RIGINAL ARTICLESanshinone IIA protects neonatal rat cardiomyocytesrom adriamycin-induced apoptosis
IE GAO, GUOQING YANG, RONGBIAO PI, RUIFANG LI, PING WANG, HUIJIE ZHANG,ANG LE, SHAORUI CHEN, and PEIQING LIU
UANGZHOU, CHINA
Tanshinone IIA (TSN) is a monomer extracted from the Chinese herb Danshen. In thisstudy, we examined the effect of Tanshinone IIA on adriamycin (ADR)-inducedapoptosis in neonatal rat cardiomyocytes and underlying molecular mechanisms.Primary cultured cardiomyocytes were treated with 1 �mol/L of adriamycin for 24 hwith or without pretreatment with Tanshinone IIA (0.5–2 �mol/L) for 2 h. 3-(4,5-dimethyl thiazol-2yl)-2,5-diphenyltetrazolium bromide (MTT) assay, Hoechst stain-ing, and flow cytometry measurement were used to assess cell viability and apo-ptosis. Fluorescent probes 2’,7’-dichlorofluorescein diacetate and dihydroethidiumwere used to detect the production of reactive oxygen species. Western blottingwas used to evaluate the expression of Bcl-2 and Bax proteins. Adriamycin signif-icantly induced apoptosis in cardiomyocytes. Tanshinone IIA (0.5–2 �mol/L) ame-liorated apoptosis induced by adriamycin in a dose-dependent manner. Tanshi-none IIA (2 �mol/L) markedly attenuated adriamycin-induced reactive oxygenspecies production. Western blotting revealed that Tanshinone IIA prevented theadriamycin-mediated reduction of the ratio of Bcl-2/Bax. In conclusion, TanshinoneIIA significantly inhibits adriamycin-induced cardiomyocyte apoptosis in a dose-dependent manner, and this effect is at least partly caused by its antioxidantproperties. (Translational Research 2008;151:79-87)
driamycin (ADR) is also known as doxorubi-cin, which is an anthracycline derivative; it isone of the most frequently used antineoplastic
gents for treating hematologic malignancies and solid
rom the Laboratory of Pharmacology and Toxicology, School ofharmaceutical Sciences, Sun Yat-sen University, Guangzhou, Peo-le’s Republic of China.
upported by the National Natural Science Fund of China30672459), Major program in Key Field of the People’s Governmentf Guangdong province, People’s Republic of China (2003A 30904),ey Program of Ministry of Education, People’s Republic of China
104146), and Key Program of Zhuhai City, PR of China (2005-
205088). d
umors.1,2 Unfortunately, its effective use in clinic isimited because of the dose-dependent cardiac toxicityhat may lead ultimately to a severe and irreversibleorm of cardiomyopathy known as “adriamycin cardio-
ubmitted for publication July 20, 2007; revision submitted Novem-er 19, 2007; accepted for publication November 26, 2007.
eprint requests: Peiqing Liu, Laboratory of Pharmacology and Tox-cology, School of Pharmaceutical Sciences, Sun Yat-sen University,hongshan No.2 Road #74, Guangzhou 510080, People’s Republicf China; e-mail: [email protected].
yopathy.” It has been well documented that ADR cannduce apoptosis in cultured rat cardiomyocytes3–5 andhat myocardial apoptosis might be a common mecha-ism of acute and chronic myocyte loss; thus, it may because of heart failure.6–8
Several mechanisms have been reported to be possi-ly involved in ADR-induced cardiomyocyte apopto-is, such as intercalation into DNA with consequentnhibition of macromolecular biosynthesis, free radicalormation with consequent induction of DNA damage,ipid peroxidation, DNA binding and alkylation, andirect membrane effects, etc.9 Among them, most stud-es support the view that oxidative stress plays a crucialole in the pathogenesis of adriamycin cardiomyopathynd that scavengers of free radicals can inhibit ADR-nduced cardiomyocyte apoptosis.10,11
Danshen (Radix Salvia miltiorrhiza), which is a well-nown traditional Chinese herb, has been used fre-uently in China for the treatment of cardiovasculariseases, including coronary heart disease, hyperten-ion, and chronic heart failure. Tanshinone IIA (TSN)hich is one of the major lipid-soluble pharmacologic
onstituents of Danshen, is the most abundant form ofanshinone extracted from Danshen and possesses theost characteristic structure. Figure 1 shows the struc-
ural formula of TSN. Recently, our laboratory hashown in a rat model that TSN attenuates atheroscle-otic calcification by inhibition of oxidative stress.12
thers have also reported that TSN has antioxidativeroperties and can protect against oxidative stress in
AT A GLANCE COMMENTARY
Background
The use of the antineoplastic agent adriamycin islimited by its cardiac toxicity known as “adriamy-cin cardiomyopathy.” Oxidative stress has beenreported to be a common mechanism participatingin it. Danshen, which is a traditional Chinese herb,has long been used in China for the treatment ofcardiovascular diseases. Tanshinone IIA, which isextracted from Danshen, possesses antioxidantproperties and anti-tumor activity.
