MOLECULAR CANCER THERAPEUTICS | SMALL MOLECULE THERAPEUTICS

ANovel Salicylanilide Derivative Induces Autophagy CellDeath in Castration-Resistant Prostate Cancer via ERStress-Activated PERK Signaling PathwayChia-Ling Hsieh1,2, Hsu-Shan Huang3, Kuan-Chou Chen4,5,6, Teigi Saka4, Chih-Ying Chiang1,Leland W.K. Chung7, and Shian-Ying Sung1,2,8,9

ABSTRACT◥

Metastatic castration-resistant prostate cancer (CRPC) is cur-rently incurable. Cancer growth and progression is intimatelyaffected by its interaction with hostmicroenvironment. Cotargetingof the stroma and prostate cancer is therefore an emerging ther-apeutic strategy for metastatic CRPC. Cancer-induced osteoclas-togenesis is known to contribute to CRPC bone metastasis. Thisstudy is to extend pharmacologic value of our synthesized LCC03, aderivative of 5-(20,40-difluorophenyl)-salicylanilide that has pre-viously testified for its osteoclastogenesis activity, by exploring itsadditional cytotoxic properties and underlying mechanism inCRPC cells. LCC03 was chemically synthesized and examined forcell growth inhibition in a serial of CRPC cell lines. We dem-onstrated that LCC03 dose-dependently suppressed proliferationand retarded cell-cycle progression in CRPC cells. The classicalautophagy features, including autophagosome formation andLC3-II conversion, were dramatically shown in LCC03-treated

CRPC cells, and it was associated with the suppressed AKT/mTOR signaling pathways, a major negative regulator of autop-hagy. Moreover, an expanded morphology of the endoplasmicreticulum (ER), increased expression of the ER stress markersGRP78 and PERK, and eIF2a phosphorylation were observed.Blockage of autophagy and PERK pathways using small moleculeinhibitors or shRNA knockdown reversed LCC03-inducedautophagy and cell death, thus indicating that the PERK–eIF2apathway contributed to the LCC03-induced autophagy. Further-more, treatment of tumor-bearing mice with intraperitonealadministered LCC03 suppressed the growth of CRPC xenograftsin mouse bone without systemic toxicity. The dual action of5-(20,40-difluorophenyl)-salicylanilide on targeting both theosteoclasts and the tumor cells strongly indicates that LCC03is a promising anticancer candidate for preventing and treatingmetastatic CRPC.

IntroductionDespite high response rates to androgen deprivation therapy in

men with advanced prostate cancer, nearly all types of prostatecancer are eventually progression to the androgen-independentstage, which is termed castration-resistant prostate cancer (CRPC).CRPC is an incurable stage of prostate cancer, in which approx-imately 90% of patients develop metastases, mainly in the skele-

ton (1). Bone metastasis and skeletal complications are the majorcontributing factors to morbidity and mortality in patients withprostate cancer. Over time, the conventional treatment is ineffec-tive, and the patients die of CRPC; the survival time is less than19 months. Although new therapies (and drugs), including tubulintargeting chemotherapy (cabazitaxel), immunotherapy (sipuleucel-T),the steroidogenesis inhibitor (abiraterone), AR antagonist (enzalu-tamide), and a-emitting radiotherapy (radium-223), have shownpromising results in impairing the tumor growth and extendingsurvival, a considerable proportion of patients with CRPC becomeunresponsive or become resistant to these treatment after a shortperiod (2). The development of novel therapeutics that targetdistinct mechanisms of action is necessary to overcome resistancein patients with CRPC.

The endoplasmic reticulum (ER) plays pivotal roles in cell homeo-stasis and survival, which involved in biosynthesis of lipids, regulationof intracellular calcium concentration and metabolism of carbohy-drates, and synthesis and folding of proteins. The accumulation ofmisfolded proteins within the ER lumen induces ER stress that triggersthe unfolded protein response (UPR), an evolutionarily adaptivemechanism for restoring ER homeostasis; thus, UPR protects cellsagainst the toxic accumulation of misfolded proteins (3). Underenvironmental stress, such as nutrient deprivation or hypoxia, malig-nant cells are particularly prone to protein misfolding and UPRactivation, which lead to tumor progression and survival (4). However,in contrast to cytoprotection, recent studies have highlighted the roleof severe or unresolved ER stress in cell death (5); unsolved ER stresshas been previously associated with various human diseases, includingatherosclerosis (6), neurodegenerative disorders (7), and type 2 dia-betes (8). Therefore, the pharmacologic modulation of cellularresponses toward ER stress-induced cell death pathway may prevent

1The Ph.D. Program for Translational Medicine, College of Medical Science andTechnology, Taipei Medical University, Taipei, Taiwan. 2TMU Research Center ofCancer Translational Medicine, Taipei Medical University, Taipei, Taiwan. 3Grad-uate Institute of Cancer Molecular Biology and Drug Discovery, College ofMedical Science and Technology, Taipei Medical University, Taipei, Taiwan.4Graduate Institute of Clinical Medicine, College of Medicine, Taipei MedicalUniversity, Taipei, Taiwan. 5Department of Urology, College of Medicine, TaipeiMedical University, Taipei, Taiwan. 6Department of Urology, Taipei MedicalUniversity-Shuang Ho Hospital, New Taipei City, Taiwan. 7Department of Med-icine, Cedars-Sinai Medical Center, Los Angeles, California. 8Joint ClinicalResearch Center, Office of Human Research, Taipei Medical University, Taipei,Taiwan. 9Clinical Research Center, Taipei Medical University Hospital, TaipeiMedical University, Taipei, Taiwan.

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

Corresponding Author: Shian-Ying Sung, The Ph.D. Program for TranslationalMedicine, College of Medical Science and Technology, Taipei Medical University,250 Wu-Hsing Street, Taipei 11031, Taiwan. Phone: 88-62269-72035, ext. 105;E-mail: ssung@tmu.edu.tw

Mol Cancer Ther 2019;XX:XX–XX

doi: 10.1158/1535-7163.MCT-19-0387

�2019 American Association for Cancer Research.

AACRJournals.org | OF1

on April 21, 2020. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst September 17, 2019; DOI: 10.1158/1535-7163.MCT-19-0387

tumor development and growth, thus representing an attractivetherapy for different cancers (9), including prostate cancer.

Cancer growth and progression is intimately affected by its inter-action with the adjacent host microenvironment, that comprised ofvarious stromal cell types, growth factors, and extracellular matrices,and thus presents an attractive target for therapeutic intervention (10).Salicylanilides are a key class of aromatic pharmacophore, comprisingamides of salicylic acid and aniline. Salicylanilides derivatives havebeen reported to possess myriad pharmacological activities, such asantimicrobial [against bacterial (11), mycobacteria (12), viruses (13),and fungi (14)], anti-inflammatory (15), and antitumor proper-ties (16). Recently, a novel function of 5-(20 ,40-difluorophenyl)-sali-cylanilide derivatives as potent inhibitors of osteoclastogenesis has alsobeen discovered (17–19). Because pathological osteoclast activation isassociated with an increased risk of tumor progression (20), subse-quent skeletal complications and death due tometastatic CRPC, in thisstudy, the additional pharmacological value of 5-(20,40-difluorophe-nyl)-salicylanilide was examined by determining its effect and eluci-datingmechanism of action underlying its direct cytotoxicity to CRPCcells.

