PARP Inhibition Induces Enrichment of DNA Repair Proﬁcient ... · DNA Damage and Repair PARP Inhibition Induces Enrichment of DNA Repair–Proﬁcient CD133 and CD117 Positive Ovarian

DNA Damage and Repair

PARP Inhibition Induces Enrichment of DNARepair–Proficient CD133 and CD117 PositiveOvarian Cancer Stem CellsChiara Bellio1,2, Celeste DiGloria1,2, Rosemary Foster1,2,3, Kaitlyn James4,Panagiotis A. Konstantinopoulos5,Whitfield B. Growdon1,2,3, and Bo R. Rueda1,2,3

Abstract

PARP inhibitors (PARPi) are FDA-approved monotherapy agentsfor the treatment of recurrent ovarian cancer in patients with andwithout a BRCAmutation. Despite promising response rates, notall patients derive benefit, and the majority develop resistance.PARPi treatment in vitro and in vivo induced an enrichment ofCD133þ and CD117þ ovarian cancer stem cells (CSC). This effectwas not affected by BRCA mutation status. In the CSC fractions,PARPi induced cell-cycle arrest in G2–M with a consequentaccumulation of gH2AX, RAD51, and uniquely DMC1 foci. DNAdamage and repair monitoring assays demonstrated that CSCsdisplay more efficient DNA repair due, in part, to activation ofembryonic repair mechanisms which involved the RAD51 homo-logue, DMC1 recombinase. Preserved and induced homologousrepair (HR) could be a mechanism of an inherent resistance ofCSCs to the synthetic lethality of PARPi that likely promotesdisease recurrence.

Implications: Treatment with PARPi fails to significantly affectovarian cancer CSC populations, likely contributing to recurrentdisease. Ovarian cancer CSCs stabilize genomic integrity afterPARPi treatment, due to a more efficient inherent DNA repaircapacity. PARPi-induced DMC1 recombinase and HR proficiencyprovide CSCs the opportunity to repair DNA damage moreefficiently.

Visual Overview: http://mcr.aacrjournals.org/content/molcanres/17/2/431/F1.large.jpg.

IntroductionOvarian cancer is the fifth leading cause of cancer-related death in

women and the most lethal gynecologic cancer in the United States

(1). In 2018, approximately 22,240 women will be diagnosed withovarian cancer, and about 14,070 women will succumb to theirdisease (2). Although the majority of women with advanced-stagedisease initially respond to platinum-based chemotherapy, tumorresurgence is observed in a majority of women rendering theirdisease difficult to control durably (3). Novel therapies to bothprevent recurrence and actively treat recurrent disease are currentlybeing investigated for women with ovarian cancer in order toprolong survival (4). Whole-genomic analyses have demonstratedthat deficient homologous repair (HR) is a prevalent signature inhigh-grade serous ovarian carcinoma among women with andwithout germline breast cancer (BRCA) gene mutations (5). Nota-bly, recent clinical trials have demonstrated that PARP inhibitors(PARPi) significantlyprolong the time to recurrence andprogressionin all women with ovarian cancer, with the most robust antitumoractivity observed in those women harboring a somatic or germlineBRCAmutation (6). These clinical data have led to FDA approval ofmultiple PARPi as both maintenance and third-line therapies.

1Department of Obstetrics and Gynecology, Massachusetts General Hospital,Boston, Massachusetts. 2Harvard Medical School, Boston, Massachusetts. 3Divi-sion of Gynecologic Oncology, Department of Obstetrics and Gynecology,Massachusetts General Hospital, Boston, Massachusetts. 4Deborah Kelly Centerfor Outcomes Research, Department of Obstetrics and Gynecology, Massachu-setts General Hospital, Boston, Massachusetts. 5Dana-Farber Cancer Institute,Boston, Massachusetts.

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

Corresponding Author: Bo R. Rueda, Massachusetts General Hospital, 55 FruitStreet, Boston, MA 02114. Phone: 617-724-2825; Fax: 617-726-0561; E-mail:[email protected]

doi: 10.1158/1541-7786.MCR-18-0594

�2018 American Association for Cancer Research.

MolecularCancerResearch

www.aacrjournals.org 431

on June 11, 2020. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst November 6, 2018; DOI: 10.1158/1541-7786.MCR-18-0594

http://crossmark.crossref.org/dialog/?doi=10.1158/1541-7786.MCR-18-0594&domain=pdf&date_stamp=2019-1-24

http://mcr.aacrjournals.org/content/molcanres/17/2/431/F1.large.jpg

http://mcr.aacrjournals.org/content/molcanres/17/2/431/F1.large.jpg

http://mcr.aacrjournals.org/

Currently, Olaparib (Lynparza, AstraZeneca), Niraparib(Junuvia, Tesaro), and Rucaparib (Rubraca, Clovis Oncology) areFDA-approved PARPi treatments for advanced ovarian cancerpatientswith andwithout deleterious germline and somaticBRCAmutations (7–9). Phase II and III clinical trials with PARPi inovarian cancer have resulted in notable antitumor activity andincreased progression-free survival. Although PARPi representeffective therapies, acquired resistance occurs in the overwhelm-ing majority of patients and represents a major clinical hurdle(10, 11). This acquired resistance in ovarian cancer patients hasprompted investigators to explore why patients, even those withsomatic or germline BRCA mutation, develop resistance to thisonce effective therapy. Chemoresistance has been suggested as apotential mechanism by which the cancer cells are eluding thisdouble shut down of the DNA repair pathway. A subset ofchemoresistance has been attributed to secondary genetic andepigenetic events that restore functional HR pathway inHR-deficient tumors or activate other DNA repair pathways orincreased expression of p-glycoprotein efflux transporter mediat-ing multidrug resistance (12).

Alternatively, ovarian tumors may be composed of several cellpopulations that respond to PARPi in different ways due tofluctuating innate levels of HR efficiency. Given the highly het-erogeneous nature of ovarian cancer, it is possible that a subset ofthe bulk tumor cells is less responsive to this treatment strategy.Cancer researchers inmultiple solid and liquid tumor disease siteshave long speculated that small subpopulations with stem-likefeatures—often called cancer stem cells (CSC)—can repopulatetumors (13). Such rare cell populations are thought to turn overless frequently and are therefore more resistant to conventionaltherapies targeting the rapidly dividing cells. From a clinicalperspective, CSCs are hypothesized to persist in tumors as adistinct population following standard cytotoxic therapy therebycontributing to disease relapse and distant metastasis (14).

In this study, we hypothesized that differential antitumor activ-ity of PARPi in ovarian CSC and non-CSC populations contributessignificantly to the development of PARPi resistance, independentof BRCA mutational status. We suspected that the antitumoractivity of PARPi is due to the preferential targeting of a non-CSCpopulation of cells. Although the CSC population manifestsrelative resistance leading to repopulation of tumor cells with theinnate ability to overcome synthetic lethality, we sought to utilizeBRCA-mutated and wild-type models to test this hypothesis andelucidate specific mechanisms that allow CSCs to manifest resis-tance to PARPi.Weprovide evidence that PARPi treatment resultedin the induction of an enrichment of cell populations expressingantigens linked to stem phenotype in ovarian cancer includingCD133, CD117, and aldehyde dehydrogenase (ALDH). Further-more, our results suggest that ovarianCSCs andnon-CSCs responddifferentially toDNA stress, and that CSCsmayhavemore efficientDNA repair pathways due to an activation of embryonic repairmechanisms that may confer a survival advantage, contributing totreatment resistance and eventual recurrence.

