Doxorubicin-antioxidant co-drugs

Article abstract of DOI:10.1016/j.bmcl.2013.07.070

Doxorubicin-antioxidant multitarget compounds 6 and 7 were obtained by combining doxorubicin (DOX) with caffeic and ferulic acids through an ester linkage at C-14. The products were studied in in vitro models of cardiomyocytes and breast cancer cells, characterized by different degrees of resistance to DOX, due to different expressions of ATP binding cassette (ABC) transporters. Compound 7 was found to be less toxic than DOX in cardiomyocytes and to display the same possibly higher toxicity against the resistant breast cancer cells. This result shows that appropriate DOX-antioxidant co-drugs can limit the onset of cardiac damage, a significant side-effect of DOX, without impairing the antitumor activity of the parent antibiotic.

Full text of DOI:10.1016/j.bmcl.2013.07.070

Accepted Manuscript
Doxorubicin-antioxidant co-drugs
Konstantin Chegaev, Chiara Riganti, Barbara Rolando, Loretta Lazzarato, Elena
Gazzano, Stefano Guglielmo, Dario Ghigo, Roberta Fruttero, Alberto Gasco
PII:
S0960-894X(13)00926-8
DOI:
http://dx.doi.org/10.1016/j.bmcl.2013.07.070
BMCL 20735
Reference:
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date:
Revised Date:
Accepted Date:
14 June 2013
26 July 2013
30 July 2013
Please cite this article as: Chegaev, K., Riganti, C., Rolando, B., Lazzarato, L., Gazzano, E., Guglielmo, S., Ghigo,
D., Fruttero, R., Gasco, A., Doxorubicin-antioxidant co-drugs, Bioorganic & Medicinal Chemistry Letters (2013),
doi: http://dx.doi.org/10.1016/j.bmcl.2013.07.070
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Graphical Abstract
Doxorubicin-antioxidant co-drugs
Konstantin Chegaeva, Chiara Rigantib, Barbara Rolandoa,
Loretta Lazzaratoa, Elena Gazzanob, Stefano Guglielmoa,
Dario Ghigob, Roberta Frutteroa,* Alberto Gascoa.

Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com
Doxorubicin-antioxidant co-drugs
Konstantin Chegaeva, Chiara Rigantib, Barbara Rolandoa, Loretta Lazzaratoa, Elena Gazzanob, Stefano
Guglielmoa, Dario Ghigob, Roberta Frutteroa,* Alberto Gascoa.
aDepartment of Science and Drug Technology, University of Torino, via Pietro Giuria 9, 10125, Torino, Italy
bDepartment of Oncology, University of Torino, via Santena 5/bis, 10126, Torino, Italy
ARTICLE INFO
ABSTRACT
Article history:
Received
Revised
Accepted
Available online
Doxorubicin-antioxidant multitarget compounds 6 and 7 were obtained by combining
doxorubicin (DOX) with caffeic and ferulic acids through an ester linkage at C-14. The
products were studied in in vitro models of cardiomyocytes and breast cancer cells,
characterized by different degrees of resistance to DOX, due to different expressions of ATP
binding cassette (ABC) transporters. Compound 7 was found to be less toxic than DOX in
cardiomyocytes and to display the same possibly higher toxicity against the resistant breast
cancer cells. This result shows that appropriate DOX-antioxidant co-drugs can limit the onset
of cardiac damage, a significant side-effect of DOX, without impairing the antitumor activity
of the parent antibiotic.