Translational Significance
In this study, the protective effect of the monomerTanshinone IIA implicates that the use of Tanshi-none IIA might be a new solution to adriamycincardiomyopathy in clinic.
itro and in vivo.13–16 However, whether TSN protects c
DR-induced cardiomyocyte apoptosis has never beeneported.
The purpose of this article is to investigate the effectf TSN on ADR-induced apoptosis in primary culturedat cardiomyocytes in vitro and underlying molecularechanisms.
ETHODS
Cell culture and treatment. Cardiomyocytes were ob-ained from neonatal Spragure-Dawley rats (purchased fromhe Center of Experimental Animal of Sun Yat-sen Univer-ity, Guangzhou, People’s Republic of China) using the op-imized repetitive trypsinization method established in ouraboratory.17 After differential adhesion, cardiomyocytesere cultured in DMEM supplied with 10% (vol/vol) heat-
nactivated fetal bovine serum in a humidified atmosphereontaining 5% CO2 for 48 h. Cell cultures were confirmedositive (�95%) for sarcomeric �-actin (Sigma Corporation,t. Louis, Mo) immunostaining using this modified cultureethod. For all experiments, cells were starved for 24 h inulbecco’s modified Eagle’s medium (DMEM) supple-ented with 0.5% fetal bovine serum.Cells were incubated with 1 �mol/L ADR (Sigma) for 24 h
ith or without TSN (kindly provided by Professor Lianquanu, School of Chemistry and Chemical Engineering, Sunat-sen University) pretreatment for 2 h except where indi-
ated otherwise. TSN was dissolved in dimethyl sulfoxideDMSO; Sigma), and the DMSO content in all groups was.1%. The doses and treatment period of ADR and TSN wereetermined on the basis of our preliminary experiments (dataot shown). Meanwhile, the dose of ADR was chosen be-ause it is the plasma drug concentration that is achievable inlinical use.9,18
All animal experimental procedures were performed inccordance with the Guide for the Care and Use of Labora-ory Animals (National Institutes of Health, Bethesda, Md)nd were approved by the Ethical Committee for Animalesearch of Sun Yat-sen University.3-(4,5-dimethyl thiazol-2yl)-2,5-diphenyltetrazolium bro-ide (MTT) assay. In the current study, cardiomyocytes were
lanted in 96-well plates. MTT assay was performed asescribed previously.19 Briefly, MTT (Sigma) was added into
Fig 1. Structural formula of TSN.
ell cultures at a final concentration of 0.5 mg/mL and was
iwts(wu
dswtl3wDlwpp
AcAe7cc1mt
o2lpdwif�HeD(flctwt
damccsw
w
mmc1CabtkpSvTai�igssfiNtHn
�muhA
R
FcitrtwaHiit1(dAcw
Translational ResearchVolume 151, Number 2 Gao et al 81
ncubated for 4 h at 37°C. Subsequently, the culture mediumas removed, and DMSO was added to each well to dissolve
he resulting formazan crystals. The absorbance was mea-ured at a wavelength of 570 nm using a microplate readerBio-Tek Instruments Inc., Richmond, Va). Percent viabilityas defined as the relative absorbance of treated versusntreated control cells.
Hoechst staining. Apoptotic cells were identified by theistinctive condensed or fragmented nuclear structure in cellstained with the chromatin dye Hoechst 33342 (Sigma). Cellsere fixed with 4% paraformaldehyde for 10 min at room
emperature and were washed twice with phosphate buffer so-ution (PBS). Cells were then incubated with 5-�g/mL Hoechst3342 for 15 min. After that, the cells were washed 3 times andere photographed using fluorescence microscope (LeicaFC500 System; Leica Microsystems, Bannockburn, Ill). At
east 500 nuclei from randomly selected fields in each groupere analyzed for each experiment, and the percentage of apo-totic cells was calculated as the ratio of the numbers of apo-totic cells versus total cells counted.