Materials and MethodsCell lines and in vitro culture

Human ovarian cancer cell lines (Supplementary Table SA)UWB1.289 MUT (BRCA1 mutant, 2594delC germline mutationin exon 11 and deletion of the wild-type allele; RRID:CVCL_B079) and UWB1.289 WT (RRID: CVCL_B078) were pur-

chased from the American Type Culture Collection (ATCC).Human ovarian cancer cell line PEO1 (BRCA2-mutated, homo-zygous mutation 5193C>G; RRID: CVCL_2686) was purchasedfrom Sigma Aldrich. OVCAR4 (RRID: CVCL_1672) and OVCAR3(RRID: CVCL_0465) were purchased from National Cancer Insti-tute-Developmental Therapeutics Program.OVCAR3 All cell lineswere routinely tested forMycoplasma, cultivated 37�C in 5% CO2

humidity, and passaged until passage 12. UWB1.289 MUT andUWB1.289WTweremaintained in 50%RPMI 1640 (GIBCO, LifeTechnologies), 50% MEGM (Mammary Epithelial Growth Medi-um, MEGM Bullet Kit CC-3150; Lonza), 10% FBS (GIBCO, LifeTechnologies), and 1% Pen/Strep (Thermo Fisher Scientific).UWB1.289 WT is a stable cell line derived from UWB1.289 MUT(ATCC CRL-2945), in which BRCA1 function is restored throughtransfection with a plasmid carrying the wild-type BRCA1 gene.Selection is maintained by culturing the cells in 200 mg/mLof G-418 (Life Technologies). OVCAR3 cells were maintainedin RPMI 1640 (GIBCO), 10% FBS (GIBCO), 1% Pen/Strep(Thermo Fisher Scientific), and 0.01 mg/mL of bovine insulin(Sigma-Aldrich). OVCAR4 cells were maintained in RPMI 1640(GIBCO), 10% FBS (GIBCO), and 1% Pen/Strep (Thermo FisherScientific). PEO1 cells were maintained in RPMI 1640 (GIBCO),10% FBS (GIBCO), 1% Pen/Strep (Thermo Fisher Scientific), and2 mmol/L Sodium Pyruvate (GIBCO).

Drug treatment (olaparib, rucaparib, carboplatin)—MTT andcell counting

Olaparib and rucaparib were purchased from Selleckchem,and carboplatin was obtained from Sigma. IC50 values (theconcentration required to kill and/or inhibit growth of cellsby 50% as compared with untreated control wells) of thesedrugs were estimated from concentration–response curves byusing MTT colorimetric assay [MTT: 3(4,5-dimethylthiazol-2-yl)-2,5-dyphenyltetrazolium bromide] analysis. All the cell lineswere plated in 96-well plates (10,000 cells/well in 150 mLmedia).After overnight incubation, olaparib (0.1, 1, and 10 mmol/L),rucaparib (5, 10, and 50 mmol/L), and carboplatin (10, 25, and50 mmol/L) were added in six replicates to each population. Cellviability was measured after 72 hours (carboplatin) or 7 days(olaparib and rucaparib) using theMTT assay. A calibration curvewas prepared using the data obtained fromwells that contained aknown number of cells. Cell proliferation was assessed by trypanblue (Thermo Fisher Scientific) staining and cell counting fol-lowing treatment with the indicated concentrations of olaparib orrucaparib. Olaparib and rucaparib treatment was 7 days withretreatment every 2 days due to the half-lives of the drugs.

Flow cytometry (CD133, CD117, cell viability, and cell cycle)and cell sorting

Cells were stained with Live-Dead (Pacific Blue, 1:600; Invitro-gen) to discriminate living cells. Apoptotic and necrotic cells werediscriminated using the Annexin/PI staining Kit (Roche) followedby flow cytometry. The following anti-human monoclonalantibodies were used to discriminate CD133þ and CD117þ cells:anti-CD133 [CD133/2 clone 293C3, 1:10, phycoerythrin (PE)-conjugated; Miltenyi Biotec; RRID: AB_2661207] and anti-CD117 [cloneA3C6E2, 1:10, allophycocyanin (APC)-conjugated;Miltenyi Biotec; RRID: AB_2660103]. Following incubation withFcR-blocking reagent (Miltenyi Biotec), cells were resuspended inPBS buffer (PBS, 2% FBS, and 1 mmol/L EDTA) and stained withthe relevant antibodies. Respective IgG isotype antibodies were

Bellio et al.

Mol Cancer Res; 17(2) February 2019 Molecular Cancer Research432




included as negative controls in the first analysis and antibodytitration analyses. In subsequent experiments, unstained cells wereincluded as a control for background fluorescence. Cells wereanalyzed using a Fluorescent Activated Cell Sorting (FACS) LSRIIcytofluorimeter (BD Biosciences). Data were collected from at least1 � 105 live cells/sample and analyzed with FlowJo 10.1 version(TreeStar). ALDH activity was evaluated by using the ALDEFLUORKit (Stemcell Technologies). Briefly, cells were harvested, washed,counted, and suspended in ALDEFLUOR buffer at a density of 1�106 cells/mL. Cells were divided into two aliquots and incubatedwith 1.5mmol/LALDEFLUOR substrate in the presence or absencesof the ALDH inhibitor DEAB. Following a 45-minute incubation at37�C, the samples were analyzed using the FACS LSRII cytofluori-meter. For cell-cycle analysis cells were treatedwith either vehicle or10 mmol/L. The cells were then fixed in 70% ethanol 72 hours aftertreatment, incubated with 10 mg/mL RNase A (Sigma-Aldrich) atroom temperature, and stained with 1 mg/mL DAPI (ThermoFisher). The DNA content of the cells (100,000 cells per experi-mental group) was quantified using a FACS LSRII cytofluorimeterand analyzed using cell-cycle software of FlowJo 10.2 version. Forcell separation, cell suspensions stained with human anti-CD133and anti-CD117 antibodies (Miltenyi Biotec) were separated usinga FACSAria sorter (BD Biosciences) to obtain the double-negativeCD133� CD117� cell populations, and live cells were discrimi-nated with propidium iodide (PI; 1 mg/mL, Thermo Fisher). Toacquire a pure CD133þ population, the AutoMacs cell separator(Miltenyi Biotec) was used. Specifically, cells stained with PE-conjugated anti-CD133 antibody (Miltenyi Biotec) were magnet-ically labeled with anti-PE MicroBeads UltraPure (Miltenyi Biotec)and loaded on a Macs column which was placed in the magneticfieldof anAutoMacs separator. Inall experiments, post-sort analysisusing a FACS LSRII cytofluorimeter confirmed a population purity>90%. In the analysis where it was possible to discriminatethe different cell populations by flow cytometry expression-dependent gates, CD133þCD117þ and CD133�CD117� cellswere compared (cell viability with Live-Dead staining and cell-cycle analysis). Conversely, in the analysis where it was impor-tant to obtain a relatively pure population of the different cellsubtypes with limited DNA damage, the cells were sorted withthe AutoMacs bead system and CD133þ and CD133� cells werecompared. Specifically, the bead separation system was favoredinstead of flow cytometric cell sorting for the analysis of theDNA damage induced by PARPi treatment (immunofluores-cence and comet assay analysis). This bead system reducedbackground of DNA damage induced by flow cytometric cellsorting. It was important to obtain the highest purity >98% ofthe CD133þ and CD133�CD117� cells for the gene cardanalysis. Although it was not possible to obtain this level ofpurity in the CD133þCD117þ–specific sorted population insufficient numbers.

