This Month in Investigative Urology

Zoledronic Acid Induces Autophagic CellDeath in Human Prostate Cancer Cells

Lin et al (page 1490) from Taipei, Taiwan studiedwhether zoledronic acid (ZA) inhibits prostate cancercell growth by autophagy or type II programmed celldeath. Autophagy (self-eating) is a phenomenon ofsequestration and degradation of intracellular or-ganelles in lysosomes. Induction of autophagy wasinvestigated in the PC-3, DU-145, LNCaP andCRW22Rv1 cell lines upon ZA treatment. A charac-teristic of activation of the autophagic pathway is anincrease of LC3-II protein formation and the forma-tion of this protein was detected by Western blot-ting. LC3-II incorporation into autophagosomes wasdetected by immunofluorescent staining. Formationof acidic organelles was detected by acridine orangestaining. Rescue experiments using an apoptosis in-hibitor and/or an autophagy inhibitor were per-formed by MTT assays. Induction of autophagy wasdetected by the formation of the LC3-II protein afterexposure to 100 �M ZA. The investigators foundthat LC3-II and caspase 3 processing could be de-tected at 6 days after treatment. Acidic organelleswere detectable by acridine orange staining, andimmunofluorescent staining showed round-up andcondensed staining of LC3-II, suggesting the forma-tion of autophagosomes in the cytoplasm during au-tophagic cell death. The rescue of cell growth oc-curred only with the administration of an apoptosisand autophagy inhibitor during ZA treatment, indi-cating that ZA induces prostate cancer death byapoptotic and autophagic cell death.

Thus, the authors confirmed previous studies thatZA induces apoptosis in prostate cancer cells, but forthe first time were able to demonstrate that ZA alsoinduces autophagic cell death in prostate cancer celllines regardless of their sensitivity to androgen. Theauthors concluded that future work to identify thelinks between autophagy and apoptosis induced by ZA

is important, and may shed light on the development

0022-5347/11/1854-1176/0THE JOURNAL OF UROLOGY®

© 2011 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.

1176 www.jurology.com

of effective combination therapy to regulate autophagyalong with ZA treatment for prostate cancer.

Histotripsy Fractionation of Prostate Tissue

To characterize the local effects and systemic responseafter histotripsy treatment of prostate tissue in an invivo canine model, Hempel et al (page 1484) from AnnArbor, Michigan applied histotripsy transabdominallyto the prostate in 18 intact male dogs under generalanesthesia. Acoustic bursts (4 microseconds) were de-livered at a 300 Hz pulse repetition rate from a highlyfocused 750 kHz piezoelectric ultrasound transducer.The prostate and surrounding structures were har-vested at prescribed points (0, 7, 28 or 56 days) afterthe procedure. Clinical evaluation, incorporating a val-idated veterinary pain scale, was performed daily, andblood and urine parameters were assessed periodi-cally. Conventional transrectal ultrasound imaging fa-cilitated targeting of the focal volume and providedreal-time assessment of cavitation activity.

Fractionation of the targeted volume and clear-ance of the resultant debris with urination produceda treatment cavity within each prostate. No acousticcollateral damage was seen and urothelialization ofthe treatment cavity occurred within 28 days oftreatment. Only transient laboratory value abnor-malities and minimal hematuria were noted aftertreatment. Pain scores revealed only mild posttreat-ment discomfort, but as the authors emphasize, rel-evant perception of pain associated with histotripsycan only be assessed in human studies. Thus, theauthors demonstrated that histotripsy produced con-sistent tissue fractionation and prostate debulkingwithout collateral acoustic injury or clinical side ef-fects, and that it was well tolerated in the caninemodel. Their findings prompt further evaluation anddevelopment of histotripsy.

Karl-Erik Andersson

Section Editor

Vol. 185, 1176, April 2011Printed in U.S.A.