Keywords:
Doxorubicin
Ferulic Acid
Antioxidant
Cardiotoxicity
ABC transporters
2009 Elsevier Ltd. All rights reserved.
Doxorubicin (DOX) (1, Figure 1) also known as Adriamycin,
is a potent broad-spectrum antineoplastic antibiotic belonging to
the anthracycline family. It is widely used, as single agent or in
combination with other anticancer drugs, in treating a variety of
cancers including solid tumors, soft-tissue sarcomas, lymphomas,
and leukemias.1 DOX displays a number of clinical toxicities, of
which cardiomyopathy is the most important. Two kinds of
cardiomyopathies can occur: an acute form and a chronic,
cumulative dose-related form. The former kind is rarely a serious
problem, while the latter can lead to congestive heart failure that
is unresponsive to digitalis.2 The mortality rate in patients with
congestive heart failure is close to 50%. The classic molecular
mechanisms underlying both the anticancer and the toxic effects
of DOX operate at two distinct levels: by modifying DNA, and
by inducing oxidative stress. Other possible mechanisms include
altered metabolism of Ca++ ions or prostaglandins.3
leads to cell damage, consequent on the alteration of enzymes,
proteins, and DNA, and to lipid peroxidation, with the final
formation of reactive electrophilic aldehydes.4,5 The heart is very
sensitive to oxidative stress, because of its strongly oxidative
metabolism and poor antioxidant defenses. DOX can generate
free radicals in a number of different ways: it is reduced by
various biological systems to a semiquinone free radical giving
•
rise to a number of ROS, including peroxide anions (O2 ),
hydrogen peroxide (H2O2), and hydroxyl radicals (OH·). DOX
activates NAD(P)H oxidases (NPXs), which generate free
radicals; it also produces ROS through a non-enzymatic
mechanism involving iron. Lastly, DOX undergoes metabolic
transformation to doxorubicinol, a secondary alcohol that induces
the release of iron from cytoplasmic aconitase, with consequent
ROS generation.3,6 It is known that the heart is very rich in
mitochondria, which contain a phospholipid called cardiolipin,
for which DOX displays great affinity;7 DOX undergoes redox
cycling at complex I of the mitochondrial electron transport
chain. The final result is cardiomyocyte damage, mainly deriving
from the impairment of mitochondrial functioning.8,9
The use of a variety of natural and synthetic antioxidants to
prevent DOX’s cardiotoxicity has been considered,1,3,10-12 and
combinations of DOX with agent(s) capable of blocking its ROS
mediated cardiotoxicity effect have been investigated. To the best
of our knowledge, no study has yet reported the combination of
DOX and the antioxidant in a single molecule. The use of such
polyvalent molecules can show some advantages compared with
a cocktail of drugs, including a lower risk of drug–drug
interactions, improved compliance by the patient, and a more
predictable pharmacokinetic profile. As a development of our
Figure 1. Doxorubicin (1) and hybrid compounds general structure.
Oxidative stress is the consequence of a disturbed prooxidant /
antioxidant intracellular balance, due to abnormal production of
reactive oxygen (ROS) and nitrogen (RNS) species and/or to
depletion of antioxidant defenses. Continuative oxidative stress

studies on semi-synthetic doxorubicins,13 we here report
preliminary results obtained with the first two DOX hybrids of
general structure 2, in which the antibiotic is linked with selected
antioxidant moieties through appropriate spacers.
induced cytotoxicity (Figure 2A, right panel). By contrast, 6 was
as toxic as DOX in H9c2 cells and MCF7 cells, but was
ineffective against MDA-MB-231 (Figure 2A, left panel).
Interestingly, 7 accumulated in larger amounts than did the parent
drug DOX in both types of breast cancer cells, but not in
cardiomyocytes (Figure 1B, right panel), whereas 6 had the same
rate of accumulation as DOX in all cell lines tested (Figure 2B,
left panel). Since 6 showed neither reduced toxicity/accumulation
in H9c2 cells, nor increased toxicity/accumulation in MCF7 or
MDA-MB-231 cells, it was not investigated further.
