Doxorubicin has in vivo toxicological effects on ex vivo cultured ... · proliferation was assessed by plating 1×103 cells/cm2 in 6-well plates (day 0). For each one of the four

Page 1

Dm

MDa

b

h

••••

a

ARR1AA

KDMCC

1

t(d

pmcnds

0h

Toxicology Letters 224 (2014) 380– 386

Contents lists available at ScienceDirect

Toxicology Letters

jou rn al hom ep age: www.elsev ier .com/ locate / tox le t

oxorubicin has in vivo toxicological effects on ex vivo culturedesenchymal stem cells

aira Souza Oliveiraa,∗, Juliana Lott Carvalhob, Ana Carolina De Angelis Camposb,awidson Assis Gomesb, Alfredo Miranda de Goesb, Marília Martins Meloa

College of Veterinary Medicine, Universidade Federal de Minas Gerais, Av. Antônio Carlos 6627, Caixa Postal 567, 30123-970 Belo Horizonte, MG, BrazilInstitute of Biological Sciences, Universidade Federal de Minas Gerais, Av. Antônio Carlos 6627, Caixa Postal 567, 30123-970 Belo Horizonte, MG, Brazil

i g h l i g h t s

Mesenchymal stem cells (MSCs) were isolated from rats that received doxorubicin.Doxorubicin had toxic effects on MSCs.MSCs grew up in lower rates and had decreased alkaline phosphatase production.MSCs failed to express connexin 43 and troponin T.

r t i c l e i n f o

rticle history:eceived 15 October 2013eceived in revised form9 November 2013ccepted 20 November 2013vailable online 28 November 2013

eywords:rug toxicity

a b s t r a c t

Doxorubicin (dox) is an effective chemotherapeutic agent that leads to cardiotoxicity. An alternative treat-ment for dox-cardiotoxicity is autologous mesenchymal stem cells (MSCs) transplantation. It remainsunclear if dox has deleterious effects on MSCs from subjects under chemotherapy, therefore this studyaimed to evaluate dox in vivo toxicological effects on ex vivo cultured MSCs, inferring whether autolo-gous transplantation may be an alternative treatment in patients who are exposed to the drug. Wistar ratsreceived either dox or saline. Following treatments, animals were sacrificed and bone marrow MSCs wereisolated, characterized for cell surface markers and assessed according to their viability, alkaline phos-phatase production, and proliferation kinetics. Moreover, MSCs were primed to cardiac differentiation

esenchymal stem cellsell therapyardiac differentiation

and troponin T and connexin 43 expressions were evaluated. Compared to control, undifferentiated MSCsfrom dox group kept the pattern for surface marker and had similar viability results. In contrast, theyshowed lower alkaline phosphatase production, proliferation rate, and connexin 43 expression. PrimedMSCs from dox group showed lower troponin T levels. It was demonstrated a toxic effect of dox in hostMSCs. This result renders the possibility of autologous MSCs transplantation to treat dox-cardiotoxicity,

table

which could be a non-sui

. Introduction

One of the most effective chemotherapeutic agents used for

reatment of hematological and solid tumors is doxorubicin (dox)Butany et al., 2009) However, it causes a dose-dependent car-iotoxicity that may lead to irreversible heart failure by different

Abbreviations: AP, alkaline phosphatase; BCIP, 5-bromo-4-chloro-3-indolylhosphate; BSA, bovine serum albumin; DMEM, Dulbecco’s modified Eagle’sedium; dox, doxorubicin; FBS, fetal bovine serum; MSCs, mesenchymal stem

ells; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide; NBT,itroblue tetrazolium salt; PBS, phosphate buffered saline; PVDF, polyvinylideneifluoride; SDS, sodium dodecyl sulfate; TBS, tris buffered saline; TBST, tris bufferedaline with Tween.∗ Corresponding author. Tel.: +55 31 3409 2274; fax: +55 31 3409 2230.

E-mail address: maira [email protected] (M.S. Oliveira).

378-4274/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved.ttp://dx.doi.org/10.1016/j.toxlet.2013.11.023

option for subjects receiving such antineoplastic agent.© 2013 Elsevier Ireland Ltd. All rights reserved.

pathogenic mechanisms which have not been completely eluci-dated (Gianni et al., 2008). Production of reactive oxygen species,calcium imbalance and apoptosis are considered among the patho-logical mechanisms for dox-induced cardiotoxicity (Gianni et al.,2008; Tan et al., 2010; Zhang et al., 2011).

A progressive reduction of left ventricular function, detected bydifferent echocardiography approaches, either in animal models(Oliveira et al., 2013), or in a significant proportion of pediatric(Lipshultz et al., 2012), adult (Tassan-Mangina et al., 2006) andelderly (Aapro et al., 2011) patients, is observed during the courseof the anti-cancer therapy, leading to a life-threatening conges-tive heart failure. While cardiotoxicity can develop at no specific

time once the treatment has begun, identifying cardioprotectivestrategies to minimize the long-term damage caused by anthracy-clines is imperative. In this context, many alternatives have beeninvestigated either in experimental models or in clinical trials in