DOI:10.1016/j.juro.2011.01.008

Zoledronic Acid Induces Autophagic Cell Death in HumanProstate Cancer Cells

Ji-Fan Lin,* Yi-Chia Lin,* Yi-Hsuan Lin, Te-Fu Tsai, Kuang-Yu Chou, Hung-En Chenand Thomas I-Sheng Hwang†From the Central Laboratory, Shin-Kong Wu Ho-Su Memorial Hospital (JFL, YHL), Division of Urology, Department of Surgery, Shin-Kong WuHo-Su Memorial Hospital (YCL, TFT, KYC, HEC, TISH), Division of Urology, School of Medicine, Fu-Jen Catholic University (YCL, TFT, KYC,TISH) and Department of Urology, Taipei Medical University (TISH), Taipei, Taiwan

Abbreviations

and Acronyms

3-MA � 3-methyladenine

AO � acridine orange

AVO � acidic vesicular organelle

Baf A1 � bafilomycin A1

mTOR � mammalian target ofrapamycin

ZA � zoledronic acid

Z-VAD-FMK � benzyloxycarbonyl-val-ala-asp (ome)fluoromethylketone

Submitted for publication July 29, 2010.Study received institutional review board ap-

proval.Supported by Shin-Kong Wu Ho-Su Memorial

Hospital Grants SKH-8302-98-DR-14 (YCL), SKH-8302-98-DR-16 and SKH-8302-97-NDR-08 (TISH).

* Equal study contribution.† Correspondence: Division of Urology, De-

partment of Surgery, Shin Kong Wu Ho-Su Me-morial Hospital, Taipei 101, Taiwan (telephone:886-2-28332211, extension 2065; FAX: 886-2-28389404; e-mail: M001009@ms.skh.org.tw).

Purpose: Bisphosphonates are potent inhibitors of bone resorption. In vitrostudies show that zolendronic acid inhibits prostate cancer cell growth by acti-vating apoptosis. We investigated whether zolendronic acid also inhibits prostatecancer cell growth by autophagy (type II programmed cell death).Materials and Methods: We investigated the induction of autophagy in the PC-3,DU-145, LNCaP and CRW22Rv1 cell lines upon zolendronic acid treatment.LC3-II protein formation was detected by Western blot. LC3-II incorporation intoautophagosomes was detected by immunofluorescence staining. Acidic organelleformation was detected by acridine orange staining. Rescue experiments using anapoptosis inhibitor and/or an autophagy inhibitor were performed by MTT assay.Results: Autophagy induction was detected by formation of the LC3-II proteinafter exposure to 100 �M zolendronic acid. LC3-II and caspase-3 processing wasdetected 6 days after treatment. Acidic organelles were detectable by acridineorange staining and immunofluorescence showed round-up and condensed stain-ing of LC3-II, suggesting autophagosome formation in the cytoplasm duringautophagic cell death. Cell growth was rescued only by administering an apopto-sis and autophagy inhibitor during zolendronic acid treatment, indicating thatzolendronic acid induces prostate cancer death by apoptotic and autophagic celldeath.Conclusions: To our knowledge we report the first study showing that zolen-dronic acid markedly inhibits human prostate cancer cell growth through au-tophagic cell death. Zolendronic acid shows its anticancer activity via apoptosisand autophagy. These findings can potentially contribute to the beneficial use ofzolendronic acid for prostate cancer treatment.

Key Words: prostate, prostatic neoplasms, zoledronic acid,

autophagy, apoptosis

1490 www.jurology.com

PROSTATE cancer is among the mostprevalent cancers in men worldwide.1

Treatment of prostate cancer withmetastasis is a real challenge for cli-nicians. Current effective treatmentfor hormone sensitive prostate cancerwith metastasis mainly focuses on hor-monal therapy with a luteinizing hor-

mone-releasing hormone agonist and/or

0022-5347/11/1854-1490/0THE JOURNAL OF UROLOGY®

© 2011 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RES

antiandrogens.2 However, prostatecancer eventually becomes hormoneresistant and the only effective treat-ment under this condition is chemo-therapy with docetaxel.3 Bones, espe-cially the central bones (spine, ribsand pelvis), are the most common or-gans to be involved in patients with

metastatic prostate cancer.4 Patients

Vol. 185, 1490-1496, April 2011Printed in U.S.A.