Sphere formation and ELDAThe extreme limiting dilution assay (ELDA) was performed to

analyze the sphere-forming efficiency after olaparib and rucaparibtreatment in vitro. Cells were treated under monolayer cultureconditions, harvested after 7 days, and plated in ultra-low-attachment 96-wells plates (Corning, Inc.) at different cell con-centrations (24 wells with 1 cell/well, 24 wells with 10 cells/well,24wellswith 50 cells/well, and24wellswith 100 cells/well). Cellswere maintained for 10 days in DMEM-F12 without serum, andbFGF and EGF growth factors were added to the culture medium

every 3 days. Counting of spheres was performed after 10 days,and statistical analysis was performed using the ELDA software(http://bioinf.wehi.edu.au/software/elda/).

In vivo experiment (olaparib treatment)Using an Institutional Review Board–approved protocol

(protocol number: 2005N000273) and in compliance with theInstitutional Care and Use Committee, the animal experimentswere conducted at our institution Massachusetts General Hos-pital. Twelve-week-old NOD/SCID mice at 12 weeks of agewere s.c. injected with either 3 � 106 OVCAR3 cells or 5 � 106

PEO1 cells in 1:1L PBS:Matrigel (Corning Matrigel, BD Bio-sciences). After injection of tumor cells, the sizes of the resultingtumors were measured every other day using a caliper; the bodyweight of each mouse was also determined twice per week. Thetumor volume was calculated using the following formula:(width2 � height)/2. When the tumor volume reached 150 to200 mm3, the mice were randomly divided into 2 groups(15 mice per group in OVCAR3 experiment, 7 mice per groupin PEO1 experiment). Treatment regimen for olaparib includedchronic daily administration with 50 mg/kg of olaparib for14 consecutive days via i.p. injection. The control group wastreated with a 4% DMSO/30% polyethylene glycol/ddH2Osolution alone. In OVCAR3 experiment, an acute treatment of6 hours was performed, and tumors were formaldehyde-fixedand paraffin-embedded or frozen for further analysis. The othermice were euthanized with CO2 when the tumor volumereached 900 mm3. Tumor volume was measured every 3 days.At the end of the treatment study, mice were euthanized andxenografts were harvested. Portions of each xenograft were snapfrozen as well as formaldehyde-fixed and paraffin-embeddedfor further analyses. Tumors were processed following a pre-viously described protocol (15), and H-2Kdþ mouse cells wereremoved using a FITC-conjugated antibody (BD Biosciences)and Macs LD columns (Miltenyi Biotec) as per the manufac-turers' recommendations. H-2Kd� cells were stained with Live-Dead (Pacific Blue, 1:600; Invitrogen), anti-CD133 (CD133/2clone 293C3, 1:10, PE-conjugated; Miltenyi Biotec), and anti-CD117 (clone A3C6E2, 1:10, APC-conjugated; Miltenyi Biotec)and analyzed using FACS LSRII cytofluorimeter (BD Bios-ciences). Data were collected from at least 1 � 105 live cells/sample and analyzed with FlowJo 10.1 version (TreeStar).

Comet assayUWB1.289 WT and UWB1.289 MUT cell lines were treated

either with 10 mmol/L olaparib or with vehicle for 72 hours.Following the treatment, the media were changed, and the cellswere allowed to recover for an additional 24 hours withouttreatment. CD133� andCD133þ fractionswere isolated followingeach timepoint using the AutoMacs Microbead system(Miltenyi Biotec). After cell separation,CD133� andCD133þweresubjected to an alkaline comet assay using the Trevigen CometAssayKit (Trevigen) following themanufacturer's protocol. Resultsfrom three experiments were statistically quantified by countingthe number of cells with andwithout comet tails in aminimumof100 cells per sample. DNA damage induced by olaparib wasdetermined following comparisons between comet tail frequencyin BRCA1-wild-type– and mutant-derived CD133þ and CD113�

cell populations 72 hours after vehicle or olaparib treatment.Parallel analyses after the 24-hour recovery period determined therelative DNA repair efficiency of CSCs versus non-CSCs.

Ovarian CSCs Mediate Resistance to PARP Inhibition

www.aacrjournals.org Mol Cancer Res; 17(2) February 2019 433



http://bioinf.wehi.edu.au/software/elda/


Homologous recombination reporter assayUWB1.289 MUT, UWB1.289 WT, and OVCAR4 cell lines were

stably transfected with the plasmid pDR-GFP (a gift from MariaJasin; Addgeneplasmid26475)using FugeneHD(Roche). TheHRreporter assaywas performed as previously described (16). Briefly,the plasmid pCBA-SceI (Addgene plasmid 26477) was used toexogenously express and induce a double-strand break (DSB).Cells were then analyzed 48hours later forGFPfluorescence usinga LSRII flow cytometer. The results were analyzed using FlowJosoftware.

Statistical analysisData were analyzed with GraphPad Prism 6.0 (GraphPad

Software). Data from replicate experiments are shown as meanvalues � standard error. Comparisons between groups wereanalyzed by a two-tailed Student t test or two-way ANOVA, asappropriate. A P value < 0.05 was considered statisticallysignificant.

ResultsPARP inhibitionwith olaparib and rucaparib in vitro induces anenrichment of CD133- and CD117-positive cells

We hypothesized that PARPi induces an enrichment ofCD133þ, CD117þ, and ALDHþ cell populations that have beenshown to manifest CSC-like properties (17). To test this hypoth-esis, we used five different ovarian cancer cell lines with varyingBRCA1/2 mutation status (see Supplementary Table SA): (1)UWB1.289 MUT (BRCA1 mutant); (2) UWB1.289 WT in whichBRCA1 function has been restored via stable transfection; (3)PEO1 (BRCA2 mutant); (4) OVCAR4 (BRCA1/2 wild type); and(5) OVCAR3 (BRCA1/2 wild type). Cells were treated with eitherolaparib or rucaparib, and CD133/CD117 expression levels andcell viability were measured 7 days after treatment by flowcytometry. The dose-response curves were similar for olaparibacross all cell lines, whereas the response to rucaparib was cell linedependent (Supplementary Fig. SB1). Both PARPi showed spec-ificity in reducing PAR levels (Fig. 1A). Olaparib treatment in allthe cell lines was associatedwith a decrease of cell viability by LiveDead or Annexin/PI staining (Fig. 1B, left; Supplementary Fig.SB2) when compared with the vehicle-treated cells. The efficacy ofolaparib at the lower concentrations wasmore pronounced in theBRCA1-mutant UWB1.289 cell line compared with the BRCA1restored wild-type line; however, at the highest concentration,there was no difference between the two cell lines. Olaparibtreatment at 10 mmol/L concentration showed an enrichment ofCD133þ andCD117þ cells (Fig. 1B, P value� 0.001) in all the celllines, with the exception of PEO1 cell line in which 90% of thecells expressed CD133 before the treatment. RT-PCR analysisconfirmed the increase of PROM1 (CD133) and KIT (CD117)expression in all cell lines as well as increased expression of thestem-related genes SOX2 and POU5F1 (OCT4; SupplementaryFig. SC and Supplementary Table S1). We also investigated theactivity ofALDH, adetoxifying enzyme recognized asCSCmarker.Olaparib treatment of all cell lines was associated with an enrich-ment of cells with elevated ALDH activity, further supporting itsrole in chemoresistance and survival mechanisms in CSCs (Sup-plementary Fig. SD). Enrichment of CSC-related marker expres-sion was drug dose-dependent and not specific to one PARPi.Rucaparib showed a similar enrichment of CD133þ and CD117þ

