identiﬁcation of selective lead compounds for...

Small Molecule Therapeutics

Identification of Selective Lead Compounds forTreatment of High-Ploidy Breast CancerAlka Choudhary1,2, Brittany Zachek1,2, Robert F. Lera1,2, Lauren M. Zasadil1,3,Amber Lasek1,2, Ryan A. Denu1,2, Hyunjung Kim1,2, Craig Kanugh4, Jennifer J. Laffin4,5,Josephine M. Harter6, Kari B.Wisinski1,2, Sandeep Saha1,7, Beth A.Weaver1,3, andMark E. Burkard1,2

Abstract

Increased ploidy is common in tumors but treatments fortumors with excess chromosome sets are not available. Here, wecharacterize high-ploidy breast cancers and identify potentialanticancer compounds selective for the high-ploidy state. Among354 human breast cancers, 10% have mean chromosome copynumber exceeding 3, and this is most common in triple-negativeand HER2-positive types. Women with high-ploidy breast can-cers have higher risk of recurrence and death in two patientcohorts, demonstrating that it represents an important groupfor improved treatment. Because high-ploidy cancers are aneu-ploid, rather than triploid or tetraploid, we devised a two-stepscreen to identify selective compounds. The screen was designedto assure both external validity on diverse karyotypic back-grounds and specificity for high-ploidy cell types. This screenidentified novel therapies specific to high-ploidy cells. First,

we discovered 8-azaguanine, an antimetabolite that is activat-ed by hypoxanthine phosphoribosyltransferase 1 (HPRT1),suggesting an elevated gene-dosage of HPRT1 in high-ploidytumors can control sensitivity to this drug. Second, we discovereda novel compound, 2,3-diphenylbenzo[g]quinoxaline-5,10-dione (DPBQ). DPBQ activates p53 and triggers apoptosis ina polyploid-specificmanner, but does not inhibit topoisomeraseor bind DNA. Mechanistic analysis demonstrates that DPBQelicits a hypoxia gene signature and its effect is replicated, inpart, by enhancing oxidative stress. Structure–function analysisdefines the core benzo[g]quinoxaline-5,10 dione as being nec-essary for the polyploid-specific effects of DPBQ. We concludethat polyploid breast cancers represent a high-risk subgroup andthat DPBQ provides a functional core to develop polyploid-selective therapy. Mol Cancer Ther; 15(1); 48–59. �2015 AACR.

IntroductionCancers are commonly aneuploid, with irregular karyotypes. In

some cases, during tumor evolution, cancers gain entire chromo-some sets and the resulting cells are polyploid (1, 2). Polyploidyconfers characteristic alterations to cellular physiology that couldbe exploited for cancer therapy (3–5), but clinical characteristicsof these tumors is unclear. One confounding factor in analysis is

that few cancers harbor precise triploid or tetraploid chromosomesets. Instead, tumors may double their genome at an incipientpoint in oncogenic transformation, by cell–cell fusion, endore-duplication, or cytokinesis failure (6–9). This genome doublingproduces tetraploid cells that are buffered from haploinsuffi-ciency, thereby permitting chromosome losses. In addition topermitting genomic abnormalities, the mechanism of genomedoubling will generally double centrosomes in concert. Theresulting supernumerary centrosomes can promote tumor evo-lution by enhancing chromosomal gains/losses (10). By bufferingagainst haploinsufficiency and enhancing chromosomal instabil-ity, the tetraploid state permits selection of high-ploidy aneuploidstates most permissive of tumor growth and resistance (11, 12). Itis unsurprising, therefore, that a genome doubling event canpromote oncogenic transformation in model systems (13, 14).