Page 2

ogy Le

ozp2teaaatsft(asHptcwi

2

2

Mtepphmsnp(PoDetttitvcaipotmi

2

wmri

M.S. Oliveira et al. / Toxicol

rder to prevent or diminish cardiac dysfunction, such as dexra-oxane (Lipshultz et al., 2010), carvedilol (Machado et al., 2008),lant extracts (Du and Lou, 2008) and stem cell therapy (Chen et al.,010). Nevertheless, none of those are really established in rou-ine medical practice to date. Between 1966 and 2010, Van-Dalent al. (2011) investigated randomized controlled trials in whichny cardioprotective agent was compared to no additional ther-py or placebo in cancer patients (children and adults) receivingnthracyclines. Although dexrazoxane had showed the potentialo prevent heart damage, the authors found no definitive conclu-ions about a real cardioprotective agent. A promising alternativeor prevention or treatment of tissues and organs is the regenera-ive medicine using stem cells. Autologous mesenchymal stem cellMSCs) transplantation is an obvious first option when thinking of

routine practice, given the ease of cell isolation and in vitro expan-ion required for therapy (Mayhall et al., 2004; Wang et al., 2012).owever, the autologous MSCs transplantation for cardiac repair inatients under chemotherapy must be re-thought. It is paramounto firstly understand the effects of dox in such cell population beforeonsidering its use. For this reason, the aim of the present researchas to study whether MSCs niches are affected by dox treatment

n subjects under chemotherapy.

. Materials and methods

.1. Dox cytotoxicity testing

Neonatal ventricular cardiomyocytes, adipose tissue derivedSCs and breast cancer MDA-MB 231 (ATCC® HTB-26TM) cell cul-

ures were evaluated for dox cytotoxicity. Cardiomyocytes (Ottt al., 2008) and MSCs (Oliveira et al., 2013) were obtained asreviously described. The cell cultures were seeded on 24-welllates at 5 × 104 cells/cm2. Thereafter, they were incubated in aumidified atmosphere with 5% CO2 at 37 ◦C using Dulbecco’sodified Eagle’s medium (DMEM) high glucose (Sigma–Aldrich)

upplemented with 10% fetal bovine serum (FBS) (Cripion Biotec-ologia LTDA), 5 mM sodium bicarbonate (Cinética Química Ltda),enicillin (100 U/mL), streptomycin (0.1 mg/mL), amphotericin B0.25 mg/mL) (Sigma–Aldrich), and gentamicin (60 mg/L; Schering-lough). After 24 h, media was changed and was added 5 �mol/lf dox [dox hydrochloride (Adriblastine®, Pfizer)] on the plates.ox concentration was used as previously described (Spallarossat al., 2004). The length of incubation was determined by evaluatinghe time where 50% of cardiomyocytes died. Then, this incubationime was performed for all cell cultures. Afterward, to determinehe cytotoxic effect of dox on the different cell types, cell viabil-ty was assessed by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyletrazolium bromide (MTT) assay (Cat. M-6494; Invitrogen) as pre-iously described by Paula et al. (2013). Briefly, 24 h after plating theells, medium was changed to 170 �l of the MTT solution (5 mg/ml)nd 210 �l of basal media (DMEM/10% FBS). Cells were, then,ncubated in a humid 5% CO2 atmosphere at 37 ◦C. 2-h later, therecipitated formazan crystals were solubilized by adding 210 �lf sodium dodecyl sulfate (SDS)-10% HCl. After 18 h, 100 �l of solu-ion was transferred to a 96-well plate, and the optical density was

easured at 595 nm using a microplate reader. Data is from threendependent experiments.

.2. Animals and experimental groups

Sixteen male 6–8 weeks old Wistar rats weighing 300–350 g

ere studied. Animals were kept in climate-controlled environ-ent under a 12-h light/dark cycle with free access to standard

odent chow and water. All experimental protocols were performedn accordance with the guidelines of the Institutional Animal Care

tters 224 (2014) 380– 386 381

and Use Committee in place at the University (protocol number176/2010).

Animals were allocated into two groups containing eight ratseach: control group [saline intraperitoneal injection (i.p.)] and dox-treated group [dox hydrochloride (Adriblastine®, Pfizer) 5 mg/kg;i.p.]. Animals were weighted weekly and the dox dosage was prop-erly calculated to each animal. The injections were given once aweek for four weeks. After 48 h of the last dox/saline adminis-tration, animals were anaesthetized and sacrificed by overdose ofisoflurane anesthesia. The cumulative dose of dox used was previ-ously demonstrated to be effective and to promote cardiotoxicityand heart failure (Oliveira et al., 2013). In order to confirm theoccurrence of cardiac dysfunction, animals from both groups wereevaluated by echocardiography examination before (baseline) andafter dox treatment.

2.3. Bone marrow derived MSCs isolation and culture

Following the euthanasia, bone marrow MSCs were isolatedfrom rats of both control and dox groups, as previously described(Assis et al., 2010) with minor modifications. Briefly, bone mar-row was flushed out from tibias and femurs of both control anddox-treated rats, using DMEM high glucose supplemented withsodium bicarbonate, penicillin, streptomycin, amphotericin B, andgentamicin. This content was centrifuged, and the pellet contain-ing MSCs was put into suspension in the aforementioned mediumadded with 10% FBS and, then, plated in 75 cm2 flasks. Cell cul-tures were kept in a humidified atmosphere with 5% CO2 at 37 ◦Cfor 24 h before the first medium change. The medium was thenchanged every 2–3 days. Cell adherence to cell culture flask guar-anteed mesenchymal cell population enrichment. The cells wereallowed to grow near confluence, and were split in a 1:3 ratio using0.25% trypsin–EDTA. Fourth passage bone marrow derived MSCswere used in all experiments.