cells (Supplementary Fig. SE2, P value � 0.001). To demonstrate

that olaparib specifically targeted the CD133�CD117� cells, weperformed posttreatment flow cytometric analysis to determinethe cell viability of CD133þ, CD117þ, CD133þCD117þ, andCD133�CD117� cell fractions (Fig. 1C). The graphs reveal thatolaparib treatment at 10 mmol/L had a negative effect on thesurvival of CD133�CD117� cells, with a >50% decrease inviability 7 days after treatment. In contrast, CD133þ, CD117þ,and CD133þCD117þ cell populations did not show any changein viability, confirming their ability to survive the PARPi treat-ment. To corroborate these data, we confirmed theCSCpropertiesin the cells enriched with olaparib and rucaparib treatment. Theirsphere-forming efficiency in vitrowas tested in a limiting-dilutionassay and analyzed by the ELDA software. After olaparib andrucaparib treatment, the enriched CD133þ and CD117þ cellfractions demonstrated a corresponding increase in theirsphere-forming capacity (Fig. 2A; Supplementary Table SE4).Interestingly, the enrichment of CD133þ and CD117þ cells inresponse to olaparib or rucaparib in vitro was not different whencomparing BRCA-mutated cell lines (UWB1.289MUT and PEO1)with BRCA-wild-type lines (UWB1.289, OVCAR3, andOVCAR4),suggesting a potential resistance mechanism to PARPi therapiesindependent of BRCA mutation status.

PARP inhibition with olaparib in vivo induces an enrichment ofCSC markers

The enrichment of CD133þ and CD117þ cells after olaparibtreatment in vitro led us to investigate whether we would observe asimilar effect in vivo. OVCAR3 and PEO1 cells were injected s.c. inNOD/SCID mice, and a 2-week course of daily olaparib treatmentbegan when tumor volume reached approximately 150 to200 mm3. Olaparib alone blunted tumor growth (P value< 0.001) compared with vehicle control in both models (Fig. 2B,top plot). In both in vivo experiments, flow cytometric analysis oftumors harvested at the end of the treatment period revealed anincreased frequency of CD133þ and CD117þ cells in olaparib-treated tumors compared with vehicle-treated controls (Fig. 2B,bottom plot, P value < 0.01). This enrichment of CD133þ andCD117þ cell populations is similar to what we observed in ourin vitro analyses and confirms PARPi-mediated enrichment ofCSCs.

Olaparib treatment results inDNADSBs,G2–Mcell-cycle arrest,and RAD51 foci formation in CD133þ cells

Many studies suggest that normal stem cells responddifferentlyto genotoxic injury by activating a more efficient DNA repairmechanism as compared with differentiated cells (18–20). Grow-ing evidence suggests that CSCs may also utilize these samemechanisms of DNA repair to mediate resistance to both che-motherapy and radiotherapy (21–23). To understand better howovarian CSCs may escape the negative effects of PARPi, weassessed cell-cycle progression, DNA damage, and DNA repair inthe CSC and non-CSC fractions. Analysis of cell-cycle distributionin purified fractions after olaparib treatment showed a greaterpercentage of cells in G2–M in CD133þCD117þ cells comparedwith CD133�CD117� cells in all cell lines, suggesting thatolaparib induces a G2–M cell-cycle arrest in CSCs (Fig. 3, P value< 0.0001). Cells arrest in G2–M phase to repair DNA DSBsthrough HR (24). Histone H2AX is phosphorylated in responseto DNA damage and is an early driver of the recruitment andlocalization of DNA repair proteins (25). Nuclear gH2AX (phos-phorylated H2AX) and RAD51 foci formation was monitored byimmunofluorescence in the UWB1.289 BRCA1-wild-type and

Bellio et al.





BRCA1-mutant cell lines following vehicle or olaparib treatment(Fig. 4A). Cells treated with carboplatin were included as acontrol. Confocal imaging confirmed that both gH2AX and

RAD51 foci accumulated in the nuclei after olaparib treatmentof theBRCA1-wild-type and -mutant cell lines, and this increase inRAD51 foci was statistically significant (Fig. 4A, P value < 0.0001).

Figure 1.

Treatment with the PARPi olaparib in vitro induces an enrichment of CD133þ and CD117þ cells in ovarian serous cancer cell lines. A, Western analysis ofPolyADP-Ribose (PAR) levels in cells treated in vitro with the indicated concentrations of olaparib for 7 days to confirm the dose-dependent on target effectof olaparib. B, Flow cytometric determination of the frequency of viable (left plots), CD133þ (center plots), and CD117þ (right plots) cells in the indicated cell lines7 days following treatment with increasing concentrations of olaparib in vitro. C, Relative cell viability of CD133�CD117�, CD133þ, CD117þ, and CD133þCD117þ cells7 days following in vitro treatment with the indicated concentrations of olaparib as determined by flow cytometry. The graph for each cell line shows eitherthe mean percentage of expression � SEM (B) or mean viable cell number � SEM (C) calculated from three independent experiments; P value < 0.001.






The levels of gH2AX and RAD51 were assessed by Westernblotting following treatment with vehicle or PARPi in UWB1.289WT and MUT cell lines (Supplementary Fig. SF). gH2AX andRAD51 protein levels were similarly increased after olaparibtreatment compared with vehicle, regardless of BRCA gene status.

To explore HR pathway activity in CSC and non-CSC cellpopulations, CD133þ and CD133� subfractions were isolatedfrom UWB1.289 BRCA1-wild-type and -mutant cell lines, andtreated with either vehicle or olaparib. Confocal imaging ofRAD51 and gH2AX foci formation demonstrated an increase inthe number of RAD51 foci in CD133þ cells compared withCD133� cells after olaparib treatment. For this experiment, wechose to minimize any potential background damage from sort-ing. Therefore, the cells were sorted with the AutoMacs beadsystem (see Materials and Methods). Quantitative comparisonof 100 cells showed an increase of RAD51 expression in the nucleiof CD133þ cells (Fig. 4B, bottom plot) in both BRCA1-mutantand BRCA1-wild-type cells (P value < 0.0001), suggesting adifferent olaparib-induced DNA damage response in CSCs thatis independent of BRCA1 gene status.

