combination of conventional chemotherapeutics with...

Small Molecule Therapeutics

Combination of Conventional Chemotherapeutics withRedox-Active Cerium Oxide Nanoparticles—A Novel Aspectin Cancer Therapy

Maren Sack1, Lirija Alili1, Elif Karaman1, Soumen Das2, Ankur Gupta2, Sudipta Seal2, and Peter Brenneisen1

AbstractNanotechnology is becoming an important field of biomedical and clinical research and the application of

nanoparticles in disease may offer promising advances in treatment of many diseases, especially cancer.

Malignant melanoma is one of the most aggressive forms of cancer and its incidence is rapidly increasing.

Redox-active cerium oxide nanoparticles (CNP) are known to exhibit significant antitumor activity in cells

derived fromhuman skin tumors in vitro and in vivo, whereas CNP is nontoxic and beyond that even protective

(antioxidative) in normal, healthy cells of the skin. As the application of conventional chemotherapeutics is

associated with harmful side effects on healthy cells and tissues, the clinical use is restricted. In this study, we

addressed the question ofwhetherCNP supplement a classical chemotherapy, thereby enhancing its efficiency

without additional damage to normal cells. The anthracycline doxorubicin, one of the most effective cancer

drugs, was chosen as reference for a classical chemotherapeutic agent in this study. Herein, we show that CNP

enhance the antitumor activity of doxorubicin in human melanoma cells. Synergistic effects on cytotoxicity,

reactive oxygen species generation, and oxidative damage in tumor cells were observed after co-incubation. In

contrast to doxorubicin, CNP do not cause DNA damage and even protect human dermal fibroblasts from

doxorubicin-induced cytotoxicity. A combination of classical chemotherapeutics with nongenotoxic but

antitumor active CNP may provide a new strategy against cancer by improving therapeutic outcome and

benefit for patients. Mol Cancer Ther; 13(7); 1740–9. �2014 AACR.

IntroductionIn recent years, nanotechnology has become an impor-

tant field of biomedical and clinical research formingthe subject area of nanomedicine. Application ofnanoparticles in disease offers promising possibilities fordiagnostics (nanoimaging) and drug delivery systems(nanocarrier) as well as the pharmaceutical use of nano-particles itself (nanopharmaceuticals; refs. 1–3). Nanopar-ticle applications offer advances in treatment of manydiseases especially cancer, which is the second mostcommon cause of death in the United States and Europefollowing cardiovascular diseases (4). Malignant melano-ma is one of the most aggressive types of cancer. Earlystages ofmelanoma canbe cured by surgery; however, the

treatment of metastasizing forms is still difficult and thesurvival rates of 5% are really poor. The incidence of skincancer is rapidly growing, suggesting a doubling of therate each decade. Hence, more effective therapies withless harmful effects are required (5, 6).

Recent studies have shown that redox-active ceriumoxide nanoparticles (CNP) exhibit cytotoxic and anti-invasive effects in several cancer cells (7, 8) and are ableto sensitize tumor cells to radiation, while protecting thenormal cells in the tumor surrounding stroma (9–11).

The use of dextran-coated and oxygen vacancies con-tainingCNPwith a size of about 5 nm indiameter resultedin cell killing of the squamous skin carcinoma cell lineSCL-1 and the human melanoma cell line A375 andlowered the invasive capacity (12). In a xenograft mousemodel with A375 melanoma cells, tumor growth wassignificantly inhibited by CNP, which was the first studythat showed an antitumor activity in vivo (13). The cyto-toxic effect of CNP in tumor cells was mediated by aprooxidant activity of CNP, which significantly increasedthe reactive oxygen species (ROS) level, especially H2O2,and thereby leading to apoptosis of tumor cells. In con-trast, in normal cells (e.g., stromal fibroblasts), CNPexerted antioxidant properties. This bifunctional modeof action is mediated by a pH-dependent redox activity ofCNP (9). Tumor cells show an increased glycolysis rate("Warburg effect") compared with healthy cells resulting

Authors' Affiliations: 1Institute of Biochemistry & Molecular Biology I,Medical Faculty, Heinrich-Heine-University, D€uesseldorf, Germany; and2Advanced Materials Processing and Analysis Center, Nanoscience andTechnology Center (NSTC), Mechanical, Materials, and Aerospace Engi-neering (MMAE), University of Central Florida, Orlando, Florida

M. Sack and L. Alili contributed equally to this work.

Corresponding Author: Maren Sack, University of D€uesseldorf, Medicalfaculty, Heinrich-Heine-University, D€usseldorf 40225, Germany. Phone:49-211-81-12834; Fax: 49-211-81-12833; E-mail:[email protected]

doi: 10.1158/1535-7163.MCT-13-0950

�2014 American Association for Cancer Research.

MolecularCancer

Therapeutics

Mol Cancer Ther; 13(7) July 20141740

on June 8, 2018. © 2014 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst May 13, 2014; DOI: 10.1158/1535-7163.MCT-13-0950

http://mct.aacrjournals.org/

in high production of lactate and a slight acidification oftumor cells and the extracellular space (14–16). This dif-ference in pH of tumor cells and normal cells is thedecisive factor that makes CNP either working as a pro-or antioxidant (13).Taken together, these findings implicate a promising

potential for a clinical use ofCNP in cancer therapy. In thisstudy, the question was addressed of whether CNP couldsupplement a classical chemotherapeutical approach.Because of the proapoptotic and anti-invasive propertiesin tumor cells, CNP could enhance the therapeutic out-come of chemotherapies. In contrast to conventional che-motherapeutics, which are often accompanied with adamage of healthy cells and tissues (17, 18), CNP isnontoxic and even protective in stromal cells of the skin(12).The anthracycline doxorubicin, an "evergreen" of the

chemotherapeutic agents, was chosen as a reference sub-stance for this study. It belongs to themost effective cancerdrugs ever developed, however the clinical use of doxo-rubicin is restricted because of its diverse toxic effects inhealthy cells and tissues (19, 20). The antitumor activity ofdoxorubicin is mainly mediated by several interactionswith genomic DNA leading to DNA damage, cell-cyclearrest and, subsequently, to apoptosis. Furthermore,doxorubicin is known to generate ROS via a redox cyclingprocess, thus contributing to its toxicity (19). Herein, apotential synergistic effect of CNP enhancing the antitu-mor activity of doxorubicin was investigated in humanmelanoma cells. As CNP is antioxidative and protectiveagainst exogenous noxes in normal cells, this nanoparticlemay lower the side effects of doxorubicin, therebyimproving the therapeutic outcome. In addition, theimpact of CNP to protect human dermal fibroblast (HDF),being the most frequently stromal cells of the skin, fromdoxorubicin-induced toxicity was assessed.