The tumor suppressor, p53 places an important restraint ongenome doubling. After genome-doubling event, human cellstypically elicit a p53-mediated cell-cycle arrest (15). This activa-tion can promote apoptosis and cell-cycle arrest. Additionally, afraction of cellsmay undergo interphase cytofission to resolve to adiploid state (16). Nevertheless, some tetraploid cells enter thecell cycle and continue to proliferate (17). Loss of p53 is permis-sive of continued cell-cycle progression after genome doubling(18), but this loss is not strictly required (17, 19). P53-mediatedarrest can also be bypassed by sustainedmitogenic signaling or bydownregulation of the hippo tumor suppressor pathway (15).The loss or bypass of p53 effect can further enhance permissive-ness of chromosomal instability. Thus, genome-doubling events

1University of Wisconsin Carbone Cancer Center, University of Wis-consin,Madison,Wisconsin. 2Hematology/OncologyDivision, Depart-ment of Medicine, University of Wisconsin School of Medicine andPublic Health, Madison, Wisconsin. 3Department of Cell and Regen-erativeBiology,UniversityofWisconsin School ofMedicine andPublicHealth, Madison,Wisconsin. 4Wisconsin State Laboratory of Hygiene,University of Wisconsin, Madison, Wisconsin. 5Department ofPediatrics, University of Wisconsin, University of Wisconsin Schoolof Medicine and Public Health, Madison, Wisconsin. 6Department ofPathology, University ofWisconsin, University ofWisconsin School ofMedicine and Public Health, Madison, Wisconsin. 7Department ofBiostatistics and Medical Informatics, University of Wisconsin Schoolof Medicine and Public Health, Madison,Wisconsin.

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

A. Choudhary and B. Zachek contributed equally to this article.

Corresponding Author: Mark E. Burkard, Wisconsin Institutes for MedicalResearch, Room 6059, 1111 Highland Avenue, Madison, WI 53705. Phone:608-262-2803; Fax: 608-265-6905; E-mail: [email protected]

doi: 10.1158/1535-7163.MCT-15-0527

�2015 American Association for Cancer Research.

MolecularCancerTherapeutics

Mol Cancer Ther; 15(1) January 201648

on March 12, 2019. © 2016 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst November 19, 2015; DOI: 10.1158/1535-7163.MCT-15-0527

http://mct.aacrjournals.org/

can be oncogenic, enhance chromosomal instability, and biologicrestraints are incomplete.

Here, our goal is to elucidate the clinical characteristics ofpolyploid breast cancer. We accomplish this by analyzing theploidy status of two sets of breast cancer, and characterizing theirclinical behavior. Additionally, we develop laboratory models ofgenome-doubled human cells and compare with the tumors.Finally, we develop a two-step screen to identify compounds thatare specific for high-ploidy cell types, and effective on cells withdiverse karyotypes. Our findings provide an important basis forthe idea of developing polyploid-selective cancer therapies.

Materials and MethodsPatient samples and FISH

The tissue microarray (TMA) used in this analysis has beendescribed previously (20). Briefly, primary tumor samples werearrayed and annotated under protocol OS10111 [InstitutionalReview Board (IRB) approval 2010-0405]. For FISH, slides weredeparaffinized, and treated with 0.2N HCl, 1 mol/L sodiumthiocyanate, Protease I, 10% formalin, and dehydrating ethanolseries. In situ DNA and probe were denatured by heating at 73�Cfor 2 minutes. Hybridization was performed over two nights at37�C with human-specific alpha-satellite-labeled centromereprobes for chromosomes 3 (D3Z1), 4 (D4Z1), 7 (D7Z1), 9(D9Z1), 10 (D10Z1), and 17 (D17Z1; Abbott Molecular). Slideswere mounted with Vectashield with DAPI (Vector LaboratoriesInc.). Chromosomes 4, 10, and 17 were probed on one section,and chromosomes 3, 7, and 9 on a second section. Chromosomeswere counted by observers blinded to patient conditions in aminimumof10 cells per case. A small fractionof sampleswere notevaluable due to loss of tissue, insufficient cellularity, or othertechnical issues and were excluded. Similarly, a subset of sampleshad a single probe that was poorly visualized; these were evalu-ated if two probes per slide were visualized. FISH data were linkedto de-identified clinical data by position on the TMAand analyzedusing the R statistical package and Excel (Microsoft).