CD133þ cells efficiently recover from DNA damageThe accumulation of RAD51 foci in the nucleus suggested to

analyze the DNA damage and repair rate using the comet assay,

which can discriminate between a cell with DNA damage (asevidenced by a comet tail) and a cell without DNA damage (nocomet tail). UWB1.289 BRCA1 WT and MUT-derived CD133þ

and CD133� cells were isolated using the AutoMacs bead systemto avoid FACS-induced DNA damage. The comet assay wasperformed after 72 hours after olaparib treatment to confirm theinduction of DNA damage. At that time, olaparib was removedfrom the culturemedium, and cells were allowed to recover for anadditional 24 hours. A second comet assay was then assessed todetermine the relative DNA repair efficiency following recoveryfrom the olaparib treatment. Confocal imaging showedan increase of cells with comet tails after olaparib treatment(Supplementary Fig. SG; P value < 0.001), suggesting DNA dam-age in the treated cells and consistentwith the observed increase ingH2AX foci formation in both CD133þ and CD133� cells(Fig. 4B). After the 24-hour recovery period, confocal imagingshowed an increased number of CD133þ cells (P value < 0.0001)without a comet tail relative to that of CD133� cells (P value <0.01; Fig. 5A), suggesting that the CSC population is better able torepair DNA damage within 24 hours after cessation of olaparibtreatment. Quantification of these data (Fig. 5B) showed thatalthough both cell populations have a statistically significantincrease in cells undergoing active DNA repair, a greater percent-age of the CD133þ CSC population manifested a more efficient

Figure 2.

PARPi olaparib induces enrichmentof stemness phenotype and in vivoenrichment of CD133þ and CD117þ

expression. A, ELDA of the indicatedcell lines 7 days following treatmentwith 10 mmol/L olaparib treatmentin vitro. The data are expressed assphere-forming frequency of cellsexposed to olaparib normalized to thecorresponding vehicle control andsuggest the relative frequency of CSCsin each sample. Three separateiterations of the analysis wereperformed with each cell line, andthe numbers represent the meansphere-forming frequency � SEM(P value < 0.001). B, Mice harboringeither OVCAR3-derived xenografts(left plots) or PEO1-derived xenografts(right plots) were treated with eithervehicle or olaparib as described inMaterials and Methods. CD133þ andCD117þ cell frequencies in tumorsharvested at the end of the treatmentperiod were determined by flowcytometry. Error bars represent SEM;� , P < 0.01.

Bellio et al.





DNA repair rate, which was comparable with that of untreatedcells. This differential DNA repair activity in CSCs was not cor-related with BRCA1 mutation status.

CD133þ cells exhibit unique DNA DSB repair gene expressionafter olaparib treatment

Expression analysis of a subset of genes involved in DSB repairwas performed to identify the specific pathways associated withPARPi resistance in CSCs. Although previous work demonstrated

that drug inhibition or gene silencing of the RAD51 family ofproteins affected response to PARPi treatment (26), there is noinformation regarding the relative efficiency ofDNADSB repair inCSCs. This lack of data stresses the importance of the HR pathwayin DNA repair and PARPi resistance. We utilized custom PCRarrays comprising genes associated with DSB repair and BRCA-ness to analyze the relative expression of each gene in UWB1.289WT- and MUT-derived CD133þ and CD133� CD117� cell popu-lations. Olaparib treatment induced an increase in the expression

Figure 3.

CD133þCD117þ cells shift to G2–M after olaparibtreatment in vitro. The cell-cycle distribution ofCD133�CD117� and CD133þCD117þ cells treated witheither vehicle or 10 mmol/L olaparib was assessed 72hours after treatment by measurement of DNA contentvia DAPI staining. The figure shows themean values fromthree different iterations of this analysis in each cell line.Measurement of G2–M phase after vehicle and olaparibtreatment show statistical difference in CD133þCD117þ

population (P < 0.0001) compared with theCD133�CD117� population, which is still significant butnot as robust as CD133þCD117þ population.






Figure 4.

Olaparib treatment induces an accumulation of RAD51 foci in CD133þ cells. A, Confocal images of RAD51 foci formation in UWB1.289 WT and UWB1.289MUT cells 72 hours following treatment with 5 mmol/L carboplatin or 10 mmol/L olaparib. Graphs show RAD51 foci frequency (n¼ 3, mean� SEM, P value < 0.0001).B, RAD51 foci formation in CD133þ and CD133� enriched fractions from vehicle- and olaparib-treated UWB1.289 WT and UWB1.289 MUT cells. The histogramfor each cell line shows a statistically significant (P < 0.005) olaparib induced increase in foci formation in CD133þ cells (n ¼ 3, mean � SEM).

Bellio et al.





Figure 5.

CD133þ cells show enhanced DNA repair efficiency. A, UWB1.289 WT (top plot0) and UWB1.289 MUT (bottom plot) cells were analyzed 96 hours aftertreatment with vehicle or 72 hours after treatment with 10 mmol/L olaparib followed by a 24-hour no treatment recovery period and then subjected tomagnetic bead separation to isolate CD133þ and CD133� cells. DNA damage in the purified cell fractions was assessed by the comet assay. Representativeresults are shown. B, Quantitation of the comet assay results (see details in Materials and Methods) determined that the DNA repair efficiency of CD133þ

cells is significantly enhanced compared with their CD133� counterparts (n ¼ 3, mean � SEM, P value < 0.0001).






of the RAD51 gene in both CD133þ (Fig. 6A; SupplementaryTable S2) and CD133�CD117� cell populations (Fig. 6B; Sup-plementary Table S2), confirming the important role of RAD51proteins in the DNA damage response. Olaparib treatment wasalso associated with marked alteration in the expression of theDMC1 genewhich encodes an embryonic RAD51 homolog that isnot expressed in somatic tissue. DMC1 expression in CD133þ

cells derived from both UWB1.289 WT and MUT cell lines wasincreased more than 2 times when compared with expression invehicle-treated cells (Fig. 6A; P value < 0.0001). An olaparib-

induced increase in relative DMC1 expression was also observedin the CD133� CD117� cell populations but was much lessrobust than the one detected in CD133þ cells (Fig. 6C). Appre-ciating the challenge of obtaining sufficient numbers of CD133þ-sorted cells to analyze DMC1 protein expression byWestern blot,we chose to use the immunofluorescence approach to confirmchanges nuclearDMC1protein expression. CD133þ andCD133�

cell fractions were isolated fromUWB1.289WT andMUT cells byAutoMacs bead system to reduce cell sorting–correlated DNAdamage. DMC1 foci formation was analyzed in the samples

Figure 6.

PARPi olaparib inducesoverexpression of DNA repair genes inCD133þ cells. CD133þ cells and CD133�

CD117� cells were FACS purified fromUWB1.289 WT and UWB1.289 MUT 72hours following treatment with eithervehicle or 10 mmol/L olaparib.Expression of DSB repair genes wasassessed by RT-PCR array analysis.DMC1 expression is upregulated inresponse to olaparib treatment inCD133þ cells (A). DMC1 induction inCD133�CD117� cells (B) was lowerthan that observed in CD133þ cells. C,Comparison of DMC1 gene expressionin UWB1.289 WT (left) and UWB1.289MUT (right) cell lines showing themean fold changeof expression�SEMrelative to the vehicle samples.(Continued on the following page.)

