Dockrell, J. Immunol. Methods 82 (1985)
161–167.
[41] W.C. Earnshaw, L.M. Martins, S.C. Kaufmann, Annu. Rev. Biochem. 68 (1999)
383–424.
[42] U.N. Das, M.E. Begin, G. Ells, Y.S. Huang, D.F. Horobin, Biochem. Biophys. Res.
Commun. 145 (1987) 15–24.
[43] M.A. Chowdhury, F. Huq, A. Abdullah, P. Bearle, K. Fisher, J. Inorg. Biochem. 99
(2005) 1098–1112.
[11] Z.S. Yoshimura, Y. Banno, S. Nakashima, K. Takenaka, H. Sakai, Y. Nishimura,
N. Sakai, S. Shimizu, Y. Eguchi, Y. Tsujimoto, Y. Nozawa, J. Biol. Chem. 273
(1998) 6921–6927.
[12] G. Minotti, P. Menna, E. Salvatorelli, G. Cairo, L. Gianni, Pharmacol. Rev. 56
(2004) 185–229.
[13] L.J. Marnett, J.N. Riggins, J.D. West, J. Clin. Invest. 111 (2003) 583–593.
[14] N. Lavignac, J.L. Nicholls, P. Ferruti, R. Duncan, Macromol. Biosci. 9 (2009) 480–
487.