Geranylgeranylacetone Blocks Doxorubicin-Induced Cardiac...

Small Molecule Therapeutics

Geranylgeranylacetone Blocks Doxorubicin-InducedCardiac Toxicity and Reduces Cancer Cell Growth andInvasion through RHO Pathway Inhibition

PolinaSysa-Shah1, Yi Xu1, XinGuo1, Scott Pin1, DjahidaBedja1, Rachel Bartock1, AllisonTsao1, AngelaHsieh1,Michael S. Wolin5, An Moens2, Venu Raman3, Hajime Orita4, and Kathleen L. Gabrielson1

AbstractDoxorubicin is a widely used chemotherapy for solid tumors and hematologic malignancies, but its use is

limited due to cardiotoxicity. Geranylgeranylacetone (GGA), an antiulcer agent used in Japan for 30 years, has

no significant adverse effects, and unexpectedly reduces ovarian cancer progression in mice. Because GGA

reduces oxidative stress in brain and heart, we hypothesized that GGA would prevent oxidative stress of

doxorubicin cardiac toxicity and improve doxorubicin’s chemotherapeutic effects. Nudemice implantedwith

MDA-MB-231 breast cancer cells were studied after chronic treatment with doxorubicin, doxorubicin/GGA,

GGA, or saline. Transthoracic echocardiography was used to monitor systolic heart function and xenografts

evaluated. Mice were euthanized and cardiac tissue evaluated for reactive oxygen species generation,

TUNEL assay, and RHO/ROCK pathway analysis. Tumor metastases were evaluated in lung sections. In

vitro studies using Boyden chambers were performed to evaluate GGA effects on RHO pathway activator

lysophosphatidic acid (LPA)–induced motility and invasion. We found that GGA reduced doxorubicin

cardiac toxicity, preserved cardiac function, prevented TUNEL-positive cardiac cell death, and reduced

doxorubicin-induced oxidant production in a nitric oxide synthase–dependent and independent manner.

GGA also reduced heart doxorubicin-induced ROCK1 cleavage. Remarkably, in xenograft-implanted mice,

combined GGA/doxorubicin treatment decreased tumor growth more effectively than doxorubicin

treatment alone. As evidence of antitumor effect, GGA inhibited LPA-induced motility and invasion

by MDA-MB-231 cells. These anti-invasive effects of GGA were suppressed by geranylgeraniol suggesting

GGA inhibits RHO pathway through blocking geranylation. Thus, GGA protects the heart from doxoru-

bicin chemotherapy-induced injury and improves anticancer efficacy of doxorubicin in breast cancer.

Mol Cancer Ther; 13(7); 1717–28. �2014 AACR.

IntroductionDoxorubicin (Adriamycin) is one of the most widely

used anticancer agents and it is currently a first-choicechemotherapeutic drug for the treatment of primary,recurrent, and metastatic breast cancer (1–4). Unfortu-nately, the use of doxorubicin in breast cancer chemo-therapy is frequently limited due to its severe cumulativecardiac toxicity (5). Clinical signs of cardiac toxicity mayoccur during weeks, months, or even years after chemo-

therapy. Strategies that reduce cardiac toxicity couldpotentially allow higher dosages of doxorubicin to beused with an improved long-term survival in patientswith improved quality of life.

A prevention strategy that reduces oxidative stress, acommon mechanism of doxorubicin cardiac toxicity (5),led us to consider potential benefits of geranylgeranyla-cetone (GGA), an acylic polyisoprenoid (Selbex), whichhasbeenused since 1984as anantiulcerdrug in Japanwithno adverse reactions (6, 7). In multiple animal models ofischemia and reperfusion, GGA prevents oxidative stressin liver, heart, brain, kidney, and retina (8–14). A secondmechanism linked to GGA’s mechanism of action is itsability to inhibit the activation of the RHO family ofGTPases through reduction of protein geranylgeranyla-tion, a posttranslational modification required for RHOfamilymembrane targeting and signaling (15, 16). Target-ing the RHO/ROCK pathway with specific inhibitors isbeneficial in a number of cardiovascular diseases, includ-ing hypertension, angina, ischemia-reperfusion injury,cardiac hypertrophy, chronic heart failure (17–23),all conditions with some level of oxidative stress.

Authors' Affiliations: Departments of 1Molecular andComparative Patho-biology, and 2Cardiology, Johns Hopkins Medical Institutions; 3Depart-ment of Radiology, Johns Hopkins University; 4Department of Pathology,Johns Hopkins University School of Medicine, Baltimore, Maryland; and5Department of Physiology, New YorkMedical College, Valhalla, New York

Current address for HajimeOrita: Juntendo ShizuokaHospital, Departmentof Surgery, Izunokuni, Shizuoka, Japan.

Corresponding Author: Kathleen L. Gabrielson, Johns Hopkins MedicalInstitutions, MRB 807, 733 N. Broadway, Baltimore, MD 21205. Phone:443-287-2953; Fax: 443-287-2954; E-mail: [email protected]

doi: 10.1158/1535-7163.MCT-13-0965

�2014 American Association for Cancer Research.

MolecularCancer

Therapeutics

www.aacrjournals.org 1717

on July 14, 2018. © 2014 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst April 15, 2014; DOI: 10.1158/1535-7163.MCT-13-0965

http://mct.aacrjournals.org/

Interestingly, this molecular activity of GGA might alsomake it useful as a cancer therapeutic because RHOGTPases, and their downstream target, RHO-associatedkinases (ROCK), are implicated in a variety of physiologicfunctions associated with cancer-related changes in theactin cytoskeletal assembly, such as cell adhesion, motil-ity, and migration (24).

Because GGA may reduce a variety of molecular func-tions important for cancer cell growth andmigration, andat the same time, may reduce oxidative stress in the heart,we undertook an investigation to determine whetherGGAcan inhibit the cardiac adverse effects of doxorubicinwhile simultaneously inhibiting cancer cell growth. Spe-cifically, we developed a breast cancer mouse model ofchronic doxorubicin injury to the heart and investigatedcellular and molecular responses to various treatmentsinvolving GGA and doxorubicin.

Materials and MethodsReagents and materials

GGA (Lot # 17022802) was provided by Eisai Co Ltd..GGAwas dissolved in 100% ethanol for in vitro studies. 1-Oleoyl LPA (18:1 LPA; Cat. # 857130) was obtained fromAvanti Polar Lipids. Doxorubicin (Cat. # NDC 55390-238-01) was from Bedford Laboratories. Matrigel (Cat. #354234) and cell culture inserts (Cat. # 353182) for invasionand motility assay were obtained from BD Biosciences.RHO-associated coiled-coil protein kinase 1 (ROCK1)primary antibody (Cat. # A300-457A) was from BethylLaboratories. RPMI 1640 (Cat. # 11835-030) was fromGibco (Life Technologies), FBS (Cat. # SH30088.03) wasfrom HyClone (Thermo Fisher Scientific), penicillin–streptomycin solution (Cat. # 30-001-CI)was fromCellgro.

Cell lineThe human breast cancer cell lineMDA-MB-231 (ATCC)

was obtained in 2010, and was used both for xenograftin vivo studies and in cell culture for in vitro experiments.Cells were grown in RPMI medium 1640, supplementedwith 10% (v/v) FBS, penicillin (10 U/mL)–streptomycin(10 U/mL) at 37�C in humidified 5% CO2 atmosphere.

