an epigenetic reprogramming strategy to resensitize radioresistant ... · an epigenetic...

Molecular and Cellular Pathobiology

An Epigenetic Reprogramming Strategy toResensitize Radioresistant Prostate Cancer CellsClaudia Peitzsch1, Monica Cojoc1, Linda Hein1, Ina Kurth1, Katrin M€abert1,Franziska Trautmann1, Barbara Klink2,3,4, Evelin Schr€ock2,3,4, Manfred P.Wirth5,Mechthild Krause1,3,4,6,7, Eduard A. Stakhovsky8, Gennady D.Telegeev9,Vladimir Novotny5,Marieta Toma10, Michael Muders10, Gustavo B. Baretton10, Fiona M. Frame11,Norman J. Maitland11, Michael Baumann1,3,4,6,7, and Anna Dubrovska1,3,4,7

Abstract

Radiotherapy is a mainstay of curative prostate cancer treat-ment, but risks of recurrence after treatment remain significantin locally advanced disease. Given that tumor relapse can beattributed to a population of cancer stem cells (CSC) thatsurvives radiotherapy, analysis of this cell population mightilluminate tactics to personalize treatment. However, thisdirection remains challenging given the plastic nature of pros-tate cancers following treatment. We show here that irradiatingprostate cancer cells stimulates a durable upregulation of stemcell markers that epigenetically reprogram these cells. In bothtumorigenic and radioresistant cell populations, a phenotypic

switch occurred during a course of radiotherapy that wasassociated with stable genetic and epigenetic changes. Specif-ically, we found that irradiation triggered histone H3 meth-ylation at the promoter of the CSC marker aldehyde dehydro-genase 1A1 (ALDH1A1), stimulating its gene transcription.Inhibiting this methylation event triggered apoptosis, promot-ed radiosensitization, and hindered tumorigenicity of radio-resistant prostate cancer cells. Overall, our results suggest thatepigenetic therapies may restore the cytotoxic effects of irra-diation in radioresistant CSC populations. Cancer Res; 76(9);2637–51. �2016 AACR.

IntroductionProstate cancer is themost common cancer amongmen and the

third leading cause of cancer-related deaths in men worldwide(1). A significant proportion of prostate cancer patients arediagnosed with potentially curable localized tumor that can be

treated with surgery and radiotherapy alone or in combinationwith androgen ablation. Tumor radioresistance can be a limitingfactor of the efficacy of radiotherapy for somepatientswith a high-risk prostate cancer (2, 3).Depending on the stage of disease, up to45% of prostate cancers can relapse after radiotherapy (4–6).Tumor relapse after radiotherapy has been attributed to thepopulation of cancer stem cells (CSC) or tumor-initiating cells(7–9). The CSC hypothesis argues that cancer cells are hierar-chically organized according to their tumorigenic potential, andtumors are initiated and maintained by the populations of CSCs(10). Decades of radiobiologic research have demonstrated thatthe frequency of tumor cells with CSC characteristics and theirintrinsic radiosensitivity varies between tumors (7, 11, 12).Radiobiologic studies provided evidence for the importance ofthese cells for local tumor control and suggested that efficienttumor treatment by irradiation might require eradication of theentire CSC population. The role of CSCs in tumor developmentand recurrence has recently motivated investigations of CSC-specific biomarkers for the analysis of CSC populations in tumorpretreatment biopsies for the prediction of clinical outcome andselection of the optimal treatment strategy (13, 14). It is alsoassumed that the combination of radiotherapy with agents thattarget or radiosensitize theCSCpopulationmight bebeneficial forthe treatment refinement (14). However, compelling evidencesuggests a high diversity of CSCs and plasticity of their featuresimposed by tumor treatment, which could challenge the devel-opment of CSC-related predictive biomarkers and CSC-targetedtherapy. Furthermore, CSCs, which are selected or induced fol-lowing chemo- or radiotherapy, are very difficult to detect beforetreatment and might acquire treatment resistance, resulting intumor relapse (15).

1OncoRay-National Center for Radiation Research in Oncology, Fac-ultyofMedicine andUniversity Hospital Carl GustavCarus,TechnischeUniversit€at Dresden and Helmholtz-Zentrum Dresden-Rossendorf,Dresden, Germany. 2Institute of Clinical Genetics, Faculty of Medicineand University Hospital Carl Gustav Carus, Technische Universit€atDresden, Dresden, Germany. 3German Cancer Consortium (DKTK),Dresden,Germany. 4German Cancer ResearchCenter (DKFZ), Heidel-berg, Germany. 5Department of Urology, Faculty of Medicine andUniversityHospital CarlGustavCarus,TechnischeUniversit€atDresden,Dresden, Germany. 6Department of Radiation Oncology, Faculty ofMedicine and University Hospital Carl Gustav Carus, Technische Uni-versit€at Dresden, Dresden, Germany. 7Helmholtz-Zentrum Dresden-Rossendorf, Institute of Radiation Oncology, Dresden, Germany.8National Cancer Institute, Kyiv, Ukraine. 9Institute of Molecular Biol-ogy and Genetics NAS of Ukraine, Kyiv, Ukraine. 10Department ofPathology, Faculty of Medicine and University Hospital Carl GustavCarus,Technische Universit€at Dresden, Dresden, Germany. 11YCR Can-cer Research Unit, Department of Biology, University of York, York,United Kingdom.

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

M. Cojoc and L. Hein contributed equally to this article.

Corresponding Author: Anna Dubrovska, Technische Universit€at Dresden,Fetscherstr. 74, Dresden 01307, Germany. Phone: 4935-1458-7150; Fax:4903-5145-87311; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-15-2116

�2016 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 2637

on May 18, 2020. © 2016 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst March 16, 2016; DOI: 10.1158/0008-5472.CAN-15-2116

http://cancerres.aacrjournals.org/

In this study, we show that irradiation induces genetic andepigenetic alterations that affect prostate cancer cell tumorigenic-ity and radioresistance. The epigenetic changes driven by irradi-ation are mediated by histone methylation, which induces theexpression of ALDH1A1 gene regulating the maintenance of theprostate CSCs and their radioresistance. Targeting of the histone 3methylationwithDZNep, an inhibitor of lysinemethyltransferaseenhancer of zeste homologue 2 (EZH2), leads to the downregula-tion of ALDH1A1 expression and tumor cell radiosensitization.

Furthermore, irradiation causes long-term alterations in theexpression of stem cell markers and induces tumor cell repro-gramming. To our knowledge, this is the first study demonstratingthat radioresistant cell populations within prostate cancer cellsundergo a phenotypic switch during the course of irradiation. Wefound that in contrast to nonirradiated cells, ALDH activity is notindicative of a radioresistant cell subset in the cells surviving thecourse of fractionated irradiation. Moreover, our study demon-strates that radioresistant prostate cancer cells are more sensitiveto the DZNep treatment, which induces apoptosis and abrogatestumorigenicity in combination with X-ray irradiation.

Our findings suggest that radioresistant properties of cancercells are dynamic in nature and that a combination of radiother-apy with drugs that prevent tumor cell reprogramming may bebeneficial for eradication of tumor-initiating and radioresistantcell populations. Future research must extend this work by usingadditional experiments on primary cultures and biopsies, and bycombination of DZNep treatment with radiotherapy in xenograftmouse models of human prostate cancer.

Materials and MethodsCell lines and culture condition

Prostate cancer cell lines DU145, PC3, and LNCaP werepurchased from the ATCC and cultured according to the manu-facturer's recommendations in a humidified 37�C incubatorsupplemented with 5% CO2. DU145 and PC3 cell lines weremaintained in DMEM (Sigma-Aldrich) and LNCaP cells inRPMI1640 medium (Sigma-Aldrich) containing 10% FBS (PAALaboratories) and 1mmol/L L-glutamine (Sigma-Aldrich). Radio-resistant (RR) cell lines were established as described previously(16). The RR cells were kept in culture up to 6months after the lastirradiation and their radioresistance was confirmed by radiobio-logic cell survival assays. The corresponding age-matched nonir-radiated parental cells were used as controls for RR cell lines.Human papillomavirus 18 (HPV 18) immortalized, nontumori-genic RWPE1 cells were purchased from the ATCC and main-tained in the complete keratinocyte growthmedium, K-SFM (LifeTechnologies) supplemented with 50 mg/mL of bovine pituitaryextract and 5 ng/mL of EGF. All cell lines were genotyped usingmicrosatellite polymorphism analysis and tested formycoplasmadirectly prior to experimentation.

Human prostate tumor primary cells were obtained from N.J.Maitland's laboratory at the YCR Cancer Research Unit, Universityof York. Primary cell culture 311/13 was established from 67-year-old prostate cancer patient with Gleason score 3 þ 4 ¼ 7, pT2c,preoperative PSA 8.7 ng/mL. Primary cell culture 312/13 wasestablished from 66-year-old prostate cancer patient with Gleasonscore 3 þ 4 ¼ 7, pT2c, preoperative PSA 6.6 ng/mL. Epithelial cellcultures were prepared and characterized as described previously(17), andmaintained in stem cellmedium containing keratinocytegrowth medium supplemented with 5 ng/mL of EGF, bovine

pituitary extract (Life Technologies), 2ng/mL leukemia inhibitoryfactor (Peprotech)2ng/mL stemcell factor (Peprotech),100ng/mLcholera toxin (Sigma-Aldrich), and 1ng/mL granulocyte macro-phage colony-stimulating factor (GM-CSF;MiltenyiBiotec; ref. 17).Cells were cultured in the presence of irradiated (60Gy) STO(mouse embryonic fibroblast) cells. After expansion, the primaryepithelial cellswereplatedatdensity of8�104 cells/mL incollagenI–coated chamber slides (Corning) without feeder cells. Cells werepretreatedwithDMSO(control cells)orDZNepat concentrationof1 mmol/L for 24 hours, irradiated with 4 Gy or left nonirradiatedandfixedwith4%formaldehyde24hours after irradiation.Primaryprostate cellswereused for the experiments at the earlypassages (upto 6 passages for the cells frompatient 312/13 and up to 5 passagesfor the cells from patient 311/13).

Human tissue samplesClinical material was collected at the Department of Urology,

University Hospital Carl Gustav Carus, Technische Universit€atDresden and Kyiv National Cancer Institute with informed con-sent and approval from the local ethics committee (InstitutionalReview Board of the Faculty of Medicine, Technische Universit€atDresden, EK49022015 and Ethics Committee of Kyiv NationalCancer Institute, protocol no. 44, respectively). The tumor spec-imen PT1 was taken from a lymph node of a 69-year-old patient(pT3bpN1L1V0R0G3a,Gleason score 4þ3¼7, preoperative PSAlevel 6.58 ng/mL) with newly diagnosed prostate cancer, whounderwent primary surgery (bilateral lymphadenectomy andpros-tate vesiculectomy). The tumor specimen PT2 is a radical prosta-tectomy tissue taken from a 64-year-old patient (pT2c N0 M0,Gleason score 3þ 4¼ 7, preoperative PSA level 19.8 ng/mL). Thetumor specimen PT3 is a radical prostatectomy tissue taken from48-year-old patient (pT3apN0L0V0Pn1, Gleason score 4 þ 3 ¼ 7,preoperative PSA level 10.7 ng/mL).

Human tumor tissue ex vivo cultureTumor tissue samples were cut into small pieces of approx-

imately 3 mm � 3 mm and transferred to 12-well ultra-lowattachment plates (Corning) in MEBM (Lonza) supplementedwith 4 mg/mL insulin (Sigma-Aldrich), B27 (Invitrogen), 20ng/mL EGF (Peprotech), 20 ng/mL basic fibroblast growth factor(FGF; Peprotech). Penicillin-streptomycin solution was addedto 100 mL of cell culture media for a final concentration of 50to 100 I.U./mL penicillin and 50 to 100 (mg/mL) streptomycin.The tissues were treated with 2 mmol/L of DZNep for 24 hours,consequently irradiated with 4 Gy of X-ray and fixed 24 hoursafter irradiation. Tissue fragments incubated under the sameconditions but treated with DMSO and sham irradiation wereused a control. The presence of tumor cells in the tissue pieceswas analyzed by hematoxylin and eosin staining, by fluorescentimmunostaining with PSMA and CK5/14 antibody, and wasconfirmed by a pathologist.

Animals and in vivo tumorigenicity assayThe animal facilities and the experiments were approved

according to the institutional guidelines and the German animalwelfare regulations [protocol number 24(D)-9168.TVV2014/30].The experiments were performed using 8- to 12-week-old maleNMRI (nu/nu) mice that were bred in-house (ExperimentalCentre, Faculty of Medicine, Technische Universit€at Dresden). Toimmunosuppress the nudemice further, themice underwent totalbody irradiation 1–3days before tumor transplantationwith 4Gy

Peitzsch et al.