2.4. Cellular characterization by indirect immunofluorescence

Cells were fixed in 4% paraformaldehyde for 10 min at roomtemperature, permeabilized in 0.1% Triton X-100 (Sigma) for 1 h atroom temperature. Samples were blocked with 1% bovine serumalbumin (BSA) and 5% goat serum in phosphate buffered saline(PBS) solution and incubated for 1 h at room temperature fol-lowed by two washes with PBS. Primary antibodies [CD45 (cat.#610266), CD54 (cat.# 554967), CD73 (cat.# 551123) and CD90(cat.# 554895), all from BD Biosciences, USA] were added in 1%BSA in PBS solution and incubated overnight at 4 ◦C at 1:100 dilu-tion. Samples were washed with PBS twice. Secondary antibodies[Alexa Fluor dye conjugated (Invitrogen, USA)] were diluted (1:500)in the same solution as the primary antibodies and incubated atroom temperature for 1 h in dark. Samples were washed with PBStwice. Coverslips were sealed with hydramount aqueous media(Cat. HS-106; National diagnostics) and analyzed with a Zeiss LSM510—Meta confocal microscope. Nine random fields were collectedfrom each experimental group. Data is from three independentexperiments.

2.5. Cell viability assay

Cell viability was assessed by the MTT assay as aforementioned.In these studies, cultures of 5 × 104 cells/cm2 were seeded on 24-well plates. Data is from three independent experiments.

2.6. Alkaline phosphatase (AP) activity

AP activity was evaluated with the 5-bromo-4-chloro-3-indolylphosphate (BCIP)/nitroblue tetrazolium salt (NBT) Kit assay as

Page 3

3 ogy Le

dc2ibwmi

2

6wc3gw

2

ao2aorWtdmc

2

wlpst4pmerbTbfa21atpfcw

2

dw

(p = 0.0001). Considering that both groups had either same cellseeding counts or MTT results, the data of NBT–BCIP test indicatesthat MSCs from dox group had AP production deficiency (Fig. 3).

Table 1Echocardiography variables obtained from Wistar rats following four weeks of eitherdoxorubicin or saline treatments, showing an effective induction of cardiac failureon animals exposed to the drug.

Echocardiography variables Groups p-Value

Control Doxorubicin

Heart rate (bpm) 332 (9.9) 332 (49.9) 0.9067

82 M.S. Oliveira et al. / Toxicol

escribed by the manufacturer (Cat. 00-2209; Invitrogen). Briefly,ultures of 5 × 104 cells/cm2 were seeded in 24-well plates and,4 h later, 200 �l of the BCIP–NBT solution was added. After a 2-h

ncubation period, the insoluble purple precipitate was solubilizedy adding 210 �l of SDS-10% HCl. After 18 h, 100 �l of solutionere transferred to a 96-well plate, and the optical density waseasured at 595 nm using a microplate reader. Data is from three

ndependent experiments.

.7. Cell proliferation study

Cell proliferation was assessed by plating 1 × 103 cells/cm2 in-well plates (day 0). For each one of the four time points, cellsere seeded in three independent wells. Following cell attachment,

ells from both control and dox groups were harvested on days 1,, 5, 7 and counted in triplicates using a hemocytometer. Lastly,rowth curves were determined. Three independent experimentsere considered.

.8. Cardiomyogenic differentiation induction

Plasticity of MSCs toward cardiomyogenic phenotype wasssessed after inducing the cells to differentiate into cardiomy-cytes using a standard induction medium (Planat-Bénard et al.,004). This medium basically consisted of: DMEM/20% FBSdded with IL-3, IL-6, SCF, 10−4 M of 2-mercaptoetanol, 2 mMf l-glutamine, 200 �g/ml human apo-transferrin, and 10 �g/mlecombinant human insulin. Cells were cultured on 25 cm2 flasks.

hen cells reached approximately 80% confluence, they were cul-ured with the induction media (4 ml/flask changing every twoays). MSCs from both dox and control groups were kept in basaledium, which consisted of DMEM supplemented with 10% FBS as

ontrol.

.9. Western blotting

Briefly, bone marrow derived MSCs from dox and control groupsere washed twice with ice-cold PBS, harvested by scraping and

ysed in a lysis buffer (150 mM NaCl, 1 mM EDTA, 20 mM Tris–HCLH 8.0, 0.5% Nonidet P-40). After incubation on ice, the cells of eachample were homogenized by vortex and sonicated. The respec-ive homogenates were centrifuged at 16,100 × g for 20 min at◦C. Protease inhibitors (Sigma, USA) were added to each sam-le. The protein amount was assessed by the Bradford assay. Thirtyicrograms of protein of each group were subjected to SDS-PAGE

lectrophoresis and, then, transferred to a polyvinylidene difluo-ide (PVDF) membrane (n = 3 in each group). The membrane waslocked with 5% skim milk in Tris buffered saline (TBS) withween (TBST; TBS plus 0.1% Tween 20) for 60 min and then incu-ated with primary antibody. Commercially available antibodiesor connexin-43 (cat.# 11370), cardiac troponin-T (cat.# 8295), andlpha-actinin (cat.# 9465) (all from Abcam, UK) and GAPDH (cat.#5778, from Santa Cruz Biotechnology, USA) were used at 1:8000,:1000, 1:1000, and 1:250 dilutions, respectively. GAPDH was useds an input control. Incubations were carried out overnight. Afterhree washes with TBST, the membranes were incubated witheroxidase-conjugated secondary antibody (1:5000) (Sigma, USA)or 1 h at room temperature. Blots were visualized by enhancedhemiluminescence and quantitatively analyzed using Image J soft-are. Three independent experiments were done.

.10. Statistical analysis

All experiments were repeated three times in triplicate, and theata were presented as mean (standard deviation). The variablesere submitted to both normality and homoscedasticity analyses.

tters 224 (2014) 380– 386

For datasets with normal distribution, statistical significance wasdetermined by Student’s t-test for two groups or one-way analysisof variance (ANOVA) for multiple groups with post hoc test usingTukey method. Variables with non-normal distribution were ana-lyzed with Mann–Whitney post test. Significance was consideredfor 5% (p < 0.05). Analyses were done in R software program (2.11version).