Animal studiesFive- to six-week-old female athymic nude-Foxn1nu

mice (Harlan Laboratories) were exposed to 500 cGy ofradiation. The next day themicewere anesthetized and anincision was made near the right flank to expose themammary fat pad. A total of 1 � 106 cells (MDA-MB-231 breast cancer cells) were injected using a Hamiltonsyringe into the mammary fat pad. Tumor developmentwas followedandwhenxenografts averaged4mmin eachdimension (length, width, and height, approximately 3weeks after the implantation), mice with comparablesized tumors were randomly divided between the fourtreatment groups: DOX 9 mg/kg, DOX 9 mg/kg andGGA, GGA, and saline. Doxorubicin has been reportedto induce cardiotoxicity in awide range of dosages (4 to 25

mg/kg; refs. 25–29); we selected an intermediate dosage,which would allow for gradual cardiotoxicity develop-ment with multiple doxorubicin injections (25, 26). Doxo-rubicin was administered via tail vein injection every 2weeks for a total of 4 injections. GGA treatment (1 mg/gbodyweight)was given 48hour before doxorubicin per osmethod (pipette). GGA was previously shown to elicit aprotective response when given 24 to 48 hours beforestressor (30–35) in dosages from 200 to 1,000 mg/kg(orally or intraperitoneally; refs. 34–36).

Tumor progression was evaluated by palpation andtumor size measurements with calipers. Tumorvolumes were calculated by the following formula:(1/2 � L � W � H; ref. 37), in which L is the length,W is the width, and H is the height. All mice werehoused under a 12-hour light-dark cycle with free accessto food and water. This study was performed in accor-dance with the "Guide for the Care and Use of Labora-tory Animals" (2011) of the NIH. The protocol wasapproved by the Animal Care and Use Committee ofthe Johns Hopkins Medical Institutions (Baltimore, MD;Animal Welfare Assurance # A-3273-01).

EchocardiographyTransthoracic echocardiography was performed on

conscious mice using Acuson Sequoia C256 ultrasoundmachine (Siemens Corps) equipped with the 15-MHzlinear array transducer. The mouse heart was imaged ina two-dimensional mode followed byM-mode using theparasternal short axis view at a sweep speed of 200mm/sec. Measurements were acquired using the lead-ing-edge method, according to the American Echocar-diography Society guidelines (38). Left ventricle wallthickness and left ventricle chamber dimensions wereacquired during the end diastolic and end systolicphase, including interventricular septum (IVSD), leftventricular posterior wall thickness (PWTED), left ven-tricular end diastolic dimension (LVEDD), and leftventricular end systolic dimension (LVESD). Three tofive values for each measurement were acquired andaveraged for evaluation. The LVEDD and LVESD wereused to derive fractional shortening (FS) to measure leftventricular performance by the following equation: FS(%) ¼ [(LVEDD � LVESD)/LVEDD] � 100.

NecropsyMice were euthanized (CO2) and received postmortem

examination and weighed. The hearts were immediatelyexcised, rinsed in cold PBS, weighed, and sectioned at thelevel of attachment of papillary muscles. Sections of leftventricle, right ventricle, and septumwere frozen in liquidnitrogen formolecular studies. The remainder of the heartwas fixed in 10% formalin for histopathology.

Western blotThe tissue was homogenized in radioimmunoprecipi-

tation assay buffer [25 mmol/L Tris-HCl; 150 mmol/LNaCl; 1% NP-40; 1% sodium deoxycholate; 0.1% SDS; 1�

Sysa-Shah et al.

Mol Cancer Ther; 13(7) July 2014 Molecular Cancer Therapeutics1718




proteinase inhibitor (Roche; cat.#11697498001); 1� phos-phatase inhibitors (Sigma; cat.#P5726 and P0044)] andcentrifuged at 12,000 � g at 4�C for 15 minutes. Proteinmeasurements were performed using a Bio-Rad proteinassay (BioRad); equal amounts of total protein (40 mg)were used per lane. Proteins were denatured in SDS gel-loading buffer (0.125 mol/L Tris, pH 6.8, 20% glycerol,5% mercaptoethanol, 4% SDS, and 0.002% bromophenolblue) at 100�C for 5minutes and separated on a 4% to 12%SDS-PAGE gradient gel using an XCell SureLock Mini-Cell Electrophoresis System with Kaleidoscope-pre-stained molecular weight standards (Bio-Rad). Afterelectrophoresis, the proteins were transferred to a poly-vinylidene difluoridemembrane. Blotswere blockedwith5% nonfat milk in Tris Buffered Saline with 0.1% Tween20 (TBST; 50mmol/LTris-HCl, pH7.4, 150mmol/LNaCl,and 0.1% Tween 20) and probed with primary antibody(diluted in 5% nonfat milk-TBST). Immunoblots wereprocessed with horseradish peroxidase–conjugatedanti-rabbit immunoglobulin G; bound antibodies weredetected using aWestern blot chemiluminescence reagentkit (Pierce).

HistologyHistopathology was assessed in each treatment group

as described previously (39). The hearts were fixed in10% phosphate-buffered formalin, embedded in paraf-fin, sectioned at a thickness of 5 mm, and stained withhematoxylin–eosin (H&E). Lungs were inflated with10% formalin after euthanasia, allowed to fix for 48hours, embedded in paraffin, sectioned at a thicknessof 5 mm, and stained with H&E. Lung sections with 5lobes were scanned by Aperio and evaluated for thepresence of tumor metastases.

Terminal Deoxynucleotidyl Transferase dUTP NickEnd LabelingStaining was performed using DeadEnd Fluorometric

TUNEL (terminal deoxynucleotidyl transferase dUTPnick end labeling) system (Promega) according to themanufacturer’s instructions, as described previously(40). Counterstainingwith 40,6-diamidino-2-phenylindole(DAPI) aided in the morphologic evaluation of normaland apoptotic nuclei, inwhich normal nuclei were stainedas blue and apoptotic nuclei as green. The number ofTUNEL-positive cells within a 2.5-mm2 field in left ven-tricle free wall was counted, and eight randomly selectedfields per slide and five sections per hearts were averagedfor statistical analysis.

Measurement of myocardial superoxide generationFresh-frozen left ventricular myocardium was homo-

genized on ice and sonicated. After centrifugation (30seconds, 4,000 RPM), the supernatant was added to alucigenin (5 mmol/L) solution containing NADPH(100 mmol/L). Superoxide generation was measuredusing lucigenin-enhanced chemiluminescence (BeckmanLS6000IC) and corrected for the baseline value. Superox-

ide generation was expressed as counts per minute(cpm)/mg tissue. Nitric oxide synthase (NOS)–depen-dent superoxide generation was measured by addingL-NAME (100 mmol/L) to the solution (41).

Invasion and motility assayInvasion studies were performed as described previ-

ously (16). MDA-MB-231 cells were plated in serum-freemedium (2 � 104/well) onto 12-well Transwell platescontaining Matrigel. Complete medium (10% FBS) wasadded into the bottomwell and the cellswere incubated at37�C. Lysophosphatidic acid (LPA; 25 mmol/L) wasapplied to stimulate the invasion of the cancer cellsthrough Matrigel and LPA-induced invasiveness wascompared with the basal level of cell invasiveness. Sev-enty-two hours later, the number of cells in the bottomwellwas counted.Motility assayswere performed similarto invasion assay, but no Matrigel was used.