Materials and MethodsCell lines and cell culture

Human prostate cancer cell lines: PC3, DU145, C4-2, andCWR22Rv1 were originally obtained from ATCC. Human benignprostatic hyperplasia (BPH) epithelial cell line BPH-1was a kind gift ofDr. Su-Hwa Lin (The University of Texas MD Anderson CancerCenter). No further authentication was performed by authors. Thecells were maintained in RPMI1640 medium (Thermo Fisher Scien-tific) containing 10% FBS (GeneDireX), 100 U/mL of penicillin, and100mg/mL of streptomycin (Thermo Fisher Scientific) at 37�Cunder 5% CO2 for less than 25 passages after thawing and routinely testingnegative for Mycoplasma contamination using a PCR MycoplasmaDetection Kit (Abcam) to conduct described experiments.

ChemicalsN-(3,4-difluorophenyl)-20,40-difluoro-4-hydroxy[1,10-biphenyl]-

3-carboxamide (C19H11F4NO2; LCC03) was chemically synthesizedusing previously published procedures (21). The chemical structureof LCC03 is shown in Fig. 1A. 3-Methyladenine (3-MA, PubChemCID:135398661), an autophagy inhibitor was purchased fromSigma-Aldrich. Selective PERK inhibitor GSK2606414 (PubChemCID:53469448) was purchased from EMD Millipore.

Cell proliferation and viability assayThe cytotoxic effects of LCC03 was assayed using a WST-1 Cell

Proliferation Assay Kit (Roche Applied Science) following the man-ufacturer's protocol. The IC50 was calculated as the concentration ofsample needed to reduce 50% of the absorbance compared with thedimethyl sulfoxide (DMSO)-treated control. The viability of the PC3and C4-2 cells treated with LCC03 in the absence or presence of 3-MA(2 mmol/L) for 72 hours was assessed through trypan-blue exclusion(Thermo Fisher Scientific). All assays were done in 6 replicates on atleast 4 independent experiments.

Western blot analysisProtein expression was analyzed as performed previously (22). The

primary antibodies were listed in Supplementary Table S1. Afterincubation with an HRP-conjugated secondary antibody (1:5,000; GEHealthcare Life Sciences), the corresponding bands were detected

using an ECL Kit (Advansta) and an AI 600 chemiluminescentimaging and analysis system (GE Healthcare Life Sciences). Each blotwas performed at least twice.

Gene knockdownSmall hairpin RNA (shRNA) expression plasmids, including

pLKO.1-shPERK (TRCN0000262373, target sequence TGCATCT-GCCTGGTTACTTAA), pLKO.1-shAtg7 (TRCN0000007584, targetsequence GCCTGCTGAGGAGCTCTCCAT; TRCN0000007587,target sequence CCCAGCTATTGGAACACTGTA), and a mam-malian nontargeting shRNA control pLKO.1-shGFP (sh-ctr,TRCN0000072178, target sequence, CAACAGCCACAACGTCTA-TAT)were obtained from theNational RNAiCore Facility (Institute ofMolecular Biology). The recombinant lentivirus was generated bytransfecting shRNA plasmids along with the packaging plasmidspCMV-cR8.91(Gag/Pol/Rev) and pMD.G (VSV-G envelope) in293FT cells and used for infecting the PC3 and C4-2 cells accordingto a previously described protocol (23). The extent of gene knockdownwas determined through an immunoblotting assay.

Transmission electron microscopySample preparation for TEM were prepared with aid from the

Imaging Core at Taipei Medical University. Briefly, the cells grown onplastic chamber slides were fixed using 2% paraformaldehyde (PFA)and 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer (pH 7.4)for 30minutes at room temperature and then subjected to postfixationwith 2% osmium tetroxide. After dehydration in ascending grades ofethanol and propylene oxide, the samples were embedded in Epon.Epon polymerization was conducted at 62 �C for 48 hours. Ultrathinsections (approximately 70 nm in thickness) were obtained using anultramicrotome (LeicaUltracutUCT; LeicaMicrosystemsGmbH) andwere collected on 100 mesh copper grids. After double staining withuranyl acetate and lead citrate, the sections were examined under aHitachH-600 Transmission ElectronMicroscope operating at 100 kV.

Immunofluorescence stainingThe PC3 and C4-2 cells transfected with the EGFP-LC3 (24)

plasmid (a gift from Karla Kirkegaard; Addgene plasmid #11546)were grown on coverslips for overnight and treated with DMSO orLCC03 combined with GSK2606414 for 72 hours. After removing theculture media, the cells were fixed using 4% PFA and washed thriceusing PBS. Subsequently, the cells were stained with DAPI and thenmounted in ProLong Gold Antifade reagent (Thermo Fisher Scien-tific). Images of the cells were acquired using a Carl Zeiss LSM 510META confocal microscope (Carl Zeiss) equipped with a Plan Apoc-hromat 63� oil/1.4 NA DIC objective.

Animal studiesAnimal experimentswere approved by the Institutional Animal Care

and Use Committee of Taipei Medical University (LAC-2016-0345)and complied with their regulations. Six-week-old male nude mice(BALB/cAnN.Cg-Foxn1nu/CrlNarl) were purchased from theNationalLaboratory Animal Center (Taipei, Taiwan) and maintained under thespecific pathogen-free conditions and cared for according to the criteriaoutlined in the National Academy of Sciences Guide for the Care andUse of Laboratory Animals. To initiate experimental prostate bonemetastasis in the xenograft models, the mice were anesthetized byZoletil andRompunmixture, and subjected to an intraosseous injectionof 2� 105 PC3-Luc cells into the mouse tibia following our previouslyestablished protocol (23). One week after cell injection, the tumor-bearing mice were randomized and given the following intraperitoneal

Hsieh et al.

Mol Cancer Ther; 2019 MOLECULAR CANCER THERAPEUTICSOF2

on April 21, 2020. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst September 17, 2019; DOI: 10.1158/1535-7163.MCT-19-0387

treatments twice per week for 4 weeks (n ¼ 10 for each group): (i)untreated, (ii) DMSO, (iii) LCC03 (20 mg/kg), and (iv) LCC03(40 mg/kg). The tumor growing in the bone were monitored throughbioluminescent imaging (BLI) weekly by using the IVIS 2000 systemand the Living Image software (Caliper Life Sciences) for quantificationof luminescence intensity. The body weights of all the mice weremeasured thrice per week for 4 weeks. The mice were sacrificed at5 weeks after the initial treatment and the heart, lung, kidney, liver,spleen, and tibia were excised for histopathologic analyses. Before theanimals were sacrificed, blood samples were collected to analyzebiochemical parameters with aid from the TMU Animal Care Facility.