EARCH, INC. DOI:10.1016/j.juro.2010.11.045

ZOLEDRONIC ACID INDUCES AUTOPHAGY IN HUMAN PROSTATE CANCER CELLS 1491

with bony metastasis may experience skeletal re-lated events, which can cause disability such as spi-nal cord compression. Decreased bone health mayresult from cancer metastasis or from treatment,especially androgen deprivation therapy.5

Bisphosphonates are recognized as effective med-ical treatment for tumor associated bone diseasessuch as breast cancer,6 myeloma7 and prostate can-cer.8 Of the bisphosphonates ZA decreases the inci-dence of skeletal related events, improving qualityof life in patients with bony metastasis from pros-tate cancer.9 Recent basic studies showed that ZAinduces apoptosis,10 inhibits cell adhesion11 and hasantimetastatic potential12 in prostate cancer cells invitro. The beneficial effects of combining ZA withconventional therapeutic agents may possiblyachieve cancer control in patients with metastaticprostate cancer.

Autophagy (self-eating) is a phenomenon of se-questration and degradation of intracellular organ-elles in lysosomes. Autophagy was first observed incells exposed to starvation and is thought to be aprotective mechanism for cells to survive difficultsituations.13 However, autophagy participates inthe process of programmed cell death.14 The mech-anism of the dual effects has not been totally re-vealed. Efforts have been made to apply this mech-anism to cancer therapy.

We report a novel response to ZA in PC3, DU145,CWR22Rv1 and LNCaP human prostate cancer cellsinvolving the formation of AVOs and autophagy. Toour knowledge this is the first description indicatingthe induction of autophagy as a cell death mecha-nism in prostate cancer cells by a bisphosphonate.

MATERIAL AND METHODS

ChemicalsA stock solution of ZA was prepared at a concentration of1 mM in distilled water and aliquots were stored at �20C.Unless otherwise mentioned, all other chemicals were ob-tained from Sigma®.

Cell CultureFour human prostate cancer cell lines were maintained at37C under 5% CO2. PC3 (CRL-1435) and DU145 (HTB-81)cells (ATCC®) were cultured in Dulbecco’s modified Ea-gle’s medium and minimum essential medium (Invitro-gen™), respectively. CWR22Rv1 (CRL-2505) and LNCaP(CRL-1740) cells (ATCC) were cultured in RPMI-1640 me-dium (Invitrogen). Medium was supplemented with 2 mMGlutaMAX™-I, 100 U/ml penicillin, 10% fetal bovine se-rum and 100 �g/ml streptomycin (Invitrogen).

LC3 Immunoblotting and ImmunofluorescenceCancer cell death by autophagy (type II programmed celldeath) treated with several anticancer drugs was recentlyproposed.15 Therefore, we examined autophagy in pros-

tate cancer cells treated with ZA. For immunoblot analy-

sis cells subjected to the indicated treatment were harvestedand lysed, and protein concentration was determined bybicinchoninic acid protein assay (Pierce, Rockford, Illi-nois). Proteins were separated by 14% sodium dodecylsul-fate polyacrylamide gel electrophoresis and transferred topolyvinylidene difluoride membranes (Millipore, Billerica,Massachusetts) before probing with antibodies, includingMAP1LC3B (Abgent, San Diego, California) and caspase-3 (ac-tive) (Epitomics®). Subsequently immunoblotting wasdone using a chemiluminescence procedure (Millipore) ac-cording to manufacturer instructions. To perform immu-nofluorescence detection of LC3 cells were seeded inchamber slides (BD™) and subjected to the indicatedtreatments. After removing medium cells were washedtwice with phosphate buffered saline and fixed withice-cold methanol at �20C, followed by permeabiliza-tion with 0.25% Triton X-100 at 4C. The primary anti-body to LC3 (MAP1LC3B) was incubated at 4C over-night before incubation with fluorescein isothiocyanatesecondary antibody (Jackson ImmunoResearch Labora-tories, West Grove, Pennsylvania). Analysis was per-formed using a fluorescence microscope (Olympus®)equipped with the appropriate excitation and emissionfilters.

AO Staining to Detect AutophagyCells were seeded in 6-well plates and treated with ZAfor 72 hours before AO staining according to a publishedprotocol.16 In the study of Baf A1 (Santa Cruz Biotech-nology, Santa Cruz, California) cotreatment cells weretreated with 200 nM Baf A1 for 40 minutes beforeadding AO.