Cell viability/proliferation quantification with MTTassay

MTT (3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazo-lium bromide) assay is based on the MTT conversion intoformazan crystals by live cells, proportionally to mito-chondrial activity, thus allowing to estimate the numberof viable cells. MDA-MB-231 cells were plated in 96-wellplates (0.3� 104/well) and treated with specified concen-trations of doxorubicin, GGA, or both drugs for 24 hours.MTT stock solutionwasprepared [5mgofMTT (SigmaM-5655) to 1 mL of the growth media], further diluted 1:10with growth media, and the cells were incubated in MTTsolution for 2 to 3 hours, until visible formation of blueformazan crystals was confirmed under light microscope.After that, the formazan crystals were solubilizedwith MTT solubilization solution (10% Triton X-100, 0.1N HCL in 200 mL isopropanol) and absorbance was readat 570 nm.

Statistical methodsGraphPad Prism software (GraphPad) was used to

perform statistical analysis. After determining mean andSDs, the unpaired Student t test or one-wayANOVAwereperformed to compare two or three or more unrelatedgroups as appropriate, with a P value of <0.05 deemedsignificant.

ResultsGGA prevents cardiac dysfunction induced bydoxorubicin

To our knowledge, GGA has not been used previous-ly in the setting of doxorubicin therapy. However,because GGA is protective in models of oxidative stress(8–14), we questioned whether GGA would be cardio-protective in mice with oxidative stress in the heartinduced by doxorubicin. To evaluate cardiac function,echocardiography was performed in conscious micefrom four treatment groups: saline control, GGA alone,doxorubicin (9 mg/kg) alone, and doxorubicin

GGA Exerts Cardioprotective and Anticancer Effects

www.aacrjournals.org Mol Cancer Ther; 13(7) July 2014 1719




combined with GGA. We chose to use the dosingschedule, in which GGA treatment is initiated 48 hoursbefore the oxidative stress, because a 48-hour pretreat-ment likely induces the induction of gene expression,and pretreatment is required for maximal protectiveeffect (33, 34). Echocardiography of animals at 6 weeksafter initiation of doxorubicin treatment demonstratedsignificantly decreased systolic function, as evidencedby lower FS percentage (Fig. 1A and B). Remarkably,however, pretreatment with GGA significantly pro-tected animals from the doxorubicin-induced decreasein cardiac function. This protective effect of GGA wasunlikely due to a direct stimulatory effect on cardiac

function, because GGA treatment alone had no effect onsystolic function compared with saline treatment. GGAdid not have a significant effect on the weight loss seenwith doxorubicin treatment (Fig. 1C)

GGA blocks doxorubicin-induced cell death in theheart

Extending these findings of GGA-induced protection ofcardiac function to a cellular level,we investigatedwhetherGGA can reduce doxorubicin-induced cell death in theheart, usingTUNELstaining to evaluateheart sections.Thepercentage of TUNEL-positive nuclei in left ventricles wascompared among four treatment groups: saline control,

Figure 1. Effect of GGA on cardiacfunction in murine model ofdoxorubicin-inducedcardiotoxicity. A, representativeM-mode echocardiograms of themice from saline, GGA,doxorubicin, and doxorubicin incombination with GGA treatmentgroups. Doxorubicin (9 mg/kg)given once every 2 weeks for 4cycles induces contractility deficitat 6 weeks from initial injection.GGA administration (48 hoursbefore doxorubicin) preservescardiac function in the doxorubicintreatment group; GGA alone doesnot have an effect. B, scatter plot ofFS percentage. Doxorubicincauses reduction of FS percentage(P < 0.0001), and addition of GGAto the treatment improves FSpercentage (P < 0.0001); GGAalone had no effect on FSpercentage.C, histogramofmousebodyweights in grams. Data,mean� SD; n ¼ 9 to 12 mice per group.��, P < 0.001; ��, P < 0.0001.

Sysa-Shah et al.





GGAalone, doxorubicin (9mg/kg) alone, anddoxorubicincombinedwithGGA.Wefound thatdoxorubicin treatmentresulted in significant increase in cell death in the hearts,

whereas GGA pretreatment significantly reduced thedoxorubicin-induced cell death (Fig. 2A and B), consistentwith the protective effects seen with measurements of

Figure 2. The effects of GGA treatment on cardiac cell death and ROCK1 cleavage in the hearts of doxorubicin-treatedmice. Histogram (A) and representativeimages (B) of TUNEL-positive nuclei. Mice were euthanized at 6 weeks after the initial injection of doxorubicin. Doxorubicin (9 mg/kg) was given once every2 weeks for 4 cycles. GGA administration was given 48 hours before doxorubicin. Total number of cells per �20 field was quantified by DAPI staining andTUNEL-positive nuclei were expressed as a percentage of total cells (nuclei) in each heart with five fields per heart analyzed. Cell death in the hearts of theanimals treated with doxorubicin was significantly higher than in controls (P ¼ 0.0491). In the group that received combinatory therapy, cell death wassignificantly lower than in the doxorubicin group (P¼ 0.0124). Cell death in theGGAgroupwas not significantly different from the control group. Data,mean�SD; n ¼ 3 to 8 mice per group. �, P < 0.05. C, representative H&E-stained heart sections from mice treated with saline, GGA, DOX, or combination of DOXand GGA. DOX induces fine vacuolization with cardiomyocytes cytoplasm and GGA/DOX combination hearts have fewer vacuoles. The presence ofcytoplasmic vacuoles is used in human and animal studies to assess degree of cardiac toxicity (65–67). D, Western blot of ROCK1. ROCK1 is cleaved in thehearts of doxorubicin-treated mice, but not in the hearts of mice treated both with doxorubicin and GGA. ROCK1 cleavage is further reduced in the hearts ofGGA-treatedmice. ROCK1 full length�160 kDa, cleavedROCK1�130 kDa. E, densitometry of cleavedROCK1 expressed as a fold change from the cleavedROCK1 levels in saline-treated mice hearts. Doxorubicin treatment significantly increased cleaved ROCK1 levels (P ¼ 0.0054), whereas addition of GGA todoxorubicin therapy reduced ROCK1 cleavage (P ¼ 0.0002). GGA treatment did not affect cleaved ROCK1 levels. Data, mean � SD, n ¼ 3 mice per group.��, P < 0.01; ��, P < 0.001.






systolic function (Fig. 1) in corresponding treatmentgroups. H&E sections of heart from each treatment groupreveal that GGA blocks the characteristic cytoplasmicvacuolization induced by doxorubicin (Fig. 2C).

GGA blocks doxorubicin-induced ROCK1 cleavagein the heart

GGA is known to inhibit RHO activation in cancer cells(15, 16), although the role ofGGAonRHO inactivation andROCK1 cleavage in the heart has not been previouslyaddressed.Normally, ROCK1 is folded and auto-inhibited,but RHO–GTP binding prevents that folding and auto-inhibition, thus RHO activates ROCK1. ROCK1 cleavageoccurs during apoptosis and is mediated by activatedCaspase-3 (42). To explore the effects of GGA on thedownstream effectors of RHO signaling, we evaluated theROCK1 cleavage in all four treatment groups by Westernblotting. The cleaved product was reported previously tobe absent in normal hearts, whereas it is present in myo-cardium from heart failure patients (42). ROCK 1 cleavagehas been inducedbydoxorubicin in cell culture, yet it is notknown if ROCK1 cleavage product is present in the myo-cardium after doxorubicin treatment in vivo. A represen-tativeblot shows that cleavedROCK1(130kDa) levelswereincreased in the doxorubicin treatment group, accompa-nied by decreased full-length ROCK1 (top band; 160 kDa),compared with the saline group. The combination of GGAwith doxorubicin treatment significantly reduced the levelof ROCK1 cleavage comparable with saline treatment.Hearts of mice treated with GGA alone showed furtherdecrease of cleaved ROCK1 (Fig. 2D and E).