The co-drugs 6 and 7 derive from the combination of DOX,
respectively, with caffeic acid, and with ferulic acid, through an
ester linkage at C-14. These two acids belong to the class of
polyphenols, which are important antioxidants widely distributed
in the human diet, particularly in fruits, cereals vegetables and
beverages (coffee, beer, wine, fruit juices). The phenolic acid are
rarely present in the free form, more frequently occurring as
esters.14 The co-drugs were prepared from a mixture of 14-
bromo / 14-chlorodaunorubicin 3 by reaction with the sodium
salt of the appropriate acid (4, 5) in boiling acetone (Scheme 1).
After purification by flash chromatography, the products were
isolated as hydrochlorides.
To clarify the differential accumulation pattern of 7 in
cardiomyocytes and breast cancer cells, expression of the ABC
transporters involved in DOX efflux was analyzed: as shown in
Figure 3A, H9c2 cells had undetectable levels of all transporters,
MCF7 cells showed very low amounts of Pgp and BCRP,
whereas MDA-MB-231 had high expression levels of Pgp,
MRP1 and BCRP. This pattern of expression is in agreement
with the DOX accumulation found, which followed this order:
H9c2 > MCF7 > MDA-MB-231 cells (Figure 2B). In keeping
with these data, the Vmax of DOX efflux was highest in MDA-
MB-231 cells, followed by MCF7 and then H9c2 cells (Figure
3B, upper panel). Of note, the Vmax of 7 was not different from
that of DOX in H9c2 cells, and was reduced in MCF7 cells and
MDA-MB-231 cells (Figure 3B, lower panel, Table 1). The Km
of DOX and 7 were relatively uniform in all three cell types
(Figure 3B, Table 1). Overall, these results suggest that 7 was
effluxed less markedly by the ABC transporters present in breast
cancer cells, leading to its being accumulated to a greater extent
than was DOX, in MCF7 and MDA-MB-231 cells; conversely, 7
was not retained more abundantly than DOX in cardiomyocytes,
which had undetectable levels of ABC transporters.
Scheme 1 i) acetone, reflux.
Compounds 6 and 7 were then evaluated in in vitro models of
cardiomyocytes and breast cancer cells, cell lines that are
characterized by different degrees of resistance to DOX, due to
their different expressions of ATP binding cassette (ABC)
transporters, namely P-glycoprotein (Pgp/ABCB1), multidrug
resistance related protein (MRP1/ABCC1) and breast cancer
resistance protein (BCRP/ABCG2), all of which efflux DOX.
Figure 3. A: Expression of Pgp, MRP1, and BCRP, in H9c2, MCF7 and
MDA-MB-231 cells.17 B: Efflux of 7 from cardiomyocytes and from breast
cancer cells.18
Table 1. Compounds efflux from cardiomyocytes and from
breast cancer cells.18
Cells
H9c2
Km
1.96 ± 0.26
Vmax
11.89 ± 0.53
DOX
MCF7
MDA
H9c2
1.88 ± 0.21
1.35 ± 0.53
2.01 ± 0.18
1.83 ± 0.24
1.25 ± 0.31
18.90 ± 0.29
33.11 ± 1.90
11.70 ± 0.61
11.10 ± 0.55
16.53 ± 0.70
Figure 2. Cytotoxicity and intracellular accumulation of DOX, 6 and 7.15
7
MCF7
MDA
H9c2 cells, which are characterized by several cardiomyocyte-
like properties,16 and the human breast cancer cell lines MCF7
and MDA-MB-231 were chosen to check the in vitro cytotoxicity
of ferulic acid-DOX (7) and caffeic acid-DOX (6) conjugates. 7
was significantly less toxic than DOX in H9c2 cells and retained
the same toxicity as DOX in MCF7 cells; of note, it also induced
cell damage in MDA-MB-231, which were refractory to DOX-
DOX and 7 had the same potency as topoisomerase II
inhibitors (Supplementary, Figure S2) and a similar intracellular
distribution: most of 7 was found within nuclei, a small
percentage in mitochondria (Supplementary, Figure S3). In
nuclear extracts of breast cancer cell lines, the amount of 7 was

higher than that of DOX (Supplementary, Figure S3), in line with
its lower efflux rate from whole cells.
with the same affinity as DOX, but it is transported outside at a
lower rate, probably due to the different sterical and physico-
chemical properties of 7. On the other hand, the different
structure of 7 did not affect other “peculiar” properties of DOX,
such as the predominant intranuclear accumulation and the
effective inhibition of topoisomerase II.