Proliferation AssayThe effect of ZA on cell growth was assayed by the MTTmethod, as previously described.10 For proliferation stud-ies done with the autophagy inhibitor 3-MA and thebroad-spectrum caspase inhibitor Z-VAD-FMK (Promega,Madison, Wisconsin) cells were administered ZA in thepresence of 1 mM 3-MA or 20 �M Z-VAD-FMK for theindicated durations. Cell viability was determined as apercent of control. Each condition was performed in 6replicate wells and data were obtained from at least 3separate experiments.

Statistical AnalysisAll experiments were done in triplicate. Results are shownas the mean � SD. SigmaPlot®, version 10 was used fordata analysis and graphic presentation.

RESULTS

ZA Induced Autophagic

Prostate Cancer Cell Death

Increased LC3-II protein in PC3, CWR22Rv1 and

LNCaP cells. A characteristic of activating the au-tophagic pathway is increased LC3-II protein forma-tion.17 Processing of LC3-II protein was observed inZA treated PC3, CWR22Rv1 and LNCaP cells at 48hours, and it further increased at 72 hours (fig. 1). In

DU145 cells no LC3-II was detectable with or with-

ZOLEDRONIC ACID INDUCES AUTOPHAGY IN HUMAN PROSTATE CANCER CELLS1492

out ZA treatment (fig. 1), even at higher ZA concen-trations or with longer incubation (data not shown).

LC3-II processing and caspase-3 activation at 6

days. ZA significantly inhibited the growth of all 4cell lines in a time and dose dependent manner byactivating apoptosis.10,12 In our previous study wealso detected the activation of caspase-3 activity,and the expression of pro-apoptosis and anti-apop-tosis genes in cells treated with ZA.18 As reported by

Figure 1. Increased autophagy marker LC3-II protein in PC3,CWR22Rv1 and LNCaP cells treated with ZA. Cells were treatedwith vehicle (�) or 100 �M ZA (�) for 24, 48 and 72 hours beforeharvest. Total cell extracts were probed with antibodies to LC3and �-actin as internal control. Data are from 1 experimentrepeated in triplicate with similar results.

Figure 2. Autophagy induction coexisted with apoptosis 4, 5and 6 days after treatment in ZA treated PC3 cells. ProcessedLC3-II and cleaved caspase-3 were analyzed by Western blottingin PC3 cells treated with vehicle (�) or 100 �M ZA (�) for 4, 5 and

6 days with �-actin as internal control.

Dumon et al,10 we noted substantial apoptotic ef-fects on day 2, which doubled by day 6 in ZA treatedPC3 cells. To test whether autophagy was only atransient effect we measured the induction of LC3-IIprocessing and caspase-3 cleavage on days 4, 5 and6 after treatment. Autophagy induction coexistedwith apoptosis in ZA treated PC3 cells (fig. 2).

LC3 aggregation in vesicular structures of treated

cells. To further analyze signal autophagosome for-mation we used anti-LC3 immunofluorescence in acell based detection system in PC3 and CWR22Rv1cells. In ZA treated cells LC3 protein increased intotal abundance and was aggregated in granular/vesicular structures (fig. 3, A). We also quantifiedthe autophagosomes in ZA and vehicle treated cellsto determine the fraction of cells with increased LC3aggregation levels since different cell types showvarying degrees of autophagocytosis at baseline.19

Results revealed that ZA treatment significantly in-

Figure 3. LC3 aggregation in ZA treated cells. Fluorescence mi-croscopy reveals PC3 and CWR22Rv1 cells labeled with 4=,6-diamidino-2-phebylindole for DNA (blue areas) and fluoresceinisothiocyanate secondary antibody against MAP1LC3B (greenareas) (A). Number of cells with significant autophagy, definedas punctuate LC3 positive vacuoles in 4 � 200 fields (B). Data are

from 1 experiment repeated 3 times with similar results.

ZOLEDRONIC ACID INDUCES AUTOPHAGY IN HUMAN PROSTATE CANCER CELLS 1493

creased the cellular abundance of autophagosomes41.47% vs 0.8% in PC3 cells and 25.07% vs 1.33% inCWR22Rv1 cells (fig. 3, B).