GGA reduces doxorubicin-induced oxidative stressin the heart

One of the described mechanisms of doxorubicin car-diac toxicity is the generation of highly reactive oxygenspecies (ROS), such as superoxide radical (5, 43). GGAtreatment is known to reduce oxidative stress in variousanimalmodels (8–14), yet GGAhas never been used in thesetting of doxorubicin toxicity. To evaluate the contribu-tion of ROS generation in doxorubicin-induced cardiactoxicity and whether GGA pretreatment would reduceoxidative stress, lucigenin chemiluminescence assay wasperformed tomeasure superoxide in freshly frozenmousecardiac tissue (41). In doxorubicin-treated mice, superox-ide production from the heart was significantly increased,as comparedwith saline controls. GGAtreatment reducedthe basal level of superoxide production and when givenbefore doxorubicin, significantly abolished the doxorubi-cin-induced superoxide production (Fig. 3).

Doxorubicin induces uncoupling of NOS enzymes sothat doxorubicin undergoes redox activation by theenzyme to form a doxorubicin semiquinone and super-oxide (44). Endothelial NOS (eNOS)–deficient mice areless susceptible to doxorubicin-induced cardiac toxicity,suggesting a role of eNOS in the mechanism of oxidantgeneration and cell damage (43). To elucidate the role ofGGA in inhibiting NOS and doxorubicin-induced super-

oxide production, we used L-NAME, a potent inhibitor ofNO synthases. L-NAME did not have a significant effecton basal superoxide generation in controls or in GGA-only–treated animals, but significantly reduced produc-tion of superoxide in doxorubicin-treated animals. L-NAME did not cause further significant change in super-oxide production in doxorubicin–GGA-treated hearts,providing a NOS-mediated mechanism for GGA’s pro-tection against doxorubicin toxicity. L-NAME treatmentdid not result in a complete reduction of superoxideproduction in the doxorubicin-treated mice, suggestingthat other sources of superoxide production exist in doxo-rubicin-treated hearts not related to NOS (Fig. 3).

GGA enhanced doxorubicin-induced chemotherapyeffects in breast cancer xenografts

In light of the cardioprotective effects of GGA duringdoxorubicin therapy, we next askedwhether GGAwouldalso attenuate or enhance the antineoplastic effects ofdoxorubicin. Previously, inactivation of the RHO path-way by GGA was found to inhibit ovarian tumor cellgrowth and invasion (15, 16), leading us to hypothesizethat GGA might actually contribute to antineoplasticactivity during doxorubicin chemotherapy, while at thesame time have a cardioprotective effect. To test thishypothesis, MDA-MB-231 breast cancer cells (45) (witha high level of RHOexpression)were implanted in fat padof athymic nude female mice. When xenograft tumorsbecame palpable, tumors were measured and mice wererandomly assigned to treatment groups. After treatment,compared with saline controls, GGA control mice did nothave a significant difference in tumor volumes. But

Figure 3. Effects of doxorubicin and GGA treatment on total and eNOS-dependent superoxide generation. Doxorubicin treatment significantlyincreased superoxide levels inmurine left ventricles (P¼ 0.0159). L-NAMEhad no effect on superoxide generation in controls, but significantlyreduced doxorubicin-mediated overproduction of superoxide in treatedanimals (P ¼ 0.0159). GGA had no effect on superoxide production byitself, but abolished doxorubicin-induced superoxide production (P ¼0.0077). L-NAME did not result in further reduction of superoxideproduction in doxorubicin-treated compared with controls. Data, mean �SD; n ¼ 3 to 5 mice per group. �, P < 0.05; ��, P < 0.01.

Sysa-Shah et al.





remarkably, GGA given with doxorubicin significantlyenhanced the antitumor effect of doxorubicin comparedwith doxorubicin treatment alone (Fig. 4A).Additionally, we compared metastatic lung tumors

occurrence in all treatment groups, using H&E-stainedlung sections evaluation, evaluating the average numberof tumor nodules per lung section (Fig. 4B). Representativelung sections are provided (Fig. 4C). GGA alone slightlyinhibited the number of tumor nodules per lung section,although the differences did not reach statistical signifi-cance (P¼ 0.2446) due to high individual variability of thenumber of tumor nodules per lung section. Furthermore,these tumorswere highly responsive to doxorubicin alone,and thus any added effect of GGA on decreasing tumorgrowth or metastasis could not be readily appreciated atthe doses of doxorubicin used in our experiments.

GGA inhibited LPA-induced tumor cell motility andinvasion in breast cancer cells

To explore possible cellular mechanisms for the anti-cancer effects of GGA, we examined effects of GGA ontumor cell migration and invasiveness using LPA RHOsignaling stimulus. LPA exerts its biologic effects throughthe LPA receptors in cancer cells (46) and is associatedwith breast cancer cellmigration and invasiveness (47, 48).LPAas an inducer ofRHOsignaling and tumor invasion isused in in vitro systems to study cancer cells invasion (49,50). GGA reduced invasion of ovarian cancer cells in vitro(16), although its role in motility and invasion in breastcancer cells has not been evaluated. Therefore, we per-formed in vitro experiments to assess GGA effects onmotility and invasiveness of MDA-MB-231 breast cancercells that express a high level of RHOA (45). In motility

Figure 4. Effects of GGA onxenograft tumor volumes indoxorubicin-treated mice. A,effects of doxorubicin and GGAtreatment on the average xenografttumor volume. The mice wereinjected with MDA-MB-231 breastcancer cells into the mammary fatpad. Tumor development wasfollowed and themicewere dividedinto four treatment groups: DOX 9mg/kg, DOX 9 mg/kg and GGA,GGA, and saline. Doxorubicin(9 mg/kg) was given once every 2weeks for 4 cycles. GGAadministration was given 48 hoursbefore doxorubicin. Mice wereeuthanized at 6 weeks after theinitial injection of doxorubicin. GGAalone did not affect tumor volumes,but significantly enhancedantitumor effect of doxorubicin(P ¼ 0.0012). Data, mean � SD;n ¼ 5 to 7 mice per group.��, P < 0.01. B and C, lungmetastases were evaluated usingH&E-stained lung sections,comparing the average number oftumor nodules (black arrows) perlung section. GGA treatment (aloneor in combinationwith doxorubicin)did not significantly change thenumber of metastatic nodules.Data, mean � SD; n ¼ 9 to 13 pergroup.






experiments, GGA reduced cellmotility in the presence ofLPA (Fig. 5A). GGA treatment alone reduced the basallevel of invasion, whereas GGA in combination with LPAfurther reduced invasion of MDA-MB-231 cells (Fig. 5B).