Materials and MethodsCell culture medium Dulbecco’s modified Eagle medi-

um (DMEM)was purchased from Invitrogen and the fetalcalf serum (FCS gold) from Biochrom. All chemicals,including protease as well as phosphatase inhibitorcocktail 1 and 2, were obtained from Sigma or MerckBiosciences unless otherwise stated. The Protein AssayKit (Bio-RadDC, detergent compatible)was fromBio-RadLaboratories. The Oxyblot Protein Oxidation DetectionKit was from Millipore, whereas the 20,70-dichlorofluor-escin diacetate was provided from Sigma. The enhancedchemiluminescence system (SuperSignal West Pico/Femto Maximum Sensitivity Substrate) was supplied byPierce.Monoclonalmouse antibody raised against humanglyceraldehyde 3-phosphate dehydrogenase (GAPDH)and monoclonal mouse antibody raised against humana-tubulin were supplied by Sigma. The monoclonal anti-body raised against and poly(ADP-ribose) polymerase(PARP) was obtained from Cell signaling. The polyclonalrabbit a-hapten antibody directed against oxidized thiolgroups (sulfenic acid) was a gift from Kate S. Carrol’s

group from TSRI (Department of Chemistry, Jupiter,Florida; ref. 21). The following secondary antibodies wereused: polyclonal horseradishperoxidase (HRP)–conjugat-ed rabbit anti-mouse IgG antibody (DAKO) and goat anti-rabbit immunoglobulin G antibodies were from Dianova.Doxorubicin was obtained from Sigma and dissolved indimethyl sulfoxide (DMSO; 0.25% final concentration).

Cell cultureThe human malignant melanoma cell line A375, orig-

inally derived from a 54-year-old woman, was purchasedfrom ATCC (22). HDFs were established by outgrowthfrom foreskin biopsies of healthy human donors with anage of 3 to 6 years. Cells were used in passages 2 to 12,corresponding to cumulative population-doubling levelsof 3 to 27 (23). Human melanoma cells and HDFs werecultured as described (24).

Cell viabilityThe cell viabilitywasmeasured by 3-(4,5-Dimethylthia-

zol-2-yl)-2,5-diphenyltetrazoliumbromid bromide assay(MTT; ref. 25). The reduction of MTT (Sigma) by mito-chondrial dehydrogenases to formazan indicates themet-abolic activity of cells and is an indicator of cellularviability.

Briefly, serum-free medium containing MTT (0.5mg/mL) was added to the cells after incubation withdifferent concentration of CNP or doxorubicin. Afterincubation with MTT cells were washed with PBS andlysed in DMSO. The formation of the blue formazan wasmeasured at 570 nm. The results were presented as per-centage of untreated controls, which were set at 100%.

Synthesis of CNPCNPs were synthesized in dextran (molecular weight:

1,000 Da) using previously described methods (26). Brief-ly, stoichiometric amounts of dextran were at first dis-solved in deionized water followed by cerium nitratehexahydrate. The solution was stirred for 2 hours fol-lowed by addition of ammonium hydroxide (30%, w/w).The pH of the solution was kept below 9.5 to avoidprecipitation of cerium hydroxide. At a final concentra-tion of 150 mmol/L CNP in DMEM, the cells wereincubated in 0.9% ammonium hydroxide. At this concen-tration, ammonium hydroxide belongs to the GenerallyRecognized As Safe (GRAS) substances as suggested bythe Food and Drug Administration (FDA). The resultingdextran-coated CNP were analyzed using UV-visiblespectroscopy for determining the oxidation state of nano-particles and transmission electron microscopy for parti-cle size.

Synthesis of FITC-conjugated CNPThe FITC tagged dextran-coatedCNPwere prepared as

described earlier (13). Briefly, dextran coating on thesurface of CNPs were oxidized with 10 mmol/L sodiumperiodate. Oxidized dextran coated CNPs were then dia-lyzed extensively against distilled water to remove

Redox-Active Cerium Oxide Nanoparticles in Cancer Therapy

www.aacrjournals.org Mol Cancer Ther; 13(7) July 2014 1741




any trace amount of sodium periodate. Then aminationbetween oxidized dextran and FITC were carried out inbicarbonate buffer at pH 8.5. Then, FITC-conjugatednanoparticles were washed several times with distilledwater to remove any free FITC. Finally, nanoparticleswere reconstituted using distilled water.

Cellular uptake of nanoparticlesHumanmelanoma cells inDMEMwere treatedwith 150

mmol/LFITC-labeledCNP for 4 hours or untreated. There-after, cells were washed twice with PBS and fixed withmethanol. ProLong Gold (Invitrogen) was used for visu-alization, a reagent that simultaneously stains the nucleiwithDAPI. Thefluorescencemicroscopic examinationwasdone with a Zeiss Axiovert 100TV and the documentationwith a digital camera system (Hamamatsu C4742-95).

SDS-PAGE and Western blottingSodium dodecyl sulfate-polyacrylamide gel electro-

phorese (SDS-PAGE) was performed according to thestandard protocols published elsewhere (27) with minormodifications. Briefly, cells were lysed after incubationwith CNP or doxorubicin in 1% SDS with 1:1,000 pro-tease inhibitor cocktail (Sigma). After sonication, theprotein concentration was determined by using a mod-ified Lowry method (Bio-Rad DC). SDS-PAGE samplebuffer (1.5M Tris-HCl pH 6.8, 6 mL 20% SDS, 30 mLglycerol, 15 mL b-mercaptoethanol, and 1.8 mg bromo-phenol blue) was added, and after heating, the samples(20–30 mg total protein/lane) were applied to 12% (w/v)SDS-polyacrylamide gels. After electroblotting, immu-nodetection was carried out (1:1,000 dilution of primaryantibodies; 1:20,000 dilution of anti-mouse/rabbit anti-body conjugated to HRP). Antigen–antibody complexeswere visualized by an enhanced chemiluminescencesystem. a-tubulin or GAPDH was used as internalcontrol for equal loading.