Finally, we checked the effects of 7 on preventing oxidative
stress, and analyzed intracellular levels of ROS, and of
malonyldialdehyde (MDA), a marker of lipid peroxidation and
oxidative damage. Compared with DOX, 7 significantly reduced
ROS and MDA in H9c2 cardiomyocytes; the reduction was
smaller in MCF7 cells. In MDA-MB-231 cells, DOX caused no
increase in ROS or MDA, and 7 did not produce any significant
changes (Figure 4). The intramitochondrial ROS level was
similar for DOX and 7, and followed the same trend of ROS in
whole cells (Supplementary, Figure S4).
Within H9c2 cells, the ferulic acid co-drug 7 was found to
reduce the oxidative stress induced by DOX, a result that is in
accordance with its decreased cytotoxicity. In the DOX-sensitive
breast cancer MCF7 cells, 7 did not show any superior benefit
compared to DOX: although 7 induced less oxidative stress, it
was also effluxed to a lesser extent than was DOX, due to the
lower content of Pgp and BCRP. 7 accumulated more markedly
than DOX in MCF7 cells; this increased accumulation likely
compensates for the decreased oxidative damage, leading to the
same cytotoxicity as has DOX. The greatest benefit of 7 was seen
in DOX-resistant tumor cells, i.e. MDA-MB-231: DOX was
strongly effluxed from these cells, eliminating its cytotoxicity.
Conversely 7, which was effluxed less markedly and reached
sufficient intracellular concentration to induce appreciable
cytotoxic effects. Notably, at this concentration 7 did not elicit
any cell damage on cardiomyocytes.
The most critical limitation on the use of DOX against breast
tumors is the onset of cardiotoxicity.23 This preliminary study
shows that the compound 7, obtained by combining DOX with
ferulic acid, is more effective in resistant breast cancer cell, and
less toxic against cardiomyocytes, making it worthy of future in
vitro and in vivo investigations in DOX-resistant tumors. It may
be speculated that 7 might induce the same cytotoxic effects
against resistant breast tumors, even if administered at lower
doses than DOX: since the chronic form of cardiotoxicity is
dependent on the cumulative dose of DOX,23 the use of 7 instead
of DOX might lead to a reduced risk of cardiac damage in
patients bearing resistant breast tumors.
Figure 4. Effects of 7 on ROS production and lipid peroxidation. A: ROS
levels. B: Lipoperoxidation assessment.19
In this study we show preliminary in vitro data on the efficacy
of DOX conjugated with antioxidants, designed to reduce drug-
induced cardiotoxicity caused by oxidative damages. Oxidative
stress, however, is one of the multiple biochemical mechanisms
by which DOX exerts its anticancer effects:3 thus it is mandatory
to design DOX-antioxidant derivatives that reduce oxidative
stress in cardiomyocytes, but do not lose their cytotoxicity
against tumor cells. Ferulic acid and caffeic acid lipophilic
derivatives have recently been shown to exert cytotoxic effects in
breast cancer cells, including MCF7 and MDA-MB-231 cells;20
moreover, both compounds are known to be antioxidant
molecules.21 In the light of these observations, ferulic acid-DOX
(7) and caffeic acid-DOX (6) derivatives could be suitable co-
drugs, able on one hand to reduce oxidative stress, while on the
other hand retaining DOX’s cytotoxic effects against breast
Acknowledgments
This work was supported by the Italian Association for Cancer
Research (grant: MFAG 11475) and the Italian Ministry for
Universities and Research (grant: RBFR12SOQ1 FIRB 2012).
cancer cells. Indeed,
7 was less toxic than DOX in
References and notes
cardiomyocytes in vitro, whereas its toxicity against breast cancer
cells was equal to or higher than that of the parent drug.