Because GGA inhibits the RHO pathway in ovariancancer, we hypothesized that GGA would inhibit RHOactivation in breast cancer cells by inhibiting the RHOfamily member activation through reduced geranylger-anylation. To confirm the role of RHO family geranyl-geranylation in GGA-induced inhibition of invasion ofMDA-MB-231 cells, we applied GGOH (geranylgera-niol) to a subset of experiments testing the mechanismof GGA’s inhibition on LPA-induced migration.GGOH, added to the cells before migration, is metab-olized to geranylgeranylpyrophosphate, which subse-quently aids RHO geranylgeranylation, to rescue theinhibitory effects of GGA (16). In our experiments,GGOH reduced GGA anti-invasive effect, providingevidence that GGA inhibits RHO family geranylgera-nylation in MDA-MB-231 cells suggesting a mechanismfor the role of GGA in reducing tumor cell invasion(Fig. 5B). GGOH also reduced but did not completely

eliminate GGA anti-invasive effect in LPA-treated can-cer cells (similarly to the findings in ovarian cancercells), suggesting additional mechanisms in LPA-induced invasiveness.

GGA decreases cell viability/proliferation in breastcancer cells

To evaluate the effects of GGA alone or in combinationwith doxorubicin on cancer cells viability in vitro, weperformed MTT assay with various concentrations ofGGA and doxorubicin. GGA alone significantly reducedMTT absorbance in MDA-MB-231 cells treated for 24hours, whereas GGA in combination with doxorubicindid not result in a significant change compared withdoxorubicin alone (Fig. 5C).

DiscussionThe optimal cardiac toxicity prevention strategy for

doxorubicin would include an agent that improves theefficacy of the doxorubicin-based cancer therapy andprevents cardiac toxicity. In our experiments, pretreat-ment with GGA blocked doxorubicin cardiac toxicity by

Figure 5. Effects of GGA on invasiveness of MDA-MB-231 breast cancer cells in vitro. Invasion studies were performed with MDA-MB-231 cells. LPA wasapplied to stimulate the motility or invasion of the cancer cells, and LPA-induced motility and invasiveness were compared with the basal level ofcell motility and invasiveness. GGOH treatment was used in invasion studies to counteract theGGA inhibitory effect on cancer cells invasion. A,GGAdoes notaffect basal motility of MDA-MB-231 breast cancer cells, but reduces motility in LPA presence (P ¼ 0.0195). Data, mean � SD; n ¼ 3 wells per group.B, GGA treatment decreases invasiveness of MDA-MB-231 cells in vitro. GGA treatment reduced invasion (P ¼ 0.0423), and GGA in combination with LPAfurther reduced invasion of MDA-MB-231 cells (P¼ 0.0121). GGOH abolishes GGA anti-invasive effect (P¼ 0.0550). Bar graph represents invasion of MDA-MB-231 cells as a total number of cells used in the assay. Data, mean� SD; n¼ 3 wells per group. �, P < 0.05; ��, P < 0.01. C, GGA decreases MDA-MB-231cells viability in vitro, while not significantly altering doxorubicin effects in MTT assay. Data, mean � SD; n ¼ 7 wells per group. ��, P < 0.0001.

Sysa-Shah et al.





maintaining systolic function and decreasing cell death inthe heart. Most remarkably, GGA also contributed todoxorubicin’s chemotherapy efficacy in MDA-MB-231xenografts in parallel with protecting the heart. GGA’santineoplastic effect is likely due to its inhibition of RHOfamily proteins in both the heart and cancer cells, and weselected MDA-MB-231 for these experiments because ofthe endogenous high RHO activity in these cells. BecauseGGA has been used in Japan since 1984 to prevent stom-ach ulcers and has a long history of safety and lack ofadverse effects, we suggest that this novel approach toprevention of doxorubicin toxicity should be furtherinvestigated.By comparison, dexrazoxane, an iron chelator used to

reduce superoxide formation in doxorubicin toxicity (51),currently has limited use due to European MedicinesAgency and Food and Drug Administration restrictions.In 2011, dexrazoxane was restricted for use only inpatients with breast cancer when doxorubicin doseexceeds 300 mg/m2 or epirubicin exceeds 540 mg/m2.This restriction was based on clinical trials that reportedcases of acute myeloid leukemia and myelodysplasticsyndrome in children receiving dexrazoxane (52, 53).Furthermore, there are anecdotal reports of reduced anti-cancer therapeutic efficacy for dexrazoxane, GGA, bycontrast, offers an attractive alternative to dexrazoxane,withmolecular activity that both prevents cardiac toxicityand increases effectiveness of doxorubicin chemotherapy.The efficacy of cardiac protection by GGA may be

related to multiple mechanisms. First, GGA induced areduction in NOS-dependent superoxide production,with overall reduction in oxidative stress in the heart, animportant mechanism of doxorubicin-induced cardio-myocyte cell death and toxicity. Second, GGA also pre-vented doxorubicin-induced ROCK1 cleavage, thusblocking the proapoptotic RHO/ROCK1 pathway in theheart, resulting in less TUNEL-positive cells in the myo-cardium. In the heart, the RHO/ROCK pathway is cur-rently emerging as a potential target for inhibition incardiovascular disease (17–23). In multiple settings ofcardiac disease (17), specific RHO/ROCK inhibitors haveshown promise in disease prevention yet RHO/ROCKinhibitors have never been used to prevent doxorubicincardiac toxicity, although ROCK1 was found to be acti-vated with doxorubicin treatment in vitro (42). This find-ing suggested to us that GGA treatment may be beneficialto induce RHO/ROCK inhibition in the heart and thusreduce cardiac toxicity. We hypothesized that ROCKactivation occurs in hearts in mice treated with doxoru-bicin. In the current study, we demonstrate that ROCKactivation does occur in the heart of mice treated withdoxorubicin and this provides rationale to use GGA as acardioprotective agent. In total, our study demonstratesthat GGA can inhibit doxorubicin-induced RHO activa-tion in the heart, oxidative stress, and cardiac toxicity.The connection between oxidative stress and ROCK1

activation has recently been studied in erythroid cells (54),and we observed that GGA reduced both in the current

study. Doxorubicin treatment induced an elevation ofROCK1 cleavage product, a downstream effect or proteinindicative of RHO pathway activation. In addition, doxo-rubicin induced an increase of superoxide formation. TheL-NAME experiments demonstrated that a portion ofsuperoxide induced by doxorubicin treatment wasNOS-dependent. We suggest that the non–NOS-depen-dent superoxide is related to RHO activation in the doxo-rubicin-treated mice, because GGA further inhibited theextent of superoxide production when comparing doxo-rubicin versus doxorubicin and GGA treatment groups.GGA may also inhibit uncoupled eNOS-dependentsuperoxide production, because L-NAME did notdecrease superoxide production neither in the GGA-onlytreatment group nor in the doxorubicinþGGA treatmentgroup. This mechanism may be possible due to positiveGGA effects on HSP90 expression (55). RHOA-RHOkinase is also a well-documented inducer of oxidativestress, and oxidative stress is known to induce RHOfamily activation (56–58).