Cancer Res; 76(9) May 1, 2016 Cancer Research2638




(200 kV X-rays, 0.5 mm Cu filter, 1 Gy/minute). Mice wereexamined once a week and the relative tumor volumes (mm3)based on caliper measurements were calculated as length�width� height/2. At a tumor volume of 100 mm3, mice were scored astumor bearing to calculate tumor uptake. Kaplan–Meier plotswere analyzed using the log-rank test (GraphPad Prism 5). Anal-ysis of the tumor-initiating cell numbers was performed using theweb-based ELDA (Extreme Limiting Dilution Analysis) statisticalsoftware at http://bioinf.wehi.edu.au/software/elda/index.html,which uses the frequency of tumor positive and negative animalsat each transplant dose to determine the number of tumor-initiating cells within the injected cell populations.

Microarray analysis of the prostate cancer cell linesGene expression profiling of the DU145, DU145-RR, ALDHþ

DU145, ALDH� DU145, ALDHþ DU145-RR, ALDH� DU145-RR cells was performed using SurePrint G3 Human Gene Expres-sion 8 � 60 K v2 Microarray Kit (Design ID 039494, AgilentTechnologies) according to the manufacturer's recommenda-tions. Total RNA was isolated from cell pellets using the RNeasyKit (Qiagen). Sample preparation for analysis was carried outaccording to the protocol detailed by Agilent Technologies(Santa Clara). Briefly, first and second cDNA strands weresynthesized, double-stranded cDNA was in vitro transcribedusing the Low Input Quick Amp Labeling Kit, and the resultingcRNA was purified and hybridized to oligonucleotide arraysrepresenting about 60,000 features, including 27,958 EntrezGene RNAs and 7,419 lincRNAs. Arrays were processed usingstandard Agilent protocols. Probe values from image files wereobtained using Agilent Feature Extraction Software. The datasetwas normalized using GeneSpring software, and the list ofdifferentially regulated genes with fold change > 2 and P <0.05 was further analyzed using the web-based Panther PathwayAnalysis tool (http://www.pantherdb.org/tools/index.jsp).

Data deposition. All data are MIAME compliant and the raw datahave been deposited in the Gene Expression Omnibus (GEO)database, accession no GSE53902 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc¼GSE53902).

Immunofluorescence microscopyCells were plated onto Millicell EZ SLIDE 8-well glass slides

(Millipore) in a regular culture medium containing 10% serum.After 18 hours, the cells were fixed for 30 minutes with 3.7%formaldehyde at room temperature and permeabilized with0.125% Triton X-100 for 10 minutes followed by washing withPBS, and blocked by incubation with 10% BSA in PBS. The cellswere then incubated for 12 hours at 4�C with primary antibodyagainst phospho-histone H2AX (S139; Millipore), b-catenin(#D10A8, Cell Signaling Technology), ALDH1A1 (clone H-4,Santa Cruz Biotechnology), cleaved PARP (Asp214; #5625, CellSignaling Technology), or ALDH1A3 (HPA046271, Sigma-Aldrich) diluted in 3% BSA in PBS as 1/100 and washed tentimes with PBS. Cells were then incubated for 0.5 hour with asecondary antibody conjugated with Alexa Fluor 488 or 555(Invitrogen) diluted 1/500 in 3% BSA in PBS. After appropriatewashes with PBS, cells were stained with DAPI and examined byfluorescent microscopy. For quantification, 200 to 1,000 cells percondition in at least three randomly selected fields were counted.Nuclear intensity of gH2A.X and cleaved PARP signal was eval-uated using ImageJ software.

Clonogenic cell survival assayCells were plated at a density of 500–2,000 cells/well depend-

ing on the cell line in 6-well plates and irradiated with doses of 2,4, 6, and 8Gy of 200 kV X-rays (Yxlon Y.TU 320; dose rate 1.3 Gy/minute at 20mA)filteredwith 0.5mmCu. The absorbeddosewasmeasured using a Duplex dosimeter (PTW). After 10 days, thecolonies were fixed with 10% formaldehyde (VWR) and stainedwith 0.05% crystal violet (Sigma-Aldrich). Colonies containing>50 cells were counted using a stereo microscope (Zeiss). Theplating efficiency (PE) was determined as the ratio between thegenerated colonies and the number of cells plated. The survivingfraction (SF) was calculated as the PE from the irradiated cellsdivided by the PE from the nonirradiated control. For three-dimensional (3D) colony formation analysis, a cell suspension(500 cells/ml) in 0.2% low-melting SeaPlaque CTG agarose(Cambrex Bio Science Rockland, Inc.) with MEBM medium(Lonza) supplemented with 4 mg/mL insulin (Sigma-Aldrich),B27 (Invitrogen), 20 ng/mL EGF (Peprotech), and 20 ng/mL FGF(Peprotech) was overlaid into 24-well low attachment plates(Corning). All samples were plated in triplicates. The plates wereincubated at 37 �C, 5%CO2 in a humidified incubator for 14 days.

Sphere formation assayTo evaluate the self-renewal potential, cells were grown as

nonadherent multicellular cell aggregates (spheres). Cells wereplated at a density of 1,000 cells/2 mL/well in 6-well ultra-lowattachment plates (Corning) in MEBM medium (Lonza) supple-mented with 4 mg/mL insulin (Sigma-Aldrich), B27 (Invitrogen),20 ng/mL EGF (Peprotech), and 20 ng/mL FGF (Peprotech).Media containing supplements were refreshed once a week andspheres with a size > 100 mm were assayed after 14 days usingAxiovert 25 microscope (Zeiss) or were automatically scannedusing the Celigo S Imaging Cell Cytometer (Brooks). Spherenumber and size were calculated using ImageJ software.

Cell proliferation and cytotoxic activityCell proliferation and cytotoxicity was analyzed using the MTT

assay (Sigma-Aldrich). A total of 3 � 103 cells/well were seededinto 96-well plates and quantified after 1–4 days at 560 nmabsorbance, with reference at a wavelength of 690 nm using amicroplate reader (Tecan).

Transwell migration assayProstate cancer cells were starved for 24 hours under serum-free

conditions. A total of 104 cells were plated on top of the mem-branewith 8 mmpore size (Boyden chamber, BD Biosciences) in a24-well plate. After 24 hours, membranes were fixed with 10%paraformaldehyde and stained with 0.04% crystal violet. Abso-lute migration was evaluated by counting migrating cells on themembrane using Axiovert25 microscope (Zeiss).

Chemical inhibitors and Western blot analysisXAV939, BIX01294, and DZNep inhibitors were purchased

from Biomol GmbH. DMSO was used at concentration �1%v/vDMSOas a drug solvent, and corresponding concentrations ofDMSO were used as controls in all the experiments, whichincluded cell treatment. Cells were lysed in RIPA buffer (SantaCruz Biotechnology) and protein concentrations were deter-mined using the bicinchoninic acid assay according to manufac-turer's recommendations (Pierce). All antibodies were used atconcentrations recommended by the manufacturer followed by

An Epigenetic Targeting for Prostate Cancer Radiosensitization

www.aacrjournals.org Cancer Res; 76(9) May 1, 2016 2639




the incubation with a 1:5,000 dilution of appropriate HRP-conjugated secondary antibody (Santa Cruz Biotechnology). Thesignal was visualized using the enhanced chemiluminescencedetection reagent (GE Healthcare). Semiquantitative analysis ofprotein expression relative to the loading control b-actin orGAPDH was performed with ImageJ software.

Flow cytometry analysisFor single cell analysis and multi-color staining, cells were

dissociated with Accutase (PAA) and stained at 4�C in PBS bufferwith 1 mmol/L ethylenediaminetetraacetic acid (EDTA) and 5%FBS according to the manufacturer's protocol. Aldehyde dehy-drogenase activity was analyzed using Aldefluor assay (Stem CellTechnologies) according to manufacturer's protocol. Sampleswere analyzed with the BD LSRII flow cytometer (Beckton Dick-inson). A minimum of 100,000 viable cell events were collectedper sample. Data were analyzed using FlowJo software (version7.6.2) and gates were set according to the individual isotypecontrols. To purify ALDHþ cell population, cells were stainedusing the Aldefluor assay (Stem Cell Technologies). Cells incu-bated with the specific ALDH inhibitor diethylaminobenzalde-hyde (DEAB) served as negative control. Dead cells were excludedby 1 mg/mL DAPI staining and doublets were excluded using theFSC-W&H and SSC-W&H function of the BD FACSDiva 6.0software. ALDHþ and ALDH� populations were sorted on a BDFACS Aria III (Beckton Dickinson) using 100 mm nozzle with anaverage purity of >90%.

Chromatin immunoprecipitation assayChromatin immunoprecipitation assays were performed using

the ChIP Assay Kit (17-295, Millipore) according to the manu-facturer's instructions. DNA shearing was performed using Cov-aris S2 ultrasonicator under the following conditions: intensity 4,duty cycle 10%, cycles per burst 200, treatment time 55 seconds.The following antibodies were used to immunoprecipitate DNA:anti-H3 (D2B12, Cell Signaling Technology), anti-H3K36me3(D5A7, Cell Signaling Technology), and normal rabbit IgG (CellSignaling Technology).

Luciferase reporter assayALDH1A1-promoter luciferase reporter was purchased from

Switchgear genomics. M50 Super 8x TOPFlash andM51 Super 8xFOPFlash (TOPFlash mutant) were a gift from Randall Moon(Addgene plasmid # 12456 and # 12457; Supplementary ref. 1).Transient transfectionof plasmidDNA inDU145 andLNCaP cellswas performed using Lipofectamine 2000 (Life TechnologiesGmbH) according to the manufacturer's instructions. Luciferaseactivity was normalized to FOP-Flash reporter. Luciferase assaywas conducted using LightSwitch Assay (Switchgear genomics) orBright-Glo luciferase assay system (Promega).

HistologyHuman tumor specimens were fixed by immersion in 4%

paraformaldehyde, cryoprotected in 30% sucrose, and embeddedin TissueTek O.C.T compound (Sakura Finetek). The 10-mm thicktissue sections (Microm HM 560, Cryo-Star Cryostat) were col-lected on Superfrost plus slides (Thermo Fisher Scientific). Slideswere blocked for 1 hour in PBS buffer containing 10% of horseserum and 0.4% TritonX100 and then incubated overnight at 4�Cwith the primary antibodies against EpCAM (sc-21792, SantaCruz Biotechnology), gH2A.X (05-656, Millipore), and cleaved

PARP (Asp214; #5625, Cell Signaling Technology). Staining withantibody against CK5/14 (#C1857C01, DCS Innovative Diag-nostik-Systeme, clone XM26/SFI-3) was used to distinguishbenign prostate glands (CK5/14 positive) from cancer glands(CK5/CK14 negative). Bound antibodies were detected withappropriate secondary antibodies conjugated with AlexaFluor488 or 555 (Molecular Probes) at room temperature for 1 hour.Slides were analyzed using the Meta Confocal Microscope(LSM510, Zeiss). Images were analyzed using ImageJ software.For quantification, two independently treated tissue pieces vali-dated for the presence of tumor cells were analyzed for eachcondition.

Determinationof glutathione and reactive oxygen species levelsTo analyze glutathione (GSH) level, cells were stained with

40 mmol/L monochlorobimane (mBCI; Life Technologies) for30 min at 37�C, washed with PBS at the end of culture andanalyzed by Celigo cytometer. To measure the level of reactiveoxygen species (ROS), cells were incubatedwith 5 mmol/L 5-(and-6)-carboxy-20,70-dichlorodihydrofluorescein diacetate acetyl ester(CMH2DCFDA, Invitrogen Molecular Probes) for 15 minutes at37�C. Samples were analyzed with the BD LSRII flow cytometer(Beckton Dickinson). A minimum of 100,000 viable cell eventswere collected per sample. Data were analyzed using FlowJosoftware (version 7.6.2) and gates were set according to theindividual controls.

Array comparative genome hybridizationArray comparative genome hybridization was performed on

Agilent's SurePrint G3HumanCGHþSNPMicroarray Kit 2� 400K (Design ID 028081, Agilent) according to the manufacturer'sinstructions with the exception that the labeling of reference andtest DNA was reversed. Arrays were scanned using an Agilentmicroarray scanner. Agilent's CytoGenomics Editions 2.7 and 2.9were used for extraction and processing of Raw data (using theintegrated Feature Extraction software) and to determine deletedand amplified regions based on the draft of the reference humangenome (GRCh37/hg19) using the Default Analysis Method -CGH v2. All results were additionally checked by eye and wereevaluated by a board-certified medical geneticist (Barbara Klink).CNVs that were commonly found in the Database of GenomicVariants (http://dgv.tcag.ca/) and therefore can be considered aspolymorphisms that aremost likely germline were excluded fromfurther analysis.