Bellio et al.





following treatment with either vehicle or olaparib treatment.Confocal imaging revealed specific DMC1 foci formation inCD133þ cells compared with CD133� cells (Fig. 6D) after ola-parib treatment. Quantitative comparison of 100 cells revealed anincrease of DMC1 expression in the nuclei of CD133þ cells (Fig.6D, bottom plot) derived from both the BRCA1-wild-type andthe BRCA1-mutant cell lines (P value < 0.0001). Specifically,DMC1 foci accumulation in CD133þ cells was PARPi treatment-

dependent and not induced by carboplatin, a more traditionalcytotoxic drug (Fig. 6E; Supplementary Fig. SH). These datasuggest that the differential DNA repair mechanism in CSCsinvolves the DMC1 recombinase. RAD51 and DMC1 are impor-tant regulators in the initiation of HR (27). The olaparib inducedincrease of RAD51 and DMC1 foci in the nucleus of CD133þ cellpopulations suggests that CSCs can circumvent PARPi treatmentby manifesting inherent HR processes. To provide some proof of

Figure 6.

(Continued. ) D, Representativeconfocal images of nuclear RAD51(green) and DMC1 (red)immunofluorescence in UWB1.289WTand UWB1.289 MUT-derived CD133�

and CD133þ cells following treatmentwith vehicle or olaparib. Nuclei werestained with DAPI. Histogramsshowing themean frequency� SEMofDMC1 foci from three individualexperiments in CD133� and CD133þ

cells from UWB1.289 WT (left) andUWB1.289 MUT (right), normalized tothe vehicle CD133� samples (n ¼ 3,P value < 0.0001). E, Analysis andquantification of DMC1 foci formation72 hours following treatment withvehicle, rucaparib, or carboplatin werecarried out as described in A (n ¼ 3,P value < 0.0001).






Figure 7.

A, Western analysis of DMC1 protein levels in UWB1.289 WT (left plot) and UWB1.289 MUT (right plot) cells transfected with either a DMC1 expressionvector (DMC1) or the corresponding control vector (CTRL). Exogenous DMC1 expression was analyzed 2 days after transfection and again 3 days after thetransfected cells had undergone treatment with vehicle or olaparib (5-day samples). The relative DMC1 protein level in each sample was determined bycomparing DMC1 expression to b-tubulin level. Results were normalized to the DMC1 level in cells transfected with the control plasmid. B, UWB1.289 WT(left plot) and UWB1.289 MUT (right plot) cells were assessed 96 hours after treated with either vehicle or 72 hours after treatment with 10 mmol/L olaparibfollowed by a 24-hour recovery period in the absence of olaparib. CD133� cells were then isolated by magnetic bead separation. DNA damage in the purifiedcell fractions was assessed by the comet assay. Representative results are shown. Quantitation of the comet assay results (see details in Materials and Methods)determined that the DNA repair efficiency of CD133� cells is significantly enhanced in the DMC1-transfected cells compared with the CTRL-transfected cells (n¼ 3,mean � SEM, P value < 0.0001). (Continued on the following page.)

Bellio et al.





concept for the role of DMC1 recombinase as a potential con-tributor to DNA repair efficiency in the CSCs, we induced over-expression of DMC1 gene in the CD133� sorted cells by plasmidtransfection to observe whether the nonstem cells would displaythe differential phenotype observed in the CSCs. The DNA repairrate was analyzed after 24-hour recovery period after olaparibtreatment. We confirmed the DMC1 protein overexpression byWestern blotting after the transfection and before the 24-hourrecovery period in both UWB1.289 WT and MUT cell lines (Fig.7A). CD133þ and CD133� cells were isolated after a 24-hourrecovery period, and the DNA repair efficiency was quantifiedwith the comet assay. DMC1 overexpression resulted in enhancedDNA repair in the CD133� cells after olaparib treatment andprovided a protective role from inherent DNA damage in thevehicle-treated cells, compared with the CTRL vector-transfectedcells (Fig. 7B).

CD133þ and CD117þ cells are inherentlymore efficient at DNArepair

To determine the inherent DNA repair capacity of CD133þ andCD117þ cells relative to that of their CD133� and CD117�

counterparts, we utilized a reporter assay which assesses HRactivity (Fig. 7C). UWB1.289 BRCA1 WT and MUT cells stablyexpressing theDR-GFP reporterwere transiently transfectedwith aplasmid encoding the endonuclease I-SceI to generate a specificDSB in the GFP coding region. In this assay, functional GFP canonly be restored if the DSB is repaired using the downstream GFPfragment as a template for HR (Fig. 7C). GFP expression detectedafter 48 hours of the endonuclease-induced DSB damage revealedamore efficient DNA repair capacity in the CD133þ and CD117þ

cell populations, as evidenced by a statistically significant increasein GFP expression (Fig. 7D, P value < 0.05) when compared withthe one in the CD133� CD177� population. These data suggest

Figure 7.

(Continued. ) C,CD133þandCD117þ cells showenhancedDNA repair efficiency followingDSBDNAdamage. Schematic representationof homologous recombinationassay. Exogenous expression of SceI in cells transfected with pDR-GFP results in a DSB in GFP and disruption of the open reading frame. The efficiency of inherentDNA repair in the transfected cells can be assessed by flow cytometric analysis of restored GFP fluorescence. D, Level of GFP expression in UWB1.829 WT- andUWB1.289 MUT-gated CD133þ, CD117þ, CD133�, and CD117� cell fractions (n ¼ 3, mean � SEM, P value < 0.05).






that CSCs harbor inherent DNA DSB repair mechanisms thatallow them to adapt to selective pressure and manifest HR.

DiscussionRecent clinical trials have confirmed the clear clinical benefits of

PARPi in the treatment of women with ovarian cancer, yet inves-tigators acknowledge its limited impact on the frequent acquiredresistance in the overwhelming majority of patients (28). Wehypothesized that treatment with a PARPi would result in anenrichment of CSC populations that persist beyond therapy andeventually serve to seed tumor resurgence. Utilizing in vitro andin vivomodels, we demonstrated that olaparib treatment enrichedovariancancer stem–likeCD133þ andCD117þ cell populations, aswell as inducing cell death in the CD133� CD117� non–stem-likepopulation, regardless ofBRCAmutation status. A similar responseto the PARPi rucaparib suggested this result was not limited to asingle PARPi but was more likely to be a class effect. The CD133þ

cell fractions displayed an extended G2–M phase which was morepronounced following PARPi treatment, suggesting that the CSCpopulation may be undergoing DNA repair. Gene expressionanalyses utilizing arrays focused on HR and non-homologous endjoining (NHEJ) genes revealed a preferential expression of DMC1in the CD133þ fractions that again was also more pronounced inresponse to olaparib. Assessment of gH2AX alongwith RAD51 andDMC1 foci revealed that RAD51 and DMC1 foci were abundant inthe CSC population when compared with the non-CSC popula-tion. RAD51 and DMC1 are both associated with HR repair (29).Further analysis using the comet assay and an HR reporter assay tobetter determine DNA repair efficiency confirmed the capacity ofovarian CSC to respond to and repair DNA damage. Collectively,these data reveal the enhanced DNA repair capability of stem-likeovarian cancer cells which likely enables them to escape the PARPi-induced disruption in their genomic integrity leading to theirsurvival and driving of disease recurrence.