IHC stainingBone specimens were fixed in 10% neutral buffered formalin for

24 hours. They were then decalcified using 14% EDTA (pH 7.2) for14 days at room temperature and embedded in paraffin. IHCanalyses were performed using the Novolink Polymer DetectionSystem (Leica Microsystems) with the following antibodies: mousemonoclonal anti-Ki-67 antibody (1:100; Clone GM010; GenemedBiotechnologies) and rabbit polyclonal anti-human LC3B (1:50;LS-B9807; LifeSpan BioSciences). The extent of apoptosis andosteoclastogenesis was evaluated using an Apo-BrdU-IHC In SituDNA Fragmentation Assay Kit (BioVision, Inc.) and a TRACP &

Figure 1.

Effect of LCC03 on cell growth and cell-cycle progression in prostate cancer cell lines.A, Rational design synthesis and chemical structures of small molecule LCC03.B, Proliferation of benign (BPH) andmalignant prostate cancer cell lines treated with a serial of doses of LCC03 for 48 hours was determined using theWST-1 assay.Relative cell proliferation is presented as themean� SD percentage of absorbance (at 450 nm) comparedwith the DMSO-vesicle–treated control (0 mmol/L LCC03)for at least 4 independent experiments (n � 4). C, Bar graphs of the percentages of G1, S, and G2–M populations in the PC3 and C4-2 cells treated with the DMSOvesicle control and the indicated concentrations of LCC03 for 48 hours. The results of 1 of 3 independent experiments are shown.D,Western blotting analysis of cell-cycle checkpoint markers. b-Actin is shown as the protein loading control. The expected molecular weight of each protein is indicated on the left side of the figure.

Salicylanilide as an Autophagy Inducer for CRPC Therapy

AACRJournals.org Mol Cancer Ther; 2019 OF3

on April 21, 2020. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst September 17, 2019; DOI: 10.1158/1535-7163.MCT-19-0387

ALP Double-Stain Kit (TaKaRa), respectively following the man-ufacturer's instructions.

Statistical analysisStatistical analysis was performed using GraphPad Prism 6 (Graph-

Pad Software). Data are presented as mean� SD and % change versuscontrols. Difference between 2 groups was measured by Student t test.A probability value of P < 0.05 was considered statistically significant.

ResultsLCC03 induces growth retardation and cell-cycle arrest inprostate cancer cells

Treating patients with CRPC remains a substantial clinical chal-lenge. Thus, we assessed whether LCC03, a novel derivative of 5-(20,40-difluorophenyl)-salicylanilide (Fig. 1A), is potentially toxic to a seriesof CRPC-like cell lines, including androgen receptor (AR)-positive celllines C4-2 and CWR22rv1 and the AR-negative cell lines PC3 andDU145. A cell proliferation assay revealed that the incubation ofprostate cancer cells with LCC03 for 48 hours caused a marked anddose-dependent growth inhibition (Fig. 1B). The IC50 in different celllines was 0.69 to 4.8 mmol/L (Table 1). By contrast, the cell prolifer-ation of benign hyperplastic prostatic epithelial cells (BPH-1) wasunaffected by LCC03 at a concentration of <5 mmol/L. The IC50 valueof LCC03 for BPH-1 (15.7 mmol/L) cells was 3 to 20 times higher thanthat for the cancerous cells, thus indicating that LCC03 is selectivelymore toxic to cancerous cells than to noncancerous cells. The highestIC50 value of LCC03 was observed in the PC3 cell line, which wasoriginally derived from patient bone metastases. The lowest IC50 valueof LCC03was observed in the C4-2 cells; thus, the C4-2 cells weremostsensitive to LCC03. These 2 cell lines were selected for subsequentinvestigation of the anticancer properties of LCC03.

We analyzed the cell-cycle distribution in the PC3 and C4-2 cellsafter 48 hours of incubation with LCC03 to characterize the anti-proliferative properties of LCC03 in detail. We observed dose-dependent accumulation of the PC3 cells in the G2–M phase onLCC03 exposure (Supplementary Fig. S1A and S1C, PC3). Theincrease in the (G2–M phase) was coupled with a reduction inpercentage of cells in the G0–G1 phase. In the C4-2 cells, LCC03treatment increased the proportion of cells in theG1 phase and reducedthe proportion of cells in the G2 phase (Supplementary Fig. S1A andS1C, C4-2). The effect of LCC03 on cell-cycle progression was alsosupported by the result of mRNA array-based gene profiling analysis,which showed a marked change in the cell-cycle pathway in the PC3cells after treatment with LCC03 (Supplementary Fig. S1B). Western

blotting analysis further confirmed the time- and dose-dependentrepression of cyclin D1 as well as the cdc2/cyclin B complex activity byLCC03 in both cell lines (Fig. 1D), thus indicating a strong effect ofLCC03 on the cell-cycle checkpoints.

LCC03 causes cytotoxic activity through autophagy but not viaan apoptosis-dependent pathway

The cytotoxic effect of LCC03 on prostate cancer cell survival wasassessed through the trypan blue assay after incubating the cells for3 days with LCC03. LCC03 treatment reduced cell viability in a dose-dependent manner (Fig. 2A), which was similar to the effect ofLCC03 on cell proliferation. We then assessed whether the cytotoxiceffect of LCC03 on prostate cancer cells was correlated with increasedapoptosis. Notably, the percentage of annexin V-positive cellsincreased only slightly among the PC3 cells and decreased amongthe C4-2 cells, when the cells were treated with LCC03 (up to theirrespective IC50 concentrations) for 72 hours (SupplementaryFig. S2A). Moreover, Western blotting analysis revealed that onlyminimal cleavage of PARP and caspase 3 was detectable in LCC03-treated cells (Supplementary Fig. S2B). Collectively, these data indi-cate that LCC03-induced cytotoxicity was not closely associated withthe apoptotic cell death.