AO staining of acidic autophagolysosomes in

treated cells. Another characteristic of autophagyis AVO formation, which is detected and measured byvital staining with AO.17,20 Vital staining of PC3 andCWR22Rv1 cells with AO showed the accumulation ofAVOs in the cytoplasm of cells treated with 100 �MZA (fig. 4, B and F). This provides further evidencethat the autophagic process was activated by ZAtreatment. The H� adenosine triphosphatase inhib-itor Baf A1 prevents the transition of autophago-somes to autophagolysosomes by disrupting lyso-some fusion to autophagosomes.20 When Baf A1 wasco-administered, the accumulation of AVOs inducedby ZA was markedly decreased (fig. 4, D and H).Results indicate that the autophagosomes formed byZA treatment undergo the same fusion process withlysosomes.

Figure 4. Acidic vesicle formation. PC3 (A to D) and CWR22Rv1(E to H) cells treated with vehicle (A, C, E and G) or 100 �M ZA(B, D, F and H) for 72 hours were stained with AO in absence (A,B, E and F) or presence (C, D, G and H) of Baf A1 cotreatment.Green areas indicate cytoplasm and nucleoli. Red and orangered areas indicate acidic compartments. Data are from 1 exper-

iment repeated 3 times with similar results.

Rescue of ZA Treated Prostate Cancer Cell Death

Rescue of cell death in ZA treated prostate cancercells only occurred by the simultaneous inhibition ofautophagy and apoptosis. The known inhibitor ofautophagy 3-MA, which acts by inhibiting class IIIphosphatidylinositol 3-kinase,17 was used to assesswhether the induction of autophagy in ZA treatedcells contributed to the cell death observed. We firstdetected LC3-II protein levels in cells treated with3-MA plus ZA and compared that with levels in cellstreated with ZA alone. Adding 3-MA decreased thelevel of ZA induced LC3-II protein in PC3,CWR22Rv1 and LNCaP cells (fig. 5). This suggeststhat autophagy in ZA treated cells is regulated bythe class III phosphatidylinositol 3-kinase. Inhibit-ing LC3-II protein in DU145 cells treated with ZAand 3-MA was not evaluated since LC3-II was notdetected in this cell line (fig. 2).

The data presented indicate that ZA treatmentinduced apoptosis and autophagy in the tested celllines. These data raise the question of whether au-tophagy is an important mechanism of cancer celldeath due to ZA treatment in cells with the ability toundergo apoptosis. To address this question we usedthe autophagy inhibitor 3-MA and the broad-spec-trum caspase inhibitor Z-VAD-FMK to perform arescue experiment in cells treated with ZA. The vi-ability of PC3, DU145, CWR22Rv1 and LNCaP cellstreated with 100 �M ZA was decreased to a mean of25.37% � 2.38%, 20.70% � 2.84%, 19.13% � 0.69%and 30.00% � 1.03%, respectively (fig. 6). Viabilitywas only rescued in cells treated with the simulta-neous addition of 3-MA and Z-VAD-FMK, including106.57% � 4.01% in PC3, 135.89% � 3.59% inDU145, 111.86% � 2.22% in CWR22Rv1 and 73.78% �

Fig. 5. LC3-II formation was inhibited in cells treated with ZAplus autophagy inhibitor 3-MA. Immunoblot analysis of LC3-IIand �-actin in cells treated for 72 hours with 100 �M ZA (lanes 1,3 and 5) or 100 �M ZA plus 0.5 mM 3-MA (lanes 2, 4 and 6). Dataare from 1 experiment repeated 3 times with similar results.

9.54% in LNCaP cells.

ZOLEDRONIC ACID INDUCES AUTOPHAGY IN HUMAN PROSTATE CANCER CELLS1494

DISCUSSION

To our knowledge the current study provides thefirst experimental evidence of the activation of anautophagic program in human prostate cancer cellstreated with ZA. In PC3, CWR22Rv1 and LNCaPcells ZA mediated autophagy was characterized byLC3-II induction, followed by the detection of AVOformation and the recruitment of processed LC3-IIto autophagosomes. Also, ZA induced cell death wasassociated with autophagic and apoptotic processes.Inhibiting either pathway alone failed to rescue thecell death caused by ZA treatment.