3. Results

3.1. Dox showed cytotoxic effect for breast cancer cells,cardiomyocytes and MSCs

As expected for dox, which is a chemotherapy agent thatleads to cardiotoxicity, the cellular viability for breast cancer cellsand cardiomyocytes decreased after incubation with the drug.MTT metabolization for cancer cells was 69(12.26)% of its con-trol and for cardiomyocytes was 65(9.54)% of its control. Moreover,MTT metabolization for MSCs incubated with dox also decreased,corresponding to 83(12.74)% of control, indicating an in vitro dox-cytotoxic effect on MSCs (Fig. 1).

3.2. In vivo exposure to dox promoted cardiotoxicity

Animals under dox treatment showed cardiac dysfunction. Bothcontrol and dox groups had similar baseline echocardiographyparameters. At the end of the experiment, compared to control,dox group presented significant decreased values of ejection frac-tion, fractional shortening, longitudinal strain, and radial strain,indicating an effective induction of a cardiotoxicity model (Table 1).

3.3. MSCs from dox group kept the cell surface marker pattern

MSCs showed fibroblast-like morphology and the cell culturewas enriched following the passages. Similar to control group, MSCsisolated from rats that received dox treatment were positive forCD54, CD73, CD90, and negative for CD45 expression, as expectedfor undifferentiated MSCs according to the International Society forCellular Therapy (Dominici et al., 2006) (Fig. 2).

3.4. MSCs from dox group were viable but with reduction on APproduction

The values detected for MTT metabolization were similar forboth groups (p = 0.0516). On the other hand, the values detectedfor NBT–BCIP test were statistically different between the groups

Ejection fraction (%) 68.9 (1.9) 54.4 (4.4) 0.0056Fractional shortening (%) 36.9 (2.1) 28.8 (6.1) 0.0084Longitudinal strain (%) 15.2 (4.2) 7.1 (2.0) 0.0036Radial strain (%) 17.1 (6.1) 8.1 (1.1) 0.0148

Page 4

M.S. Oliveira et al. / Toxicology Letters 224 (2014) 380– 386 383

Fig. 1. Doxorubicin (dox) cytotoxicity in breast cancer cells (MDA-MB 231), cardiomyocytes and mesenchymal stem cells. Cellular viability was decreased in all cell culturesafter 10 h of incubation with 5 �mol/l doxorubicin, compared to the respective control (N = 9, *p < 0.01, **p < 0.05).

Fig. 2. Immunophenotype characterization for bone marrow mesenchymal stem cells derived from either control (a, left side of the panel) or doxorubicin (b, right side of thepanel) groups showed the expected pattern of expression, being positive for CD54, CD73, CD90 and negative for CD45. Cells incubated only with secondary antibody wereused as a negative control. Scale = 10 �m. Data shown is from one of the three independent experiments that were done (N = 9).

Fig. 3. Cellular viability and alkaline phosphatase (AP) production for bone marrow mesenchymal stem cells derived from either control or doxorubicin (dox) groups. Cellsfrom dox group showed decreased AP production compared to control (N = 9, *p < 0.01).

Page 5

384 M.S. Oliveira et al. / Toxicology Le

Fig. 4. Doxorubicin (dox) inhibits mesenchymal stem cell proliferation. Mesenchy-mal stem cells derived from dox group show decreased counting values comparedt

3

aa(

3e

4bsdgd

4

hpd2si22

(Henninger et al., 2012). Dox induced nephrotoxicity was related

FWa(

o control after day five of plating (N = 9, *p < 0.05).

.5. Dox induced lower cell proliferation rate of MSCs in culture

The cellular growth curve pointed out differences from controlnd dox groups (Fig. 4). MSCs from dox group had lower prolifer-tion rates at day 5 and day 7 after seeding, compared to controlp < 0.05).

.6. Alterations on connexin-43 and cardiac troponin Txpression were detected on cells from dox group

Undifferentiated cells from dox group had decreased connexin-3 expression compared to control, when evaluated by westernlotting (p < 0.05). Priming MSCs to cardiomyogenic lineage wasupposed to induce cardiac troponin T expression. Indeed it wasetected for MSCs from control group. Otherwise, cells from doxroup did not express cardiac troponin T following cardiomyogenicifferentiation induction (Fig. 5).

. Discussion

Dox remains among the most effective agents againstematopoietic and solid tumors, and for that reason, it has beenrescribed worldwide, despite the possible occurrence of car-iotoxicity (Aapro et al., 2011; Butany et al., 2009; Gianni et al.,008; Lipshultz et al., 2012; Tassan-Mangina et al., 2006). Eveno, there is no consensus about a real effective treatment for dox-

nduced cardiotoxicity in clinical practice, to the date (Du and Lou,008; Lipshultz et al., 2010; Machado et al., 2008; Van-Dalen et al.,011).

ig. 5. The expression of both troponin T and connexin 43 proteins was decreased on meestern blotting was used to study troponin T and connexin 43 expression after doxorub

positive control, as seen on the left side of the panel (a). Quantification of troponin T anb and c) (N = 3, *p < 0.05). Dox, doxorubicin; MSC, mesenchymal stem cell; pMSC, primed

tters 224 (2014) 380– 386

Among the alternatives that have been investigated to preventor repair cardiac injuries, stem cell transplantation may be con-sidered as an option, due to promising data from experimentalstudies involving several heart failure models. Various undiffer-entiated stem cell sources have been proposed for regenerativemedicine, each having their advantages and drawbacks. Embryonicstem cells are characterized by pluripotency and an unlimited self-renewal capacity, but they may present high tumorigenic potentialand their use is associated with major ethical concerns. In con-trast, MSCs are not ethically restricted, but show limited capacityto proliferate and differentiate into different cell lineages (mul-tipotency) (Barreto Filho and Oliveira, 2012). Even though MSCsdo not terminate differentiation toward cardiomyogenic lineage,these cells are employed in cardiac repair due to their paracrineeffects. MSCs have been shown to promote the preservation ofheart function in acute as well as chronic lesions, mainly myocar-dial infarction (Carvalho et al., 2013; Mias et al., 2009) and chronicischemic heart failure (Lilyanna et al., 2013), and even in doxinduced cardiotoxicity models (Chen et al., 2010; Oliveira et al.,2013).