Recent studies have identified autophagy as themajormechanism ofcell death in addition to apoptosis in response to cellular stress. Hence,we examined whether autophagy is induced by LCC03 and whether ithas a critical role in mediating cell death. Cyto-ID green fluorescentdye was used to stain autophagic vacuoles in the cells and thefluorescence intensity was quantified through flow cytometry. Similarto the treatment with the autophagy inducers tunicamycin andrapamycin, an increase in fluorescent intensity was observed in thePC3 and C4-2 cells after treatment of LCC03 for 72 hours, thussuggesting the induction of autophagy (Fig. 2B). To confirm thisresult, TEM was performed for ultrastructural analysis. Numerousautophagic vacuoles and empty vacuoles were observed in the LCC03-treated PC3 cells, andmost of the autophagic vacuoles contained intactlamellar structure and/or residual digested materials (Fig. 2C). TheDMSO-vesicle treated control exhibited only a few autophagic fea-tures. TheWestern blot analysis results (Fig. 2D) revealed a time- anddose-dependent increase in the conversion of LC3B-I to LC3B-II,which is a hallmark of macroautophagy. LCC03 also induced ahigher amount of the autophagy factor Atg12-Atg5 conjugatebetween 48 and 72 hours after treatment, and its induction wasassociated with increased LC3B conversion in both cell lines.Notably, Beclin-1 protein, a key regulator complexed with class IIIPI3K for autophagosome formation through dissociation from theanti-apoptotic protein Bcl2, was upregulated concomitantly with adecreased in Bcl2 in the C4-2 cells after treatment with LCC03.However, no significant changes were observed in the PC3 cells,thus suggesting that LCC03 elicited a canonical or noncanonicalautophagic pathway in a cell-type–dependent manner. Because Bcl-2 is a prosurvival protein, the increased expression of Bcl-2 seen inPC3 cells after the longer exposure (72 hours) of LCC03 mightcontribute to helping cell surviving under treatment, by which PC3cells acquire more resistance than C4-2 cells to the cytotoxic effectsof LCC03. In addition, blockage of autophagosome formation byusing 3-MA, a PI3K inhibitor, or genetically knocking downautophagy-related 7 (ATG7) expression significantly attenuated theinduction of autophagosome marker expression and restored theviability of cells treated with LCC03 (Fig. 2E and F), thus con-firming that LCC03 induced cytotoxicity in prostate cancer cellsthrough activation of autophagy.

Table 1. Summary of IC50 values for LCC03 in 1 human benignprostate hyperplasia (BPH) and 4 prostate cancer cell lines after48 hours' treatment.

Cell lineMean IC50 � SD

(mmol/L)a

BPH-1 15.7322 � 8.944C4-2 0.688 � 0.061CWR22r-v1 2.57 � 0.55PC3 4.48 � 2.58DU145 2.64 � 0.83

aAll values were presented as mean � SD of 6 replicates in 4 independentexperiments.

Hsieh et al.

Mol Cancer Ther; 2019 MOLECULAR CANCER THERAPEUTICSOF4

on April 21, 2020. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst September 17, 2019; DOI: 10.1158/1535-7163.MCT-19-0387

LCC03 activates ER-stress–induced PERK–eIF2a–ATF4signaling pathway

To gain a better insight into the LCC03-induced autophagic path-ways, we identified autophagy-related signaling molecules. The kinasemTOR is a critical regulator of autophagy induction. The phosphor-ylation of mTOR suppresses autophagy, whereas the dephosphoryla-tion of mTOR promotes autophagy. Our results identified significantdownregulation of p-mTOR level in C4-2 and PC3 cells treated withLCC03 (Fig. 3A). This downregulation of p-mTOR was stronglyassociated with Akt inactivation by dephosphorylation alone in thePC3 cells or together with reduced expression of total Akt protein inthe C4-2 cells. The expression of AMPK, a negative regulator ofmTOR, although did not differ significantly between the LCC03- andDMSO-treated cells, the phosphorylation of AMPK-a was massivelyinduced upon LCC03 treatment. Collectively, these findings indicatethat LCC03 triggered autophagy activation in the prostate cancer cellsvia an AMPK- and Akt-dependent mTOR pathway.

Several reports have reported that ER-stress-activated UPR cantrigger several signaling pathways that cause autophagy (25, 26). In

addition to the autophagosome-like vesicles, TEM revealed that theLCC03-treated cells exhibited abnormal ER expansion (Fig. 3B).Therefore, we determined whether LCC03 could induce an ER stressresponse and subsequently regulate autophagy. The Western blotanalysis results revealed upregulation of GRP78 expression, whichindicated the activation of ER stress (Fig. 3C). Moreover, LCC03treatment considerably increased the expression of the ER stress sensorPERK, the levels of phosphorylated eIF2a and ATF4, and the ATF4downstream target genes CHOP (a transcription factor known toinduce autophagy) and TRIB3 (a negative regulator of Akt activation)in a time- and dose-dependent manner. These results demonstratedthat the PERK–eIF2a–signaling pathwaywas activated during LCC03-induced ER stress.

Blockage of the PERK activation attenuates the LCC03-inducedautophagic cell death

We investigated whether LCC03-induced autophagic cell death inthe tested cells was mechanistically related to ER stress signalingpathways by using pharmacological and genetic approaches to inhibit

Figure 2.

Effect of LCC03onautophagy activation inCRPCcells.AandF,Trypanblue staining for cell viability assayof the (A) parental and (F)ATG-(sh-ATG) andnontargetingshRNA-expressing (sh-ctr) PC3 andC4-2 cells treatedwith the indicated concentrations of LCC03, andwith orwithout autophagyblocker 3-MA for 72 hours. Data arepresented as mean � SD of 5 independent experiments (n ¼ 5). � , P < 0.05; �� , P <0.005 versus vesicle control. B, Representative flow cytometry histogram of 3independent assays (n¼ 3) for CytoID fluorescence in the PC3 cells. Cells treated with 10 mg/mL tunicamycin or rapamycin as a positive control. Cell acquisition wasperformedonAttuneNxt flow cytometer (Thermo Fisher Scientific) and analyzed using FlowJo 10.2 software.C, TEM imageswith differentmagnifications of the PC3cells treatedwithDMSOvesicle control and LCC03. Images are representative of 3 independent treatments.D andE,Western blot analyses of the proteins involved inautophagy from the total cell extracts of the (D) parental and (E) shRNA-expressing PC3 and C4-2 cells treated with the indicated concentrations of LCC3. b-Actin isshown as the protein loading control. The expected molecular weight of each protein is indicated on the left side of the figure. The cells were treated with LCC03 for72 hours in all assays unless specified otherwise.

Salicylanilide as an Autophagy Inducer for CRPC Therapy

AACRJournals.org Mol Cancer Ther; 2019 OF5

on April 21, 2020. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst September 17, 2019; DOI: 10.1158/1535-7163.MCT-19-0387

ER stress sensor PERK activity. Stable transfection of EIF2AK3-targeting shRNA (sh-PERK) significantly suppressed PERK expres-sion, blocked LCC03-induced upregulation of LC3B conversion, andinhibited eIF2a phosphorylation compared with cells transfected withnegative control shRNA (Fig. 4A). Similarly, inhibition of PERKkinase activity by using the compound GSK2606414 slightly increasedthe basal levels of PERK; however, it prevented the phosphorylation ofeIF2 a and the conversion of LC3B-I to LC3B-II when the cells were

exposed to LCC03 (Fig. 4B). We further analyzed autophagosomeformation in the PC3 and C4-2 cells that were transfected with aplasmid expressing autophagosome-associated LC3 protein fused togreen fluorescent protein (GFP-LC3). In the GFP-LC3 transfectedcells, treatment with LCC03 resulted in an increase in the redistribu-tion of green fluorescence from diffused to a punctum pattern in theperinuclear region. Treatment with GSK2606414 also suppressed theLCC03-induced autophagosome formation, which was indicated by

Figure 3.