Inhibiting the RHO family of proteins has a differentrole in cancer cells, which might explain how GGA couldhave divergent effects on cardiac cells (nonmotile, non-dividing cells) and cancer cells. In ovarian cancer cells,GGAwas shown to inhibit both RHO and RAS activation,possibly through inhibitinggeranylation (15, 16).Multipletypes of human cancers (including breast cancer) have ahigh activity of RHO family proteins, which likely con-tributes to the invasive malignant phenotype and tumormetastasis (59, 60). Because the RHO family is responsiblefor this phenotype, this subgroup of cancers will be moresusceptible to GGA-induced RHO pathway inhibition.We selected MDA-MB-231 due to high RHO activityand found that in xenograft-implanted nude mice,GGAþdoxorubicin treatment also significantly reducedtumor mass versus doxorubicin treatment alone. Thenumbers of metastases in the lung were reduced in GGAversus control groups, although these differences did notreach statistical significance. The numbers of metastaseswere significantly reduced in both doxorubicin groupswith and without GGA. Due to the high effectiveness ofdoxorubicin (a high dose needed to produce cardiactoxicity) in this xenograft model, any added effect of GGAwas masked. However, in our in vitro studies, GGAinhibited LPA-induced Matrigel invasion by MDA-MB-231 cells, suggesting that RHO pathway inhibition is animportant target of GGA in breast cancer cell motility andinvasion. GGA anti-invasive effect was suppressed byGGOH,demonstrating thatRHOgeranylation is inhibitedby GGA.

Our finding of GGA enhancing doxorubicin antitu-mor effect in vivo can be attributed to GGA-negativeeffect on cancer cells viability, as shown in our in vitrostudy and in previously published studies (61). We didnot see an additive effect in combined GGA–doxorubi-cin treatment groups in vitro; however, in vivo multiplefactors contribute to the tumor progression, which maynot be accounted for with in vitro assays on isolated






cancer cells, but which may be affected by GGA treat-ment. For example, angiogenesis, an important factorwhich regulates tumor development, may be affected byGGA. It was shown that inhibition of geranylgeranyla-tion interferes with angiogenesis, evaluated by tubeformation by human dermal microvascular endothelialcells (62). RHO GTPases and ROCK inhibition are alsoreported to affect angiogenesis (63). In addition, theRHO/ROCK1 pathway is involved in cancer cells sur-vival and propagation through various functions,including cell adhesion, motility, and migration (24).ROCK1 is found in MDA-MB-231 cells, although in lowquantities, and GGA-negative effects on the RHO/ROCK1 pathway may also contribute to overall GGAeffects on cancer progression, observed in our study(64).

Our studies demonstrate that GGA has both impor-tant cardioprotective properties and antineoplasticproperties, which together can be therapeutically effec-tive when GGA is used in combination with doxorubi-cin. Because GGA has been used for 30 years withoutadverse effects, we suggest that this novel approach toprevention of doxorubicin cardiac toxicity should beconsidered for clinical testing in patients undertakingdoxorubicin treatment for cancer.

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

Authors' ContributionsConception anddesign:P. Sysa-Shah,D. Bedja,V. Raman,K.L.GabrielsonDevelopment ofmethodology:P. Sysa-Shah,Y. Xu,D. Bedja,A. Tsao,M.S.Wolin, A. Moens, K.L. GabrielsonAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): P. Sysa-Shah, Y. Xu, D. Bedja, R. Bartock, M.S.Wolin, A. Moens, K.L. GabrielsonAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis): P. Sysa-Shah, Y. Xu, S. Pin, D. Bedja,A. Tsao, A. Moens, V. Raman, K.L. GabrielsonWriting, review, and/or revision of the manuscript: P. Sysa-Shah, D.Bedja, R. Bartock, A. Hsieh, M.S. Wolin, A. Moens, K.L. GabrielsonAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): P. Sysa-Shah, Y. Xu, X. Guo, S. Pin,R. Bartock, H. Orita, K.L. GabrielsonStudy supervision: Y. Xu, K.L. Gabrielson

Grant SupportK.L. Gabrielson was supported by the RO1HL088649, Spore in Breast

Cancer Developmental Research Award #P50CA88843-04, Department ofDefense BC030727, and American Heart Association (AHA) SDG0335273N.

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 8, 2013; revised March 17, 2014; accepted April 3,2014; published OnlineFirst April 15, 2014.

References1. Senkus E, Kyriakides S, Penault-Llorca F, Poortmans P, Thompson A,

Zackrisson S, et al. Primary breast cancer: ESMO Clinical PracticeGuidelines for diagnosis, treatment and follow-up. Ann Oncol 2013;24Suppl 6:vi7–23.

2. Cardoso F, Harbeck N, Fallowfield L, Kyriakides S, Senkus EGroupEGW. Locally recurrent or metastatic breast cancer: ESMO ClinicalPractice Guidelines for diagnosis, treatment and follow-up. Ann Oncol2012;23 Suppl 7:vii11–9.

3. Blum RH, Carter SK. Adriamycin: a new anticancer drug with signif-icant clinical activity. Ann Intern Med 1974;80:249–59.

4. http://www.cancer.gov [Internet]. Bethesda: National Cancer Institute;c2013-2014 [updated 2014 Feb 21, cited 2014 Feb 28]. Breast CancerTreatment (PDQ� ): Stage I, II, IIIA, and Operable IIIC Breast Cancer.Available from: http://www.cancer.gov/cancertopics/pdq/treatment/breast/healthprofessional/Page6#Section_859.

5. Octavia Y, Tocchetti CG, Gabrielson KL, Janssens S, Crijns HJ,Moens AL. Doxorubicin-induced cardiomyopathy: from molecularmechanisms to therapeutic strategies. J Mol Cell Cardiol 2012;52:1213–25.

6. Niwa Y, Nakamura M, Miyahara R, Ohmiya N, Watanabe O, Ando T,et al. Geranylgeranylacetone protects against diclofenac-inducedgastric and small intestinal mucosal injuries in healthy subjects: aprospective randomized placebo-controlled double-blind cross-overstudy. Digestion 2009;80:260–6.

7. http://www.eisai.com [Internet]. Tokyo: Eisai Co. Selbex�; c2013-4[updated 2014 Jan 4, cited 2014 Feb 28]. Available from: http://www.eisai.jp/medical/products/di/EPI/SLX_C-FG_EPI.pdf.

8. Ohkawara T, Takeda H, Nishiwaki M, Nishihira J, Asaka M. Protectiveeffects of heat shock protein 70 induced by geranylgeranylacetone onoxidative injury in rat intestinal epithelial cells. Scand J Gastroenterol2006;41:312–7.

9. Tanito M, Kwon YW, Kondo N, Bai J, Masutani H, Nakamura H, et al.Cytoprotective effects of geranylgeranylacetone against retinal pho-tooxidative damage. J Neurosci 2005;25:2396–404.

10. Suzuki S, Maruyama S, Sato W, Morita Y, Sato F, Miki Y, et al.Geranylgeranylacetone ameliorates ischemic acute renal failure viainduction of Hsp70. Kidney Int 2005;67:2210–20.

11. Hirota K, Nakamura H, Masutani H, Yodoi J. Thioredoxin superfamilyand thioredoxin-inducing agents. Ann N YAcad Sci 2002;957:189–99.

12. Hirota K, Nakamura H, Arai T, Ishii H, Bai J, Itoh T, et al. Geranylger-anylacetone enhances expression of thioredoxin and suppressesethanol-induced cytotoxicity in cultured hepatocytes. Biochem Bio-phys Res Commun 2000;275:825–30.