Flow cytometry measurement of hypo-diploid nuclei.poptotic cells can be recognized and distinguished from ne-
rotic cells using flow cytometric analysis of the cellular DNA.s described previously,20 both floating and trypsinized adher-
nt cells were collected, washed with PBS, and fixed in ice-cold0% ethanol at a density of 1 � 106 cells/mL. After fixation,ells were washed twice with PBS and stained with DNA spe-ific fluorochrome propidium iodide (PI) for 20 min. Of the2,000 cells from each sample, the PI fluorescence of nuclei waseasured using flow cytometry (Beckman Coulter Inc., Fuller-
on, Calif).2’,7’-dichlorofluorescein (DCF) detection. The analysis
f fluorescence intensities of DCF formed by the reaction of’,7’-dichlorofluorescein diacetate (H2DCF-DA) with intracel-ular reactive oxygen species (ROS) was performed as describedreviously.5 Briefly, cells were plated on coverslips in 35-mmishes (1 coverslip/dish). After serum starvation for 24 h, cellsere incubated with 10 �mol/L H2DCF-DA (Sigma) for 30 min
n serum-free medium, washed twice with PBS, and then addedresh serum-free medium. Cells were then treated with ADR (10mol/L) for 4 h with or without pretreatment with TSN for 1 h.igher dose and shorter incubation time of ADR was used in this
xperiment because H2DCF-DA was not stable in cells. TheCF fluorescence was detected by fluorescence microscope
Leica DFC500 System; Leica Microsystems) equipped with auorescein isothiocyanate filter, and images were taken using aonstant exposure time. The DCF fluorescence intensity parallelso the amount of intracellular ROS. The fluorescence intensityas quantified using Leica QWin software (Leica Microsys-
ems).Dihydroethidium (DHE) fluorescence intensity. After
rug treatment, cells were detached by trypsin and resuspendedt a concentration of 1 � 106 cells/mL in PBS that contains 10mol/L EDTA. Then, DHE (Sigma) was added at a final con-
entration of 10 �mol/L and incubated for 30 min at 37°C. Theells were then washed with PBS and maintained in 1-mLerum-deprived culture medium. Cellular fluorescence intensity
as determined by flow cytometry.21 The fluorescence values e
ere normalized to the number of cells in each sample.Western blotting. Cardiomyocytes were lysed on ice for 15in with lysis buffer that contained protease inhibitors (50mol/L Tris-HCl, pH 7.4, 1% NP-40, 0.25% sodium deoxy-
ells were then scraped off the dishes, centrifuged at 14000 � gt 4°C for 20 min, and the supernatant was collected. Westernlotting was performed as described previously.22 Briefly, pro-ein concentration was determined using the BCA protein assayit (Pierce Chemical Co, Rockford, Ill). An equal amount ofrotein samples from different groups was separated on 12%DS-polyacrylamide gels and then electrotransferred to a poly-inylidene difluoride membrane (Millipore, Billerica, Mass).he membrane was incubated with specific primary antibodiesgainst Bcl-2 (Cell signaling; 1:1000 dilution), Bax (Cell signal-ng; 1:1000 dilution), and �-tubulin (Sigma; 1:10000 dilution).-Tubulin was used as loading control. The membrane was then
ncubated with the corresponding horseradish peroxidase conju-ated secondary antibodies (Cell signaling; 1:2000 dilution). Theignals were visualized by the enhanced chemiluminescenceystem (ECL, Pierce Chemical Co.) and then exposed to X-raylms (Kodak X-Omat BT Film; Eastman Kodak, RochesterY). Quantitation was performed by scanning and determining
he optical densities of the bands by using National Institutes ofealth (NIH) Image software (version 1.62; NIH). Data wereormalized against those of the corresponding �-tubulin.
Statistical analysis. Data were expressed as the mean valuestandard error of the mean (SEM) of 3 independent experi-
ents. Statistical significance between groups was assessed bysing one-way analysis of variance, followed by Dunnett’s postoc test with the SPSS 11.5 program (SPSS Inc., Chicago, Ill).
P value of �0.05 was considered statistically significant.
ESULTS
ADR induced apoptotic cell death in cardiomyocytes.irst, we examined whether ADR-induced apoptosis inultured cardiomyocytes of neonatal rats under our exper-mental conditions. When cardiomyocytes were exposedo 1 �mol/L ADR for 24 h, cell viability was shown to beeduced to 80.4 � 3.5% (P � 0.05) compared with that ofhe control cells using MTT assay (Fig 2). To determinehether the cell death induced by ADR was mediated by
poptosis, the nuclear morphology was analyzed usingoechst staining assay. The apoptotic cells exhibited typ-
cal fragmented nuclei and condensed chromatin on stain-ng with Hoechst 33342. The percentage of apoptotic cellso the total number of cells was increased from 3.6 �.2% to 21.4 � 2.8% after 1 �mol/L of ADR treatmentFig 3). To confirm even more that ADR-induced celleath is apoptosis, flow cytometric analysis was used.fter incubation with 1 �mol/L ADR for 24 h, the per-
entage of hypo-diploid cells (designated apoptotic cells)as increased from 5.3 � 0.1% in cell cultures without
xposure to ADR to 19.2 � 0.3% (Fig 4).