Time and dose effect of LCC03 on ER stress activation in CRPC cells.A and C,Western blotting analysis of signalingmolecules involved in (A) autophagy and (C) ERstress activation in the PC3 and C4-2 cells. b-Actin is shown as the protein loading control. The expectedmolecular weight of each protein is indicated on the left sideof thefigure.B,Ultrastructural analysis of ER stress activation in the PC3 cells treatedwithDMSOor 10mmol/L LCC03 for 48 hours through TEM.Asterisks indicate theenlarged ER. Images are representative of 3 independent treatments.

Hsieh et al.

Mol Cancer Ther; 2019 MOLECULAR CANCER THERAPEUTICSOF6

on April 21, 2020. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst September 17, 2019; DOI: 10.1158/1535-7163.MCT-19-0387

the reduced appearance of GFP-LC3 puncta (Fig. 4C). The PERKsignaling pathway attenuated by either shRNA or small moleculeinhibitors restored cell survival in response to LCC03 treatment, thusconfirming the critical role of PERK-mediated ER-stress-signalingpathway in LCC03-induced cell death (Fig. 4D). Interestingly,knocked down of EIF2AK3 did not reverse the LCC03-mediatedmTOR dephosphorylation (Fig. 4A). Conversely, activation of mTORby MHY1485 decreased PERK expression in LCC03-treated PC3 cells(Supplementary Fig. S3), implying that PERK-mediated autophagy byLCC03, at least in part, is regulated via mTOR pathway.

LCC03 treatment suppresses the growth of osteolytic prostatecancer metastasis in mouse bone

To investigate the therapeutic potential of LCC03 for prostatecancer, particularly targeting CRPC bone metastases and the associ-ated osteoclast activation, PC3 cells that are relatively highly resistantto LCC03 and produce a pure osteolytic reaction in the bone were usedin an experimental animal model. Luciferase expressing-PC3 cells(PC3-Luc) were injected directly into the tibia of the nude mice and

allowed to form tumors for 10 days, as demonstrated through BLI. Thetumor-bearingmice received intraperitoneal administration of LCC03at a low (20 mg/kg) and high (40 mg/kg) dose or DMSO vesicle everyother day (QOD) for 4 weeks. A control group received no treatment.Simultaneously, the body weights of all the mice were also recorded.The responsiveness of the PC3-Luc tumors to this therapy wasmonitored was monitored through weekly BLI. During the 5-weekmonitoring period, the bioluminescence signal gradually increasedwith time in the untreated control group, which indicated that thetumors progressed aggressively (Fig. 5A, imaging). A similar imagingpattern was observed in the mice treated with DMSO vesicles and thelow-dose LCC03, which indicated that 20 mg/kg QOD of LCC03 wasnot effective at suppressing tumor growth. By contrast, the micetreated with high-dose (40 mg/kg) LCC03 exhibited a relativelyconstant or slowly changing profile of bioluminescence signals overtime. A quantitative analysis of the BLI data revealed that the averagesignal intensity in the mice treated with high-dose LCC03 wasapproximately 75% less than that in the untreated and DMSO-treated control mice (Fig. 5A); thus confirmed a significant anticancer

Figure 4.

Effect of PERK antagonists on LCC03-induced cell death in CRPC cells. A and B, Western blotting analysis of PERK signaling molecules involved in autophagyactivation and in the PC3 and C4-2 cells that (A) stably expressed PERK-targeting shRNA or (B) were treatedwith PERK inhibitor GSK2606414 in the presence of theindicated concentration of LCC03 for 48 hours. Nontargeting shRNA (sh-ctr) and DMSO treatment (0 mmol/L LCC03) are the corresponding negative controls. C,Representative confocal fluorescent images of the PC3 and C4-2 cells transfected with LC3-GFP followed by the treatment of PERK inhibitor GSK2606414 or DMSO-vesicle control combinedwith the indicated concentrations of LCC03 for 48 hours. The right panel is the enlarged images of the boxed regions at the left side.D, Cellviability as assessed by trypan blue-exclusion assay in PC3 and C4-2 cells that express PERK-targeting shRNA or treated with PERK inhibitor GSK2606414 in thepresence of the indicated concentration of LCC03 for 72 hours. Data are presented as mean � SD of 5 independent experiments (n ¼ 5). � , P < 0.05 versus thecorresponding nontargeting shRNA (sh-ctr) or DMSO-vesicle control.

Salicylanilide as an Autophagy Inducer for CRPC Therapy

AACRJournals.org Mol Cancer Ther; 2019 OF7

on April 21, 2020. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst September 17, 2019; DOI: 10.1158/1535-7163.MCT-19-0387

effect on tumor growth. These results confirmed dose-dependentinhibition of tumor growth by LCC03 and demonstrated that 40mg/kgQOD for 4 weeks are the efficacious dose to induce regression.

In addition, the high-dose treatment was not associated with anyobserved adverse effect in the body weight and cellular structure ofnontarget organ tissues (Fig. 5B). The biochemical parameters,including BUN (blood urea nitrogen), CREA (creatinine), ALT (ala-nine aminotransferase), ALP (alkaline phosphatase), and TBIL (totalbilirubin) were all within the reference range and no significantchanges (P > 0.1) in mice before and after treatment with LCC03(40mg/kg) for 5weeks (Supplementary Table S2), which indicated thatthe administered dose of LCC03 was safe for therapeutic use.

Histologic analysis (Fig. 5C, H&E) of the affected legs from thesacrificed mice after 5 weeks of treatment revealed healthy and packedtumor cells growing in the marrow cavity of the control groups (notreatment or DMSO-vesicle treatment). The highly proliferativenature of the tumor cells in these control groups was confirmedthrough ki-67 expression (Fig. 5C, ki-67). By contrast, extensivenecrotic regions were found in the tumors excised from the micetreated with high-dose LCC03, and only a few ki-67-stained cancer

cells were detected in these tumors. IHC staining of LC3B indicated astrong increase in LC3B expression in the tumors from themice treatedwith high-dose LCC03. TUNEL staining results of the tumors indi-cated that apoptotic cells were absent (Fig. 5C, LC3B and TUNEL).Collectively, these results confirmed that the considerable tumorregression caused by systemic LCC03 treatment was associated withautophagy but not apoptosis. Moreover, the tartrateresistant acidphosphatase (TRAP)-positive osteoclasts were distributed mainly inosteolytic bone lesions of the control groups but barely detected in theLCC03-treated tumor sections (Fig. 5C, TRAP), further confirmingthe dual targeting of prostate cancer cells and osteoclastogenesis byLCC03.

DiscussionThe NF-kB ligand (RANKL) pathways participate in the activation

and survival of osteoclasts; therefore, they represent a therapeutictarget for osteoclast-induced bone destruction in treatment- andmetastatic cancer-induced osteolysis. The RANKL inhibitor denosu-mab is recently approved by theU.S. FDA for the prevention of skeletal

Figure 5.