Statistical analysisThe results of colony formation assays, microscopy image

analysis, gene reporter assay, ChIP assay, cell migration, flowcytometry analysis, cell proliferation assays, gene expression,and siRNA-mediated gene silencing were analyzed by pairedt test. Tumor uptake rate was calculated using log-rank (Man-tel–Cox) test. Multiple comparison analysis for the datadepicted on Fig. 2D, Supplementary Fig. S9B and S9C wasperformed using one-way ANOVA analysis by GraphPad Prismsoftware. The differences between cell survival curves depictedon Figs. 3A and 4C, and Supplementary Fig. S7C were analyzedusing the statistical package for the social sciences (SPSS) v23software as described by Franken and colleagues (18) by fittingthe data into the linear-quadratic formula S(D)/S(0) ¼ exp(aDþ bD2) using stratified linear regression. A P value of <0.05 wasregarded as statistically significant. IC50 values were calculated

Peitzsch et al.





by fitting the curves with nonlinear regression using GraphPadPrism software.

ResultsIrradiation causes long-term alterations in the expression ofstem cell markers

In our previous studies, we have shown that radioresistantsublines obtained by multiple irradiation of prostate cancer celllines with 4 Gy of X-ray in vitro are characterized by an increasedexpression of stem cell markers and enhanced tumorigenicity inmice (16). Here, we analyzed the time-dependent effect of irra-diation on the expression of different CSCmarkers and activationof prosurvival pathways in parental (P) and radioresistant (RR)DU145 and LNCaP cells (Supplementary Fig. S1A). Irradiation ofDU145 and LNCaP cells with a single X-ray dose of 4 Gy led tolong-term and time-dependent changes in the cell phenotype,such as enhanced ALDH activity, increased expression of stem cellmarkers including OCT4, NANOG, BMI1, ABCG2, activation ofPI3K/AKT signaling pathway, and gain of epithelial–mesenchy-mal transition (EMT) signatures such as upregulation of theb-catenin and vimentin expression (Fig. 1A and B and Supple-mentary Fig. S1B and S1C). The expression level of these stem cellmarkers and EMT signatures continuously increased during 3 to 4weeks after irradiation and was correlated with an increase inclonogenic potential of DU145 and DU145-RR cells (Fig. 1C).

Fractionated irradiation with more than 56 Gy (14 � 4 Gy) ofX-rays resulted in the establishment of cell sublines with a stable

radioresistant phenotype, which has upregulated CSC markerexpression, increased migratory potential, and high tumorigenic-ity in vivo (Supplementary Fig. S1D; ref. 16). We and others haveshown that prostate cancer cells with a high ALDH activity(ALDHþ) have functional characteristics of tumor-initiating cells(16, 19). We found that irradiation might cause ALDHþ cellenrichment inboth, ALDHþ andALDH� cell populations isolatedby fluorescence-activated cell sorting (FACS) from DU145 pros-tate cancer cells and cultured in serum-containing medium fortwoweeks (Fig. 1D). Notably, the ALDHþ cell population did notshow an increased proliferation in response to irradiation, ascompared with ALDH� cells (Supplementary Fig. S1E). Theseresults suggest that irradiation can not only select but also inducesa prostate tumor cell population with CSC features. Taken togeth-er, these findings have shown that irradiationmight influence cellclonogenicity and radioresistance, and has a long-term effect onthe expression of stem cells markers and cell phenotype.

Irradiation induces epigenetic and genetic alterations inprostate cancer cells

Our previous studies demonstrated that regrowth of DU145xenograft tumors after chemotherapy can be attributed to cer-tain genetic aberrations, which lead to an increase in tumor-initiating cells (20). Microsatellite polymorphism analysisrevealed genetic alterations in DU145-RR cells as comparedwith the parental DU145 cells, suggesting that X-ray treatmentmight induce genomic instability in prostate cancer cells (Sup-plementary Fig. S2A).

0

5

10

15

20

25

50403020100

Plat

ing

effic

ienc

y (%

)

Time (days)

0

10

20

30

40

50

60

70

50403020100

Pla

ting

effic

ienc

y (%

)

Time (days)

3D Colony2D Colony

Time (days)

ALDH

DU145 PDU145 RR

DU145 PDU145 RR

DU145 P

NanogOCT 4

p-Akt (Ser473)

β-catenin

Total AKT

Vimentin

NI 1 3 7 14 21 28 35 42

BMI1

E-cadherin

GAPDH

LNCaP P

NI 1 3 7 14 21 28 35 42

Days after 4 Gy irradiation

A

C D

* P = 0.02

P = 0.06

0

50

100

150

200

250

NI 1 3 7 14 21 28 35 42

0

50

100

150

200

250

NI 1 3 7 14 21 28 35 42Days after 4 Gy irradiationDays after 4 Gy irradiation

Rel

ativ

e pr

otei

n ex

pres

sion

leve

l

BDU145 P LNCaP P

02468

1012141618

ALDH– ALDH+

ALD

H+

cell

popu

latio

n

(fold

cha

gne)

0 Gy2 × 4 Gy

DU145 P

2 × 4 Gy Day 14:

Day 0

analysis of ALDH

population

ALDH+

ALDH–

+/-

02468

101214161820

100 3020 40

%)

H p

ositi

ve (

DA

L

DU145 PDU145 RR

β-cateninOCT4NANOGBMI1

Figure 1.Analysis of the time-dependent effects of irradiation on the properties of prostate cancer cells. A, the delayed effect of X-ray irradiation on the expression of thestem cell and EMT markers. After irradiation with 4 Gy of X-ray, the lysates of DU145 and LNCaP cells were prepared at the indicated time points during6 weeks after irradiation and analyzed by Western blotting. NI, nonirradiated cells. The accompanying graphs display the densitometry-based analysis of theWestern blot data. B, flow cytometry analysis of the ALDH activity in DU145 and DU145-RR cells within 6 weeks after irradiation with 4 Gy of X-ray.Error bars, SEM; n ¼ 4. P, parental cells; RR, radioresistant cells. C, the colony-forming properties of DU145 cells and DU145-RR cells were analyzed within6 weeks after the last irradiation. Error bars, SEM; n ¼ 3. D, induction of the ALDHþ cell population after irradiation of ALDHþ and ALDH� cell subsets.FACS-sorted ALDHþ and ALDH� cells were plated in a 6-well plate and irradiated with 4 Gy 18 hours later. Irradiation with 4 Gy was repeated at day 7 afterthe first irradiation. The cells were analyzed at day 14 by flow cytometry. Error bars represent SEM; n ¼ 3. P, parental cells; RR, radioresistant cells.






To identify genomic aberrations in the irradiated prostatecancer cells more comprehensively, we analyzed DNA fromDU145 and DU145-RR cells using comparative genome hybrid-ization (CGH) microarray. CGH array revealed a high number ofvarious copy number variations (CNV) in parental DU145 cells,including partial chromosome deletions and duplications (e.g.,partial gains and losses on chromosome 4). There were alsomarked differences between the parental and radioresistant cells(Supplementary Fig. S2B and S2C). Noteworthy, comparativeanalysis of the gene expression profiling and CGH data revealedthat 53.5% (711/1,328) differentially regulated genes (�2 fold,P < 0.05) for the parental and radioresistant DU145 cells did notmap within the deleted or amplified chromosomal regions. Thissuggests that expression of these genes is mediated by geneticalterations in other genes or by epigenetic mechanisms (Supple-

mentary Table S1). Gene expression profiling of the parental andradioresistant prostate cancer cells and FACS-isolated ALDHþ

and ALDH� cell populations revealed changes in the expressionlevel of a number of genes involved in the epigenetic regulationof protein expression through DNA and histone modification(Fig. 2A). To determine the biologic processes that are activated inradioresistant cells, we performed a gene set enrichment analysis(GSEA) based on the 3553 features that were significantly upre-gulated by �1.5 fold (P < 0.05) in DU145-RR as compared withparental DU145 cells. Analysis of the existing gene sets depositedin the MSigDB (molecular signature database, Broad Institute)revealed that DU145-RR upregulated genes have a strong associ-ation with epithelial mesenchymal transition, stemness, activa-tion of WNT and other prosurvival pathways, and histone mod-ifications (Fig. 2B; Supplementary Table S2).

H3K36me3

H3K4me3

H3β-Actin

0 40 4

00.5

11.5

22.5

33.5

4

00 4 4

H3K4me3H3K36me3 ***

Rel

ativ

e pr

otei

n ex

pres

sion

leve

l (a.

u.)

DU145 P DU145 RR

Dose, Gy

DU145 P DU145 RR H3K4me3

H3K36me3

H3

GAPDH

DU145 P

LNCaP P

NI 1 3 7 14 21 28 35 42

NI 1 3 7 14 21 28 35 42

Days after 4Gy irradiation

0 Gy 4 GyH3K36me3 (i)

0

1

2

3

4

5

6

Luc

inte

nsity

, A.U

.

*

*

Input

Contro

l IgG

H3H3K

36me3

(i)

H3K36

me3 (ii

)Inp

ut

Contro

l IgG

H3H3K

36me3

(i)

H3K36

me3 (ii

)ALDH1A1Promoter

0 Gy 4 Gy

DU145 P

Fold change P

DU145 P ALDH+ vs. ALDH–

DU145 RR vs. DU145 PA

Dose, Gy

Days after 4 Gy irradiation

H3K4me3

H3K36me3

H3

GAPDH

B C D

E

F

H3K27me3

H3K27me3

1050–5

ASMTLDNMT3B

DOT1LEZH2

HDAC1HDAC11

HDAC2HDAC5HDAC6HDAC8HEMK1

LRTOMTMETTL15

METTL21AMETTL25METTL2A

METTL3NSD1

PCMTD2PNMT

SETSETD3SETD5

SETD5-AS1SETD5-AS1

SETD6SETD8SETD8SHMT1SMYD2

SUV39H2TPMT

0.0030.0000.0310.0450.0090.0080.0130.0070.0060.0060.0010.0000.0050.0200.0210.0050.0000.0120.0040.0190.0080.0030.0010.0060.0300.0060.0020.0090.0010.0010.0110.010

–4 –2 0 2 4

HDAC1

HDAC11

HDAC2

HDAC5

HDAC6

HDAC8

METTL15

MLL

MLL5

MLLT11

SUV420H2

Fold change P

0.0140.0080.0040.0070.0060.0050.0190.0180.0040.0180.020

“RR” positively correlated

“P” negatively correlated

Rank in ordered dataset Ra

nked

list m

etric

(S

igna

l to

Noise

)En

richm

ent s

core

(ES)

Enrichment plot: Hallmark Epithelial Mesenchymal Transition

P < 0.05

“RR” positively correlated

“P” negatively correlated

Rank in ordered dataset Ra

nked

list m

etric

(S

igna

l to

Noise

)En

richm

ent s

core

(ES)

Enrichment plot: Mikkelsen MEF LCP with H3K4ME3

P < 0.05

*

Rel

ativ

e in

tens

ity (a

.u.)

0123456

Figure 2.Epigenetic alterations in the irradiated prostate cancer cells. A, gene expression profiling of the DU145 parental and DU145-RR cells and FACS-isolated ALDHþ andALDH� cell populations revealed a number of genes involved in the epigenetic regulation of protein expression through DNA and histone modifications aswell asmethyltransferases involved in the variousmetabolic processes. B,GSEAenrichment analysis for thegenes upregulated inDU145-RRas comparedwithDU145cells. The "Hallmark Epithelial Mesenchymal Transition" gene set includes genes defining epithelial–mesenchymal transition, as in wound healing, fibrosis, andmetastasis. The "Mikkelsen MEF LCP with H3K4me3" gene set includes genes with low-CpG-density promoters (LCP) bearing a trimethylation mark at H3K4(H3K4me3) inMEF cells (embryonic fibroblasts). C, irradiation induces long-lasting changes in H3K4, H3K36, and H3K27 trimethylation. DU145 and LNCaP cells wereirradiatedwith 4Gyof X-ray, and the lysates of cellswere prepared at the indicated time points during 6weeks after irradiation and analyzed byWestern blotting. NI,nonirradiated cells. D, analysis of H3K4 and H3K36 methylation in response to irradiation. Densitometric analysis of the protein immunoblots is shown in theaccompanying graph. The level of H3K4 and H3K36methylationwas normalized to the total H3 level. Error bars, SEM; n¼ 3; � , P <0.05. E, the results of ChIP analysisshowed that irradiation induces a H3K36me3 transcription activation mark on ALDH1A1 promoter within two-core b-catenin/TCF binding sites (i) and asdescribed earlier (ii, ref. 16). Error bars, SEM; n ¼ 3; � , P < 0.05. F, the results of the luciferase reporter assay where luciferase gene expression is regulated by anendogenous ALDH1A1 gene promoter showed that ALDH1A1 transcription is significantly increased in DU145-RR and LNCaP-RR cells as compared with theparental cell lines. Error bars, SEM; n � 3; � , P < 0.05. P, parental cells; RR, radioresistant cells.