Conversely, 6 maintained the same toxicity profile as DOX. This
difference could be due to the different lipophilicities of these
hybrid compounds, or may be explained by the different rates of
intracellular accumulation of the two semisynthetic DOXs: while
6 had the same intracellular retention as DOX, 7, surprisingly,
accumulated in larger amounts in breast cancer cells.
1. Minotti, G.; Menna, P.; Salvatorelli, E.; Cairo, G.; Gianni, L.
Pharmacol. Rev. 2004, 56, 185, and references therein reported.
2. Chabner, A.; Amrein, P. C.; Druker, B. J.; Michaelson, M. D.;
Mitsiades, C. S.; Goss, P. E.; Ryan, D. P.; Ramachandra, S.;
Richardson, P. G.; Supko, J. G.; Wilson, W. H. Antineoplastic
Agents. In Goodman & Gilman’s, The Pharmacological Basis of
Therapeutic; Brunton, L.; Lazo, J. S.; Parker, K. L., Eds.
McGraw-Hill: New York, Chicago, 2006; 1315.
3. Granados-Principal, S.; Quiles, J. L.; Ramirez-Tortosa, C. L.;
Sanchez-Rovira, P.; Ramirez-Tortosa, M. Food Chem. Toxicol.
2010, 48, 1425, and references therein reported.
4. Eberhardt, M. K. Reactive Oxygen Metabolites: Chemistry and
Medical Consequences; CRC Press: Boca Raton, FL, 2000.
Cytotoxicity against cardiomyocytes and tumor cells is
directly related to the concentration of intracellular DOX, which
in turn is dependent on the ABC transporters Pgp, MRP1 and
BCRP.22 In our models, the accumulation of DOX was lower in
ABC transporter-positive cells, which showed the most marked
DOX efflux. By contrast, the accumulation ratio of 7 versus
DOX was higher in MDA-MB-231 cells, which have high
expressions of Pgp, MRP1 and BCRP, than it was in MCF7 or
H9c2 cells. The decrease in Vmax efflux suggests that 7 is prone to
accumulate markedly within ABC transporter-positive cancer
cells, which are usually resistant to DOX’s toxicity. The absence
of Pgp, MRP1 and BCRP in H9c2 cells explains why 7 was not
accumulated more markedly in cardiomyocytes, nor effluxed to a
lesser extent, than was DOX. According to Km and Vmax values of
DOX and 7, we suggest that the latter binds to ABC transporters
5.
Esterbauer, H.; Schaur, R. J.; Zollner, H. Free Radical Biol. Med.
1991, 11, 81.
6. Simunek, T.; Sterba, M.; Popelova, O.; Adamcova, M.; Hrdina,
R.; Gersl, V. Pharmacol. Rep. 2009, 61, 154.
7. Goormaghtigh, E.; Ruysschaert, J. M. Biophys. Acta 1984, 779,
271.
8.
Berthiaume, J. M.; Wallace, K.B. Cell Biol. Toxicol. 2007, 23, 15.
9. Doroshow J. H. ; Davies, K. J. A. J. Biol. Chem. 1986, 261, 3068.
10. Quiles, J. L.; Huertas, J. R. ; Battino, M.; Mataix, J.; Ramirez-
Tortosa, M. C. Toxicology 2002, 180, 79, and references therein
reported.

11. DeAtley, S. M.; Aksenov, M. Y.; Aksenova, M. V.; Harris, B.;
Hadley, R.; Harper, P. C.; Carney, J. M.; Butterfield, D. A. Cancer
Lett. 1999, 136, 41.