Therefore, it is reasonable to hypothesize that the application ofthose cells should preserve heart function following dox treatmentin clinical practice. Even though MSCs are not immunogenic andmay be used in allogenic transplants, the most obvious choice oftreatment is the use of autologous cells. In such scenario, it becomesparamount to assess the quality and characteristics of MSCs fromthe patient receiving dox. Here, we firstly investigated whetherdox was toxic to MSCs in vitro. Then, we performed an in vivostudy designed to promote cardiac toxicity due to dox adminis-tration, thus evaluating MSCs isolated from control and dox hostsaccording to: adherence to plastic, cell morphology and phenotype,viability, proliferation, alkaline phosphatase production and poten-tial for differentiation toward cardiomyocyte cell fate. As we founda cytotoxic effect of the drug to MSCs in vitro, it was reasonableto think that the dox treatment could interfere in the biologicalaspects of MSCs from subjects receiving chemotherapy. Anotherreason to consider a possible side effect of dox toward MSCs isthe fact that the drug is also related to other organotoxic poten-tials, raising the possibility for the occurrence of cytotoxic effectson different organs and tissues. It was demonstrated that dox pro-moted pro-inflammatory and pro-fibrotic stress responses in theliver by increasing mRNA expression of tumor necrosis factor alphaand connective tissue growth factor and leading to DNA damage

to increased serum urea nitrogen and creatinine levels, meanwhilekidney tissues studies demonstrated increases in the expression ofpro-apoptotic caspase-3, Bad and Bax proteins, reduction in the

senchymal stem cells isolated from animals that had been exposed to doxorubicin.icin treatment. GAPDH was used as a loading control and heart tissue was used as

d connexin 43 expression were shown as values normalized by GAPDH expression mesenchymal stem cells.

Page 6

ogy Le

etesre

aDatt2apwnnptpd

u2Rtilsm

btampdiiibhAms(Cgn

itcaptc(e

ocmmt

M.S. Oliveira et al. / Toxicol

xpression of anti-apoptotic Bcl-2 and Bcl-xL genes and eleva-ions in lipid peroxidation and genomic DNA fragmentation (Lahotit al., 2012). Moreover clinical trial investigations reported myolo-uppression, alopecia, nausea and vomiting in patients who haveeceived dox either alone or in combined preparations (Rafiyatht al., 2012).

With respect to cytotoxic potential of dox to MSCs, it is presum-ble that other anti-cancer drugs may show similar side effects.uring cancer treatment, combined chemotherapy preparationsre used to improve the efficacy in killing cancer cells. Amonghem, the use of dox plus cyclophosphamide followed by doce-axel plus trastuzumab to treat breast cancer (Saracchini et al.,013), or the iPAD protocol which comprises bortezomid, doxnd intermediate-dose dexamethasone to treat refractory multi-le myeloma (Takamatsu et al., 2013), or even the CHOP protocolhich comprises cyclophosphamide, dox, vincristine and pred-isone in the treatment of various cancer types including refractoryon-Hodgkin lymphomas (Yang et al., 2013). Considering the dataresented here for the use of one drug alone (dox), it is possibleo assume that such common practice of combined chemotherapyreparations probably make the side effects on MSCs even moreramatic.

There are many protocols for the induction of cardiotoxicitysing dox (Chen et al., 2010; De Angelis et al., 2010; Hazari et al.,009; Hou et al., 2009; Kumar et al., 2011; Oliveira et al., 2013).oute of administration, dose and number of applications seemso differ and even drug manufacturers generate different effectsn vivo. Before studying stem cell toxicity induced by dox, we estab-ished an in vivo model of cardiopathy induced by the drug. Ashown in echocardiography analysis, we successfully generated aodel in which cardiotoxicity by dox is observed.Next, we isolated bone marrow MSCs and evaluated a possi-

le toxic effect of dox in this cell population. In the present study,he adipose tissue was not considered as source of MSCs becausell animals from dox group showed markedly weight loss and theacroscopic appearance of the fat seemed to be degenerated, inca-

able of being used. Our results indicate that treatment with doxid not affect bone marrow MSC’s ability to adhere to plastic, allow-

ng similar isolation protocol as in the control group. In addition,solated MSCs from both groups presented similar morphology andmmunophenotype. In the present work, the MSC population fromoth groups had a CD54+/CD73+/CD90+/CD45− phenotype. Reportsave shown that MSCs express many different surface markers.lthough differences are found the phenotype presented hereatches some of the markers that have been reported by others,

uch as CD73+/CD90+/CD105+/CD11b−/CD19−/CD45−/HLA-DR−

Dominici et al., 2006), CD90+/CD71−/CD34− (Jiang et al., 2006),D54+/CD73+/CD90+/CD45− (Assis et al., 2010). It is possible to sug-est, therefore, that isolated MSCs were not altered regarding theirature during dox treatment.