Peitzsch et al.





We next investigated whether irradiation induces global chro-matin modifications. Methylation of lysine residues K4 and K36of histone H3 correlates with transcriptional activation, whereastrimethylation of histone H3 on lysine 27 enforces gene silencing(21). We found that irradiation induced long-term changes inH3K4, H3K36, and H3K27 tri-methylation in prostate cancercells, which can be observed at least 6 weeks after treatment(Fig. 2C and Supplementary Fig. S3A). In accordance with thisobservation, Western blot analysis revealed long-term and time-dependent changes in the expression of the histone methyltrans-ferases involved in H3 methylation on residues K27 (EZH2) orK36 (SETD2; Supplementary Fig. S3A and S3B). In addition,Ingenuity Pathway Analysis (IPA) revealed histone methyltrans-ferase EZH2 as one of the most significantly upregulated tran-scriptional regulators in DU145-RR cells as compared with paren-tal cells (P < 10�8) (Supplementary Fig. S3C). Interestingly, H3K4and H3K36 tri-methylation increased during TGFb-induced EMTand cell reprogramming (21) andwas significantly upregulated inthe highlymigratoryDU145-RR and LNCaP-RR cells as comparedwith their nonirradiated parental counterparts, as well as inparental DU145 cells after irradiation (Fig. 2D and Supplemen-tary Fig. S3D). In contrast, radioresistant cells showed no signif-icant change in H3K4 and H3K36 methylation in response toirradiation (Fig. 2D). In line with this observation, we found thatDU145-RR and LNCaP-RR cells display an increased expression ofEMT and stem cell markers, which was altered after irradiation toa lesser extent than in parental cell lines (Fig. 1A and B andSupplementary Fig. S4A). Moreover, irradiation induced anALDHþ cell population to a lesser degree in both DU145-RRALDHþ and DU145-RR ALDH� cell populations, as comparedwith the ALDHþ and ALDH� cells isolated from the parental cells(Supplementary Fig. S4B).

To determine whether histone H3 methylation can regulateALDH1A1 gene transcription after X-ray irradiation, a chromatinimmunoprecipitation assay (ChIP) was performed. The cross-linked DNA–protein complexes from X-ray irradiated or nonir-radiated parental and DU145-RR cells were immunoprecipitatedeither with anti-H3K36me3 or with anti-H3 and control IgGantibodies. PCR amplification was performed using primersflanking two regions of the ALDH1A1 promoter containing bind-ing sites for b-catenin/TCF transcription factor and namedALDH1A1 (i) and ALDH1A1 (ii) with sizes of 508 bp and 518bp, respectively (Fig. 2E; refs. 16, 22). The results of the ChIPanalysis showed that the H3K36 methylation mark at ALDH1A1promoter increased after irradiation with 4 Gy of X-ray (Fig. 2Eand Supplementary Fig. 4C). To perform a direct analysis of thetranscriptional regulation ofALDH1A1, we used a reporter systemwhere luciferase gene expression is regulated by the endogenousALDH1A1 gene promoter. The results of the luciferase reporterassay confirmed thatALDH1A1 transcription is highly activated inthe RR prostate cancer cell lines and irradiation of parental cellsled to a significant increase ofALDH1A1 transcription as well as toupregulation of b-catenin/TCF reporter gene expression (Fig. 2Fand Supplementary Fig. S5A). These results suggest an epigeneticregulation of ALDH1A1 transcription by irradiation, which ismediated by histone H3 methylation and is consistent with flowcytometry data for the ALDH enzymatic activation in response toirradiation (Fig. 1B). Taken together, these findings suggest thatgenetic alterations and changes in the histone methylation thathave occurred during treatment contribute to the emergence ofcell radioresistance and stem cell reprogramming.

Characterization of the properties of ALDHþ and ALDH� cellpopulations in radioresistant prostate cancer cells

Our previous study demonstrated that ALDH activity is indic-ative of prostate tumor–initiating cells with increased radioresis-tance (16). Chemical inhibition of ALDH activity as well asknockdown of ALDH1A1 expression mediated by siRNA resultedin a significant decrease in ALDH activity, spherogenicity, andradioresistance (Supplementary Fig. S5B–S5D; ref. 16), suggest-ing that ALDH activity and cell radioresistance not only correlatebut are causally linked.

Taking into account the genetic, phenotypical, and functionaldifferences between the parental and radioresistant prostate can-cer cell lines, we hypothesized that irradiation can induce plas-ticity of the tumor stem cell phenotype. To test this assumption,we performed a comparative analysis of the spherogenic, tumor-igenic, and radioresistance properties of ALDHþ and ALDH� cellsubsets isolated from the parental and DU145-RR cell lines.Similarly to the DU145 ALDHþ cells, DU145-RR ALDHþ cellsshowed an increase in sphere-forming potential over DU145-RRALDH� cells (Supplementary Fig. S5E). To analyze the tumori-genic potential of DU145-RR ALDHþ and DU145-RR ALDH�

populations, the FACS-purified cells were injected subcutaneous-ly into athymic immunodeficientNMRInu/numice at varying cellnumbers (100, 1,000, and 3,000). The DU145-RR ALDHþ pop-ulation showed a significantly higher frequency of tumor-initiat-ing cells than the DU145-RR ALDH� cell population (1/961, at95% confidence interval: 1/2,061 to 1/448 cells vs. 1/4,637, at95% confidence interval: 1/10,558 to 1/2,037 cells, respectively;P < 0.01; Supplementary Fig. S5F). When the radioresistance ofthe two populations was compared, DU145-RR ALDHþ andDU145-RR ALDH� cells showed similar responses, unlike theALDHþ and ALDH� cell populations isolated from the parentalDU145 cell line. No significant differences in the radiobiologicresponse of DU145-RR ALDHþ andDU145-RR ALDH� cells wereobserved in either the 2D or the 3D survival assays (Fig. 3A andSupplementary Fig. S5G).

The phosphorylation of histone H2A.X at Ser 139 by ATM orATR protein kinases on the sites of DNA damage is a marker ofDNA double-strand break (DSB) response. Residual nuclear fociof the phosphorylatedH2A.X (g-H2A.X),measured 24hours afterirradiation, are associated with DNA lesions, which cannot berepaired and can lead to cell death (23, 24). We have recentlyreported that ALDHþ cell population among the nonirradiatedprostate cancer cells has a lower level of DSBs and more efficientDNA repair after irradiation as compared with the ALDH� cells.We found that, in contrast to the parental DU145 cells, thenumber of residual g-H2A.X foci 24 hours after irradiation wasnot increased in DU145-RR ALDHþ cells compared with DU145-RR ALDH� cells (Supplementary Fig. S5H). Various ALDH iso-forms contribute to the high ALDHactivity in prostate cancer cellsincluding ALDH1A1 and ALDH1A3, which show abundantexpression in prostate cancer tissues (25). Both of these isoformsare expressed at high levels in the radioresistant prostate cancercell lines, and in the ALDHþ cell populations (SupplementaryFig. S6A–S6C; ref. 16). Microscopy analysis of the residual g-H2A.X foci in the ALDH1A3-positive DU145 and DU145-RR cellsconfirmed that, in contrast to the parental cells, ALDH1A3 expres-sion in DU145-RR cells is not indicative of more efficient DNADSB repair (Fig. 3B). Taken together, these data suggest that, incontrast to the parental prostate cancer cells, ALDHactivity is not amarker of more radioresistant cell population in cells with






acquired radioresistance. These findings suggest that radioresis-tant prostate cancer cell populations can change their phenotypeduring the course of irradiation.

Comparative gene expression profiles of ALDHþ and ALDH�

cell populations in parental and radioresistant prostate cancercells

To better understand the biological mechanisms that mayunderlie the lack of difference in radiosensitivity betweenALDHþ and ALDH� cell populations in cells with acquired radio-resistance, we employed comparative gene expression profiling ofALDHþ and ALDH� cells isolated from the parental DU145 cellline and from its radioresistant derivative DU145-RR cell line.Whereas the differences between DU145 ALDHþ and DU145ALDH� cell populations are largely based on the deregulation ofthe WNT and cadherin signaling pathway, the distinct character-

istics of the DU145-RR ALDHþ and DU145-RR ALDH� cellsdepend on the differential regulation of DNA replication, whichmight be reflective of a high rate of DU145-RR ALDHþ cellproliferation (Fig. 3C and Supplementary Fig. S6D). Moreover,as indicated by the number of commonly regulated genes, theset of genes that is differentially expressed between DU145ALDHþ and DU145 ALDH� cells is more similar to the geneset that is differentially regulated between parental DU145 andDU145-RR cells, as compared with the genes that are differentlyexpressed between DU145-RR ALDHþ and DU145-RR ALDH�

cells (Fig. 3C).Analysis of the genes that are involved in the WNT signaling

pathway anddifferentially regulated betweenDU145ALDHþ andALDH� population and between DU145 and DU145-RR cellsrevealed that expression of these genes does not differ betweenDU145-RR ALDHþ and DU145-RR ALDH� cells except WNT9A

Sur

viva

l fra

ctio

n

Dose (Gy)A

B

C D

E

βcateninβcateninDAPI βcatenin

βcateninDAPI

Cells with nuclear β-catenin, %

30 μm

30 μm

DU145 P ALDH+ : 1, 2, 3

DU145 RR total: 4, 9, 10DU145 RR ALDH+ : 5, 11, 12DU145 RR ALDH- : 6, 7, 8

DU145 P total : 14, 15, 16

DU145 P ALDH- : 13, 17, 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Probe Name Gene NameASCL5A_32_P122240CDH1A_23_P206359CDH15A_32_P25357CDH3A_23_P49155EPCAMA_23_P91081FGFBP1A_23_P30126INO80A_24_P39454INO80A_33_P3291325LEF1A_24_P20630MYH15A_24_P11900PCDHA11A_24_P360206PCDHA11A_33_P3351904PCDHA4A_33_P3318002PCDHAC1A_24_P191931PCDHAC2A_23_P218942PCDHB16A_23_P30200PCDHB2A_23_P213375PCDHB2A_24_P251962PCDHB5A_23_P69863PCDHB6A_23_P58464PCDHGA5A_33_P3395651PCDHGA5A_33_P3395650PCDHGC4A_23_P303101PCDHGC4A_33_P3422301PRKCHA_23_P205567PRPF18A_23_P127195SERPINA1A_23_P218111SFRP1A_23_P10121SOX9A_23_P26847TCF7L2A_23_P127013WNT10AA_23_P102117WNT9AA_33_P3384058

Max

MinAverage

ALDH+Cells

ALDH-Cells

DU145 P DU145 RR

Cytoskeletal regulation by Rho GTPase

DNA replication

p53 pathway

Inflammation mediated by chemokines and cytokines

Nicotinic acetylcholine receptor signaling

RR ALDH+ vs. RR ALDH–

Axon guidance mediated by Slit/Robo

6543210Number of genes

P1.19e-7

0.100

0.052

0.060

0.066

0.018

ActualExpected

RR vs. P

ALDH+ vs. ALDH-

417

33

1265

363

13

1282

RR ALDH+ vs.RR ALDH-

RR vs. P

8 76543210

GABA-B receptor II signalingEndogenous cannabinoid signaling

Opioid proopiomelanocortin pathwayOpioid prodynorphin pathway

Enkephalin releaseHistamine H1 receptor mediated signaling pathway

Thyrotropin - releasing hormone receptor signaling pathwayMetabotropic glutamate receptor group II pathway

Alzheimer disease-presenilin pathwayAngiogenesis

Cadherin signaling pathwayWnt signaling pathway

DU145 P ALDH+ vs. ALDH–

0.0360.0020.0550.0130.0170.0320.0920.0720.0570.0600.0340.072

P

F

DU145 P DU145 RR

γ-H2A.X (Ser139) DAPIALDH1A3

γ-H2

A.X

Foci

per c

ell

DU145 P DU145 RR25 μm25 μm

ALDH1A

3+

ALDH1A

3+

ALDH1A

3-

ALDH1A

3-

01234567

*n.s.

6420

0.001

0.01

0.1

1

Sur

viva

l fra

ctio

n

Dose (Gy)6420

Dose (Gy)

DU145 ALDH RR-

DU145 ALDH RR+

0.01

0.1

1 8 6 4 20

n.s.