Determination of oxidized (carbonylated) proteinsA375 melanoma cells were grown to subconfluence on

tissue culture dishes. After removal of serum-containingmedium, cells were cultured in DMEM supplementedwith 0.5%FCS and either untreated or treated for differenttime periods with 150 mmol/L CNP nanoparticles. Aspositive control, the cells were treated with 1 mmol/LH2O2. Thereafter, cells were lysed and carbonyl groups ofoxidized proteins were detected with the OxyBlotTMProtein Oxidation Detection Kit (Millipore) according tothe manufacturer‘s protocol. Briefly, the protein concen-tration was determined by using amodified Lowrymeth-od (Bio-Rad DC). Ten mL of the whole cell lysates withequalizedprotein concentrationswere incubatedwith 2,4-dinitrophenyl (DNP) hydrazine to form the DNP hydra-zone derivatives. Labeled proteins were separatedby SDS-PAGE and immunostained using rabbit anti-DNPantiserum (1:150) and goat anti-rabbit IgG conjugatedto HRP (1:300). Blots were developed by enhancedchemiluminescence.

Determination of oxidized thiol groups (sulfenicacids)

A375 melanoma cells were grown to subconfluenceon tissue culture dishes. After removal of serum-con-taining medium, cells were cultured in 0.5% FCS con-taining medium for different incubation times withCNP and doxorubicin. In the last 2 hours of incubation,10 mmol/L of the diketone dimedone was added (Sig-ma). As positive control, the cells were co-incubatedwith 1 mmol/L H2O2 and 10 mmol/L dimedone for2 hours. Cells were harvested, washed with PBS andlysed. Then, Western blot analysis was performed withthe a-hapten antibody directed against oxidized SHgroups (1:1,000; ref. 21).

Immunochemical stainingA375 melanoma cells were grown to subconfluence on

tissue culture dishes containing cover slips and treatedwith CNP or doxorubicin alone as well as in combinationwith both agents.After incubation cellswerefixeddirectlywith methanol for 10 minutes at �20�C, incubated withblocking solution for 1 hour and treated with a specifica-hapten (1:1,000) antibody directed against oxidized SHgroups overnight at 4�C. The secondary antibody (AlexaFluor 556; Invitrogen) was applied to the cells for 45minutes at 37�C in the dark. After removal of the second-ary antibody, coverslipswere attachedwith ProLong goldantifade reagent (Invitrogen) on microscope slides. Thesamples were dried for at least 12 hours and stored at 4�C.Samples were analyzed by fluorescence microscopy(AxioVert 100TV; Zeiss) using aHamamatsuDigital Cam-era C4742-95 and AquaCosmos version 1.2 software(Hamamatsu Photonics Deutschland GmbH). Pictures ofrandomly selected areas were taken for each sample.

Preparation of nuclear and cytoplasmic extractsSeparation of nuclear and cytoplasmic extracts of A375

melanoma cells was performedwith the NE-PERNuclearand Cytoplasmic Extraction Reagents (Themo Scientific),following the manufacture’s protocol. Protein concentra-tions of the cell lysate were assessed by using a modifiedLowry method (Bio-Rad DC) and adjusted for equal gelloading.

Comet assayThe alkaline comet assay (single-cell gel electrophore-

sis) was used to measure DNA single- and double-strandbreaks together with alkali-labile sites (28). Cells wereplated in dishes and incubated with CNP or doxorubicin.Cells were harvested immediately after incubation, cen-trifuged, and suspended in 200 mL low-melting pointagarose and kept at 37�C. The suspensionwas transferredto prepared microscope slides containing a layer of 10%agarose and then cooled for 4 minutes at 4�C. Coverslipswere gently dropped off and the microscope slides wereplaced overnight at 4�C in lysis buffer (2.5 M sodiumchloride, 100 mmol/L EDTA, 10 mmol/L Tris-Base, sodi-um hydroxide) to lyse cells and enable DNA unfolding.

Sack et al.

Mol Cancer Ther; 13(7) July 2014 Molecular Cancer Therapeutics1742




Slideswerewashedwithwater andplaced on ahorizontalgel electrophoresis chamber, which was filled with high-pH electrophoresis buffer (10 N sodium hydroxide, 200mmol/L EDTA) to submerge the slides. Slides were keptin the buffer for 25 minutes to denature the DNA beforeelectrophoresis. Electrophoresis was performed for 25minutes at 25 V and 300 mA (Bio-Rad; PowerPacHC).After electrophoresis, the slideswerewashed 3 timeswithneutralizing buffer (0.4 M Tris-Base, pH 7.5). For finalfixation, the slides were kept in ethanol (80%) for 5minutes and then dried overnight. After ethidium bro-mide (Molecular Probes, Invitrogen) staining, the cellswere subjected to fluorescencemicroscopy. DNAdamagewas evaluated bymeasuring the comet area in pixel (headand tail). At least 40 stained comets were selected ran-domly and analyzed with CometScore (TriTek Corp.).

Statistical analysisMeans were calculated from at least 3 independent

experiments, and error bars represent standard error ofthe mean (SEM). Analysis of statistical significance wasperformedby Student t test orANOVAwith �P� 0.05, ��P� 0.01, and ��P � 0.001 as levels of significance.

ResultsUptake and cellular distribution of CNPCNP showed a cytotoxic effect in tumor cells, which is

mediated via apoptosis (13). Depending on size andcharge, nanoparticles are internalized by cells or bind tosurface molecules mediating their effects via receptorsignaling (29). To elucidate the uptake and distributionof CNP in the A375 melanoma cell line, cells wereincubated with FITC-labeled CNPs (CNP-FITC) at a con-centration of 150 mmol/L as described earlier (13) andanalyzed by fluorescence microscopy at different timepoints. An uptake of CNP-FITC was confirmed after 4hours of incubation (Fig. 1). Fluorescence was observedprimarily in the cytosol of the cells with an accumulationin the perinuclear region, but not in the nuclei. In addition,24and48hours after treatment, thefluorescence of labeledCNP was merely detected in the cytosol of the cells (datanot shown), indicating that the particles are not able to

pass the nuclear membrane even after longer incubationtimes.