In contrast with the aforementioned results, dox treatmentnduced lower alkaline phosphatase activity, slower prolifera-ion rates, decreased connexin-43 production and hindered MSCsapacity to respond to cardiomyogenic differentiation stimuli. Databout stem cell transplantation to regenerate heart injuries areromising. The most discussed mechanism involved in the cardiacissue repair is related to the paracrine effect of the MSCs, as thoseells do not terminally differentiate into cardiomyogenic lineageCarvalho et al., 2013; Chen et al., 2010; Dai et al., 2005; Garbadet al., 2009; Oliveira et al., 2013).

Even though is not well established weather MSCs are capablef terminally differentiating into the cardiomyogenic lineage, those

ells are employed in a vast number of studies regarding treat-ent of heart function. Those data show promising results and theore discussed mechanism by which MSCs repair heart tissue is

heir paracrine effects, although it is unclear which components

tters 224 (2014) 380– 386 385

are really secreted by the MSCs in the host tissues. The productionof two proteins was investigated in the present study, connexin43 and troponin T. It is known that cardiomyocytes highly expressconnexin 43 and are rich in GAP junctions, which are extremelyimportant to the maintenance of the conductivity of the electricimpulses throughout the cardiomyocytes (Valiunas et al., 2000).Troponin T is a regulatory protein that is involved in the heart mus-cle contraction (Watkins et al., 1995). The results presented hereshowed that MSCs, isolated from subjects that had been exposedto dox, had decreased connexin 43 expression and troponin T aftercardiomyogenic differentiation. Regarding cell therapy it is rea-sonable to consider that part of transplanted cells may stay inan undifferentiated status, while others may prime to differenti-ation following the host tissue cells. So, the cell transplantationmay account with healthy cells which could be able to expressproteins and secret products in the niche where they will host,independently if they will follow differentiation or not. Thereaftera cell transplantation procedure using MSCs isolated from donorsthat had been exposed to dox could be unsuccessful due to a sink-ing expression of two major proteins, connexin 43 and troponin T,needed for the cardiomyocytes to do their ordinary functions.

Therefore, in addition to the already described cardiotoxicity,the results of the current study show for the first time, a cytotoxiceffect of dox on bone marrow MSCs from subjects receiving suchdrug. In addition to bone marrow, other stem cell niches mightbe compromised by the use of such drug, as demonstrated by DeAngelis et al. (2010) studying cardiac progenitor cells.

In conclusion, the present data demonstrated that dox promotesbiological impairments on MSCs, raising the possibility of autolo-gous stem cell transplantation to treat dox-induced cardiotoxicityas an improper option for subjects under chemotherapy with suchantineoplastic drug. In order to have a better comprehension of thebiological impairment of dox on MSCs, further studies are needed,such as an evaluation of in vivo transplantation of the isolated MSCs(derived from both the groups) into rats with dox-induced heartfailure, addressing the main question of the present study.

Conflict of interest statement

Authors declare that they have no conflict of interest.

Acknowledgements

“Fundac ão de Amparo à Pesquisa do Estado de Minas Gerais”(FAPEMIG) and “Conselho Nacional de Desenvolvimento Cientí-fico e Tecnológico” (CNPq) for financial support; “Coordenac ão deAperfeic oamento de Pessoal de Nível Superior” (CAPES) for thescholarship awarded to M.S. Oliveira.

Appendix A. Supplementary data

Supplementary data associated with this article can be found,in the online version, at http://dx.doi.org/10.1016/j.toxlet.2013.11.023.

References

Aapro, M., Bernard-Marty, C., Brain, E., Batist, G., Erdkamp, F., Krzemienjecki, K.R.,Lluch, L., Monfardini, A., Ryberg, S., Soubeyran, M., Wedding, P.U., 2011. Anthra-cycline cardiotoxicity in the elderly cancer patient: a SIOG expert position paper.Ann. Oncol. 22, 257–267.

Assis, A.C.M., Carvalho, J.L., Jacoby, B.A., Ferreira, R.L.B., Castanheira, P., Diniz, S.O.F.,Cardoso, V.N., Goes, A.M., Ferreira, A.J., 2010. Time-dependent migration of sys-

temically delivered bone marrow mesenchymal stem cells to the infarcted heart.Cell Transplant. 19, 219–230.

Barreto Filho, J.B., Oliveira, M.S., 2012. Placental structure and biological aspects offetal membranes cultured in vitro. In: Ceccherini-Nelli, L., Matteoli, B. (Eds.),Biomedical Tissue Culture. In Tech, Rijeka, Croatia, pp. 115–132.

Page 7

3 ogy Le

B

C

C

D

D

D

D

G

G

H

H

H

J

K

L

L

L

L

86 M.S. Oliveira et al. / Toxicol

utany, J., Ahn, E., Luk, A., 2009. Drug-related cardiac pathology. J. Clin. Pathol. 62,1074–1084.

arvalho, J.L., Braga, V.B.A., Melo, M.B., Campos, A.C.D.A., Oliveira, M.S., Gomes, D.A.,Ferreira, A.J., Santos, R.A.S., Goes, A.M., 2013. Priming mesenchymal stem cellsboosts stem cell therapy to treat myocardial infarction. J. Cell. Mol. Med. 17,617–625.

hen, Y., Liu, W., Li, W., Gao, C., 2010. Autologous bone marrow mesenchymal celltransplantation improves left ventricular function in a rabbit model of dilatedcardiomyopathy. Exp. Mol. Pathol. 88, 311–315.

ai, W., Hale, S.L., Martin, B.J., Kuang, J.-Q., Dow, J.S., Wold, L.E., Kloner, R.A.,2005. Allogeneic mesenchymal stem cell transplantation in postinfarcted ratmyocardium: short- and long-term effects. Circulation 112, 214–223.