To validatefindings, a secondpatient cohortwas obtained fromclinical laboratory FISH analysis using the chromosome 17 cen-tromeric probe. Clinical laboratory ploidy data were linked tooutcome data from the University of Wisconsin Carbone CancerCenter (UWCCC) Tumor Registry, whichwas audited for accuracyprior to removal of identifiers. This study was reviewed andapproved as OS12105 (IRB approval 2012-0196).

Data analysisKaryotypes for seven distinct cancer types were downloaded

from theMitelman database and analyzed formean chromosomenumber, and plotted in Prism (GraphPad Software).

Ploidy from theNCI-60 cell lines was obtained from themodalchromosome number (21). Drug sensitivity data were obtainedfrom the NCI Developmental Therapeutics Program (http://dtp.nci.nih.gov). The cell lines have known ploidy with modal chro-mosome numbers that range from 43 (2N; KM12 line) to 116(5N; SF-295 line; ref. 21). For each chemical, we calculated thecorrelation, r, of GI50 with cell line ploidy:

r ¼X

CellLines

ðg � �gÞðc� �cÞ

where g ¼ �log(GI50) and GI50 is the concentration that elicits50% growth inhibition, c is the modal number of chromosomes

in the cell line, and �g and �c aremean values of g and c across all celllines. Thus, r is positive for chemicals that selectively inhibitgrowth of polyploid cells. Data were analyzed using a scriptgenerated in Perl with an output of each chemical by NationalService Center (NSC) and associated value of r. These data werefurther processed and displayed graphically using Excel(Microsoft).

Cell culture, chromosome analysis, and microscopyCell lines were obtained from the ATCC in 2005 for RPE1, and

2012 for MCF10a; both were authenticated by karyotype in thisstudy. RPE1 and MCF10a cells were cultured in DMEMmixed withHam's F-12 modified medium (HyClone). Media was supplemen-ted with 100 U/mL of penicillin/streptomycin and 10% (vol/vol)FBS (RPE1) or 5% (vol/vol) horse serum (MCF10a). For MCF10a,media also included 20 ng/mL EGF, 0.5 mL/mL hydrocortisone,100 ng/mL cholera toxin, and 10 mg/mL insulin. MCF7 was obtain-ed from V.C. Jordan (University of Texas MD Anderson CancerCenter) in 2010 and verified by STR-15 analysis in 2015. MCF7cells were cultured in DMEM with 10% FBS. Chromosome spreadsand karyotypes were performed as previously described (16).

For FISH, microscopy was performed on Olympus BX41 andBX60 fluorescent microscopes using single and dual band filtersfor fluorophores DAPI, Spectrum Green, Spectrum Orange, andSpectrum Aqua. Other microscopy was performed by parafor-maldehyde cell fixation on coverslips followed by staining andindirect immunofluorescence, and analyzed on an automatedinverted Nikon Ti-E fluorescence microscope as described previ-ously (16).

Proliferation assaysCells were plated and treated with chemicals the following day.

For proliferation assays, adherent and nonadherent cells werecollected every 24 hours and counted via trypan blue exclusionwith a hemocytometer in three independent replicates. For auto-mated fluorescent analysis, cells were plated into 96-well plates,and treated the following day with chemicals. After 72 hours,SYBR green (Lonza) was added to a final volume of 140 mL, andincubated overnight. Fluorescence was read using a BioTek Syn-ergy 4 plate reader. Assays were performed in triplicate. For long-term cell proliferation assays, cells were exposed for 8 days tochemical treatment, washed with PBS, and stained with crystalviolet fixative solution for 20minutes. Cells were then rinsed anddried. A 1:1 methanol–water solution was added to each well for15 minutes. The solution was transferred to a 96-well plate andabsorbance (570 nm) was read on the Molecular Devices SpectraMax plus plate reader.