CNP versus doxorubicin cytotoxicityThe anthracycline doxorubicin belongs to the most

effective cancer drugs ever developed. The antitumoractivity of doxorubicin is based on several interactionswith DNA resulting in DNA damage, inhibition of DNAreplication, and consequently apoptosis. The toxicity ofdoxorubicin is not restricted to cancer cells, even healthy(stromal) cells are affected by doxorubicin treatment aswell (19). Thus, a therapy with doxorubicin is alwaysassociated with adverse side effects on healthy cells andwith the risk of the developing secondary cancer (30).CNPswere shown to kill tumor cells,while beingnontoxicfor stromal cells of the skin, likefibroblasts andendothelialcells. Furthermore, CNP even showed protective effectsagainst exogenous prooxidants in stromal cells (12). In thisstudy, the effect of CNP and doxorubicin on viability ofhuman melanoma cells (A375) and HDFs was evaluatedby using theMTT assay. The results (Fig. 2A) showed thatdoxorubicin exerted strong cytotoxic effects in A375 cellsat very low concentration and after short times of incu-bation. After 24-hour treatment with 0.5 mmol/L doxoru-bicin, the cell viability was decreased to about 40% to 50%in A375 compared with the untreated control, which wasset at 100%. The cytotoxicity of doxorubicin was increas-ing with incubation time, at 72-hour posttreatment with0.5 mmol/L no cells survived (data not shown). Doxoru-bicin exhibited stronger toxic effects in A375 than CNP.The cell viability was decreased in A375 cells after treat-ment with 150 mmol/L CNP to approximately 55% andwith 300 mmol/L CNP to 50% after 96 hours.

Moreover, in this study the question was addressed,whether there is a synergistic effect of CNP and doxoru-bicin on cytotoxicity in tumor cells. Therefore, A375 mel-anoma cells were incubated with 300 mmol/L CNP for 48hours or 0.5 mmol/L doxorubicin for 24 hours alone or incombination. Treatment of melanoma cells with 300mmol/L CNP for 48 hours resulted in a decrease of cellviability to 88% compared with the untreated control.After co-incubation with CNP and doxorubicin, cell

Figure 1. Cytosolic distribution ofCNP in human melanoma cell lineA375. To study the uptake andcellular distribution of CNP, A375cells were incubated with150 mmol/L fluorescein-isothiocyanate (FITC)–labeled CNPfor 4 hours and were analyzed byfluorescence microscopy. Inaddition, the nuclei were stainedwith DAPI. CNP were ubiquitouslydistributed in the cytosol. A, FITCfluorescence of labeled CNP; B,merge of FITC and DAPIfluorescence.






viabilitywas decreased to 20% comparedwith 40% to 50%after incubation with doxorubicin alone (Fig. 2B). Theseresults indicated a synergistic effect of CNP and doxoru-bicin on cytotoxicity in melanoma cells. By contrast, CNPshowed no cytotoxicity in HDFs. Doxorubicin showedless toxicity in HDF compared with melanoma cells. Aconcentration of 0.5 mmol/L doxorubicin did not signif-icantly lower the viability of HDF, whereas 25 mmol/Ldoxorubicin decreased the cell viability ofHDF to approx-imately 60% to 70% compared with the untreated control,which was set at 100% (Fig. 2C). Pre-incubation with 150mmol/L CNP 24 hours before doxorubicin treatment (25mmol/L) abrogated the cytotoxic effect of doxorubicin inHDF. The cell viability was increased after co-incubationto around 100% compared with cells that were treatedwithdoxorubicin alone. Thesedatademonstrate thatCNPmay protect HDFs from toxicity of doxorubicin (Fig. 2D).Therefore, a potential therapeutical approach based on acombination of low concentration of doxorubicin (to min-imize side effects) together with CNP (which may protectstromal cells) could be a novel tool to effectively kill tumorcells in a time-dependent manner.

ROS productionIn previous studies, CNP treatment resulted in ROS

formation in A375 melanoma cells as well as in severalother tumor cell lines (7, 8, 13, 31). Besides its directinhibitory effects on DNA replication, doxorubicin isknown to generate ROS via redox cycling. A one-electronaddition to the quinone moiety of doxorubicin results inthe formation of a semiquinone that reacts back to thequinone form thereby reducing O2 to superoxide (O2

.�;ref. 19). In this study, the questions were addressed ofwhether (i) redox-active CNP have a similar impact togenerate ROS as the classical chemotherapeutic doxoru-bicin, and (ii) a co-incubation of both substances results ina synergistic effect in context of ROS production.

The fluorescent probe H2DCF-DA was used to detectintracellular ROS. Therefore, cells were loaded with H2

DCF-DA and then incubatedwith 150 mmol/L CNP or 0.5mmol/L doxorubicin alone or co-incubated for 1.5 hours.Directly after addition of the substances, the fluorescencewas measured every 5 minutes. Incubation with CNP aswell as incubationwith doxorubicin caused an increase inthe ROS level of around 20% within 1.5 hours comparedwith the untreated control. Co-incubation of CNP anddoxorubicin resulted in an even higher ROS level, indi-cating a synergistic effect in ROS generation (Fig. 3). CNPand doxorubicin together increased the intracellular ROSlevel by 36%.

Oxidative damage of proteinsProteins are one of the major targets of ROS. Oxidative

modifications of proteins can influence the biochemicalfunctionality and activity of enzymes and transcriptionfactors (32). Sulfenic acids are a specific oxidation productof thiol groups of cysteins in protein side chains, whichsubsequently may be oxidized to sulfinic and sulfonic

Figure 2. Cytotoxicity: CNP vs. doxorubicin (DOX). Effects on cell viabilitywere assessed byMTT assay. A375melanoma cells were incubatedwithdifferent concentrations of doxorubicin for 24 hours andCNP for 96hours(A). Co-incubation with 300 mmol/L CNP for 48 hours and 0.5 mmol/Ldoxorubicin for 24 hours showed synergistic effects on cytotoxicity (B).HDFs were incubated with different concentrations of doxorubicin for 24hours and CNP for 96 hours (C). HDF were treated with 150 mmol/L CNPfor 48 hours and with 25 mmol/L doxorubicin for 24 hours as well as incombination (D). The percentage of cell viability of the untreated control,which was set on 100%, is presented. ��, P < 0.001; ��, P < 0.01;�, P < 0.05 (ANOVA, Dunnett test). Data are presented as means� SEM.