Maintenance of stem cell genomic integrity is critical for tissueand organ homeostasis (30). Consequently, it has long beensuggested that adult stem cells have a greater capacity towithstandDNA damage when compared with their non–stem cell counter-parts (31, 32). It has also been proposed that adult stem cells havethe capacity to repair their DNA damage more efficiently wheninduced by intrinsic cellular metabolism or extrinsic factors likeionizing radiation or pharmacologic agents (33). Proposedmechanisms by which stem cells evade irreparable damageinclude the stem cell's relatively quiescent state. The infrequentturnover limits endogenous DNA damage which might occur asthe result of normal cellular metabolism. Alternatively, stem cellscan often be found in protective microenvironments (34). Lastly,stem cells may rely on more efficient DNA damage response andrepair mechanisms.

In addition to an infrequent turnover rate, CSCs may employalternative mechanisms of DNA repair to mediate resistance tochemotherapy and radiotherapy (35). In our study, an extend-ed G2–M phase was evident in the CSC fractions—independentof the treatment, which suggests that these populations werenot in a fully quiescent state but more likely in a lag phaseallowing for time to respond to DNA damage. The extendedG2–M phase was more prevalent post treatment with a PARPi,further suggesting that PARPi may affect CSCs even thoughthese populations exhibited an improved ability to overcomethe deleterious effects. Evidence of DNA damage in the CSCs

was evident by the accumulation of gH2AX foci concurrent withthe accumulation of RAD51 foci. Other investigators havedemonstrated enrichment of CSCs after treating solid tumorswith cytotoxics and other pharmacologic therapies targetingbreast and ovarian cancer (36, 37).

Because olaparib extended the G2–Mphase in CSCs, we soughtto determine whether there was differential gene expressionrelated to HR and NHEJ pathways. The most striking differenceobserved was in the expression of DNA meiotic recombinase 1(DMC1), which is thought to be a meiosis-specific recombinase.Normally, genetic recombination inmeiosis contributes to genet-ic diversity through the inductionof programmedDSBs,which arerepaired throughHR. Recent data have shown thatmalignant cellsare able to adapt unique signaling pathways normally attributedto meiosis and/or embryonic development (38). DMC1 expres-sion has been detected in glioblastoma cells and was induced byradiation (39). Depletion of DMC1 in glioblastoma multiforme(GBM) cells, together with radiation treatment, blunted the DNAdamage response. Moreover, loss of DMC1 stabilized tumorgrowth and increased survival. Unlike GBM, in the ovarian cancerlines, the increase in DMC1was only induced by PARPi (olapariband rucaparib) as no evidence of DMC1 foci was observed inresponse to carboplatin. Although the GBM studies implied thatDMC1 expressionwas ubiquitous, our analyses determined that itwas primarily specific to ovarian cancer CSCs.

These data highlight DMC1 as an important mediator of HRpreservation in CSCs treated with PARPi. Interestingly, althoughcarboplatin treatment of ovarian cancer cells induced nucleargH2AX and RAD51 foci in CD133� and CD133þ cells, onlyPARPi induced an accumulation of gH2AX, RAD51, and DMC1foci exclusively in the CD133þ population. This finding suggestsan activation of alternative homologous recombination DNArepair pathways in stem-like cells. Given that two separate anal-yses provided evidence that CD133þ cells efficiently repair ola-parib-induced DNA damage, we sought to assess whether DMC1expression in CD133þ cells could augment DNA repair followingolaparib treatment. Because DMC1 expression was induced byPARPi, the gene could not be knocked down. We therefore choseto stably transfect non-CSCs with DMC1 to determine if thetransformed cells could mimic the DNA damage responseobserved in CSCs. Exogenous expression of DMC1 in the non-CSC population was sufficient tomount a DNA damage responseto reduce the impact of PARPi. Collectively, these data highlight apotential DMC1-mediated resistance mechanism to PARPi ther-apy. Whether the enhanced DNA repair capability of ovariancancer CSCs is mediated by DMC1 alone is yet to be determined.Regardless, our analyses strongly suggest that the enhanced DNArepair capability of CSCs contributes to the maintenance of theirgenomic integrity during PARPi treatment, thereby allowing themto survive and lead to the recurrence of disease.

Disclosure of Potential Conflicts of InterestP.A. Konstantinopoulos is a consultant/advisory board member for Pfizer,

AstraZeneca, Vertex, and Merck. W.B. Growdon reports receiving a commercialresearch grant from, and is a consultant/advisory board member for Tesaro.No potential conflicts of interest were disclosed by the other authors.

Authors' ContributionsConception and design: C. Bellio, W.B. Growdon, B.R. RuedaDevelopment of methodology: C. Bellio, C. DiGloria, R. Foster, B.R. RuedaAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): C. Bellio, C. DiGloria, B.R. Rueda


Bellio et al.




Analysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): C. Bellio, C. DiGloria, R. Foster, K. James,P.A. Konstantinopoulos, W.B. Growdon, B.R. RuedaWriting, review, and/or revision of the manuscript: C. Bellio, C. DiGloria,R. Foster, K. James, P.A. Konstantinopoulos, W.B. Growdon, B.R. RuedaAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): C. DiGloria, B.R. RuedaStudy supervision: B.R. Rueda

AcknowledgmentsWe are grateful to Dr. Unnati Pandya for her expertise in DNA transfection

and sequencing.

This work was funded by the Advanced Medical Research Foundation(C. Bellio, C. DiGloria, R. Foster, and B.R. Rueda) and the Vincent MemorialResearch Funds (B.R. Rueda).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received June 6, 2018; revised August 24, 2018; accepted October 18, 2018;published first November 6, 2018.

References1. Stewart SL, Harewood R, Matz M, Rim SH, Sabatino SA, Ward KC, et al.

Disparities in ovarian cancer survival in the United States (2001–2009):findings from the CONCORD-2 study. Cancer 2017;123Suppl 24:5138–59.

2. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2018. CA Cancer J Clin2018;68:7–30.

3. Narod S. Can advanced-stage ovarian cancer be cured? Nat Rev Clin Oncol2016;13:255–61.

4. JohnsonN, Liao JB.Novel therapeutics for ovarian cancer: the11thBiennialRivkin Center Ovarian Cancer Research Symposium. Int J Gynecol Cancer2017;27(9S Suppl 5):S14–S9.

5. Vanderstichele A, Busschaert P, Olbrecht S, Lambrechts D, Vergote I.Genomic signatures as predictive biomarkers of homologous recombina-tion deficiency in ovarian cancer. Eur J Cancer 2017;86:5–14.

6. Kim G, Ison G, McKee AE, Zhang H, Tang S, Gwise T, et al. FDA approvalsummary: olaparib monotherapy in patients with deleterious germlineBRCA-mutated advanced ovarian cancer treatedwith three ormore lines ofchemotherapy. Clin Cancer Res 2015;21:4257–61.

7. de Lartigue J. Olaparib for BRCA-mutated advanced ovarian cancer. JCSO2015;13:206–8.

8. Meehan RS, Chen AP. New treatment option for ovarian cancer: PARPinhibitors. Gynecol Oncol Res Pract 2016;3:3.

9. Liu JF, Konstantinopoulos PA, Matulonis UA. PARP inhibitors in ovariancancer: current status and future promise. GynecolOncol 2014;133:362–9.

10. Lupo B, Trusolino L. Inhibition of poly(ADP-ribosyl)ation in cancer: oldand new paradigms revisited. Biochim Biophys Acta 2014;1846:201–15.