For dose–response analysis, data were normalized using func-tion zi¼ [xi�min(x)]/[max(x)�min(x)]. Normalized data fromat least three independent replicates were averaged and standarddeviation was calculated. For evaluation of DAUC we estimatedAUC for both the 2N and 4N and calculated the difference. Forcalculation of dose–response curves, data were analyzed in Prism,normalized to logarithmic concentration, and fit with a four-parameter dose–response curve (fitting bottom, top, IC50, andslope). Fit curves are shown and displayed IC50 values wereobtained from the fit.

Protein analysis and immunoblottingFor Western blot analysis, 2N and 4N RPE1 cells were

treated with DMSO, 50 ng/mL doxorubicin, or 1 mmol/L

Targeting High-Ploidy Breast Cancer

www.aacrjournals.org Mol Cancer Ther; 15(1) January 2016 49




2,3-diphenylbenzo[g]quinoxaline-5,10-dione (DPBQ) for 48hours. Western blotting was performed with a SDS-PAGE Elec-trophoresis and semidry transfer. Samples were analyzed bychemiluminescence.b-actinwasused as aprotein loading control.

Flow cytometryFor apoptosis assays, 2N and 4N RPE1 cells were treated with

DMSO, 60 nmol/L staurosporine, or 1 mmol/L DPBQ for 48hours. Apoptosis was evaluated by an Annexin V-PE/7-AADapoptosis detection kit (Ebiosciences). For the flow cytometryanalysis, the percent positive cells in the upper right (Annexin Vþ/7-AADþ; late apoptotic cells) and lower right quadrants (AnnexinVþ/7-AAD; early apoptotic cells) were summed to give the totalnumber of apoptotic cells. Three independent replicates wereperformed.

Gene expression analysisRNAwas isolated fromMCF7s�6-hour treatmentwith10mmol/L

DPBQ using TRIzol. cDNA was synthesized from 200 ng RNA andisolated using the AmbionMessageAmp Premier IVT kit (37�C for 8hours). Hybridization cocktail prepared according to protocols andprocedures in the AFX Expression Analysis Technical Manual (P/N702232 Rev. 3) for the U133A 2.0 array. A total of 6.5 mg offragmented labeled aRNA applied to AFX HG U133A 2.0 array andhybridized at 45�C for 16 hours in AFX 640 Hyb oven. HumanU133A 2.0 GeneChips were postprocessed on the AFX 450 FluidicsStation according to all AFX protocols and procedures defined forthe U133A 2.0 array (FS450_0002) in the GeneChip Hybridization,Wash, and Stain Kit User Manual (P/N 702231 Rev. 3). All Gene-ChipswerescannedonanAFXGC3000G7scanner (S/N50208130).Datawere extracted fromGC3000G7 scanned images using the AFXExpression Console v 1.2.0.20 software. Sample preparation andscanning was performed at the University of Wisconsin Biotechno-logy–GeneExpressionCenter. Array datawere normalizedbyRobustMultiarray Average method and the data exported to Gene SetEnrichment Analysis (GSEA; ref. 22). Data were deposited in theGeneExpressionOmnibus archive, accessionGSE73710 (www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc¼GSE73710).

Chemicals and reagentsDPBQ and other chemicals were obtained from the NCI,

diluted in stock DMSO solutions, and used at the concentrationsshown. Other chemicals used include doxorubicin (Thermo-Fisher), etoposide (Topogen), paclitaxel (ThermoFisher/Acros),BI-2536 (Selleck Chemical), chloroquine, 17-AAG, andNutlin-3a(Fisher). Antibodies used include b-actin (ab6276, WB 1:10,000;Abcam), pericentrin (ab4448 IF, 1:1,000; Abcam), p53 (#9282,WB 1:2,000; Cell Signaling Technology), phospho-P15 p53(#9284, WB 1:2,000; Cell Signaling Technology), and p21(sc6246, WB 1:500; Santa Cruz).