Sack et al.





acids (21). Besides thiol oxidation another oxidative mod-ification of proteins is the introduction of carbonyl groupsinto several amino acids (i.e., lysine, proline, histidine,arginine). The carbonyl content is considered as the mostgeneral andwell usedbiomarker for irreversible oxidativedamage (33). To elucidate the prooxidative impact of CNPand doxorubicin more in detail, the oxidative damage ofproteins was investigated via detection of sulfenic acidsand carbonyls in proteins. Thus, cellswere incubatedwith150 mmol/L CNP or 0.5 mmol/L doxorubicin as well as incombination for 2, 4, and 24 hours.Carbonyl groups were detected via derivatization with

dinitrophenyl hdyrazine (DNPH) using the OxyblotDetection Kit (Millipore). CNP as well as doxorubicintreatment significantly increased the carbonyl amountcomparedwith untreated controls, whereasCNP inducedhigher carbonyl content increasing with incubation time(Fig. 4). The highest amount of carbonylated proteins wasobserved after a co-incubation of CNP and Doxorubin. Incontrast to the thiol oxidation, the formation of carbony-lated proteins is an irreversible damage accumulatingover time, exclusively seen with CNP.For detection of sulfenic acids, Western blot analysis

and immunochemical stainings were carried out using aspecific a-hapten antibody raised against the stablethioether product of sulfenic acid and the cell-permeable

nucleophilic diketone dimedone (21), which was addedto the cells during the last 2 hours of incubation. TheWestern blot analysis showed that CNP as well asdoxorubicin increased the amount of oxidized thiols(Fig. 5). Compared with the untreated control, doxorubi-cin-treated cells showed a 2-fold increase of sulfenic acids,whereas treatmentwithCNPs resulted in a 3-fold increaseof sulfenic acids after a 2-hour incubation. However, thelevels of oxidized thiols were decreasing with increasingincubation times, presumably a consequence of gutathio-nylation, a mechanism by which the sulfenic acids arereduced again to thiols. Treatment with CNP and doxo-rubicin in combination resulted by tendency in a higheramount of sulfenic acid compared with the single sub-stances showing again a synergistic effect. The highestcontent of sulfenic acids, a 6-fold increase compared withthe control, was detected after 2-hour co-incubation withCNP and doxorubicin (Fig. 5).

To localize the oxidized thiols in the cell, immunochem-ical stainings were performed. Figure 6A shows A375melanoma cells, which were stained with DAPI and thea-hapten antibody. Themelanoma cells treatedwith CNPalone showed a cytosolic andperinuclear staining,where-as in the nuclei a very faint fluorescence could be seen.These data suggest that ROS generation and consequentlythiol oxidation byCNPoccur primarily in the cytosol. Thisobservation matches with the uptake of CNP, showingthat CNP is distributed in the cytosol but not in thenucleus (Fig. 1). By contrast, doxorubicin-treated cellsalso showed a thiol oxidation in the nuclei, a result thatis in line with the well-described property of doxorubicinto bind to DNA. Melanoma cells that were co-incubatedwith doxorubicin andCNP revealed a strong fluorescence

Figure 3. Synergistic effect of CNP and doxorubicin (DOX) on ROSgeneration. To detect intracellular ROS, A375 melanoma cells wereloaded with H2DCF-DA and subsequently treated with 150 mmol/L CNPand 0.5 mmol/L doxorubicin (DOX) alone or in combination. Immediatelyafter adding of the substances the fluorescence was measured for1.5 hours. Presented is 1 of 3 independent experiments.

Figure 4. Irreversible protein damage after co-incubation with CNP anddoxorubicin. Carbonyl contents, as marker for irreversible proteindamage, in A375 cells were determined by Oxyblot analysis. Cells weretreated with 0.5 mmol/L doxorubicin for 2 hours or with 150 mmol/L CNPfor 2, 4, or 24 hours, aswell as in combination. H2O2was used as positivecontrol and GAPDH was used as loading control. Three independentexperiments were performed. CNP and doxorubicin both inducedcarbonyl formation, whereas cotreatment showed synergistic effects inA375.






in the cytosol as well as in the nuclei, confirming thesynergistic effect on thiol oxidation observed above(Fig. 5).

To study the localization of thiol oxidation more indetail, cytoplasmic and nuclear extracts were prepared.Western blot analysis of this extracts also showed thatCNP treatment caused thiol oxidation mainly in the cyto-sol (3-fold increase compared with the untreated cyto-plasmic control) and a smaller amount in the nucleus,whereas doxorubicin caused thiol oxidation in the cytosolas well as in the nucleus (2.5-fold increase of the cyto-plasmic fraction and about 2-fold increase of the nuclearfraction compared with the controls; Fig. 6B). After co-incubationwith both agents, thiol oxidationwas observedin cytosol and nucleus, whereas a synergistic effect wasseen only in the cytosol.

In summary, these results indicated a prooxidativeactivity of both CNP and doxorubicin, whichwas boostedby co-incubation of that substances and resulting in oxi-dative damage of proteins.

Genotoxicity of CNP and doxorubicinExcessive ROS levels result in damage of macromo-

lecules, for example DNA (34). Previous studieshave shown that CNP induce ROS-dependent intrinsicapoptosis in melanoma cells (13). Because of the

Figure 5. Synergistic effect of CNP and doxorubicin (DOX) on thioloxidation. Sulfenic acid formation in A375 cells was analyzed byWesternblot analysis.Cellswere treatedwith0.5mmol/Ldoxorubicin for 2hours orwith 150 mmol/L CNP for 2, 4, or 24 h, as well as in combination.H2O2 was used as positive control. For the last 2 hours of incubationdimedone (10 mmol/L) was added to the cells. GAPDH was used asloading control. The figure represents 1 of 3 independent experimentsthat were analyzed by densitometry with ImageJ. The x-fold increaseversus untreated controls is presented.