opiwipco2Mbmt1Wtc1(am
diBd�iocpDr
5os1�Acc
pegpBaB5
Faw(t�3
Fm(A�pptv
Translational Research82 Gao et al February 2008
TSN exerted protective effect against ADR-induced ap-ptosis in cardiomyocytes. Next, we tested whether TSNretreatment could protect cardiomyocytes from ADR-nduced apoptosis. Figures 2–4 showed the effect of TSNith different doses (0.5, 1, and 2 �mol/L) on cell viabil-
ty, nuclear morphology, and DNA content of cells ex-osed to 1 �mol/L ADR. TSN inhibited ADR-inducedell death in a dose-dependent fashion, and the viabilitiesf cardiomyocytes were increased to 85.6 � 2.0%, 89.1 �.2%, and 95.7 � 2.5%, respectively, of control cells inTT assays (Fig 2). The result was evidenced even more
y Hoechst staining and by flow cytometry assays. TSNarkedly decreased the number of apoptotic cells with
ypical fragmented nuclei induced by ADR to 18.0 �.6%, 13.8 � 0.9%, and 6.4 � 1.0%, respectively (Fig 3).hen cell apoptosis was quantified by flow cytometry,
he percentages of apoptotic cardiomyocytes were de-reased significantly to 15.6 � 0.3%, 13.3 � 0.2%, and0.0 � 0.2%, respectively, in a dose-dependent mannerFig 4). In contrast, 2 �mol/L TSN alone did not causeny significant difference in cell viability or in nuclearorphology in cultured cardiomyocytes (Figs 2–4).TSN influenced ROS production in ADR-treated car-
iomyocytes. We used DCF and DHE assays to monitorntracellular oxidant production after ADR exposure.ased on the previous results, we chose 2 �mol/L as theose of TSN used in the following experiments because 2mol/L TSN exerted the best protective effect on ADR-
nduced apoptosis. Figure 5, A showed the effect of TSNn ADR-induced peroxides production in cardiomyo-ytes. ADR increased the number of cells with increasederoxides to 412.1 � 50.2% of control, as estimated byCF fluorescence. Treatment with 2 �mol/L of TSN
ig 2. Cell viability of cardiomyocyte was examined using MTTssay. Cells were treated with 1 �mol/L ADR for 24 h with orithout TSN (0.5, 1 or 2 �mol/L) pretreatment for 2 h or with TSN
2 �mol/L) alone for 26 h. The cell viability of control was adjustedo 100%. The data presented were expressed as the mean � SEM. *P
0.05 vs control. #P � 0.05 vs ADR. The experiment was repeatedtimes with similar results.
educed it markedly to 151.4 � 39.7% of control. Figure t
, B showed the effect of TSN on ADR-induced super-xide anion production in cardiomyocytes. ADR caused aignificant increase in DHE fluorescence intensity to41.8 � 7.5% of control, whereas pretreatment with 2mol/L TSN decreased it to 119.1 � 6.3% of control.lthough the inhibitory effect of TSN on DHE fluores-
ence intensity was not as obvious as on DCF, it was stillonsidered statistically significant.
TSN had a regulatory effect on the ratio of Bcl-2 to Baxrotein expression in ADR-treated cardiomyocytes. Toxplore even more the mechanisms involved, we investi-ated the expression of the anti-apoptotic Bcl-2 and of thero-apoptotic Bax proteins and calculated the ratio ofcl-2 to Bax. The exposure to 1 �mol/L of ADR induceddecrease in the expression of Bcl-2 and an increase ofax, and the ratio of Bcl-2/Bax was decreased to 29.1 �.9% of control. TSN (2 �mol/L) pretreatment prevented
ig 3. Hoechst-stained nuclei of apoptotic myocytes were analyzedorphologically and were expressed as the percentage of total nuclei
magnification � 400). (a) Control cardiomyocyte; (b) 1 �mol/LDR; (c) 1 �mol/L ADR pretreated with 0.5 �mol/L TSN; (d) 1mol/L ADR pretreated with 1 �mol/L TSN; (e) 1 �mol/L ADRretreated with 2 �mol/L TSN; (f) 2 �mol/L TSN alone. The lowestanel was results presented in bar graph. The data were expressed ashe mean � SEM. *P � 0.05 vs control. #P � 0.05 vs ADR. (Colorersion of figure is available online.)
hese effects caused by ADR and significantly reversed
t9
D
Aa
wseise
Translational ResearchVolume 151, Number 2 Gao et al 83
he ADR-induced decrease of the ratio and went back to0. 9 � 9.3% of control (Fig 6).