Therapeutic effects of LCC03 in an experimentalmodel of prostate cancer bonemetastasis.A,Bioluminescent images of representative individualmice bearing PC3-Luc tumor at tibia 1 day before LCC03 treatment (week 1) and during the 4-week monitoring period after initial treatment. Signals were adjusted to the same colorscale for the entire time course. Quantification of photon counts of each individual at the end of the experiment (week 5) relative to that 1 day before treatment isshown on the bottom. Bars indicate the mean fold change for each (n¼ 10). Values of Pwere calculated through 1-way analysis of variance. B, Toxicity evaluation ofLCC03 in tumor-bearingmice. Mean bodyweight–time profile (top) for each group. Data are expressed as themean� SD. Hematoxylin and eosin staining ofmultipleorgans (bottom,�200) from the representativemice treatedwith DMSO vesicle control and high-dose LCC03 at the end of the experiment. C,Representative imageof histopathology (H&E), IHC for cell proliferation (Ki67), autophagy (LC3B), and apoptosis (TUNEL), and TRAP staining for osteoclasts (stained in red) in serial tumorsections of representative individual mice. The original magnification is indicated.

Hsieh et al.

Mol Cancer Ther; 2019 MOLECULAR CANCER THERAPEUTICSOF8

on April 21, 2020. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst September 17, 2019; DOI: 10.1158/1535-7163.MCT-19-0387

related events in men with metastatic CRPC, but studies on thesecompounds have not demonstrated a survival benefit (27). Moreover,recent drug discovery efforts have identified small molecules thatimpair the ER function for therapeutic intervention. For example,bortezomib, thefirst FDA-approved proteasome inhibitor that inducessevere ER stress and causes apoptosis, was used to treat patients withmultiple myeloma (28). However, the ineffectiveness of the aforemen-tioned small molecules against several solid tumors including prostatecancer emphasized the need for the development effective drugs.Autophagy can be stimulated in response to multiple forms of cellularstress, such as intracellular pathogens, nutrient and growth factordeprivation, hypoxia, damaged organelles, and ER stress (29). In thepresent study, we provided the first evidence that LCC03, a derivativeof 5-(20,40-difluorophenyl)-salicylanilides that was originated reportedas a small molecular inhibitor of RANKL-induced osteoclastogenesis,exhibits strong activity against prostate cancer cell growth through ER-stress-mediated autophagy via the PERK/eIF2a signaling pathway.Other salicylanilide derivatives, such as Niclosamide (30) and nitro-substituted hydroxybenzamides (31–33) that share structure similar-ities with gefitinib and erlotinib, a class of small molecule inhibitors ofthe epidermal growth factor receptor (EGFR) exhibited antiprolifera-tive and/or proapoptotic activities against a spectrum of human cancercell lines. However, therapies targeting the EGFR by using gefitinib,and erlotinib showed nonsignificant clinical benefit in patients withCRPC (34). The dual action of 5-(20,40-difluorophenyl)-salicylanilideson both the osteoclasts and various CRPC cell lines (both AR-positiveand AR-negative) may make it a superior anticancer candidate toEGFR tyrosine kinase inhibitors and AR-targeting agents to preventand treat metastatic CRPC.

We demonstrated that LCC03 exerts a robust antiproliferativeeffect, induces cell-cycle arrest and reduces the viability in a dose-dependent manner in various prostate cancer cell lines. The tumorsuppressor protein p53 is amaster regulator with pleiotropic effects onmetabolism, anti-oxidant defense, genomic stability, proliferation,senescence, and cell death. The activation of p53 was shown to inhibitmTOR activity and transactivate autophagy-inducing genes (35). Fourcore autophagy genes: ATG4B, ATG4D, ULK1, and ULK2 were alsorecently defined as AR-targeting genes (36). The higher susceptibilityof C4-2 to LCC03 compared with the cell lines DU145 (AR-null andp53-mutant cells), CWR22r-v1 (AR-positive and p53-mutant cells),and PC3 (AR-null and p53-null cells) may be attributable to theexpression of both functional p53 and AR to induce autophagy.Thomas and colleagues (37) has demonstrated a 2-signal model forthe regulation of cell proliferation during ER stress. The first signalrepresents inhibition of protein translation mediated via the phos-phorylation of eIF2a by PERK, leading to impaired G2–M cell cycleprogression. The second signal, which induces G1 at later stages of ERstress, requires functional p53. Moreover, autophagy is known to beassociated with the advancement of the cell cycle, with a preference forthe G1 and S phases (38). These, in theory, may explain somewhat whyLCC03 induced G1 phase arrest in the C4-2 cells but G2 phase arrestwith an increased S-phase population in the PC3 cells. Although TP53represents one of the most frequent individual mutated genes inprostate cancer, the frequency of p53 alternations is approximately10% to 15% in prostate cancer (39), which is lower than in many othercancers. Therefore, prostate cancer, regardless to the disease status,could be considered as the primary indication of LCC03.

Numerous studies have suggested that autophagy induction is a cellsurvival mechanism of chemoresistance (40, 41); therefore, blockingautophagy during cancer treatment can improve clinical outcomes.However, other studies have indicated that autophagy induction

sensitizes cancer cells to chemotherapy and radiation (42, 43), suggest-ing autophagy stimulation for tumor-targeted therapy. Because autop-hagy is negatively regulated bymTOR, the pharmacological anticanceragents, including rapamycin, everolimus, and temsirolimus, whichoperate at the level of the mTOR pathway level are currently underinvestigation for clinical induction of autophagy. In the present study,we defined LCC03 as a novel ER stressor that initiates cell death inprostate cancer cells through autophagy induction. Although class IIIPI3K activation directly affects autophagy, our results reveled that inaddition to AMPK pathway, LCC03 considerably suppressed class IPI3K/Akt activation that regulates autophagy indirectly via the mTORpathway in both AR-positive and AR-negative prostate cancer cellpopulations. Thus, it provides themechanistic rationale for the efficacyof LCC03 in autophagy induction against prostate cancer and mayextend to other tumor types.

Three signaling proteins, IRE1a, ATF6, and PERK, are involvedin the initiation of ER-stress response (5). PERK was associated withER stress-induced autophagy. Once activated, this kinase directlyphosphorylates eIF2a to block translation initiation, leading to theupregulation of autophagy-related genes and inhibition of autop-hagy suppressors. Although the involvement of the IRE1a andATF6 pathways in the anticancer property of LCC03 was notaddressed in this study, we determined that blockage of PERKexpression by sequence-specific siRNA or selective small moleculeinhibitors rendered prostate cancer cells resistant to LCC03 to asimilar degree as the untreated group, thus implicating PERK as amajor effector of LCC03-induced cell death. Notably, a basal level ofLC3-B conversion in response to LCC03 treatment was still retainedeven when the cell viability was nearly completely restored by PERKantagonist, suggesting that the other ER-stress signaling pathways,such as the IRE1a and ATF6 pathways, may contribute to cyto-protective rather than cytotoxic autophagy induction. Our currentknowledge and future work in the area of autophagy regulationshould facilitate the development of improved therapeuticapproaches for cancer.