Figure 6. Localization of oxidated thiols in A375 melanoma cells. A, afterincubationwith 150mmol/LCNPand5mmol/L doxorubicin (DOX) alone orin combination, cells were fixed for an immunochemical staining wasperformed by using the a-hapten antibody raised against the oxidationproduct of sulfenic acid and dimedone, which was added to the cells forthe last 2 hours of incubation. In addition, nuclei were stained with DAPI.Presented is 1 of 3 independent experiments. B, Western blot analysiswas carried out with cytoplasmic and nuclear cell extracts frommelanoma cells that were treated with 0.5 mmol/L doxorubicin or 150mmol/L CNP for 2 hours, as well as co-incubated. CNP caused thioloxidation mainly in the cytosol, whereas doxorubicin showed strongformation of sulfenic acids in the nuclei as well. Three independentexperiments were performed. a-Tubulin was used for the cytoplasmicextract and PARP for the nuclear extract as loading control. The figurerepresents 1 of 3 independent experiments that were analyzed bydensitometrywith ImageJ. The x-fold increaseof the untreated controls ispresented.

Sack et al.





prooxidative activity of CNP in A375 melanoma cells, itwas supposed that CNP induce apoptosis through anincrease of ROS. However, data on a CNP-mediatedDNA damage in melanoma cells were lacking untilnow.Doxorubicin is known to damage DNA via several

mechanisms, for example intercalation, alkylation, andcomplex formation with DNA and topoisomerase 2,thereby inducing cell death (19). In this study, we wereinterested in a potential CNP-mediated DNA damage inA375 tumor cells and stromal HDF. Strand breaks asa marker of DNA damage were detected with thealkaline COMET assay (ref. 28; Fig. 7). A375 cells incu-bated with 15 mmol/L doxorubicin for 4 hours showed a2.5-fold increase of the comet area compared with thecontrols, indicating DNA damage. By contrast, no DNAstrand breaks could be detected after treatment with 150mmol/L CNPs for 96 hours (Fig. 7A). CNP showed nosignificant increase in comet area compared with theuntreated controls, indicating a nongenotoxic effect ofCNP. After co-incubation with doxorubicin and CNPthe comet area was not increased compared with thetreatment with doxorubicin alone, which corroborates anongenotoxic toxicity of CNP in melanoma cells.Although being toxic, CNPs do not induce DNA dam-age in melanoma cells. A similar effect was observedin HDF (Fig. 7B). Doxorubicin treatment resulted in2.5-fold increase of the comet area of HDF, whereastreatment with CNP did not increase the comet areacompared with the untreated control. Co-incubationwith doxorubicin and CNP resulted in a comet area

that is comparable to the treatment with doxorubicinalone. In summary, these data implicate that CNP treat-ment does not induce DNA damage in both cell lines.

DiscussionNanomedicine is one of the future technologies provid-

ing revolutionary improvements and innovations fortherapy and diagnostic of many diseases, including can-cer (4, 35, 36). Cancer is still one of the most devastatingdiseases, with more than 10 million cases every year (5),and thus remains the focus of interest of basic and clinicalresearch. Conventional anticancer therapies are oftenassociated with harmful side effects on healthy cells andhold the risk of secondary cancer, in consequence theclinical application is limited (20, 30). Recent studiesshowed that redox-active CNPs exhibit a significant anti-tumor activity in several cancer cell lines (7, 8). In squa-mous cell carcinoma of the skin and melanoma, CNPexhibit proapoptotic and anti-invasive effects in a ROS-dependent manner. In contrast to conventional che-motherapeutics, CNP are nontoxic in healthy, stromalcells of the skin. It was described that CNP exert eithera pro- or antioxidant redox activity. Although CNP treat-ment increases the ROS level in tumor cells resulting inapoptosis, CNP showed antioxidant and protective prop-erties in normal cells (12). The protective and antioxidantproperties can be traced back to an inherent and pH-dependent superoxide dismutase (SOD) mimic activityof CNP (37). In that context, a medical application of CNPmay provide a promising possibility for therapy of skin

Figure 7. CNP exert no genotoxiceffects. DNA damage wasinvestigated by using the alkalinecomet assay. Melanoma cells (A)and HDFs (B) were incubated with15 mmol/L doxorubicin for 4 hoursor with 150 mmol/L CNP for96 hours. In addition, cells wereco-incubated. CNP induced noDNA strand breaks, whereasdoxorubicin caused a significantlyDNA damage. Presented are themean values of the comet area in pxof 3 independent experiments.��, P < 0.01 (ANOVA, Dunnett test).Data, means � SEM.






cancer, andmay be a valuable tool to supplement classicaltherapeutical approach.

In this study, the antitumor activity of CNP wascompared with that of the classical and very effectiveantitumor drug doxorubicin. Furthermore, it was elu-cidated whether CNP may enhance the antitumor activ-ity of doxorubicin after co-incubation, particularly withregard to a novel strategy in cancer therapy, a CNP-mediated supplementation therapy with classical che-motherapeutics. The data showed that both, CNP anddoxorubicin, exerted cytotoxic effects in the humanmelanoma cell line A375. Although other studies sup-pose melanoma to be resistant to anthracyclines (38),this study demonstrated that doxorubicin is very effec-tive in inducing cell death in A375 melanoma cells inconcentrations, which can be compared with peak orsteady-state plasma concentrations of patients after astandard infusion with doxorubicin (>1–2 mmol/L;ref. 19). After co-incubation with both agents, the cellviability of melanoma cells was even more decreasedcompared with incubation with the agents alone, indi-cating a synergistic effect on cytotoxicity in tumor cells.By contrast, in HDFs pre-incubation with 150 mmol/LCNP abolished the toxic effects of doxorubicin, showinga protective effect of CNP against the cytotoxicity ofdoxorubicin in stromal cells.

An increase in ROS level was measured after incu-bation with CNP as well as with doxorubicin. Com-pared with normal cells, cancer cells were described tohave elaborated ROS levels, which promotes their geno-mic instability and proliferation, but also make themmore susceptible for an additional increase of ROSmediated by exogenous noxes, such as redox-cyclingdrugs and other ROS-producing agents. In our study,this ROS susceptibility was exploited by the use of theantitumor agents doxorubicin and CNP. A synergisticeffect on ROS generation was detected after co-incuba-tion. The prooxidative and cytotoxic effect of CNP anddoxorubicin was confirmed by the formation of sulfenicacids and carbonylated proteins.