ISCUSSIONIn the current study, the observed cardiotoxic effects ofDR on cultured neonatal cardiomyocytes using MTT
Fig 4. Cardiomyocytes were stained with PI and ancontent were considered the apoptotic fraction. Rcardiomyocyte; (b) 1 �mol/L ADR; (c) 1 �mol/Lpretreated with 1 �mol/L TSN; (e) 1 �mol/L ADR plowest panel was results presented in bar graph. Thcontrol. #P � 0.05 vs ADR.
ssay, Hoechst staining, and flow cytometric analysis
ere in agreement with reported literature.4,5,11,23–25 Wehowed that TSN protected neonatal rat cardiomyocytesffectively against ADR-induced apoptosis in our exper-ments (Figs 2–4) and found a dose-dependent relation-hip between the concentrations of TSN and its protectiveffects.
flow cytometry. The cells with hypo-diploid DNAere expressed as percent of control. (a) Controltreated with 0.5 �mol/L TSN; (d) 1 �mol/L ADRwith 2 �mol/L TSN; (f) 2 �mol/L TSN alone. Theere expressed as the mean � SEM. *P � 0.05 vs
alyzed byesults w
ADR preretreatede data w
ADR is an effective chemotherapeutic agent; how-
ecoo
ahptcpltcictssiwoow
vs contro
FwB�pRbf0
Translational Research84 Gao et al February 2008
ver, its clinical usefulness is limited because of itsardiac toxicity. Recent studies have reported that ap-ptotic cell death contributes to the loss of cardiomy-
Fig 5. The effect of TSN on ADR-induced reactfluorescence assayed using fluorescence microscopexperiments (magnification � 200). (a) Control cpretreated with 2 �mol/L TSN. Right panel wassoftware. The DCF intensity was shown as fold of� 0.05 vs control. #P � 0.05 vs ADR. B, DHE flupanel was representative histogram of DHE fluorenumber; red line: Control cardiomyocyte; black line2 �mol/L TSN. Right panel was bar graph of mean cin control cells is expressed as 100%. *P � 0.05available online.)
ig 6. After treatment with 1 �mol/L ADR for 24 hours with orithout 2 �mol/L TSN pretreatment for 2 h, the protein levels ofcl-2 and Bax were examined using Western blotting. Blotting of-tubulin showed equal loading of proteins between each lane. Leftanel showed representative result of 3 independent experiments.ight panel was bar graph of gray intensities of the immunoreactiveands analyzed by software. The ratio of Bcl-2/Bax was shown asold of control. The data were expressed as the mean � SEM. *P �.05 vs control. #P � 0.05 vs ADR.
cyte6–8 and that ADR could induce cardiomyocyte
poptosis in vitro.3–5,11,21,23,25 Several mechanismsave been proposed; oxidative stress was the mostopular mechanism. These studies have demonstratedhe generation of superoxide and hydrogen peroxide inardiomyocytes treated with ADR5,11,21,24,25 and haveroposed that the cardiac toxicity of ADR is exerted ateast partially through increased oxidant production inhe heart.11,26–30 The nonfluorescent dye H2DCF-DAan permeate cells freely. Interacting with peroxidesntracellularly, it is hydrolyzed to fluorescent DCF thatan be detected using a fluorescence microscope. Ac-ivation of DCF is relatively specific for the detection ofome peroxides such as hydrogen peroxide. The redox-ensitive fluorophore DHE is often used to detect thentracellular superoxide anion production. Therefore,e used these 2 reagents to reflect ROS production inur study, and we confirmed the generation of super-xide and hydrogen peroxide in cardiomyocytes treatedith ADR (Fig 5).
en species production in cardiomyocyte. A, DCFanel was photos representative of 3 independentcyte; (b) 10 �mol/L ADR; (c) 10 �mol/L ADRof mean DCF fluorescence intensity assayed by
The data were expressed as the mean � SEM. *Pdensity detected using flow cytometry assay. Left
f 3 experiments with similar results. Events: cellL ADR; green line: 1 �mol/L ADR pretreated withscence intensity of DHE. The fluorescence intensityl. #P � 0.05 vs ADR. (Color version of figure is
ive oxyge. Left p
ardiomyobar graphcontrol.
orescencescence o: 1 �mol/ell fluore
TSN, which is the most abundant form of Tanshinone
eDecD
opsa
dnimaBdttfwettucmect
iuhtaciohgA
cmpyTdetds
otaimhhtmpAtpfs
nabcwptcAoo
icvpi
cZ
R
Translational ResearchVolume 151, Number 2 Gao et al 85
xtracted from the widely used traditional Chinese herbanshen, has been proved to possess antioxidant prop-
rties by many studies.12–16 In our study, TSN signifi-antly attenuated the fluorescence intensity of DCF andHE induced by ADR (Fig 5).Our findings revealed that TSN could inhibit the
ccurrence of ADR-induced apoptosis and could sup-ress the increase of oxidative stress. These resultsuggested that the anti-apoptotic action of TSN is prob-bly attributed to its antioxidant activity.Studies reveal that ADR accumulates in mitochon-
ria and is activated enzymatically to form semi-qui-one radicals and superoxide anions in it.31 Therefore,t is of great significance to explore the changes initochondria. It has been proposed that the anti-
poptotic protein Bcl-2 and the pro-apoptotic proteinax of the Bcl-2 family are associated with mitochon-ria-mediated apoptotic signaling through influencinghe permeability of the mitochondrial membrane, andhe ratio of Bcl-2 to Bax may ultimately determine theate of cells.32 Prevention of apoptosis is associatedith an increased expression of Bcl-2 and a decreased
xpression of Bax.33 Our study showed that TSN pre-reatment significantly prevented ADR-induced reduc-ion of the ratio of Bcl-2/Bax. This regulatory effect leds to presume that TSN might decrease ADR-inducedardiomyocyte apoptosis through the mitochondria-ediated apoptotic signaling pathway (Fig 6). How-
ver, additional studies such as measurement of mito-hondrial membrane potential and myocyte ATP con-ent are still needed to confirm this presumption.