11. Bitler BG, Watson ZL, Wheeler LJ, Behbakht K. PARP inhibitors: clinicalutility and possibilities of overcoming resistance. Gynecol Oncol 2017;147:695–704.

12. Gottesman MM, Ling V. The molecular basis of multidrug resistance incancer: the early years of P-glycoprotein research. FEBS Lett 2006;580:998–1009.

13. Chan KS. Molecular pathways: targeting cancer stem cells awakened bychemotherapy to abrogate tumor repopulation. Clin Cancer Res 2016;22:802–6.

14. Alvero AB, Chen R, Fu HH, Montagna M, Schwartz PE, Rutherford T, et al.Molecular phenotyping of human ovarian cancer stem cells unravels themechanisms for repair and chemoresistance. Cell Cycle 2009;8:158–66.

15. Curley MD, Therrien VA, Cummings CL, Sergent PA, Koulouris CR, FrielAM, et al. CD133 expression defines a tumor initiating cell population inprimary human ovarian cancer. Stem Cells 2009;27:2875–83.

16. Choi YE, Battelli C, Watson J, Liu J, Curtis J, Morse AN, et al. Sublethalconcentrations of 17-AAG suppress homologous recombination DNArepair and enhance sensitivity to carboplatin and olaparib in HR proficientovarian cancer cells. Oncotarget 2014;5:2678–87.

17. Foster R, Buckanovich RJ, Rueda BR. Ovarian cancer stem cells: workingtowards the root of stemness. Cancer Lett 2013;338:147–57.

18. WeedenCE, Asselin-LabatML.Mechanisms ofDNAdamage repair in adultstem cells and implications for cancer formation. Biochim Biophys Acta2018;1864:89–101.

19. Vahidi Ferdousi L, Rocheteau P, Chayot R, Montagne B, Chaker Z, FlamantP, et al. More efficient repair of DNA double-strand breaks in skeletalmuscle stem cells compared to their committed progeny. Stem Cell Res2014;13(3 Pt A):492–507.

20. Sotiropoulou PA, Candi A, Mascre G, De Clercq S, Youssef KK, Lapouge G,et al. Bcl-2 and accelerated DNA repair mediates resistance of hair folliclebulge stem cells to DNA-damage-induced cell death. Nat Cell Biol2010;12:572–82.

21. Nagel ZD, Kitange GJ, Gupta SK, Joughin BA, Chaim IA, MazzucatoP, et al. DNA repair capacity in multiple pathways predictschemoresistance in glioblastoma multiforme. Cancer Res 2017;77:198–206.

22. Wang QE. DNA damage responses in cancer stem cells: implications forcancer therapeutic strategies. World J Biol Chem 2015;6:57–64.

23. Bao S, Wu Q, McLendon RE, Hao Y, Shi Q, Hjelmeland AB, et al. Gliomastem cells promote radioresistance by preferential activation of the DNAdamage response. Nature 2006;444:756–60.

24. Branzei D, Foiani M. Regulation of DNA repair throughout the cell cycle.Nat Rev Mol Cell Biol 2008;9:297–308.

25. Mah LJ, El-Osta A, Karagiannis TC. gammaH2AX: a sensitive molecularmarker of DNA damage and repair. Leukemia 2010;24:679–86.

26. Liu Y, BurnessML,Martin-Trevino R,Guy J, Bai S, Harouaka R, et al. RAD51mediates resistance of cancer stem cells to PARP inhibition in triple-negative breast cancer. Clin Cancer Res 2017;23:514–22.

27. Gerton JL, Hawley RS. Homologous chromosome interactions in meiosis:diversity amidst conservation. Nat Rev Genet 2005;6:477–87.

28. Gadducci A, Guerrieri ME. PARP inhibitors in epithelial ovarian cancer:state of art and perspectives of clinical research. Anticancer Res2016;36:2055–64.

29. Sung P, Klein H. Mechanism of homologous recombination: mediatorsand helicases take on regulatory functions. Nat Rev Mol Cell Biol2006;7:739–50.

30. Nagaria P, Robert C, Rassool FV. DNA double-strand break response instem cells: mechanisms to maintain genomic integrity. Biochim BiophysActa 2013;1830:2345–53.

31. Mandal PK, Blanpain C, Rossi DJ. DNA damage response in adult stemcells: pathways and consequences. Nat Rev Mol Cell Biol 2011;12:198–202.

32. Vitale I, Manic G, DeMaria R, Kroemer G, Galluzzi L. DNA damage in stemcells. Mol Cell 2017;66:306–19.

33. Insinga A, Cicalese A, Pelicci PG. DNA damage response in adult stem cells.Blood Cells Mol Dis 2014;52:147–51.

34. Morrison SJ, Spradling AC. Stem cells and niches: mechanisms thatpromote stem cell maintenance throughout life. Cell 2008;132:598–611.

35. Blanpain C, Mohrin M, Sotiropoulou PA, Passegue E. DNA-damageresponse in tissue-specific and cancer stem cells. Cell Stem Cell2011;8:16–29.

36. Shafee N, Smith CR, Wei S, Kim Y, Mills GB, Hortobagyi GN, et al. Cancerstem cells contribute to cisplatin resistance in Brca1/p53-mediated mousemammary tumors. Cancer Res 2008;68:3243–50.

37. Kulkarni-Datar K, Orsulic S, Foster R, Rueda BR. Ovarian tumor initiatingcell populations persist following paclitaxel and carboplatin chemother-apy treatment in vivo. Cancer Lett 2013;339:237–46.

38. Dreesen O, Brivanlou AH. Signaling pathways in cancer and embryonicstem cells. Stem Cell Rev 2007;3:7–17.

39. RiveraM,WuQ,Hamerlik P,HjelmelandAB, Bao S, Rich JN. Acquisition ofmeiotic DNA repair regulators maintain genome stability in glioblastoma.Cell Death Dis 2015;6:e1732.






2019;17:431-445. Published OnlineFirst November 6, 2018.Mol Cancer Res Chiara Bellio, Celeste DiGloria, Rosemary Foster, et al. CD133 and CD117 Positive Ovarian Cancer Stem Cells

Proficient−PARP Inhibition Induces Enrichment of DNA Repair

Updated version

10.1158/1541-7786.MCR-18-0594doi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2018/11/06/1541-7786.MCR-18-0594.DC1

Access the most recent supplemental material at:

Overview

Visual

http://mcr.aacrjournals.org/content/17/2/431/F1.large.jpgA diagrammatic summary of the major findings and biological implications:

Cited articles

http://mcr.aacrjournals.org/content/17/2/431.full#ref-list-1

This article cites 39 articles, 7 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://mcr.aacrjournals.org/content/17/2/431To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-18-0594

http://mcr.aacrjournals.org/content/suppl/2018/11/06/1541-7786.MCR-18-0594.DC1

http://mcr.aacrjournals.org/content/17/2/431/F1.large.jpg

http://mcr.aacrjournals.org/content/17/2/431.full#ref-list-1

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/17/2/431


Documents

PARP Inhibition Induces Enrichment of DNA Repair Proﬁcient ... · DNA Damage and Repair PARP Inhibition Induces Enrichment of DNA Repair–Proﬁcient CD133 and CD117 Positive Ovarian