12. Hideg, K.; Kalai, T. Cardiovasc. Toxicol. 2007, 7, 160.
13. a) Chegaev, K.; Riganti, C.; Lazzarato, L.; Rolando, B.;
Guglielmo, S.; Campia, I.; Fruttero, R.; Bosia, A.; Gasco, A. ACS
Med. Chem. Lett. 2011, 2(7), 494; b) Riganti, C.; Rolando, B.;
Kopecka, J.; Campia, I.; Chegaev, K.; Lazzarato, L.; Federico, A.;
Fruttero, R.; Ghigo, D. Mol. Pharmaceutics. 2012, 10(1), 161.
14. Poquet, L.; Clifford, M.N.; Williamson, G. Bioavailability of
flavanols and phenolic acids. In Plant Phenolics and Human
Health; Fraga C.G., Ed.; Wiley: New Jersey, 2010, pp 51.
15. Cells were incubated for 24 h in fresh medium (ctrl) or with 5
µmol/L DOX, 6 and 7. A: LDH release was evaluated in duplicate
in the cell culture supernatant and in the cell lysate, and the former
was expressed as percentage of the total LDH activity
(extracellular + intracellular). Data are presented as means ± SD
(n = 4). Vs. ctrl: *p < 0.01; 7 vs. DOX: °p < 0.005. B: The amount
of intracellular DOX was measured in duplicate in cell lysates.
Data are presented as means ± SD (n = 3). 7 vs. DOX: * p < 0.05.
16. Hescheler, J.; Meyer, R.; Plant, S.; Krautwurst, D.; Rosenthal, W.;
Schultz, G. Circ. Res. 1991, 69, 1476.
17. Cells were lysed and the whole cell lysate was subjected to
Western blotting. Expression of actin, as product of a
housekeeping gene, was used as control for equal protein loading.
The figure is representative of three experiments with similar
results.
18. Cells were incubated for 20 min with increasing concentrations (0-
100 μmol/L) of DOX or 7, then washed and tested for intracellular
drug content. The procedure was repeated on a second series of
dishes, incubated in the same experimental conditions and
analysed after a further 10 minutes. Measurements (n = 3) were in
duplicate and data (means ± SD) are presented as rate of drug
efflux (dc/dt) plotted versus initial drug concentration. Vmax
(nmol/min/mg cell proteins) and Km (nmol/mg cell proteins) were
calculated using the Enzfitter software package.
19. Cells were incubated for 24 h in fresh medium (ctrl), or in 5
µmol/L DOX or 7. A: ROS levels. Cells were incubated with the
ROS-sensitive probe DCFDA-AM, then analysed in duplicate.
Data are presented as means ± SD (n = 3). Vs. ctrl: *p < 0.001; 7
vs. DOX: °p < 0.005. B: Lipoperoxidation assessment. The
amount of MDA was measured in triplicate in the cell lysates.
Data are presented as means ± SD (n = 3). Vs. ctrl: *p < 0.05; 7 vs.
DOX: °p < 0.01.
20. Serafim, T.L.; Carvalho, F.S.; Marques, M.P.; Calheiros, R.; Silva,
T.; Garrido, J.; Milhazes, N.; Borges, F.; Roleira, F.; Silva, E.T.;
Holy, J.; Oliveira, P.J. Chem. Res. Toxicol. 2011, 24,763.
21. Kancheva, V.D.; Saso, L.; Angelova, S.E.; Foti, M.C.; Slavova-
Kasakova, A.; Daquino, C.; Enchev, V.; Firuzi, O.; Nechev, J.
Biochimie 2012, 94, 403.
22. Gottesman, M.M.; Fojo, T.; Bates, S.E. Nat. Rev. Cancer 2002, 2,
48.
23. Leonard, R. C. F.; Williams, S.; Tulpule, A.; Levine, A. M.;
Oliveros, S. Breast 2009, 18, 218.
Supplementary Material
Supplementary data associated with this article (analytical
data of synthesized compounds and experimental details of
synthesis and additional biological tests) can be found in the on-
line version.