Besides protein damage, oxidative stress is known tocause DNA damage. CNP was shown to generate ROSand to induce apoptosis via the intrinsic pathway in A375cells, but putative DNA damage by CNP was not mea-sured until now. Surprisingly, this study demonstratesthat CNP do not cause DNA damage at a concentrationbeing toxic in A375 cells. These data may be explained bythe uptake and cellular distribution of CNP in A375 cellsdisplaying a localization of CNP in the cytosol, but not inthe nuclei. In that context it was shown recently that thediffusion of H2O2, a potential genotoxic agent, across thecytoplasm was strongly limited, as well. It provides evi-dence that H2O2 acts locally inside cells (39). Correspond-ingly, it was shown that oxidation of thiols occurredmainly in the cytosol. Even though CNP show an antitu-mor activity, no genotoxic activity was detected in A375cells. In HDFs, no DNA-damaging effect was observed aswell. This would be a beneficial aspect in a cancer

therapy as the use of nongenotoxic antitumor agentsdecreases the risk of secondary cancer. In contrast,treatment with doxorubicin resulted in a significantincrease of the comet area in A375 and HDF, represent-ing DNA damage and a genotoxic activity of doxoru-bicin. Co-incubation with CNP did not increase theDNA damage compared with cells that were treatedwith doxorubicin alone. However, CNP did not protectHDF from doxorubicin-mediated DNA damage, butinterestingly CNP counteracts the doxorubicin-inducedcell killing in HDF. In contrast to another recent studywith CNPs with a size of 16 to 22 nm (40), DNAdamaging effects were found in other tumor cell lines,indicating that the mode of action of the nanoparticles isstrongly depending on size and cell type.

In summary, this study demonstrates that CNP maybe qualified to supplement conventional chemothera-peutic drugs, such as doxorubicin. CNP enhanced theantitumor activity of doxorubicin in A375 melanomacells, in context of cytotoxicity and ROS formation aswell as oxidative damage. However, CNP protectedHDFs from doxorubicin-induced cytotoxicity. Despitethe antitumor activity of CNP, no genotoxic effects ofCNP were detected in melanoma cells as well as inHDFs. The supplementation of conventional che-motherapies with CNP may offer a novel strategy intreatment of cancer providing a better therapeutic out-come and a higher benefit for patients, by enhancingantitumor activity and lowering the damaging sideeffects of classical chemotherapeutics such as the modelsubstance doxorubicin on healthy cells.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: M. Sack, L. Alili, P. BrenneisenDevelopment of methodology: M. Sack, L. Alili, E. Karaman, S. Das,A. GuptaAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): M. Sack, L. Alili, E. Karaman, A. GuptaAnalysis and interpretation of data (e.g., statistical analysis, biostatis-tics, computational analysis):M. Sack, L. Alili, E. Karaman, P. BrenneisenWriting, review, and or revision of the manuscript: M. Sack, L. Alili,S. Seal, P. BrenneisenAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): M. Sack, E. KaramanStudy supervision: M. Sack, L. Alili, P. BrenneisenOther (material preparation): S. Das

AcknowledgmentsThis work was part of the master thesis of E. Karaman at the Heinrich-

Heine-University ofD€usseldorf. The authors thankC.Wyrich for excellenttechnical assistance. S. Seal acknowledges the National Science Founda-tion (NSF) to partially fund the nanotechnology research under NSFNIRT(CBET-0708172) and NSF (CBET-0930170). The authors also thank K.S.Carroll for providing the a-hapten antibody.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received November 5, 2013; revised April 29, 2014; accepted May 5,2014; published OnlineFirst May 13, 2014.


Sack et al.




References1. Sahoo SK, Parveen S, Panda JJ. The present and future of nanotech-

nology in human health care. Nanomedicine 2007;3:20–31.2. Peer D, Karp JM, Hong S, Farokhzad OC, Margalit R, Langer R.

Nanocarriers as an emerging platform for cancer therapy. Nat Nano-technol 2007;2:751–60.

3. Florence AT. "Targeting" nanoparticles: the constraints of physicallaws and physical barriers. J Control Release 2012;64:115–24.

4. Jain KK. Advances in the field of nanooncology. BMCMed 2010;8:83.5. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012. CA Cancer

J Clin 2012;62:10–29.6. Garbe C, Leiter U. Melanoma epidemiology and trends. Clin Dermatol

2009;27:3–9.7. Lin W, Huang YW, Zhou XD, Ma Y. Toxicity of cerium oxide nanopar-

ticles in human lung cancer cells. Int J Toxicol 2006;25:451–7.8. Eom HJ, Choi J. Oxidative stress of CeO2 nanoparticles via p38-Nrf-2

signaling pathway in human bronchial epithelial cell, Beas-2B. ToxicolLett 2009;187:77–83.

9. Perez JM, Asati A, Nath S, Kaittanis C. Synthesis of biocompatibledextran-coated nanoceria with pH-dependent antioxidant properties.Small 2008;4:552–6.

10. Chen J, Patil S, Seal S, McGinnis JF. Rare earth nanoparticles preventretinal degeneration induced by intracellular peroxides. Nat Nanotech-nol 2006;1:142–50.

11. Tarnuzzer RW, Colon J, Patil S, Seal S. Vacancy engineered ceriananostructures for protection from radiation-induced cellular damage.Nano Lett 2005;5:2573–7.

12. Alili L, Sack M, Karakoti AS, Teuber S, Puschmann K, Hirst SM, et al.Combined cytotoxic and anti-invasive properties of redox-activenanoparticles in tumor-stroma interactions. Biomaterials 2011;32:2918–29.

13. Alili L, Sack M, von Montfort C, Giri S, Das S, Carroll KS, et al.Downregulation of tumor growth and invasion by redox-active nano-particles. Antioxid Redox Signal 2013;19:765–78.