It has been demonstrated that myocardial apoptosiss a form of myocardial tissue loss that can be reg-lated and prevented34 and that beneficial effects onemodynamics and cardiac function could be ob-ained by inhibiting cardiomyocyte apoptosis.35 Inddition, removal of intracellular ROS protects myo-ytes from ADR-induced apoptosis.36 Because ADRnduces apoptosis in cardiomyocytes, and inductionf myocardial apoptosis alone is sufficient to induceeart failure, it seems especially necessary and ur-ent to find therapies to protect cardiomyocytes fromDR-induced apoptosis.Treatment of ADR-induced cardiotoxicity has re-
eived considerable attention for a long time. From theany mechanisms proposed to participate in the apo-
tosis induced by ADR, the exact mechanism is still notet fully clear. In the current study, we showed thatSN had a protective effect against ADR-induced car-iomyocyte apoptosis because of its antioxidant prop-rty. This finding is consistent with previous reports onhe protection of ADR-induced apoptosis by antioxi-ants such as C-phycocyanin,5 carvedilol,21 Plantaino-
ide D,23 and amlodipine.25 Because of its potent anti-
xidative property, we have good reasons to suggesthat TSN will become an effective drug to preventdriamycin cardiomyopathy and will help ADR to exertts full potential in clinical use. Furthermore, TSNight be superior to some other antioxidants because it
as pro-apoptotic activity in some cancer cells.37–40 Weave proved this in our experiments using human hepa-ocarcinoma Bel-7402 cells and K562 cells (Supple-entary Figs. 1, A and B). Recently, other signaling
athways such as the MAPK pathway and the PI3K/kt/GSK-3� pathway have also been reported to par-
icipate in anthracycline-induced cardiomyocyte apo-tosis.41,42 Additional studies will be needed in theuture to examine whether other mechanisms exist be-ides antioxidants.
However, one limitation of our study is that we didot test whether TSN also exerts its protective effects inn in vivomodel. This finding should be importantefore TSN’s final application in preventing “adriamy-in cardiomyopathy” in clinic. It is still controversialhether cardiomyocyte apoptosis plays a definiteathogenetic role in ADR-induced cardiomyopa-hy,43–46 although some researchers insist that they doonfirm the presence of cardiomyocyte apoptosis inDR cardiomyopathy in vivo using combined meth-ds.4,47,48 This discrepancy guarantees the necessity ofur future study in an in vivo model.In conclusion, our current work confirms that ADR
nduces apoptosis in cultured neonatal rat cardiomyo-ytes, demonstrates that TSN dose-dependently pre-ents ADR-induced apoptosis, and suggests that thisrotective effect is at least partly caused by the antiox-dant property of TSN.
We thank Mr. Qingyu Kong for his technical support in flowytometry. We also appreciate the assistance of Professor Wenhuaheng for his critical reading of this manuscript.
EFERENCES
1. Ettinghausen SE, Bonow RO, Palmeri ST, et al. Prospectivestudy of cardiomyopathy induced by adjuvant doxorubicin ther-apy in patients with soft-tissue sarcomas. Arch Surg 1986;121:1445–51.
2. Schlumberger M, Parmentier C, Delisle MJ, Couette JE, Droz JP,Sarrazin D. Combination therapy for anaplastic giant cell thyroidcarcinoma. Cancer 1991;67:564–6.
3. Kumar D, Kirshenbaum L, Li T, Danelisen I, Singal P. Apoptosisin isolated adult cardiomyocytes exposed to adriamycin. Ann NYAcad Sci 1999;874:156–68.
4. Ueno M, Kakinuma Y, Yuhki K, et al. Doxorubicin inducesapoptosis by activation of caspase-3 in cultured cardiomyocytesin vitro and rat cardiac ventricles in vivo. J Pharmacol Sci2006;101:151–8.
5. Khan M, Varadharaj S, Shobha JC, et al. C-phycocyanin ame-liorates doxorubicin-induced oxidative stress and apoptosis inadult rat cardiomyocytes. J Cardiovasc Pharmacol 2006;47:9–
20.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
Translational Research86 Gao et al February 2008
6. Saraste A, Pulkki K, Kallajoki M, Henriksen K, Parvinen M,Voipio-Pulkki LM. Apoptosis in human acute myocardial infarc-tion. Circulation 1997;95:320–3.