e Angelis, A., Piegari, E., Cappetta, D., Marino, L., Filippelli, A., Berrino, L., Ferreira-Martins, J., Zheng, H., Hosoda, T., Rota, M., Urbanek, K., Kajstura, J., Leri, A., Rossi,F., Anversa, P., 2010. Anthracycline cardiomyopathy is mediated by depletionof the cardiac stem cell pool and is rescued by restoration of progenitor cellfunction. Circulation 121, 276–292.

ominici, M., Le Blanc, K., Mueller, I., Slaper-Cortenbach, I., Marini, F., Krause, D.,Deans, R., Keating, A., Prockop, D., Horwitz, E., 2006. Minimal criteria for definingmultipotent mesenchymal stromal cells. The International Society for CellularTherapy position statement. Cytotherapy 8, 315–317.

u, Y., Lou, H., 2008. Catechin and proanthocyanidin B4 from grape seeds preventdoxorubicin-induced toxicity in cardiomyocytes. Eur. J. Pharmacol. 591, 96–101.

arbade, J., Dhein, S., Lipinski, C., Aupperle, H., Arsalan, M., Borger, M., Barten, M.J.,Lehmann, S., Walther, T., Mohr, F.-W., 2009. Bone marrow-derived stem cellsattenuate impaired contractility and enhance capillary density in a rabbit modelof doxorubicin-induced failing hearts. J. Card. Surg. 24, 591–599.

ianni, L., Herman, E.H., Lipshultz, S.E., Minotti, G., Sarvazyan, N., Sawyer, D.B.,2008. Anthracycline cardiotoxicity: from bench to bedside. J. Clin. Oncol. 26,3777–3784.

azari, M.S., Haykal-Coates, N., Winsett, D.W., Costa, D.L., Farraj, A.K., 2009. Con-tinuous electrocardiogram reveals differences in the short-term cardiotoxicresponse of Wistar-Kyoto and spontaneously hypertensive rats to doxorubicin.Toxicol. Sci. 110, 224–234.

enninger, C., Huelsenbeck, J., Huelsenbeck, S., Grosch, S., Schad, A., Lackner, K.J.,Kaina, B., Fritz, G., 2012. The lipid lowering drug lovastatin protects againstdoxorubicin-induced hepatotoxicity. Toxicol. Appl. Pharmacol. 261, 66–73.

ou, X.-W., Jiang, Y., Wang, L.-F., Xu, H.-Y., Lin, H.-M., He, X.-Y., He, J.-J., Zhang,S., 2009. Protective role of granulocyte colony-stimulating factor against adri-amycin induced cardiac, renal and hepatic toxicities. Toxicol. Lett. 187, 40–44.

iang, W., Ma, A., Wang, T., Han, K., Liu, Y., Zhang, Y., Dong, A., Du, Y., Huang, X., Wang,J., Lei, X., Zheng, X., 2006. Homing and differentiation of mesenchymal stem cellsdelivered intravenously to ischemic myocardium in vivo: a time-series study.Eur. J. Physiol. 453, 43–52.

umar, S.N., Konorev, E.A., Aggarwal, D., Kalyanaraman, B., 2011. Analysis of pro-teome changes in doxorubicin-treated adult rat cardiomyocyte. J. Proteomics74, 683–697.

ahoti, T.S., Patel, D., Thekkemadom, V., Beckett, R., Ray, S.D., 2012. Doxorubicin-induced in vivo nephrotoxicity involves oxidative stress-mediated multiple proand anti-apoptotic signaling pathways. Curr. Neurovasc. Res. 9, 282–295.

ilyanna, S., Martinez, E.C., Vu, T.D., Ling, L.H., Gan, S.U., Tan, A.L., Phan, T.T., Kofidis, T.,2013. Cord lining-mesenchymal stem cells graft supplemented with an omentalflap myocardial revascularization and ameliorates cardiac dysfunction in a ratmodel of chronic ischemic heart failure. Tissue Eng. A 19, 1303–1315.

ipshultz, S.E., Miller, T.L., Scully, R.E., Lipsitz, S.R., Rifai, N., Silverman, L.B., Colan,S.D., Neuberg, D.S., Dahlberg, S.E., Henkel, J.M., Asselin, B.L., Athale, U.H., Clavell,L.A., Laverdière, C., Michon, B., Schorin, M.A., Sallan, S.E., 2012. Changes in cardiacbiomarkers during doxorubicin treatment of pediatric patients with high-riskacute lymphoblastic leukemia: associations with long-term echocardiographicoutcomes. J. Clin. Oncol. 30, 1042–1049.

ipshultz, S.E., Scully, R.E., Lipsitz, S.R., Sallan, S.E., Silverman, L.B., Miller, T.L., Barry,

E.V., Asselin, B.L., Athale, U., Clavell, L.A., Larsen, E., Moghrabi, A., Samson, Y.,Michon, B., Schorin, M.A., Cohen, H.J., Neuberg, D.S., Orav, E.J., Colan, S.D., 2010.Assessment of dexrazoxane as a cardioprotectant in doxorubicin-treated chil-dren with high-risk acute lymphoblastic leukaemia: long-term follow-up of aprospective, randomised, multicentre trial. Lancet Oncol. 11, 950–961.

tters 224 (2014) 380– 386

Machado, V., Cabral, A., Monteiro, P., Gonc alves, L., Providencia, L.A., 2008. Carvedilolas a protector against the cardiotoxicity induced by anthracyclines (doxoru-bicin). Rev. Port. Cardiol. 27, 1277–1296.

Mayhall, E.A., Paffett-Lugassy, N., Zon, L.I., 2004. The clinical potential of stem cells.Curr. Opin. Cell Biol. 16, 713–720.