13. Yoda M, Sakai T, Mitsuyama H, Hiraiwa H, Ishiguro N. Geranylger-anylacetone suppresses hydrogen peroxide-induced apoptosis ofosteoarthritic chondrocytes. J Orthop Sci 2011;16:791–8.

14. Nagano Y,Matsui H, ShimokawaO, Hirayama A, Nakamura Y, TamuraM, et al. Bisphosphonate-induced gastrointestinal mucosal injury ismediated by mitochondrial superoxide production and lipid peroxida-tion. J Clin Biochem Nutr 2012;51:196–203.

15. Hashimoto K, Morishige K, Sawada K, Ogata S, Tahara M, Shimizu S,et al. Geranylgeranylacetone inhibits ovarian cancer progression invitro and in vivo. Biochem Biophys Res Commun 2007;356:72–7.

16. Hashimoto K,Morishige K, Sawada K, TaharaM, Shimizu S, SakataM,et al. Geranylgeranylacetone inhibits lysophosphatidic acid-inducedinvasion of human ovarian carcinoma cells in vitro. Cancer 2005;103:1529–36.

17. Shimokawa H, Rashid M. Development of Rho-kinase inhibitors forcardiovascular medicine. Trends Pharmacol Sci 2007;28:296–302.

18. Lin G, Craig GP, Zhang L, Yuen VG, Allard M, McNeill JH, et al. Acuteinhibition of Rho-kinase improves cardiac contractile function in strep-tozotocin-diabetic rats. Cardiovasc Res 2007;75:51–8.

19. Kobayashi N, Horinaka S,Mita S, Nakano S, Honda T, Yoshida K, et al.Critical role of Rho-kinase pathway for cardiac performance andremodeling in failing rat hearts. Cardiovasc Res 2002;55:757–67.

20. BaoW,HuE, TaoL,BoyceR,MirabileR, ThudiumDT, et al. Inhibition ofRho-kinase protects the heart against ischemia/reperfusion injury.Cardiovasc Res 2004;61:548–58.

Sysa-Shah et al.





21. Wolfrum S, Dendorfer A, Rikitake Y, Stalker TJ, Gong Y, Scalia R, et al.Inhibition of Rho-kinase leads to rapid activation of phosphatidylino-sitol 3-kinase/protein kinase Akt and cardiovascular protection. Arter-ioscler Thromb Vasc Biol 2004;24:1842–7.

22. Noma K, Oyama N, Liao JK. Physiological role of ROCKs in thecardiovascular system. Am J Physiol Cell Physiol 2006;290:C661–8.

23. Surma M, Wei L, Shi J. Rho kinase as a therapeutic target in cardio-vascular disease. Future Cardiol 2011;7:657–71.

24. Hall A. Rho GTPases and the actin cytoskeleton. Science 1998;279:509–14.

25. Pacher P, Liaudet L, Bai P, Mabley JG, Kaminski PM, Virag L, et al.Potentmetalloporphyrin peroxynitrite decomposition catalyst protectsagainst the development of doxorubicin-induced cardiac dysfunction.Circulation 2003;107:896–904.

26. Kang YJ, Sun X, Chen Y, Zhou Z. Inhibition of doxorubicin chronictoxicity in catalase-overexpressing transgenic mouse hearts. ChemRes Toxicol 2002;15:1–6.

27. D'Alessandro N, Candiloro V, Crescimanno M, Flandina C, DusonchetL, Crosta L, et al. Effects of multiple doxorubicin doses on mousecardiac and hepatic catalase. Pharmacol Res Commun 1984;16:145–51.

28. Riad A, Bien S, Westermann D, Becher PM, Loya K, Landmesser U,et al. Pretreatment with statin attenuates the cardiotoxicity of Doxo-rubicin in mice. Cancer Res 2009;69:695–9.

29. Doroshow JH, Locker GY, Ifrim I, Myers CE. Prevention of doxorubicincardiac toxicity in the mouse by N-acetylcysteine. J Clin Invest1981;68:1053–64.

30. Ishii Y, Kwong JM, Caprioli J. Retinal ganglion cell protection withgeranylgeranylacetone, aheat shockprotein inducer, in a rat glaucomamodel. Invest Ophthalmol Vis Sci 2003;44:1982–92.

31. Uchida S, Fujiki M, Nagai Y, Abe T, Kobayashi H. Geranylgeranyla-cetone, a noninvasive heat shock protein inducer, induces proteinkinase C and leads to neuroprotection against cerebral infarction inrats. Neurosci Lett 2006;396:220–4.

32. Fujiki M, Hikawa T, Abe T, Uchida S, Morishige M, Sugita K, et al. Roleof protein kinaseC in neuroprotective effect of geranylgeranylacetone,a noninvasive inducing agent of heat shock protein, on delayedneuronal death caused by transient ischemia in rats. J Neurotrauma2006;23:1164–78.

33. Abe E, Fujiki M, Nagai Y, Shiqi K, Kubo T, Ishii K, et al. The phospha-tidylinositol-3 kinase/Akt pathway mediates geranylgeranylacetone-induced neuroprotection against cerebral infarction in rats. Brain Res2010;1330:151–7.

34. Nagai Y, Fujiki M, Inoue R, Uchida S, Abe T, Kobayashi H, et al.Neuroprotective effect of geranylgeranylacetone, a noninvasive heatshock protein inducer, on cerebral infarction in rats. Neurosci Lett2005;374:183–8.

35. ZhuZ, Takahashi N,Ooie T, Shinohara T, YamanakaK, SaikawaT.Oraladministration of geranylgeranylacetone blunts the endothelial dys-function induced by ischemia and reperfusion in the rat heart.J Cardiovasc Pharmacol 2005;45:555–62.

36. ZhangK,ZhaoT,HuangX, LiuZH,XiongL, LiMM, et al. PreinductionofHSP70 promotes hypoxic tolerance and facilitates acclimatization toacute hypobaric hypoxia in mouse brain. Cell Stress Chaperones2009;14:407–15.

37. Tomayko MM, Reynolds CP. Determination of subcutaneous tumorsize in athymic (nude) mice. Cancer Chemother Pharmacol 1989;24:148–54.

38. Sahn DJ, DeMaria A, Kisslo J, Weyman A. Recommendations regard-ing quantitation in M-mode echocardiography: results of a survey ofechocardiographic measurements. Circulation 1978;58:1072–83.

39. Gabrielson KL, Hogue BA, Bohr VA, Cardounel AJ, NakajimaW, KoflerJ, et al. Mitochondrial toxin 3-nitropropionic acid induces cardiac andneurotoxicity differentially in mice. Am J Pathol 2001;159:1507–20.

40. Gabrielson KL, Mok GS, Nimmagadda S, Bedja D, Pin S, Tsao A, et al.Detection of dose response in chronic doxorubicin-mediated celldeath with cardiac technetium 99m Annexin V single-photon emissioncomputed tomography. Mol Imaging 2008;7:132–8.

41. Moens AL, Takimoto E, Tocchetti CG, Chakir K, Bedja D, Cormaci G,et al. Reversal of cardiac hypertrophy and fibrosis from pressure

overload by tetrahydrobiopterin: efficacy of recoupling nitric oxidesynthase as a therapeutic strategy. Circulation 2008;117:2626–36.