100%50%0%

DU145 RR ALDH+DU145 RR ALDH–

0.001

0.01

0.1

Sur

viva

l fra

ctio

n

1

LNCaP PLNCaP P + 1 μmol/L XAV939 LNCaP RR LNCaP RR + 1 μmol/L XAV939

**

DU145 PDU145 P + 1 μmol/L XAV939 DU145 RR DU145 RR + 1 μmol/L XAV939

*n.s.

Figure 3.Irradiation-induced plasticity of the mechanisms of radioresistance in prostate cancer cells. A, clonogenic cell survival assay showing that, in contrast to ALDHþ andALDH� cell populations isolated from parental DU145 cells, DU145-RR ALDHþ and DU145-RR ALDH� cells do not differ in their radioresistance properties.Error bars, SEM; n � 5. B, immunofluorescence analysis of the residual g-H2A.X foci in ALDH1A3þ and ALDH1A3� cells 24 hours after irradiation with 4 Gyof X-ray. Error bars, SEM; n¼ 5; � ,P <0.05. C, comparative gene expression profiling of ALDHþ andALDH� cell populations in the parental DU145 andDU145-RR cells.Venn diagrams show an overlap in the sets of genes differently regulated between DU145 and DU145-RR cells in comparison with the DU145 ALDHþ andDU145 ALDH� cells or to DU145-RR ALDHþ and DU145-RR ALDH� cells. PANTHER analysis was used to identify the pathways overrepresented among the genesthat are differentially regulated in DU145 ALDHþ and DU145 ALDH� cells and DU145-RR ALDHþ and DU145-RR ALDH� cells. D, hierarchical clustering analysisof the WNT signaling pathway genes differentially regulated between DU145 and DU145-RR, DU145 ALDHþ and DU145 ALDH�, as well as DU145-RR ALDHþ

and DU145-RR ALDH�. E, immunofluorescence analysis of the intracellular distribution of b-catenin in DU145 RR ALDHþ and DU145 RR ALDH� cell subsets. Thedata for DU145 ALDHþ and DU145 ALDH� are included for comparison (16). Error bars, SEM; n ¼ 2. F, the inhibition of WNT signaling pathway with XAV939inhibitor does not affect radiosensitivity of LNCaP-RR andDU145-RR cells. The data for LNCaP andDU145 cells are included for comparison (16). The cellswere serumstarved in DMEM with 3% FBS for 24 hours followed by treatment with XAV939 antagonist at concentration 1 mmol/L for 72 hours. Error bars, SEM; n � 3;� , P < 0.01. P, parental cells; RR, radioresistant cells.

Peitzsch et al.





(wingless-typeMMTV integration site family,member 9A), whichis significantly downregulated in DU145-RR ALDHþ cells ascompared with the DU145-RR ALDH� cell population (Supple-mentary Tables S3–S5). Consistent with this observation, hierar-chical clustering analysis of theWNT signaling pathway genes thatwere significantly up- or downregulated by�1.5 fold (P < 0.05) inDU145-RR as compared with DU145 cells and in ALDHþ versusALDH� cells demonstrated that DU145-RR ALDHþ and DU145-RR ALDH� cells cluster together. In contrast, DU145 ALDHþ cellsformed a cluster separated from DU145 ALDH� cells (Fig. 3D).

Activation of the canonicalWNT signaling pathway leads to thestabilization and nuclear accumulation of b-catenin, which inter-acts with T-cell factor/lymphoid enhancing factor-1 (TCF/LEF)proteins to regulate gene transcription. We found that ALDHþ

cells have a higher level of nuclear b-catenin, which is consistentwith an activation of the WNT signaling pathway in this cellpopulation (16). In contrast to the parental DU145 cells, ALDHþ

and ALDH� cell populations isolated from DU145-RR cells bothshowed a high nuclear accumulation of b-catenin and no differ-ences in its intracellular distribution (Fig. 3E). WNT signaling,which is overrepresented in both radioresistant prostate cancer

cells and ALDHþ tumor-initiating cell population is a potentinducer of EMT and cell migration (26). We next analyzed themigratory properties of ALDHþ and ALDH� cell populationsisolated from the parental and radioresistant DU145 and LNCaPcell lines. The results of the Boyden chamber based cell migrationanalysis demonstrated that in contrast to ALDHþ and ALDH� cellsubsets isolated from the radioresistant cell lines, ALDHþ cellsisolated from the parental cells lines have a higher migratorypotential as compared with ALDH� cell population (Supplemen-tary Fig. S6E). Inhibition of the WNT signaling by tankyraseinhibitor XAV939, which stimulates b-catenin degradation, ledto the inhibition of cell clonogenicity and increases in cell radio-sensitivity, with a more pronounced effect for the parental celllines LNCaP and DU145, compared with their radioresistantcounterparts (Fig. 3F and Supplementary Fig. S6F). Recent studieshave demonstrated that overexpression of LEF1might lead to cellresistance to tankyrase inhibition (27). Noteworthy, DU145-RRcells express a significantly higher level of LEF1 as compared withthe parental DU145 cells, which contribute to XAV939 resistance(Supplementary Fig. S6G). Therefore changes in the phenotype ofthe radioresistant cell population during treatment are associated

Cel

l via

bilit

y, %

Log [DZNep], mol/L

LNCaP PLNCaP RRDU145 PDU145 RR

LNCaP P LNCaP RR

Dose Dose

*

Control

A

C

B

Pla

ting

effic

ienc

y, %

LNCaP RRLNCaP PDU145 RRDU145 P DZNeP, mol/L5.99E-071.708E-062.311E-075.705E-07IC

50

24 h Treatment150

100

50

0−8 −7 −6 −5 −4 −3

150

100

50

0−9 −8 −7 −6 −5 −4

72 h Treatment

LNCaP PLNCaP RRDU145 PDU145 RR

Log [DZNep], mol/LLNCaP RRLNCaP PDU145 RRDU145 P DZNeP, mol/L

3.1699E-057.041E-052.259E-054.466E-05IC50C

ell v

iabi

lity,

%

05

101520253035404550

0

10

20

30

40

50

60

BIX 1 μmol/LBIX 2 μmol/L BIX 5 μmol/L

DU145 P DU145 RR

DU145 P DU145 RR

0

5

10

15

20

25

30

35

Control

DZNep 1 μmol/L

DZNep 2 μmol/L

DZNep 10 μmol/L

LNCaP P LNCaP RR

Pla

ting

effic

ienc

y, %

Control

DZNep 1 μmol/L

DZNep 2 μmol/L

DZNep 10 μmol/L

Pla

ting

effic

ienc

y, %

* **

**

**

0.0002

0.002

0.02

0.2

6 Gy4 Gy2 Gy0 Gy

Ctrl., n = 7

DZNep 1 μmol/L, n = 7, P = 0.021

DZNep 2 μmol/L, n = 3, P = 0.002

DZNep 5 μmol/L, n = 3, P = 0.001

DZNep 10 μmol/L, n = 2

Sur

viva

l fra

ctio

n

0.0002

0.002

0.02

0.2

6 Gy4 Gy2 Gy0 Gy

Ctrl., n = 6

DZNep 1 μmol/L, n = 6, P < 0.0005

DZNep 2 μmol/L, n = 3, P < 0.0005

DZNep 5 μmol/L, n = 3, P < 0.0005

DZNep 10 μmol/L, n = 2

*

0.0002

0.002

0.02

0.2

Dose

6 Gy4 Gy2 Gy0 Gy

Ctrl., n = 7

BIX 1 μmol/L, n = 7, P = 0.199

BIX 5 μmol/L, n = 3, P < 0.0005

DZNep 1 μmol/L, n = 4, P = 0,006

DU145 P DU145 RR

0.0002

0.002

0.02

0.2

Dose

6 Gy4 Gy2 Gy0 Gy

Ctrl., n = 7

BIX 1 μmol/L, n = 7, P = 0.972

BIX 5 μmol/L, n = 3, P = 0.063

DZNep 1 μmol/L, n = 4, P < 0.0005

Sur

viva

l fra

ctio

n

Sur

viva

l fra

ctio

n

Sur

viva

l fra

ctio

n

Figure 4.Inhibition of the histone methylation activity decreases prostate cancer cell viability and radioresistance. A, DZNep, a global histone methylation inhibitor,is more potent for the inhibition of viability of LNCaP-RR and DU145–RR cells as compared with their parental counterparts. IC50 values were determined after24 hours (n � 3) and 72 hours (n � 2) of treatment with the drugs. B, plating efficacy of the LNCaP, LNCaP-RR, DU145, and DU145-RR cells pretreated withDZNep or BIX-01294 at the different concentrations for 72 hours. Error bars, SEM; n � 3; � , P < 0.05. C, cells were pretreated with DZNep or BIX-01294 at thedifferent concentrations in DMEM with 3% FBS for 72 hours and their susceptibility to X-ray irradiation was evaluated by clonogenic cell survival assay.Error bars, SEM. � , P < 0.05. P, parental cells; RR, radioresistant cells.






with deregulation of multiple intracellular signaling pathways,which can be attributed to the epigenetic and genetic alterationsacquired by cells after irradiation.

Alteration of histone methylation affects cell radioresistanceand tumorigenicity

Next, we determined whether epigenetic regulation of geneexpression controlled byhistone lysinemethylation plays a role inthe regulation of prostate cancer cell radioresistance. Our currentstudies suggest that irradiation induces activation of a number ofproteins involved in the regulation of epigenetic modification ofDNA and histones. As shown above, IPA-based interrogation ofthe gene expression data revealed histone H3 lysine methyltrans-ferase EZH2 as one of the most significantly upregulated tran-scriptional regulators in DU145-RR cells, compared with theirparental counterparts (Supplementary Fig. S3C). To determinewhether modulation of histone H3 methylation contributes toprostate cancer cell radioresistance, we examined the effect ofinhibitors of lysine methyltransferases EZH2 and G9a, DZNep,and BIX01294, respectively, on the regulation of the viability andradiosensitivity of the parental and radioresistant DU145 cells.Noteworthy, DZNep is a broad-spectrum methyltransferaseinhibitor that affects both inhibitory and activating histonemethylation marks, thus allowing a more comprehensive epige-netic resetting. Treatment with DZNep was significantly morepotent for the inhibition of the viability of radioresistant LNCaPas comparedwith parental LNCaP cells and radioresistant DU145cells as comparedwith parental DU145 cells (Fig. 4A). In contrast,BIX01294 showed no significantly different effect in the radio-resistant and parental LNCaP and DU145 cells (SupplementaryFig. S7A). Moreover, treatment of prostate cancer cells withDZNep decreased clonogenicity and radiosensitivity of the paren-tal and radioresistant LNCaP and DU145 cells, with more pro-nounced effects on the radioresistant cell lines (Fig. 4B and C andSupplementary Fig. S7B). In contrast toDZNep, BIX01294had nosignificant inhibitory effects on the cell clonogenicity and radio-resistance of the radioresistant and parental cell lines at the samerange of concentrations (Fig. 4B and C and Supplementary Fig.S7C). Analysis of the residual g-H2A.X foci, 24 hours after irra-diation, showed that pretreatment of cells with DZNep signifi-cantly increased the number of unrepaired DSBs compared withthe untreated cells. This effect was greater in the radioresistant celllines than in the parental cells. Noteworthy, treatment of the cellswith DZNep alone at a concentration of 1 mmol/L for 72 hoursalso led to significant accumulation of DNADSBs as indicated byg-H2A.X analysis (Fig. 5A).

To understand the molecular mechanism of the more pro-nounced inhibitory effect ofDZNep treatment onDU145-RR cellsin comparison with their parental counterparts, DZNep wasapplied alone or in combination with X-irradiation, and the cellswere analyzed by Western blotting for the activation of theapoptotic markers. Interestingly, DU145-RR cells had a higherbasal expression level of PARP, which is involved in DNA repair,genomic instability, and transcription regulation, including pre-vention of the H3K4me3 demethylation (28). Recent evidencepoints to the potential role of PARP in prostate cancer progressionand therapy resistance, inferring that targeting of PARP could serveas important therapeuticmodality inhormone-refractory prostatecancer (28). In contrast to the parental DU145 cells, treatment ofDU145-RR cells with DZNep induced caspase cascade and cleav-age of PARP at Asp214, which mediates the apoptotic program

(Supplementary Fig. S8A and S8B). Induction of the PARP cleav-age in the DU145-RR and LNCaP-RR cells after treatment withDZNep was confirmed by fluorescent microscopy (Supplemen-tary Fig. S8C).