7. Olivetti G, Abbi R, Quaini F, et al. Apoptosis in the failinghuman heart. N Engl J Med 1997;336:1131–41.
8. Saraste A, Pulkki K, Kallajoki M, et al. Cardiomyocyte apoptosisand progression of heart failure to transplantation. Eur J ClinInvest 1999;29:380–6.
9. Gewirtz DA. A critical evaluation of the mechanisms of actionproposed for the antitumor effects of the anthracycline antibioticsadriamycin and daunorubicin. Biochem Pharmacol 1999;57:727–41.
0. Horenstein MS, Vander Heide RS, L’Ecuyer TJ. Molecular basisof anthracycline-induced cardiotoxicity and its prevention. MolGenet Metab 2000;71:436–44.
1. Kotamraju S, Konorev EA, Joseph J, Kalyanaraman B.Doxorubicin-induced apoptosis in endothelial cells and cardio-myocytes is ameliorated by nitrone spin traps and ebselen. Roleof reactive oxygen and nitrogen species. J Biol Chem 2000;275:33585–92.
2. Tang F, Wu X, Wang T, et al. Tanshinone IIA attenuates ath-erosclerotic calcification in rat model by inhibition of oxidativestress. Vascul Pharmacol 2007;46:427–38.
3. Meng XF, Zou XJ, Peng B, Shi J, Guan XM, Zhang C. Inhibitionof ethanol-induced toxicity by tanshinone IIA in PC12 cells. ActaPharmacol Sin 2006;27:659–64.
4. Lin R, Wang WR, Liu JT, Yang GD, Han CJ. Protective effectof tanshinone IIA on human umbilical vein endothelial cellinjured by hydrogen peroxide and its mechanism. J Ethnophar-macol 2006;108:217–22.
5. Zhang HS, Wang SQ. Nrf2 is involved in the effect of Tanshi-none IIA on intracellular redox status in human aortic smoothmuscle cells. Biochem Pharmacol 2007;73:1358–66.
6. Xia WJ, Yang M, Fok TF, et al. Partial neuroprotective effect ofpretreatment with Tanshinone IIA on neonatal hypoxia-ischemiabrain damage. Pediatr Res 2005;58:784–90.
7. Fu JJ, Gao J, Pi RB, Liu PQ. An optimized protocol for cultureof cardiomyocyte from neonatal rat. Cytotechnology 2005;49:109–16.
8. Mross K, Maessen P, van der Vijgh WJ, Gall H, Boven E, PinedoHM. Pharmacokinetics and metabolism of epidoxorubicin anddoxorubicin in humans. J Clin Oncol 1988;6:517–26.
9. Zhang H, Pi R, Li R, et al. PPAR�/� activation inhibits angio-tensin II-induced collagen type I expression in rat cardiac fibro-blasts. Arch Biochem Biophys 2007;460:25–32.
0. Hishikawa K, Oemar BS, Tanner FC, Nakaki T, Fujii T, LuscherTF. Overexpression of connective tissue growth factor geneinduces apoptosis in human aortic smooth muscle cells. Circu-lation 1999;100:2108–12.
1. Spallarossa P, Garibaldi S, Altieri P, et al. Carvedilol preventsdoxorubicin-induced free radical release and apoptosis in cardi-omyocytes in vitro. J Mol Cell Cardiol 2004;37:837–46.
2. Wang P, Tang F, Li R, et al. Contribution of different Noxhomologues to cardiac remodeling in two-kidney two-clip reno-vascular hypertensive rats: effect of valsartan. Pharmacol Res2007;55:408–17.
3. Kim DS, Woo ER, Chae SW, et al. Plantainoside D protectsadriamycin-induced apoptosis in H9c2 cardiac muscle cells viathe inhibition of ROS generation and NF-kappaB activation. LifeSci 2007;80:314–23.
4. L’Ecuyer T, Allebban Z, Thomas R, Vander Heide R. Glutathi-one S-transferase overexpression protects against anthracycline-induced H9C2 cell death. Am J Physiol Heart Circ Physiol
2004;286:H2057–64.
5. Yamanaka S, Tatsumi T, Shiraishi J, et al. Amlodipine inhibitsdoxorubicin-induced apoptosis in neonatal rat cardiac myocytes.J Am Coll Cardiol 2003;41:870–8.
6. Kang YJ, Chen Y, Yu A, Voss-McCowan M, Epstein PN.Overexpression of metallothionein in the heart of transgenicmice suppresses doxorubicin cardiotoxicity. J Clin Invest 1997;100:1501–6.