42. Chang J, Xie M, Shah VR, Schneider MD, Entman ML, Wei L, et al.Activation of Rho-associated coiled-coil protein kinase 1 (ROCK-1) bycaspase-3 cleavage plays an essential role in cardiac myocyte apo-ptosis. Proc Natl Acad Sci U S A 2006;103:14495–500.

43. Neilan TG, Blake SL, Ichinose F, Raher MJ, Buys ES, Jassal DS, et al.Disruption of nitric oxide synthase 3protects against the cardiac injury,dysfunction, and mortality induced by doxorubicin. Circulation2007;116:506–14.

44. Vasquez-Vivar J, Martasek P, Hogg N, Masters BS, Pritchard KA Jr,Kalyanaraman B. Endothelial nitric oxide synthase-dependent super-oxide generation from adriamycin. Biochemistry 1997;36:11293–7.

45. Pille JY,DenoyelleC, Varet J,Bertrand JR,Soria J,OpolonP, et al. Anti-RhoA and anti-RhoC siRNAs inhibit the proliferation and invasivenessof MDA-MB-231 breast cancer cells in vitro and in vivo. Mol Ther2005;11:267–74.

46. Choi JW, Herr DR, Noguchi K, Yung YC, Lee CW, Mutoh T, et al. LPAreceptors: subtypes and biological actions. Annu Rev PharmacolToxicol 2010;50:157–86.

47. Panupinthu N, Lee HY, Mills GB. Lysophosphatidic acid productionand action: critical new players in breast cancer initiation and pro-gression. Br J Cancer 2010;102:941–6.

48. LiuS,Umezu-GotoM,MurphM, LuY, LiuW, ZhangF, et al. Expressionof autotaxin and lysophosphatidic acid receptors increases mammarytumorigenesis, invasion, and metastases. Cancer Cell 2009;15:539–50.

49. Stahle M, Veit C, Bachfischer U, Schierling K, Skripczynski B, Hall A,et al. Mechanisms in LPA-induced tumor cell migration: critical role ofphosphorylated ERK. J Cell Sci 2003;116:3835–46.

50. Jeong KJ, ChoKH, Panupinthu N, KimH, Kang J, Park CG, et al. EGFRmediates LPA-induced proteolytic enzyme expression and ovariancancer invasion: inhibition by resveratrol. Mol Oncol 2013;7:121–9.

51. Jones RL. Utility of dexrazoxane for the reduction of anthracycline-induced cardiotoxicity. Expert Rev Cardiovasc Ther 2008;6:1311–7.

52. SalzerWL, DevidasM, Carroll WL,Winick N, Pullen J, Hunger SP, et al.Long-term results of the pediatric oncology group studies for child-hood acute lymphoblastic leukemia 1984-2001: a report from thechildren's oncology group. Leukemia 2010;24:355–70.

53. Tebbi CK, London WB, Friedman D, Villaluna D, De Alarcon PA,Constine LS, et al. Dexrazoxane-associated risk for acute myeloidleukemia/myelodysplastic syndrome and other secondary malig-nancies in pediatric Hodgkin's disease. J Clin Oncol 2007;25:493–500.

54. Vemula S, Shi J, Mali RS, Ma P, Liu Y, Hanneman P, et al. ROCK1functions as a critical regulator of stress erythropoiesis and survival byregulating p53. Blood 2012;120:2868–78.

55. Fujimura N, Jitsuiki D,Maruhashi T, Mikami S, Iwamoto Y, KajikawaM,et al. Geranylgeranylacetone, heat shock protein 90/AMP-activatedprotein kinase/endothelial nitric oxide synthase/nitric oxide pathway,and endothelial function in humans. Arterioscler Thromb Vasc Biol2012;32:153–60.

56. Soliman H, Gador A, Lu YH, Lin G, Bankar G, MacLeod KM. Diabetes-induced increasedoxidative stress in cardiomyocytes is sustainedby apositive feedback loop involving Rho kinase and PKCbeta2. Am JPhysiol Heart Circ Physiol 2012;303:H989–1000.

57. Possa SS, Charafeddine HT, Righetti RF, da Silva PA, Almeida-Reis R,Saraiva-Romanholo BM, et al. Rho-kinase inhibition attenuates airwayresponsiveness, inflammation,matrix remodeling, andoxidative stressactivation induced by chronic inflammation. Am J Physiol Lung CellMol Physiol 2012;303:L939–52.

58. Aghajanian A,Wittchen ES, Campbell SL, Burridge K. Direct activationof RhoA by reactive oxygen species requires a redox-sensitive motif.PLoS ONE 2009;4:e8045.

59. Clark EA, Golub TR, Lander ES, Hynes RO. Genomic analysis ofmetastasis reveals an essential role for RhoC. Nature 2000;406:532–5.

60. Itoh K, Yoshioka K, Akedo H, Uehata M, Ishizaki T, Narumiya S. Anessential part for Rho-associated kinase in the transcellular invasion oftumor cells. Nat Med 1999;5:221–5.






61. Okada S, YabukiM, Kanno T, Hamazaki K, Yoshioka T, Yasuda T, et al.Geranylgeranylacetone induces apoptosis in HL-60 cells. Cell StructFunct 1999;24:161–8.

62. Park HJ, Kong D, Iruela-Arispe L, Begley U, Tang D, Galper JB. 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors interferewith angiogenesis by inhibiting the geranylgeranylation of RhoA. CircRes 2002;91:143–50.

63. Bryan BA, Dennstedt E, Mitchell DC, Walshe TE, Noma K, LoureiroR, et al. RhoA/ROCK signaling is essential for multiple aspectsof VEGF-mediated angiogenesis. FASEB J 2010;24:3186–95.

64. Escudero-Esparza A, Jiang WG, Martin TA. Claudin-5 is involved inbreast cancer cell motility through the N-WASP and ROCK signallingpathways. J Exp Clin Cancer Res 2012;31:43.

65. Mettler FP, Young DM, Ward JM. Adriamycin-induced cardiotoxicity(cardiomyopathy and congestive heart failure) in rats. Cancer Res1977;37:2705–13.

66. Billingham ME, Mason JW, Bristow MR, Daniels JR. Anthracyclinecardiomyopathy monitored by morphologic changes. Cancer TreatRep 1978;62:865–72.

67. Chatterjee K, Zhang J, Honbo N, Karliner JS. Doxorubicin cardiomy-opathy. Cardiology 2010;115:155–62.


Sysa-Shah et al.




2014;13:1717-1728. Published OnlineFirst April 15, 2014.Mol Cancer Ther Polina Sysa-Shah, Yi Xu, Xin Guo, et al. RHO Pathway InhibitionToxicity and Reduces Cancer Cell Growth and Invasion through Geranylgeranylacetone Blocks Doxorubicin-Induced Cardiac

Updated version

10.1158/1535-7163.MCT-13-0965doi:

Access the most recent version of this article at:

Cited articles

http://mct.aacrjournals.org/content/13/7/1717.full#ref-list-1

This article cites 65 articles, 16 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/13/7/1717To 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-13-0965

http://mct.aacrjournals.org/content/13/7/1717.full#ref-list-1

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

mailto:[email protected]

http://mct.aacrjournals.org/content/13/7/1717


Geranylgeranylacetone Blocks Doxorubicin-Induced Cardiac...

Documents

Transcript of Geranylgeranylacetone Blocks Doxorubicin-Induced Cardiac...