An accumulation of DNA DSBs and the sensitivity of the cellsto apoptosis can be attributed to disruption of intracellularredox balance. To determine whether treatment with DZNepdoes disrupt redox balance in prostate cancer cells, GSH andROS levels were assessed by imaging cytometry or flow cyto-metry after incubation with mBCI or 20,70–dichlorofluorescindiacetate, respectively. Treatment with DZNep resulted in adecrease in GSH levels, and upregulation of ROS (Fig. 5B andC). The GSH level in DU145-RR cells was significantly moreaffected by the DZNep treatment as compared with the parentalDU145 cells (Fig. 5C). These results suggest that protective,ROS-scavenging mechanisms of the radioresistant cells mightrely on epigenetic regulation of genes involved in DNA repairand apoptosis.

Next we analyzed whether DZNep treatment can induce DNADSBs and apoptotic signaling in prostate tumor specimens andprimary prostate tumor cells. A freshly isolated lymph nodebiopsy of human hormone refractory prostate cancer (PT1) anda radical prostatectomy specimen (PT2)were dissected into piecesof approximately 3� 3 mm, treated with 2 mmol/L of DZNep for24 hours, then irradiated with 4 Gy of X-ray and fixed 24 hoursafter irradiation. Tissue culture was performed in serum-free,sphere-forming cell medium. The presence of tumor cells in thetissue pieces was confirmed by staining with hematoxylin andeosin and by fluorescent immunostaining with anti-prostate–specific membrane antigen (PSMA) and CK5/14 antibody (Fig.6A and Supplementary Fig. S9A). Microscopy analysis of thenuclear intensity of g-H2A.X and cleaved PARP 24 hours afterirradiation confirmed that inhibition of H3 methylation byDZNep led to an increase in DNA DSBs and induced activationof tumor cell apoptosis after ex vivo irradiation (Fig. 6A).Primary cell cultures 312/13 and 311/13 were established fromtargeted needle biopsies of Gleason grade 7 tumors from theprostates of two patients who underwent radical prostatectomyat Castle Hill Hospital, Cottingham, United Kingdom (kindlyprovided by Prof. Norman Maitland, University of York). Forthese experimental models, only combination of DZNep treat-ment with 4 Gy irradiation led to significant increase in thenuclear g-H2A.X intensity 24 hours after irradiation comparedwith untreated control cells, whereas no significant differencewas seen between control cells and irradiated cells withoutDZNep treatment (Supplementary Fig. S9B).

In comparison, treatment of the immortalized, nontumori-genic cell line RWPE1 with DZNep did not lead to a significantincrease in residual DNA damage as compared with the treatmentby irradiation alone, suggesting that radiosensitizing effect ofDZNep can be more pronounced in malignant cells as comparedwith normal cells (Supplementary Fig. S9C).

Next we addressed the question of whether DZNep inhibitiondirectly affects expression of stem cell markers such as ALDH1A1.Western blot analysis revealed that treatment of prostate cancercells withDZNep led to an inhibition of EZH2 andALDH1A1 andto a decrease in tri-methylation of lysine residue K36 in histoneH3 (Fig. 6B and Supplementary Fig. S9D). Inhibition ofALDH1A1expressionwas alsoobserved inDU145 cells pretreatedwithDZNep and subsequently irradiatedwith 4GyofX-ray. Thesefindings suggest that this inhibitor can be used to prevent tumor

Peitzsch et al.





cell reprogramming induced by irradiation (Supplementary Fig.S9E and S9F).

In accordance with the results of gene expression analysis andour previous findings (16), siRNA-mediated inhibition of b-cate-nin expression decreases expression of EZH2 and ALDH1A1 inprostate cancer cells, and DZNep treatment has an additive effecton b-catenin knockdown with regard to EZH2 and ALDH1A1expression (Supplementary Fig. S10A). These results suggest a linkbetween the activation of WNT/b-catenin signaling pathway aftercell irradiation and irradiation-induced epigenetic changes. Strik-ingly,ALDH1A1 and EZH2 genes are frequently co-expressedwithCTNNB1 (b-catenin) in prostate tumor specimens (Supplemen-tary Fig. S10B). Moreover, a gene signature, which includesexpression of ALDH1A1 and EZH2, correlated with reduceddisease-free survival in patients with prostate cancer, as suggestedby the analysis of the provisional The Cancer Genome Atlasdataset extracted from cBioPortal for Cancer Genomics (Supple-mentary Fig. S10C).

To analyze whether inhibition of the H3 methylation byDZNep can affect cell tumorigenicity after irradiation, we com-pared the relative in vivo growth of DU145 and DU145-RR cellsafter in vitro pretreatment with DZNep alone or in combination

with irradiation. The cells were treated with 1 mmol/L of DZNepor with DMSO as a control for 72 hours, irradiated with 6 Gy ofX-ray or left nonirradiated, embedded in Matrigel, and injectedsubcutaneously into NMRI nu/nu mice. Animals alive andtumor-free at the end of the observation period (77 days) weredefined as long-term survivors. In contrast to the parental cells,DU145-RR cells pretreated with DZNep had a significantlylower relative tumor growth rate and tumor take than controlcells, and the inhibitory effect of the DZNep treatment onDU145-RR cell tumorigenicity was significantly increased afterirradiation, leading to complete inhibition of DU145-RR celltumorigenicity (Fig. 6C and D).

DiscussionRadiotherapy is one of the most effective treatments for cancer

and is a potentially curative treatment option for patients withclinically localized prostate carcinoma (29). Recent discoverieshave provided compelling evidence that CSC population withineach individual tumor is a key contributor of radiotherapy failure,regardless of whether this population is transient or stable (30).Therefore, tumors can only be cured if all CSCs are targeted, or

A

DU145, 4 Gy

LNCaP, 4 Gy

P

RR

γH2A.X (Ser139) DAPI

10 μm

10 μm

P

RR

0

1

2

3

4

5

6

7

0

1

2

3

4

5

6

7

8

9DU145 P

DU145RR

LNCaP P

LNCaPRR

DU145 P

DU145RR

LNCaP P

LNCaPRR

Num

ber o

f γH

2A.X

(Ser

139)

per

nuc

lei

Num

ber o

f γH

2A.X

(Ser

139)

per

nuc

lei

DZNep

, 1 μm

ol/L

Contro

l

DZNep

, 1 μm

ol/L

Contro

l

DZNep

, 1 μm

ol/L

Contro

l

DZNep

, 1 μm

ol/L

Contro

l

DZNep

, 1 μm

ol/L

Contro

l

DZNep

, 1 μm

ol/L

Contro

l

DZNep

, 1 μm

ol/L

Contro

l

DZNep

, 1 μm

ol/L

Contro

l

4 Gy 4 Gy

*P = 0.035

*P = 0.029

P = 0.2

*P = 0.016

*P = 0.045

P = 0.002 **

P = 0.009

*P = 0.033

DU145

P

RR

10 μm

DZNep, 1 μmol/L

DZNep, 1 μmol/L

Control DZNep, 1 μmol/L

DZNep, 1 μmol/L

Control

Control Control

LNCaP

P

RR

10 μm C

B

0

0.2

0.4

0.6

0.8

1

1.2

DU145 P

DU145 R

RRat

io o

f mea

n gr

ay v

alue

, ± D

ZNep

DU145RR

DU145 P

DZNep, 1 μmol/LControl

GSH Level

*

**

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

4,500

5,000

DU145-RRDU145 PRel

ativ

e R

OS

leve

l (D

CFD

A, M

FI)

ControlDZNep

*

DCFDA

Cou

nt

DZNep, 1 μmol/LControl cellsUnstained cells

DU145 P cells

0

0102 103 104 105

Figure 5.Mechanisms of DZNep-mediated prostate cancer cell radiosensitization. A, analysis of gH2A.X foci in cells treated with DZNep at a concentration of 1 mmol/L for72 hours alone or in combination with 4 Gy of X-ray irradiation 48 hours after start of DZNep treatment. Residual gH2A.X foci were analyzed 24 hours afterirradiation. Error bars, SEM; n¼ 3; � , P <0.05; �� , P <0.01. B, analysis of ROS level after treatment of cells with DZNep at a concentration of 1 mmol/L for 72 hours. Cellstreated with DMSO were used as control. Error bars, SEM; n ¼ 3; � , P < 0.05. C, measurement of the GSH levels in the cells pretreated with DMSO or DZNepat a concentration of 1 mmol/L for 72 hours by incubation with monochlorobimane (mBCI) and Celigo cytometry. Error bars, SEM; n¼ 8; � , P < 0.05. P, parental cells;RR, radioresistant cells.






when the host is able to inactivate those CSC surviving anticancertreatment (7). However, development of therapeutic agents erad-icating CSCs remains challenging due to the high diversity andplasticity of CSC phenotypes. Recent advances in DNA sequenc-ing allowing global analysis of genetic landscape in individualtumor cells revealed a high heterogeneity of different subcloneswithin the same tumor specimens and has provided an insightinto the subclonal evolution, suggesting that CSCs and clonalevolution models are not mutually exclusive (15, 31). Further-more, evolved CSCs that survive after chemo- or radiotherapymight confer treatment resistance. By using CGH analysis, werevealed multiple genomic alterations and pronounced changesin the global gene expression profiling in prostate cancer cells thathave been irradiated with multiple doses of X-rays. Therefore, wehypothesized that irradiation-induced genomic instability mightlead to the evolution of the tumor-initiating populations thatchange their phenotypic features during the course of irradiation.

In our previous study, we found that prostate tumor–initi-ating cells with high aldehyde dehydrogenase activity (ALDHþ

cells) were more radioresistant than ALDH� cells, and inhibi-tion of ALDH activity radiosensitizes prostate cancer cells (16).

In this study, we addressed the question whether tumor-ini-tiating cells with acquired radioresistance can also be definedby a high ALDH activity. Surprisingly, although RR ALDHþ

population possesses higher clonogenic and tumorigenicpotential compared with RR ALDH� cells, these two popula-tions cannot be distinguished by either their radioresistanceproperties or by the activation of WNT/b-catenin signalingpathway and migratory potential. These results indicate thatthe radioresistant tumor-initiating cell populations undergo aphenotypic switching during the course of radiotherapy byanalogy with a lineage switch in leukemia patients after che-motherapy (32–34), and functional changes in ovarian andprostate CSCs following paclitaxel or docetaxel treatment (20,35). Furthermore, recent data suggest that ALDH1A1 expres-sion is predictive of poor prognosis in hormone-na€�ve prostatecancer patients. In contrast, in patients with castration-resis-tant disease, ALDH1A1 expression does not correlate withclinical outcome (25).

The previously published studies suggested that phenotypicplasticity of tumor-initiating cells might be associated not onlywith certain genetic but alsowith epigenetic changes,which canbe

C

AD

ZNep

, 2 μ

mol

/L

+ 4

Gy

4 G

y

0 7 14 21 28 35 42 49 56 63 70 77

Tum

or s

ize

(mm

3 )

Days after injection

n = 9

n = 7 n = 9

n = 8

n = 9n = 10

n = 9

n = 8

P = 0.014

Days

Per

cent

tum

or-fr

ee m

ice

B DU145 P LNCaP P LNCaP RR

GAPDH

ALDH1A1

EZH2

Contro

l

Contro

l

Contro

l

H3K36me3

H3

Contro

l

DU145 RR

BIX01

294,

1 μmol/

L

BIX01

294,

1 μmol/

L

BIX01

294,

1 μmol/

L

BIX01

294,

1 μmol/

L

DZNep, 1

μmol/

L

DZNep, 1

μmol/

L

DZNep, 1

μmol/

L

DZNep, 1

μmol/

L

Contro

lBIX

0129

4, 1 μ

mol/L

DZNep, 1

μmol/

L

Contro

lBIX

0129

4, 1 μ

mol/L

DZNep, 1

μmol/

L

Contro

lBIX

0129

4, 1 μ

mol/L

DZNep, 1

μmol/

L

Contro

lBIX

0129

4, 1 μ

mol/L

DZNep, 1

μmol/

L

D

DAPI γH2A.X PARP1 Merged H&E

Rela

tive

mea

n gr

ay v

alue

, a.u

.