CRPC is well recognized to be resistant to conventional therapies,mainly due to its typical feature of apoptosis resistance (44). Ourin vivo proof of principle study showed the clinical benefit and safety ofLCC03 treatment in CRPC bone metastases in the clinically relevantanimal models. Thus, strongly support the translational application ofLCC03 [5-(20,40-difluorophenyl)-salicylanilide derivative] for poten-tial monotherapy and combination therapy against metastatic CRPC.Additional optimization studies for therapeutic index and biologicsformulation are warranted to improve its potency and pharmacolog-ical properties.

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

Authors’ ContributionsConception and design: C.-L. Hsieh, H.-S. Huang, K.-C. Chen, L.W.K. Chung,S.-Y. SungDevelopment of methodology: C.-L. Hsieh, H.-S. Huang, L.W.K. Chung, S.-Y. SungAcquisition of data (provided animals, acquired and managed patients, providedfacilities, etc.): K.-C. Chen, T. Saka, C.-Y. Chiang,Analysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): H.-S. Huang, T. Saka, C.-Y. Chiang, S.-Y. SungWriting, review, and/or revision of the manuscript: H.-S. Huang, K.-C. Chen,S.-Y. SungAdministrative, technical, or material support (i.e., reporting or organizing data,constructing databases): H.-S. Huang,Study supervision: C.-L. Hsieh, H.-S. Huang,

Salicylanilide as an Autophagy Inducer for CRPC Therapy

AACRJournals.org Mol Cancer Ther; 2019 OF9

on April 21, 2020. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst September 17, 2019; DOI: 10.1158/1535-7163.MCT-19-0387

AcknowledgmentsWe would like to acknowledge Ms. Huei-Min Chen for her exceptional technical

support for TEMatTaipeiMedicalUniversity Core Facility. This workwas supported inpart by Ministry of Science and Technology in Taiwan (Grant No. MOST 107-2320-B-038-057 and 106-2320-B-038-056 to C.-L. Hsieh, 106-2320-B-038-055 to S.-Y. Sung,and 106-2113-M-038-003 to H.-S. Huang), Taipei Medical University (Grant No. DP2-107-21121-C-01), and the “TMU Research Center of Cancer Translational Medicine”fromThe FeaturedAreas Research Center Programwithin the framework of theHigherEducation Sprout Project by the Ministry of Education (MOE).

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 April 10, 2019; revised July 24, 2019; accepted September 12, 2019;published first September 17, 2019.

References1. Bubendorf L, Schopfer A,WagnerU, Sauter G,MochH,Willi N, et al. Metastatic

patterns of prostate cancer: an autopsy study of 1,589 patients. HumPathol 2000;31:578–83.

2. Omlin A, Pezaro C, Gillessen Sommer S. Sequential use of novel therapeutics inadvanced prostate cancer following docetaxel chemotherapy. Ther Adv Urol2014;6:3–14.

3. Rao RV, Bredesen DE. Misfolded proteins, endoplasmic reticulum stress andneurodegeneration. Curr Opin Cell Biol 2004;16:653–62.

4. Clarke HJ, Chambers JE, Liniker E, Marciniak SJ. Endoplasmic reticulum stressin malignancy. Cancer Cell 2014;25:563–73.

5. Sano R, Reed JC. ER stress-induced cell death mechanisms. Biochim BiophysActa 2013;1833:3460–70.

6. Hotamisligil GS. Endoplasmic reticulum stress and atherosclerosis. Nat Med2010;16:396–9.

7. Lindholm D, Wootz H, Korhonen L. ER stress and neurodegenerative diseases.Cell Death Differ 2006;13:385–92.

8. Back SH, Kaufman RJ. Endoplasmic reticulum stress and type 2 diabetes.Annu Rev Biochem 2012;81:767–93.

9. Luo B, Lee AS. The critical roles of endoplasmic reticulum chaperones andunfolded protein response in tumorigenesis and anticancer therapies. Oncogene2013;32:805–18.

10. Chung LW,Hsieh CL, LawA, Sung SY, Gardner TA, EgawaM, et al. New targetsfor therapy in prostate cancer: modulation of stromal-epithelial interactions.Urology 2003;62:44–54.

11. Kratky M, Vinsova J, Novotna E, Mandikova J, Trejtnar F, Stolakova J.Antibacterial activity of salicylanilide 4-(trifluoromethyl)-benzoates. Molecules2013;18:3674–88.

12. Imramovsky A, Vinsova J, Ferriz JM, Dolezal R, Jampilek J, Kaustova J, et al.New antituberculotics originated from salicylanilides with promising in vitroactivity against atypical mycobacterial strains. Bioorg Med Chem 2009;17:3572–9.

13. Kratky M, Vinsova J. Antiviral activity of substituted salicylanilides—a review.Mini Rev Med Chem 2011;11:956–67.

14. KratkyM, Vinsova J. Antifungal activity of salicylanilides and their esters with 4-(trifluoromethyl)benzoic acid. Molecules 2012;17:9426–42.

15. BrownME, Fitzner JN, Stevens T, ChinW,Wright CD, Boyce JP. Salicylanilides:selective inhibitors of interleukin-12p40 production. BioorgMedChem 2008;16:8760–4.

16. Wang B,Wang Z, Ai F, TangWK, Zhu G. Amonofunctional platinum(II)-basedanticancer agent from a salicylanilide derivative: synthesis, antiproliferativeactivity, and transcription inhibition. J Inorg Biochem 2015;142:118–25.

17. Chen CL, Lee CC, Liu FL, Chen TC, Ahmed Ali AA, Chang DM, et al. Design,synthesis and SARs of novel salicylanilides as potent inhibitors of RANKL-induced osteoclastogenesis and bone resorption. Eur J Med Chem 2016;117:70–84.

18. Chen CL, Liu FL, Lee CC, Chen TC, Ahmed Ali AA, Sytwu HK, et al. Modifiedsalicylanilide and 3-phenyl-2H-benzo[e][1,3]oxazine-2,4(3H)-dione derivativesas novel inhibitors of osteoclast differentiation and bone resorption. JMedChem2014;57:8072–85.

19. Liu FL, Chen CL, Lee CC, Wu CC, Hsu TH, Tsai CY, et al. The simultaneousinhibitory effect of niclosamide on RANKL-induced osteoclast formation andosteoblast differentiation. Int J Med Sci 2017;14:840–52.

20. SmithMR. Osteoclast targeted therapy for prostate cancer: bisphosphonates andbeyond. Urol Oncol 2008;26:420–5.

21. LeeCC, Liu FL, ChenCL,ChenTC, ChangDM,HuangHS.Discovery of 5-(20 ,40-difluorophenyl)-salicylanilides as new inhibitors of receptor activator of NF-kappaB ligand (RANKL)-induced osteoclastogenesis. Eur J Med Chem 2015;98:115–26.