Autophagy is important in normal development,in cellular responses to environmental stimuli andin numerous diseases.21 Although autophagy is rec-ognized as an important cellular process, it has now

Fig. 6. Simultaneous addition of Z-VAD-FMK and 3-MA rescued Zof dimethyl sulfoxide vehicle as control, 100 �M ZA, 20 �M Z-VZA plus 0.5 mM 3-MA and 100 �M ZA plus 20 �M Z-VAD-FMK andrepeated 3 times. Results are shown as mean � SD.

become an exciting area of cancer research with

exponential growth in the number of publications onthis topic. Several recent reviews show that au-tophagy is related to cancer and other diseases.19,22

Many anticancer agents, including tamoxifen, rapa-mycin, arsenic trioxide, temozolomide, histone deacety-lase inhibitors and ionizing radiation,23 were re-ported to induce autophagy. These reports suggestthat autophagy may be an important mechanism,besides apoptosis, in the killing of tumor cells bythese agents.19

In prostate cancer cells several cancer relevantagents also induce autophagy to protect or furthersensitize cells to treatment. Sulforaphane, a nat-urally occurring member of the isothiocyanatesfrom cruciferous vegetables, induces apoptosisand further investigation revealed that sulforaphane

ced cell death. Cell viability after 72 hours of culture in presenceK, 100 �M ZA plus 20 �M Z-VAD-FMK, 0.5 mM 3-MA, 100 �M

M 3-MA. Data represent mean of 4 determinations per condition

A induAD-FM0.5 m

causes autophagy as a defense mechanism against

ZOLEDRONIC ACID INDUCES AUTOPHAGY IN HUMAN PROSTATE CANCER CELLS 1495

apoptosis in PC3 and LNCaP cells.17 Licochal-cone-A induces autophagy related cell death, inaddition to apoptotic cell death in LNCaP cells.20

Pentagalloylglucose, which is abundantly presentin several Asian medicinal herbs, also inducesautophagy and caspase independent programmeddeath in PC3 cells.24

A characteristic marker of autophagy is cleavageof the cytosolic LC3-I protein and its further lipida-tion to form LC3-II, which is subsequently recruitedto autophagosomes. Our data revealed that ZA in-duced autophagy resulted in the formation of LC3-IIin PC3, CWR22Rv1 and LNCaP cells (fig. 1). Thissuggests that the signaling context that activatedautophagic flux was irrelevant to the androgen sen-sitivity of these cells. However, we could not detectLC3-II formation in DU145 cells. Autophagy is sup-pressed by functional p53 and certain cytoplasmicp53 mutants.25 Increased p53 transcription wasonly observed in DU145 cells treated with ZA,18

which may partially explain the suppression of au-tophagy in this cell line. However, PC3 cells, whichlack functional p53 and LNCaP cells expressingwild-type p53, were sensitive to ZA induced au-tophagy and apoptosis. Additional experiments arerequired to explore the role of p53 in ZA inducedautophagy. LC3 aggregation was detected in cellstreated with ZA. However, recent studies revealedover expression of green fluorescent protein-LC3 inthe organized smooth endoplasmic reticulum26 andprotein aggregates.27 Cell permeabilization with de-tergents can also lead to punctuate staining of cellsby green fluorescent protein-LC3.28 Thus, artifactsmay be present if LC3 is over expressed or aggre-gates are formed in cells27 when interpreting LC3localization during autophagy.

In the literature 3-MA has been widely used as anautophagy inhibitor. A recent study demonstratedthat 3-MA promotes autophagy with prolongedtreatment under nutrient rich conditions, in whichit can suppress starvation induced autophagy.29 Inour study the inhibition of LC3-II formation wasnoted when 1 mM 3-MA was administered to ZAtreated cells, suggesting an inhibitory role of 3-MAin this experiment. However, we cannot rule out thepossible induction of autophagy in prostate cancercells after increasing the 3-MA concentration. Cellrescue experiments using the cotreatment of inhib-itors with ZA in all 4 cell lines showed consistenttrends. Supplementing an autophagy inhibitor or acaspase inhibitor alone rescued a small percent ofcells but failed to fully recover cell viability. Neitherinhibitor showed a superior protective effect, sug-gesting that autophagic and apoptotic cell death oc-

curred in ZA treated cells.