25 μm

25 μm

gH2A.X PARP cleaved

PT2

PT1

4 Gy4 Gy,4 GyDZNep

2 μmol/L

4 Gy,DZNep

2 μmol/L

4 Gy,DZNep

2 μmol/L4 Gy,

DZNep2 μmol/L

4 Gy,DZNep

2 μmol/L

4 Gy,DZNep

2 μmol/L

PT1

PT1

DU145 P

DU145 RR

DU145 P, IR 6 Gy

DU145 RR, IR 6 Gy

100

50

00 20 40 60

DU145 P, DZNep

DU145 RR, DZNep

DU145 P, DZNep + IR, 6 Gy

DU145 RR, DZNep + IR, 6 GyP

< 0.0001DU145 P vs. DU145 RR< 0.0001DU145 P vs. DU145 P, DZNep< 0.0001DU145 P vs. DU145 P, IR 6 Gy< 0.0001DU145 P vs. DU145 P, DZNep + IR 6 Gy< 0.0001DU145 RR vs. DU145 RR , DZNep< 0.0001DU145 RR vs. DU145 RR, IR 6 Gy< 0.0001DU145 RR vs. DU145 RR , DZNep + IR 6 Gy< 0.0001DU145 P, IR 6Gy vs. DU145 RR, IR 6 Gy

DU145 P, DZNep + IR 6Gy vs.DU145 RR, DZNep + IR 6 Gy 0.0080.037DU145 P, IR 6 Gy vs. DU145 RR, IR 6 Gy

DU145 P DU145 RR

DU145 P, IR 6 Gy

DU145 RR, IR 6 Gy

DU145 P, DZNepDU145 RR, DZNep

DU145 P, DZNep + IR, 6 Gy

DU145 RR, DZNep + IR, 6 Gy

0

100

200

300

400

500

600

700

800

DAPI PSMA

Cleaved

100 μm

100 μm25 μm

25 μm

0

1

2

3

4

5

6

4 Gy

PT3

0

0.5

1

1.5

2

2.5

3

3.5

4

4 Gy4 Gy 4 Gy

Rela

tive

mea

n gr

ay v

alue

, a.u

.

PT1

PT3PT2EZH2ALDH1A1H3K36me3

00.20.40.60.8

11.21.41.61.8

Rel

ativ

e pr

otei

nex

pres

sion

leve

l, A

.U.

Figure 6.Regulation of histone methylation impacts prostate cancer cell tumorigenicity. A, analysis of apoptosis induction and gH2A.X increase in the specimens ofnode metastasis of prostate tumor (PT1) and radical prostatectomy (PT2 and PT3), which were treated with 4 Gy of X-ray alone or in combination with 2 mmol/L ofDZNep given for 48 hours. The tissueswere fixed 24 hours after irradiation. Error bars, SEM. H&E, hematoxylin and eosin. B,Western blot analysis of the parental andradioresistant LNCaP and DU145 cell lines treated with DZNep or BIX-01294 at a concentration of 1 mmol/L for 72 hours. Densitometric analysis of theprotein immunoblots is shown in the accompanying graph. Error bars, SEM. C, tumor uptake rates in mice injected with 5 � 104 DU145 or DU145-RR cells. The cellswere pretreated with DZNep before injection at a concentration 1 mmol/L for 72 hours alone or in combination with 6 Gy of X-ray irradiation given 48 hoursafter the start of DZNep treatment. Cells treatedwithDMSOand sham-irradiationwere used as a control; theP valuewas calculated using log-rank (Mantel–Cox) test.D, relative tumor growth of DU145 and DU145-RR cells treated as indicated. Error bars, SEM; � , P < 0.05. P, parental cells; RR, radioresistant cells.

Peitzsch et al.





essential for cell tumorigenicity and therapy resistance (15, 36).Indeed, accumulating experimental evidence suggests that tumorcell dedifferentiation, based on the induced reprogramming ofcancer cells is analogous to the reprogramming of the differen-tiated somatic cells into pluripotent cells (37). Therefore, efficientanticancer treatment must eradicate both CSCs and the tumorbulk to prevent tumor cell dedifferentiation. The populations oftumor cells with salient features of CSCs can be induced in vivo inresponse to the different microenvironmental stimuli, whichinclude hypoxia, genetic alterations that trigger EMT, as well ascancer therapies (26, 37, 38). In agreement with this, recentlypublished data showed that irradiation may induce reprogram-ming of differentiated breast and hepatocellular carcinomacells (39–41). Consistent with these findings, we have previ-ously shown that fractions of the cells positive for CSC markers,for example, ABCG2, CD133, ALDH activity were significantlyenriched after fractionated irradiation that can be potentiallyattributed to cell selection as well as to cell induction mechan-isms including both, accelerated proliferation and reprogram-ming (16, 39, 40).

Notably, our data revealed that a single 4Gy irradiation inducesa prolonged effect on the clonogenic potential and expression ofboth CSC and EMT markers, which lasts for more than 6 weeks.Thedynamics of protein expression after cell irradiation can reflecta continuous process, resulting in consequent changes of cell

phenotype and functional properties, and resembling thereprogramming process of somatic cells. Indeed we have shownthat irradiation induces methylation of H3K4 and H3K36,which is a hallmark of the reactivated transcription of theepigenetically silenced target genes. In support of the hypoth-esis that irradiation can induce prostate tumor cell populationswith CSC features, our study shows that irradiation of ALDH�

prostate carcinoma cells leads to an induction of an ALDHþ cellsubset within this population. This mechanism can potentiallycontribute to the enrichment of the cells with CSC phenotypeafter irradiation.

Heritable epigenetic regulation of gene expression, whichincludes changes in DNA methylation and histone posttransla-tional modification, plays an important role in the regulation oftumor response to the treatment. Histone modification patternshave been correlated to tumor recurrence in patients with prostatecancer (42). In support of the hypothesis that irradiation-inducedepigenetic changes can regulate tumor response to therapy, wefound that expressionof theALDH1A1 gene,which is indicative ofprostate cancer cells with increased tumorigenicity and radio-resistance, is regulated by the irradiation-induced epigeneticchanges. Our previous studies demonstrated that ALDH1A1 geneexpression is directly activated by the b-catenin/TCF transcrip-tional complex (16). Using a ChIP assay specific for the b-catenin/TCF-binding sites of the ALDH1A1 gene promoter, we revealedthat irradiation induced tri-methylation of Lys 36 in histone H3(H3K36me3) on the ALDH1A1 promoter. This modification hasbeen associated with activation of ALDH1A1 gene transcription.Our studies suggest that irradiation induced expression of thedifferent genes involved in the regulation of epigenetic modifica-tions including expression of the lysine methyltransferase EZH2.Overexpression of EZH2 has been associated with a numberof cancers, including prostate, breast cancer, and melanoma(43–46). Recent studies revealed that EZH2 enhances WNT/b-catenin–mediated gene transactivation (47). Moreover, expres-sion of EZH2 was inhibited by siRNA-mediated knockdown ofb-catenin, suggesting a direct link between irradiation-inducedactivation of the WNT/b-catenin signaling pathway and epige-netic reprogramming of the tumor cells. In accordance with thisobservation, expression of ALDH1A1 and EZH2 genes has atendency toward co-occurrence with expression of the CTNNB1(b-catenin) gene in prostate tumor specimens (48). The chemicalinhibition of EZH2 with DZNep, which acts as a broad methyl-transferase inhibitor resulted in the downregulation of ALDH1A1expression and led to a significant increase in cell radiosensitivityand inhibition of the tumorigenic properties. A compound suchas DZNep decreases H3K27me3 and H3K9me2 repressive meth-ylation mediated by EZH2 as well as the active histone markH3K4me3 and H3K36me3, therefore preventing a broad epige-netic resetting of the tumor cells (49). Noteworthy, all-transretinoid acid (ATRA), which is the only chemical drug in clinicaluse for tumor cell differentiation and is an inhibitor of ALDHactivity, was recently shown to reduce the proliferation of prostatecancer DU145 cells through downregulation of EZH2-mediatedmethylation of the HOXB13 gene, which encodes Hoxb13 anti-proliferative transcription factor (50). Several studies have dem-onstrated that the level of EZH2 expression is positively correlatedwith prostate cancer aggressiveness, and EZH2 is not only over-expressed in metastatic prostate cancer but also in the localizedcancers with a high risk of recurrence after radical prostatectomy(22, 46, 51, 52). A high EZH2 expression in esophageal squamous

ALDH

CSC

RR

ALDHCSC

RR

X

Cancer stem cell populationRadioresistant cancer stem cell populationThe course of irradiation

KK

KK

H3H4H2B H2A

H3H4H2B H2A

H3H4H2B H2A

KK

Met3Met3

Met3Met3

Met3Met3

Epigenetic changes:Cancer stem cell plasticity

Cell reprogramming Emergence of radioresistance

DZNep

Genetic changes

WNT/β-cateninsignaling

Impaired DNA repairApoptosisRadiosensitization

Figure 7.Proposed model of regulation of prostate cancer cell radioresistance byepigenetic mechanisms. Irradiation-mediated activation of the WNT/b-catenin signaling pathway leads to cell reprogramming and CSC plasticity.Inhibition of the global histonemethylation byDZNep results in impairedDNAreparation, induction of apoptosis, and increase of cell radiosensitivity.






cell carcinoma and rectal cancer was significantly correlated withthe lack of clinical complete response to chemoradiotherapy (53,54). Another study has demonstrated that EZH2 protects gliomastem cells from irradiation-induced cell death (55). Takentogether, these findings demonstrate that EZH2 bears signifi-cant potential to serve as a diagnostic and prognostic marker forprostate cancer and other cancers and as a therapeutic target fortumor radiosensitization. In support of this, recent clinicalstudies of the various epigenetic inhibitors including EZH2inhibitors EPZ-6438 and CPI-1205 for the treatment of patientswith advanced solid tumors or lymphomas (ClinicalTrials.govidentifier NCT01897571 and NCT02395601), as well as theinhibitors of histone deacetylases, which are already approvedfor clinical use or currently being clinically investigated, arepaving the way to improved effectiveness of cancer treatment bythe inhibition of tumor cell reprogramming.

This study is, to our best knowledge, the first investigationthat demonstrates that radioresistant populations within pros-tate cancer cells undergo a phenotypic switch during the courseof irradiation, rather than being selected from the preexistingcell pool. We revealed that irradiation drives histone modifica-tions on the promoter sequence of ALDH1A1 genes, whichregulate cancer cell tumorigenicity and radioresistance. Ourfindings demonstrated that treatment of tumor cells withDZNep resulted in a significant increase in DNA damage, cellapoptosis, and radiosensitivity, and cells with a higher radio-resistance are more sensitive to DZNep treatment.

Future research is needed to validate our findings by usingadditional experiments on primary cultures and tumor biopsiesand by combination of the drug with radiotherapy in patient-derivedprostate cancer xenograftmousemodels.Our data suggestthat compounds such as DZNep that inhibit EZH2 and prevents abroad resetting of histone methylation marks may be useful asepigenetic cotherapy to prevent tumor cell reprogramming intomore aggressive and therapy-resistant state (Fig. 7).

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

Authors' ContributionsConception and design: C. Peitzsch, M. Cojoc, M.P. Wirth, E.A. Stakhovsky,M. Baumann, A. DubrovskaDevelopment of methodology: C. Peitzsch, M. Cojoc, L. Hein, I. Kurth,K. M€abert, M. Krause, F.M. Frame, A. DubrovskaAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): C. Peitzsch, M. Cojoc, L. Hein, I. Kurth, K. M€abert,F. Trautmann, B. Klink, E.A. Stakhovsky, G.D. Telegeev, V. Novotny, M.I. Toma,M. Muders, G.B. Baretton, A. DubrovskaAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): C. Peitzsch, M. Cojoc, L. Hein, K. M€abert,F. Trautmann, B. Klink, E. Schr€ock, M.P. Wirth, M.I. Toma, M. Muders,A. DubrovskaWriting, review, and/or revision of the manuscript: C. Peitzsch, M. Cojoc,L. Hein, I. Kurth, K. M€abert, B. Klink, M.P. Wirth, M. Krause, G.B. Baretton,F.M. Frame, N.J. Maitland, M. Baumann, A. DubrovskaAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): C. Peitzsch, M.P. Wirth, G.D. Telegeev,V. Novotny, M.I. Toma, G.B. Baretton, N.J. Maitland, A. DubrovskaStudy supervision: A. Dubrovska

AcknowledgmentsThe authors thank Vasyl Lukiyanchuk for assistance with the analysis of the

gene expression data and Celigo imaging, Dr. Annette Linge for help with thepathologic examination of the tumor tissues, and Dr. Steffen L€ock for help withstatistical analysis.

Grant SupportThis work was supported by BMBF grants (ZIK OncoRay; A. Dubrovska,

M. Baumann, and M. Krause) and by the German Cancer Consortium (DKTK)Partnersite Dresden.

The costs of publication of this articlewere defrayed inpart by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received August 2, 2015; revised February 5, 2016; accepted February 22,2016; published OnlineFirst March 16, 2016.

References1. Ferlay J, Steliarova-Foucher E, Lortet-Tieulent J, Rosso S, Coebergh JW,

Comber H, et al. Cancer incidence and mortality patterns in Europe:estimates for 40 countries in 2012. Eur J Cancer 2013;49:1374–403.