Mias, C., Lairez, O., Trouche, E., Roncalli, J., Calise, D., Sequelas, M.H., Ordener, C., 2009.Mesenchymal stem cells promote matrix metalloproteinase secretion by cardiacfibroblasts and reduce cardiac ventricular fibrosis after myocardial infarction.Stem Cells 27, 2734–2743.

Oliveira, M.S., Melo, M.B., Carvalho, J.L., Melo, I.M., Lavor, M.S.L., Gomes, D.A.,Góes, A.M., Melo, M.M., 2013. Doxorubicin cardiotoxicity and cardiac functionimprovement after stem cell therapy diagnosed by strain echocardiography. J.Cancer Sci. Ther. 5, 52–57.

Ott, H.C., Matthiesen, T.S., Goh, S.K., Black, L.D., Kren, S.M., Netoff, T.I., Taylor, D.A.,2008. Perfusion-decellularized matrix: using nature’ platform to engineer abioartificial heart. Nat. Med. 14, 213–221.

Paula, A.C.C., Zonari, A.A.C., Martins, T.M., da, M., Novikoff, S., Silva, A.R.P., Cor-relo, V.M., Reis, R.L., Gomes, D.A., Goes, A.M., 2013. Human serum is a suitablesupplement for the osteogenic differentiation of human. Tissue Eng. A 19,277–289.

Planat-Bénard, V., Menard, C., André, M., Puceat, M., Perez, A., Garcia-Verdugo, J.-M.,Pénicaud, L., Casteilla, L., 2004. Spontaneous cardiomyocyte differentiation fromadipose tissue stroma cells. Circ. Res. 94, 223–229.

Rafiyath, S.M., Rasul, M., Lee, B., Wei, G., Lamba, G., Liu, D., 2012. Comparison ofsafety and toxicity of liposomal doxorubicin vs. conventional anthracyclines: ameta-analysis. Exp. Hematol. Oncol. 1, 1–9.

Saracchini, S., Foltran, L., Tuccia, F., Bassini, A., Sulfaro, S., Micheli, E., DelConte, A., Bertola, M., Gion, M., Lorenzon, M., Tumolo, S., 2013. Phase IIstudy of liposome-encapsulated doxorubicin plus cyclophosphamide, followedby sequential trastuzumab plus docetaxel as primary systemic therapy forbreast cancer patients with HER2 overexpression or amplification. Breast 22,1101–1107.

Spallarossa, P., Garibaldi, S., Altieri, P., Fabbi, P., Manca, V., Nasti, S., Rossettin, P.,Ghigliotti, G., Ballestrero, A., Patrone, F., Barsotti, A., Brunelli, C., 2004. Carvedilolprevents doxorubicin-induced free radical release and apoptosis in cardiomy-ocytes in vitro. J. Mol. Cell. Cardiol. 37, 837–846.

Takamatsu, Y., Sunami, K., Muta, T., Morimoto, H., Miyamoto, T., Higuchi, M., Uozumi,K., Hata, H., Tamura, K., 2013. Bortezomib, doxorubicin and intermediate-dosedexamethasone (iPAD) therapy for relapsed or refractory multiple myeloma: amulticenter phase 2 study. Int. J. Hematol. 98, 179–185.

Tan, X., Wang, D., Lu, X., Wei, H., Zhu, R., Zhu, S.S., Jiang, H., Yang, Z.J., 2010.Doxorubicin induces apoptosis in H9c2 cardiomyocytes: role of overexpressedeukaryotic translation initiation factor 5A. Biol. Pharm. Bull. 33, 1666–1672.

Tassan-Mangina, S., Codorean, D., Metivier, M., Costa, B., Himberlin, C., Jouannaud,C., Marie, A., Elaerts, J., Nazeyrollas, P., 2006. Tissue Doppler imaging and con-ventional echocardiography after anthracycline treatment in adults: early andlate alterations of left ventricular function during a prospective study. Eur. J.Echocardiogr. 7, 141–146.

Valiunas, V., Weingart, R., Brink, P.R., 2000. Formation of heterotypic gap junctionchannels by connexins 40 and 43. Circ. Res. 86, E42–E49.

Van-Dalen, E.C., Caron, H.V., Dickinson, H.O., Kremer, L.C.M., 2011. Cardioprotectiveinterventions for cancer patients receiving anthracyclines. Cochrane DatabaseSyst. Rev. 15, CD003917, http://dx.doi.org/10.1002/14651858.CD003917.pub4.

Wang, S., Qu, X., Zhao, R.C., 2012. Clinical applications of mesenchymal stem cells. J.Hematol. Oncol. 5, 1–9.

Watkins, H., McKenna, W.J., Thierfelder, L., Suk, H.J., Anan, R., O’Donoghue, A., Spirito,P., Matsumori, A., Moravec, C.S., Seidman, J.G., Seidman, C.E., 1995. Mutations inthe genes for cardiac troponin T and �-tropomyosin in hypertrophic cardiomy-opathy. N. Engl. J. Med. 332, 1058–1065.

Yang, L., Liu, H., Xu, X., Wang, X., Huang, H., Shi, W., Jiang, S., 2013. Retrospectivestudy of modified SMILE chemotherapy for advanced-stage, relapsed, or refrac-

tory extranodal natural killer (NK)/T cell lymphoma, nasal type. Med. Oncol. 30,720–725.

Zhang, Y., Kang, Y.M., Tian, C., Zeng, Y., Jia, L.X., Ma, X., Du, J., Li, H.H., 2011.Overexpression of Nrdp1 in the heart exacerbates doxorubicin-induced cardiacdysfunction in mice. PLoS ONE 6, e21104.