To further identify which ZA involved signalingpathways are responsible for autophagy we will as-sess the key pathway (the mTOR kinase complex) infuture studies. ZA inhibited prostate cancer cell pro-liferation by parallel inactivation of the extracellu-lar signal-regulated kinase/Akt pathway,30 whichprovides upstream signals to mTOR activation.mTOR inhibition increases the level of Atg8 (LC3)and kinase activity of Atg125, which are key compo-nents of autophagy. Since we observed the process-ing of LC3 after ZA treatment, we cannot rule outthe possibility that ZA targets mTOR through theextracellular signal-regulated kinase/Akt pathwayto trigger autophagy. We will examine the effects ofZA on the inhibition of mTOR kinase activity infuture studies, which may reveal the molecularmechanism underlying ZA induced autophagy.

A limitation of this study is the lack of electronmicroscopic evidence of autophagolysosomes. How-ever, we used another 2 indirect methods, namelyLC-3 II protein detection and AO staining, to con-firm autophagy. Although LC-3 II protein was foundin other organelles, such as the endoplasmic reticu-lum, combining the 2 detection methods may havedecreased bias. Several cell line studies confirmedthat ZA has anticancer effects but no clinical evi-dence has shown significant benefit. The plasmaconcentration of ZA after injecting 4 mg is around 1�M, which is insufficient to achieve an effective an-ticancer concentration. However, the concentrationin bony tissue is much higher and anticancer effectmay be more prominent in bony tissue. Further clin-ical investigations should be done to clarify thisissue.

CONCLUSIONS

We confirmed results in previous studies that ZAinduces apoptosis in prostate cancer cells. Further-more, we noted that ZA induces autophagic celldeath in prostate cancer cell lines regardless of theirandrogen sensitivity. Cell death induced by ZA re-sulted from an unknown switching between apopto-tic and autophagic cell death since cell viabilitycould only be rescued by the simultaneous additionof apoptosis and autophagy inhibitors. Future studyto identify the links between autophagy and apopto-sis induced by ZA is important and may shed lighton the development of effective combination therapyto regulate autophagy along with ZA treatment forprostate cancer.

ACKNOWLEDGMENTS

ZA was provided by Novartis, Basel, Switzerland.

ZOLEDRONIC ACID INDUCES AUTOPHAGY IN HUMAN PROSTATE CANCER CELLS1496

REFERENCES

1. Wolf AM, Wender RC, Etzioni RB et al: AmericanCancer Society guideline for the early detectionof prostate cancer: update 2010. CA CancerJ Clin 2010; 60: 70.

2. Loblaw DA, Virgo KS, Nam R et al: Initial hor-monal management of androgen-sensitive meta-static, recurrent, or progressive prostate cancer:2006 update of an American Society of ClinicalOncology practice guideline. J Clin Oncol 2007;25: 1596.

3. Petrylak DP, Tangen CM, Hussain MH et al:Docetaxel and estramustine compared with mi-toxantrone and prednisone for advanced refrac-tory prostate cancer. N Engl J Med 2004; 351:1513.

4. Coleman RE: Metastatic bone disease: clinicalfeatures, pathophysiology and treatment strate-gies. Cancer Treat Rev 2001; 27: 165.

5. Sharifi N, Gulley JL and Dahut WL: Androgendeprivation therapy for prostate cancer. JAMA2005; 294: 238.

6. Lipton A: Should bisphosphonates be utilized inthe adjuvant setting for breast cancer? BreastCancer Res Treat 2010.

7. Terpos E, Sezer O, Croucher PI et al: The use ofbisphosphonates in multiple myeloma: recom-mendations of an expert panel on behalf of theEuropean Myeloma Network. Ann Oncol 2009;20: 1303.

8. Saad F and Schulman CC: Role of bisphospho-nates in prostate cancer. Eur Urol 2004; 45: 26.

9. Saad F, Gleason DM, Murray R et al: Long-termefficacy of zoledronic acid for the prevention ofskeletal complications in patients with meta-static hormone-refractory prostate cancer. J NatlCancer Inst 2004; 96: 879.