2. Ghotra VP, Geldof AA, Danen EH. Targeted radiosensitization in prostatecancer. Curr Pharm Des 2013;19:2819–28.

3. Rukstalis DB. Treatment options after failure of radiation therapy-a review.Rev Urol 2002;4Suppl 2:S12–7.

4. Zietman AL, Bae K, Slater JD, Shipley WU, Efstathiou JA, Coen JJ, et al.Randomized trial comparing conventional-dosewith high-dose conformalradiation therapy in early-stage adenocarcinoma of the prostate: long-termresults from proton radiation oncology group/american college of radiol-ogy 95-09. J Clin Oncol 2010;28:1106–11.

5. PahlajaniN, RuthKJ, BuyyounouskiMK,ChenDY,Horwitz EM,HanksGE,et al. Radiotherapy doses of 80 Gy and higher are associated with lowermortality in men with Gleason score 8 to 10 prostate cancer. Int J RadiatOncol Biol Phys 2012;82:1949–56.

6. Johansson S, Astrom L, Sandin F, Isacsson U, Montelius A, Turesson I.Hypofractionated proton boost combined with external beam radiother-apy for treatment of localized prostate cancer. Prostate Cancer 2012;2012:654861.

7. Baumann M, Krause M, Hill R. Exploring the role of cancer stem cells inradioresistance. Nat Rev Cancer 2008;8:545–54.

8. Brunner TB, Kunz-Schughart LA, Grosse-Gehling P, Baumann M. Cancerstem cells as a predictive factor in radiotherapy. Semin Radiat Oncol2012;22:151–74.

9. Pajonk F, Vlashi E, McBride WH. Radiation resistance of cancerstem cells: the 4 R's of radiobiology revisited. Stem Cells 2010;28:639–48.

10. O'Brien CA, Kreso A, Jamieson CH. Cancer stem cells and self-renewal.Clin Cancer Res 2010;16:3113–20.

11. Baumann M, Dubois W, Suit HD. Response of human squamous cellcarcinoma xenografts of different sizes to irradiation: relationship ofclonogenic cells, cellular radiation sensitivity in vivo, and tumor rescuingunits. Radiat Res 1990;123:325–30.

12. Hill RP,Milas L. The proportion of stem cells inmurine tumors. Int J RadiatOncol Biol Phys 1989;16:513–8.

13. Butof R, Dubrovska A, Baumann M. Clinical perspectives of cancerstem cell research in radiation oncology. Radiother Oncol 2013;108:388–96.

14. Linge A, L€ock S, Gudziol V, Nowak A, Lohaus F, von Neubeck C, et al.Low cancer stem cell marker expression and low hypoxia identify goodprognosis subgroups in HPV(-) HNSCC after postoperative radioche-motherapy: a multicenter study of the DKTK-ROG. Clin Cancer Res2016 Jan 11. doi: 10.1158/1078-0432.CCR-15-1990.

15. Kreso A, Dick JE. Evolution of the cancer stem cell model. Cell Stem Cell2014;14:275–91.

16. Cojoc M, Peitzsch C, Kurth I, Trautmann F, Kunz-Schughart LA, TelegeevGD, et al. Aldehyde dehydrogenase is regulated by Beta-catenin/TCF andpromotes radioresistance in prostate cancer progenitor cells. Cancer Res2015;75:1482–94.

Peitzsch et al.





17. Frame FM, Pellacani D, Collins AT, Simms MS, Mann VM, Jones GD, et al.HDAC inhibitor confers radiosensitivity to prostate stem-like cells. Br JCancer 2013;109:3023–33.

18. Franken NA, Rodermond HM, Stap J, Haveman J, van Bree C. Clonogenicassay of cells in vitro. Nat Protoc 2006;1:2315–9.

19. van den Hoogen C, van der Horst G, Cheung H, Buijs JT, Lippitt JM,Guzm�an-Ramírez N, et al. High aldehyde dehydrogenase activity identifiestumor-initiating and metastasis-initiating cells in human prostate cancer.Cancer Res 2010;70:5163–73.

20. Dubrovska A, Elliott J, Salamone RJ, Kim S, Aimone LJ, Walker JR, et al.Combination therapy targeting both tumor-initiating and differenti-ated cell populations in prostate carcinoma. Clin Cancer Res 2010;16:5692–702.

21. McDonald OG, Wu H, Timp W, Doi A, Feinberg AP. Genome-scaleepigenetic reprogramming during epithelial-to-mesenchymal transition.Nat Struct Mol Biol 2011;18:867–74.

22. Berezovska OP, Glinskii AB, Yang Z, Li XM, Hoffman RM, Glinsky GV.Essential role for activation of the Polycomb group (PcG) protein chro-matin silencing pathway in metastatic prostate cancer. Cell Cycle 2006;5:1886–901.

23. Sharma A, Singh K, Almasan A. Histone H2AX phosphorylation: a markerfor DNA damage. Methods Mol Biol 2012;920:613–26.

24. Menegakis A, von Neubeck C, Yaromina A, Thames H, Hering S, Hennen-lotter J, et al. gammaH2AX assay in ex vivo irradiated tumour specimens: anovelmethod to determine tumour radiation sensitivity in patient-derivedmaterial. Radiother Oncol 2015;116:473–9.

25. LeMagnen C, Bubendorf L, Rentsch CA,Mengus C, Gsponer J, Zellweger T,et al. Characterization and clinical relevance of ALDHbright populations inprostate cancer. Clin Cancer Res 2013;19:5361–71.

26. Nieto MA. Epithelial plasticity: a common theme in embryonic and cancercells. Science 2013;342:1234850.

27. de la Roche M, Ibrahim AE, Mieszczanek J, Bienz M. LEF1 and B9L shieldbeta-catenin from inactivation byAxin, desensitizing colorectal cancer cellsto tankyrase inhibitors. Cancer Res 2014;74:1495–505.

28. DeshmukhD,Qiu Y. Role of PARP-1 in prostate cancer. Am JClin ExpUrol2015;3:1–12.

29. Baumann M, Krause M, Overgaard J, Debus J, Bentzen SM, Daartz J, et al.Radiation oncology in the era of precision medicine. Nat Rev Cancer2016;16:234–49.

30. Kuhlmann JD,Hein L, Kurth I,Wimberger P,DubrovskaA. Targeting cancerstem cells: promises and challenges. Anticancer Agents Med Chem 2015;16:38–58.

31. Peitzsch C, Kurth I, Kunz-Schughart L, Baumann M, Dubrovska A. Dis-covery of the cancer stem cell related determinants of radioresistance.Radiother Oncol 2013;108:378–87.

32. Mullighan CG, Phillips LA, Su X, Ma J, Miller CB, Shurtleff SA, et al.Genomic analysis of the clonal origins of relapsed acute lymphoblasticleukemia. Science 2008;322:1377–80.

33. Ding L, Ley TJ, Larson DE, Miller CA, Koboldt DC, Welch JS, et al. Clonalevolution in relapsed acutemyeloid leukaemia revealed by whole-genomesequencing. Nature 2012;481:506–10.

34. Stass S,Mirro J,Melvin S, PuiCH,Murphy SB,WilliamsD. Lineage switch inacute leukemia. Blood 1984;64:701–6.

35. Craveiro V, Yang-Hartwich Y, Holmberg JC, Sumi NJ, Pizzonia J, Griffin B,et al. Phenotypic modifications in ovarian cancer stem cells followingPaclitaxel treatment. Cancer Med 2013;2:751–62.

36. MeachamCE,Morrison SJ. Tumour heterogeneity and cancer cell plasticity.Nature 2013;501:328–37.

37. Tang DG. Understanding cancer stem cell heterogeneity and plasticity.Cell Res 2012;22:457–72.

38. Peitzsch C, Perrin R, Hill RP, Dubrovska A, Kurth I. Hypoxia as a biomarkerfor radioresistant cancer stem cells. Int J Radiat Biol 2014;90:636–52.

39. Lagadec C, Vlashi E, Della Donna L, Dekmezian C, Pajonk F. Radiation-induced reprogramming of breast cancer cells. Stem Cells 2012;30:833–44.

40. Wang Y, Li W, Patel SS, Cong J, Zhang N, Sabbatino F, et al. Blocking theformation of radiation-induced breast cancer stem cells. Oncotarget2014;5:3743–55.

41. Ghisolfi L, Keates AC, Hu X, Lee DK, Li CJ. Ionizing radiation inducesstemness in cancer cells. PLoS One 2012;7:e43628.

42. Seligson DB, Horvath S, Shi T, Yu H, Tze S, Grunstein M, et al. Globalhistone modification patterns predict risk of prostate cancer recurrence.Nature 2005;435:1262–6.

43. Yang YA, Yu J. EZH2, an epigenetic driver of prostate cancer. Protein Cell2013;4:331–41.

44. Bracken AP, Pasini D, Capra M, Prosperini E, Colli E, Helin K. EZH2 isdownstream of the pRB-E2F pathway, essential for proliferation andamplified in cancer. EMBO J 2003;22:5323–35.

45. Visser HP, Gunster MJ, Kluin-Nelemans HC, Manders EM, Raaphorst FM,Meijer CJ, et al. The Polycomb group protein EZH2 is upregulated inproliferating, cultured humanmantle cell lymphoma. Br J Haematol 2001;112:950–8.

46. Varambally S, Dhanasekaran SM, Zhou M, Barrette TR, Kumar-Sinha C,Sanda MG, et al. The polycomb group protein EZH2 is involved inprogression of prostate cancer. Nature 2002;419:624–9.

47. Jung HY, Jun S, Lee M, Kim HC, Wang X, Ji H, et al. PAF and EZH2induce Wnt/beta-catenin signaling hyperactivation. Mol Cell 2013;52:193–205.

48. Taylor BS, Schultz N, Hieronymus H, Gopalan A, Xiao Y, Carver BS, et al.Integrative genomic profiling of human prostate cancer. Cancer Cell2010;18:11–22.

49. He S, Wang J, Kato K, Xie F, Varambally S, Mineishi S, et al. Inhibition ofhistone methylation arrests ongoing graft-versus-host disease in mice byselectively inducing apoptosis of alloreactive effector T cells. Blood 2012;119:1274–82.

50. Liu Z, Ren G, Shangguan C, Guo L, Dong Z, Li Y, et al. ATRA inhibits theproliferation of DU145 prostate cancer cells through reducing the meth-ylation level of HOXB13 gene. PLoS One 2012;7:e40943.

51. Hoffmann MJ, Engers R, Florl AR, Otte AP, Muller M, Schulz WA. Expres-sion changes in EZH2, but not in BMI-1, SIRT1, DNMT1 or DNMT3B areassociated with DNA methylation changes in prostate cancer. Cancer BiolTher 2007;6:1403–12.

52. Saramaki OR, Tammela TL, Martikainen PM, Vessella RL, Visakorpi T. Thegene for polycomb group protein enhancer of zeste homolog 2 (EZH2) isamplified in late-stage prostate cancer. Genes Chromosomes Cancer2006;45:639–45.

53. He LR, Liu MZ, Li BK, Jia WH, Zhang Y, Liao YJ, et al. High expression ofEZH2 is associated with tumor aggressiveness and poor prognosis inpatients with esophageal squamous cell carcinoma treated with definitivechemoradiotherapy. Int J Cancer 2010;127:138–47.

54. Meng X, Huang Z, Wang R, Jiao Y, Li H, Xu X, et al. The prognostic role ofEZH2 expression in rectal cancer patients treated with neoadjuvant che-moradiotherapy. Radiat Oncol 2014;9:188.

55. Kim SH, Joshi K, Ezhilarasan R, Myers TR, Siu J, Gu C, et al. EZH2 protectsglioma stem cells from radiation-induced cell death in a MELK/FOXM1-dependent manner. Stem Cell Rep 2015;4:226–38.






2016;76:2637-2651. Published OnlineFirst March 16, 2016.Cancer Res Claudia Peitzsch, Monica Cojoc, Linda Hein, et al. Radioresistant Prostate Cancer CellsAn Epigenetic Reprogramming Strategy to Resensitize

Updated version

10.1158/0008-5472.CAN-15-2116doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2016/03/16/0008-5472.CAN-15-2116.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/76/9/2637.full#ref-list-1

This article cites 54 articles, 12 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://cancerres.aacrjournals.org/content/76/9/2637To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-15-2116

http://cancerres.aacrjournals.org/content/suppl/2016/03/16/0008-5472.CAN-15-2116.DC1

http://cancerres.aacrjournals.org/content/76/9/2637.full#ref-list-1

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/76/9/2637


an epigenetic reprogramming strategy to resensitize radioresistant ... · an epigenetic...

Documents