14. DenkoNC. Hypoxia, HIF1 and glucosemetabolism in the solid tumour.Nat Rev Cancer 2008;8:705–13.

15. Gatenby RA, Gillies RJ. Why do cancers have high aerobic glycolysis?Nat Rev Cancer 2004;4:891–9.

16. Kuphal S,Winklmeier A,WarneckeC,Bosserhoff AK.ConstitutiveHIF-1 activity in malignant melanoma. Eur J Cancer 2010;46:1159–69.

17. Lechner D, Weltermann A. [Pathophysiology of chemotherapy-asso-ciated thrombosis]. Hamostaseologie 2009;29:112–20.

18. PerrinoC, SchiattarellaGG,Magliulo F, Ilardi F, CarotenutoG,GargiuloG, et al. Cardiac side effects of chemotherapy: state of art andstrategies for a correct management. Curr Vascu Pharmacol 2012;12:106–16.

19. Minotti G, Menna P, Salvatorelli E, Cairo G, Gianni L. Anthracyclines:molecular advances and pharmacologic developments in antitumoractivity and cardiotoxicity. Pharmacol Rev 2004;56:185–229.

20. Du C, Deng D, Shan L, Wan S, Cao J, Tian J, et al. A pH-sensitivedoxorubicin prodrug based on folate-conjugated BSA for tumor-tar-geted drug delivery. Biomaterials 2013;34:3087–97.

21. Seo YH, Carroll KS. Profiling protein thiol oxidation in tumor cells usingsulfenic acid-specific antibodies. Proc Natl Acad Sci U S A 2009;106:16163–8.

22. Giard DJ, Aaronson SA, Todaro GJ, Arnstein P, Kersey JH, Dosik H,et al. In vitro cultivation of human tumors: establishment of cell lines

derived from a series of solid tumors. J Natl Cancer Inst 1973;51:1417–23.

23. Bayreuther K, Francz PI, Gogol J, Kontermann K. Terminal differen-tiation, aging, apoptosis, and spontaneous transformation in fibroblaststem cell systems in vivo and in vitro. Ann NY Acad Sci 1992;663:167–79.

24. StuhlmannD, Ale-AghaN, Reinehr R, Steinbrenner H, RamosMC,SiesH, et al. Modulation of homologous gap junctional intercellular com-munication of human dermal fibroblasts via a paracrine factor(s)generated by squamous tumor cells. Carcinogenesis 2003;24:1737–48.

25. Mosmann T. Rapid colorimetric assay for cellular growth and survival:application to proliferation and cytotoxicity assays. J Immunol Meth-ods 1983;65:55–63.

26. Karakoti AS, Kuchibhatla SVNT, Babu KS, Seal S. Direct synthesis ofnanoceria in aqueous polyhydroxyl solutions. J Phys Chem C2007;111:17232–40.

27. LaemmliUK.Cleavageof structural proteinsduring theassembly of thehead of bacteriophage T4. Nature 1970;227:680–5.

28. Singh NP, Stephens RE, Schneider EL. Modifications of alkalinemicrogel electrophoresis for sensitive detection of DNA damage. IntJ Radiat Biol 1994;66:23–8.

29. Albrecht C, BormPJ, Unfried K. Signal transduction pathways relevantfor neoplastic effects of fibrous and non-fibrous particles. Mutat Res2004;553:23–35.

30. Woodward WA, Strom EA, McNeese MD, Perkins GH, Outlaw EL,Hortobagyi GN, et al. Cardiovascular death and second non-breastcancer malignancy after postmastectomy radiation and doxorubicin-based chemotherapy. Int J Radiat Oncol Biol Phys 2003;57:327–35.

31. Park EJ, Choi J, Park YK, Park K. Oxidative stress induced by ceriumoxide nanoparticles in cultured BEAS-2B cells. Toxicology 2008;245:90–100.

32. Dalle-Donne I, Rossi R, Giustarini D, Milzani A, Colombo R. Proteincarbonyl groups as biomarkers of oxidative stress. Clin Chim Acta2003;329:23–38.

33. Dalle-Donne I, Giustarini D, Colombo R, Rossi R, Milzani A. Proteincarbonylation in human diseases. Trends Mol Med 2003;9:169–76.

34. Marnett LJ. Oxyradicals and DNA damage. Carcinogenesis 2000;21:361–70.

35. Bawarski WE, Chidlowsky E, Bharali DJ, Mousa SA. Emerging nano-pharmaceuticals. Nanomedicine 2008;4:273–82.

36. Cho K, Wang X, Nie S, Chen ZG, Shin DM. Therapeutic nanoparticlesfor drug delivery in cancer. Clin Cancer Res 2008;14:1310–6.

37. Heckert EG, Karakoti AS, Seal S, Self WT. The role of cerium redoxstate in the SOD mimetic activity of nanoceria. Biomaterials 2008;29:2705–9.

38. Smylie MG, Wong R, Mihalcioiu C, Lee C, Pouliot JF. A phase II, openlabel, monotherapy study of liposomal doxorubicin in patients withmetastatic malignant melanoma. Invest New Drugs 2007;25:155–9.

39. Mishina NM, Tyurin-Kuzmin PA, Markvicheva KN, Vorotnikov AV,Tkachuk VA, Laketa V, et al. Does cellular hydrogen peroxide diffuseor act locally? Antioxid Redox Signal 2011;14:1–7.

40. De Marzi L, Monaco A, De Lapuente J, Ramos D, Borras M, DiGioacchino M, et al. Cytotoxicity and genotoxicity of ceria nanopar-ticles on different cell lines in vitro. Int J Mol Sci 2013;14:3065–77.






2014;13:1740-1749. Published OnlineFirst May 13, 2014.Mol Cancer Ther Maren Sack, Lirija Alili, Elif Karaman, et al. Cancer Therapy

A Novel Aspect in−−Redox-Active Cerium Oxide Nanoparticles Combination of Conventional Chemotherapeutics with

Updated version

10.1158/1535-7163.MCT-13-0950doi:

Access the most recent version of this article at:

Cited articles

http://mct.aacrjournals.org/content/13/7/1740.full#ref-list-1

This article cites 40 articles, 3 of which you can access for free at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://mct.aacrjournals.org/content/13/7/1740To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-13-0950

http://mct.aacrjournals.org/content/13/7/1740.full#ref-list-1

http://mct.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mct.aacrjournals.org/content/13/7/1740


combination of conventional chemotherapeutics with...

Documents