10. Dumon JC, Journe F, Kheddoumi N et al: Cyto-static and apoptotic effects of bisphosphonates

on prostate cancer cells. Eur Urol 2004; 45: 521.

11. Coxon JP, Oades GM, Kirby RS et al: Zoledronicacid induces apoptosis and inhibits adhesion tomineralized matrix in prostate cancer cells viainhibition of protein prenylation. BJU Int 2004;94: 164.

12. Montague R, Hart CA, George NJ et al: Differen-tial inhibition of invasion and proliferation bybisphosphonates: anti-metastatic potential ofZoledronic acid in prostate cancer. Eur Urol 2004;46: 389.

13. Lum JJ, Bauer DE, Kong M et al: Growth factorregulation of autophagy and cell survival in theabsence of apoptosis. Cell 2005; 120: 237.

14. Eisenberg-Lerner A, Bialik S, Simon HU et al: Lifeand death partners: apoptosis, autophagy and thecross-talk between them. Cell Death Differ 2009;16: 966.

15. Hippert MM, O’Toole PS and Thorburn A: Au-tophagy in cancer: good, bad, or both? CancerRes 2006; 66: 9349.

16. Kanzawa T, Kondo Y, Ito H et al: Induction ofautophagic cell death in malignant glioma cellsby arsenic trioxide. Cancer Res 2003; 63: 2103.

17. Herman-Antosiewicz A, Johnson DE and SinghSV: Sulforaphane causes autophagy to inhibitrelease of cytochrome C and apoptosis in humanprostate cancer cells. Cancer Res 2006; 66: 5828.

18. Yi-Chia L, Ji-Fan L, Te-Fu T et al: 1197 Zoledronicacid enhances growth inhibition and apoptosis inboth hormone-refractory and hormone-sensitiveprostate cancer cell lines. J Urol 2010; 183: e463.

19. Kondo Y, Kanzawa T, Sawaya R et al: The role ofautophagy in cancer development and responseto therapy. Nat Rev Cancer 2005; 5: 726.

20. Yo YT, Shieh GS, Hsu KF et al: Licorice andlicochalcone-A induce autophagy in LNCaP pros-tate cancer cells by suppression of Bcl-2 expres-sion and the mTOR pathway. J Agric Food Chem

2009; 57: 8266.

21. Shintani T and Klionsky DJ: Autophagy in healthand disease: a double-edged sword. Science2004; 306: 990.

22. Hoyer-Hansen M and Jaattela M: Autophagy: anemerging target for cancer therapy. Autophagy2008; 4: 574.

23. Rosenfeldt MT and Ryan KM: The role of au-tophagy in tumour development and cancer ther-apy. Expert Rev Mol Med 2009; 11: e36.

24. Hu H, Chai Y, Wang L et al: Pentagalloylglucoseinduces autophagy and caspase-independent pro-grammed deaths in human PC-3 and mouseTRAMP-C2 prostate cancer cells. Mol CancerTher 2009; 8: 2833.

25. Maiuri MC, Galluzzi L, Morselli E et al: Autophagyregulation by p53. Curr Opin Cell Biol 2010; 22:181.

26. Korkhov VM: GFP-LC3 labels organised smoothendoplasmic reticulum membranes independentlyof autophagy. J Cell Biochem 2009; 107: 86.

27. Kuma A, Matsui M and Mizushima N: LC3, anautophagosome marker, can be incorporated intoprotein aggregates independent of autophagy:caution in the interpretation of LC3 localization.Autophagy 2007; 3: 323.

28. Ciechomska IA and Tolkovsky AM: Non-au-tophagic GFP-LC3 puncta induced by saponin andother detergents. Autophagy 2007; 3: 586.

29. Wu YT, Tan HL, Shui G et al: Dual role of3-methyladenine in modulation of autophagy viadifferent temporal patterns of inhibition on classI and III phosphoinositide 3-kinase. J Biol Chem2010; 285: 10850.

30. Caraglia M, Marra M, Leonetti C et al: R115777(Zarnestra)/Zoledronic acid (Zometa) cooperationon inhibition of prostate cancer proliferation isparalleled by Erk/Akt inactivation and decreasedBcl-2 and bad phosphorylation. J Cell Physiol

2007; 211: 533.