22. Hung SC, Wu IH, Hsue SS, Liao CH, Wang HC, Chuang PH, et al. Targeting l1cell adhesionmolecule using lentivirus-mediated short hairpin RNA interferencereverses aggressiveness of oral squamous cell carcinoma. Mol Pharm 2010;7:2312–23.

23. Sung SY,Wu IH,ChuangPH, Petros JA,WuHC,ZengHJ, et al. Targeting L1 celladhesion molecule expression using liposome-encapsulated siRNA suppressesprostate cancer bone metastasis and growth. Oncotarget 2014;5:9911–29.

24. Jackson WT, Giddings TH Jr, Taylor MP, Mulinyawe S, Rabinovitch M, KopitoRR, et al. Subversion of cellular autophagosomal machinery by RNA viruses.PLoS Biol 2005;3:e156.

25. OgataM,Hino S, SaitoA,MorikawaK,Kondo S, Kanemoto S, et al. Autophagy isactivated for cell survival after endoplasmic reticulum stress. Mol Cell Biol 2006;26:9220–31.

26. Senft D, Ronai ZA. UPR, autophagy, and mitochondria crosstalk underlies theER stress response. Trends Biochem Sci 2015;40:141–8.

27. Hayes AR, Brungs D, Pavlakis N. Osteoclast inhibitors to prevent bone metas-tases in men with high-risk, non-metastatic prostate cancer: a systematic reviewand meta-analysis. PloS one 2018;13:e0191455.

28. Chen D, Frezza M, Schmitt S, Kanwar J, Dou QP. Bortezomib as the firstproteasome inhibitor anticancer drug: current status and future perspectives.Curr Cancer Drug Targets 2011;11:239–53.

29. Murrow L, Debnath J. Autophagy as a stress-response and quality-controlmechanism: implications for cell injury and human disease. Annu Rev Pathol2013;8:105–37.

30. Li Y, Li PK, Roberts MJ, Arend RC, Samant RS, Buchsbaum DJ. Multi-targetedtherapy of cancer by niclosamide: a new application for an old drug. Cancer Lett2014;349:8–14.

31. Imramovsky A, Jorda R, Pauk K, Reznickova E, Dusek J, Hanusek J, et al.Substituted 2-hydroxy-N-(arylalkyl)benzamides induce apoptosis in cancer celllines. Eur J Med Chem 2013;68:253–9.

32. Kauerova T, Kos J, Gonec T, Jampilek J, Kollar P. Antiproliferative and pro-apoptotic effect of novel nitro-substituted hydroxynaphthanilides on humancancer cell lines. Int J Mol Sci 2016;17:1219.

33. Zhu XF,Wang JS, Cai LL, Zeng YX, Yang D. SUCI02 inhibits the erbB-2 tyrosinekinase receptor signaling pathway and arrests the cell cycle in G1 phase in breastcancer cells. Cancer Sci 2006;97:84–9.

34. Rathkopf DE, Larson SM, Anand A, Morris MJ, Slovin SF, Shaffer DR, et al.Everolimus combined with gefitinib in patients with metastatic castration-resistant prostate cancer: Phase 1/2 results and signaling pathway implications.Cancer 2015;121:3853–61.

35. Feng Z, Zhang H, Levine AJ, Jin S. The coordinate regulation of the p53 andmTOR pathways in cells. Proc Natl Acad Sci U S A 2005;102:8204–9.

36. Blessing AM, Rajapakshe K, Reddy Bollu L, Shi Y, White MA, Pham AH, et al.Transcriptional regulation of core autophagy and lysosomal genes by theandrogen receptor promotes prostate cancer progression. Autophagy 2017;13:506–21.

37. Thomas SE,Malzer E, Ordonez A,Dalton LE, van'tWout EF, Liniker E, et al. p53and translation attenuation regulate distinct cell cycle checkpoints duringendoplasmic reticulum (ER) stress. J Biol Chem 2013;288:7606–17.

38. Tasdemir E,Maiuri MC, Tajeddine N, Vitale I, Criollo A, Vicencio JM, et al. Cellcycle-dependent induction of autophagy, mitophagy and reticulophagy.Cell Cycle 2007;6:2263–7.

39. Kluth M, Harasimowicz S, Burkhardt L, Grupp K, Krohn A, Prien K, et al.Clinical significance of different types of p53 gene alteration in surgically treatedprostate cancer. Int J Cancer 2014;135:1369–80.

40. Garcia JA, Danielpour D. Mammalian target of rapamycin inhibition as atherapeutic strategy in the management of urologic malignancies.Mol Cancer Ther 2008;7:1347–54.

Hsieh et al.

Mol Cancer Ther; 2019 MOLECULAR CANCER THERAPEUTICSOF10

on April 21, 2020. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst September 17, 2019; DOI: 10.1158/1535-7163.MCT-19-0387

41. Thorburn J, Horita H, Redzic J, Hansen K, Frankel AE, Thorburn A. Autophagyregulates selective HMGB1 release in tumor cells that are destined to die.Cell Death Differ 2009;16:175–83.

42. Chen YR, Tsou B, Hu S,MaH, Liu X, Yen Y, et al. Autophagy induction causes asynthetic lethal sensitization to ribonucleotide reductase inhibition in breastcancer cells. Oncotarget 2016;7:1984–99.

43. Fiorini C, Cordani M, Gotte G, Picone D, Donadelli M. Onconase inducesautophagy sensitizing pancreatic cancer cells to gemcitabine and activates Akt/mTOR pathway in a ROS-dependentmanner. Biochim Biophys Acta 2015;1853:549–60.

44. Leo S, Accettura C, Lorusso V. Castration-resistant prostate cancer: targetedtherapies. Chemotherapy 2011;57:115–27.

AACRJournals.org Mol Cancer Ther; 2019 OF11

Salicylanilide as an Autophagy Inducer for CRPC Therapy

on April 21, 2020. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst September 17, 2019; DOI: 10.1158/1535-7163.MCT-19-0387

Published OnlineFirst September 17, 2019.Mol Cancer Ther   Chia-Ling Hsieh, Hsu-Shan Huang, Kuan-Chou Chen, et al.   Stress-Activated PERK Signaling PathwayDeath in Castration-Resistant Prostate Cancer via ER A Novel Salicylanilide Derivative Induces Autophagy Cell

  Updated version

  10.1158/1535-7163.MCT-19-0387doi:

Access the most recent version of this article at:

  Material

Supplementary

  http://mct.aacrjournals.org/content/suppl/2019/09/17/1535-7163.MCT-19-0387.DC1

Access the most recent supplemental material at:

   

   

   

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

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

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

  Permissions

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

.http://mct.aacrjournals.org/content/early/2019/12/11/1535-7163.MCT-19-0387To request permission to re-use all or part of this article, use this link

on April 21, 2020. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst September 17, 2019; DOI: 10.1158/1535-7163.MCT-19-0387