7. Doroshow JH, Locker GY, Baldinger J, Myers CE. The effect ofdoxorubicin on hepatic and cardiac glutathione. Res CommunChem Pathol Pharmacol 1979;26:285–95.
8. Doroshow JH. Effect of anthracycline antibodies on oxygenradical formation in rat heart. Cancer Res 1983;43:460–72.
9. Kang YJ, Chen Y, Epstein PN. Suppression of doxorubicincardiotoxicity by overexpression of catalase in the heart of trans-genic mice. J Biol Chem 1996;271:12610–6.
0. Yen HC, Oberley TD, Vichitbandha S, Ho YS, St Clair DK. Theprotective role of manganese superoxide dismutase againstadriamycin-induced acute cardiac toxicity in transgenic mice.J Clin. Invest 1996;98:1253–60.
1. Green PS, Leeuwenburgh C. Mitochondrial dysfunction is anearly indicator of doxorubicin-induced apoptosis. Biochim Bio-phys Acta 2002;1588:94–101.
2. Ferri KF, Kroemer G. Organelle-specific initiation of cell deathpathways. Nat Cell Biol 2001;3:E255–63.
3. Shihab FS, Andoh TF, Tanner AM, Yi H, Bennett WM. Expres-sion of apoptosis regulatory genes in chronic cyclosporine neph-rotoxicity favors apoptosis. Kidney Int 1999;56:2147–59.
4. Ito H, Shimojo T, Fujisaki H, et al. Thermal preconditioningprotects rat cardiac muscle cells from doxorubicin-induced apo-ptosis. Life Sci 1999;64:755–61.
5. Ruixing Y, Wenwu L, Al-Ghazali R. Trimetazidine inhibitscardiomyocyte apoptosis in a rabbit model of ischemia-reperfusion. Transl Res 2007;149:152–60.
6. Wang S, Kotamraju S, Konorev E, Kalivendi S, Joseph J, Kaly-anaraman B. Activation of nuclear factor-kappaB duringdoxorubicin-induced apoptosis in endothelial cells and myocytesis pro-apoptotic: the role of hydrogen peroxide. Biochem J 2002;367:729–40.
7. Liu JJ, Lin DJ, Liu PQ, Huang M, Li XD, Huang RW. Inductionof apoptosis and inhibition of cell adhesive and invasive effectsby tanshinone IIA in acute promyelocytic leukemia cells in vitro.J Biomed Sci 2006;13:813–23.
8. Wang J, Wang X, Jiang S, et al. Growth inhibition and inductionof apoptosis and differentiation of tanshinone IIA in human gli-oma cells. J Neurooncol 2007;82:11–21.
9. Wang X, Wei Y, Yuan S, et al. Potential anticancer activity oftanshinone IIA against human breast cancer. Int J Cancer 2005;116:799–807.
0. Yuan SL, Wei YQ, Wang XJ, Xiao F, Li SF, Zhang J. Growthinhibition and apoptosis induction of tanshinone II-A on humanhepatocellular carcinoma cells. World J Gastroenterol 2004;10:2024–8.
1. Zhu W, Zou Y, Aikawa R, et al. MAPK superfamily plays animportant role in daunomycin-induced apoptosis of cardiac myo-cytes. Circulation 1999;100:2100–107.
3. Rosenoff SH, Olson HM, Young DM, Bostick F, Young RC.Adriamycin-induced cardiac damage in the mouse: a small-animal model of cardiotoxicity. J Natl Cancer Inst. 1975;55:191–4.
4. Zhang J, Clark JR Jr, Herman EH, Ferrans VJ. Doxorubicin-
induced apoptosis in spontaneously hypertensive rats: differen-
4
4
4
4
S
Sf
Translational ResearchVolume 151, Number 2 Gao et al 87
tial effects in heart, kidney and intestine, and inhibition byICRF-187. J Mol Cell Cardiol. 1996;28:1931–43.
5. Li L, Takemura G, Li Y, et al. Preventive effect of erythropoietinon cardiac dysfunction in doxorubicin-induced cardiomyopathy.Circulation. 2006;113:535–43.
6. Li L, Takemura G, Li Y, et al. Granulocyte colony-stimulatingfactor improves left ventricular function of doxorubicin-inducedcardiomyopathy. Lab Invest 2007;87:440–55.
7. Arola OJ, Saraste A, Pulkki K, Kallajoki M, Parvinen M, Voipio-Pulkki LM. Acute doxorubicin cardiotoxicity involves cardiomy-
ocyte apoptosis. Cancer Res 2000;60:1789–92. j
8. Wu S, Ko YS, Teng MS, et al. Adriamycin-induced cardiomy-ocyte and endothelial cell apoptosis: in vitro and in vivostudies.J Mol Cell Cardiol 2002;34:1595–607.
upplementary data
upplementary data associated with this article can beound, in the online version, at doi:10.1016/