The topoisomerase assaywasperformedusing the Topo II Assaykit as per the manufacturer's instructions (TopoGEN). Circulardichroismof 0.5mg/mL salmon spermDNAwasperformedusingan AVIV Model 420 CD Spectrometer at room temperature. Toanalyze DNA binding, 100 mmol/L ethidium bromide or DPBQwas added to the sample to evaluate changes in spectrum con-sistent with intercalation.

Statistical analysisThe clinical outcomes analyzed in this study were relapse-free

survival (RFS) and overall survival (OS). RFS was defined as the

time from initial breast cancer diagnosis to recurrence or death;stage IV patients were excluded. OS was defined as the time fromdiagnosis to the date of death or last follow-up. RFS and OS wereplotted using the Kaplan–Meier method, and log-rank tests wereused to compare patientswith tumors thatwere polyploid (ploidy� 3) versus not polyploid (ploidy < 3). Cox proportional hazardsmodel included ploidy, stage, tumor grade, hormone receptorstatus, and HER2 status. Associations between these factors andeither RFS or OS were analyzed and presented as HR with 95%confidence intervals. The SAS (version 9.4, SAS Institute) and R(version 3.1.1) software packages were used for these analyses. P <0.05 was considered statistically significant.

ResultsTo characterize diversity in chromosome number across

diverse cancers, we visualized the mean chromosome numbersof diverse human cancers from the Mitelman database (23).Significant fractions of diverse tumors have greater than threechromosome sets, including breast, ovary, pancreas, lung, anddiffuse large–B-cell lymphoma (Fig. 1A). Interestingly, the lackof samples with 92 chromosomes demonstrates that true tet-raploidy is rare. We conclude that high-ploidy aneuploid statesare common and may represent a targetable feature of somecancers.

Polyploid cells are prone to genetic instability due to bothaberrations of division (10) and permissiveness to chromosomallosses (12). We therefore hypothesized that polyploid breastcancers would constitute a subtype with distinct clinical features.To address this, we employed a clinically annotated TMA of 354unique primary stage I to III breast cancer specimens frompatientstreated at the UWCCC. We performed FISH analysis of six chro-mosomes using centromeric probes (Fig. 1B, top). Polyploidcancers were defined as those that had mean chromosome num-bers of �3 and 10%met this criterion. Clinically, breast cancer iscategorized as hormone receptor positive (having estrogen orprogesterone receptor), HER2-positive (regardless of hormonereceptor status), or triple negative. High-ploidy breast cancerswere found among all three subsets. However, high ploidy iscommon in tumors of the triple-negative and HER2þ subtypes(21% of each; Fig. 1B).

Given increased permissiveness to genomic alterations, weanticipated that high tumor ploidy would confer worse clinicaloutcomes. To test this, we employed the Kaplan–Meier method(Fig. 1C–F). As predicted, patients with high ploidy had a signif-icantly higher risk of cancer relapse (P¼ 0.009) and of death (P¼0.009; Fig. 1C and D). To validate this in an additional cohort, weretrospectively reviewedoutcomes for 1,095 individuals treated atthe UWCCC in which FISH was performed for the centromere ofchromosome 17 as part of a standard clinical assay, and matchedto a database of clinical outcomes. Importantly, the ploidy ofchromosome 17 correlates well with 6-chromosome ploidy in thefirst cohort (Supplementary Fig. S1), suggesting that this chro-mosome is a reasonable surrogate for tumor ploidy. Again, hightumor ploidy is associated with higher risk of cancer relapse (P ¼0.006) and death (P¼ 0.049).Our findings led us to consider thathigh-ploidy status of a tumormight impart aggressive phenotypesleading to high tumor grade and stage in breast cancer. To test this,we performed a Cox proportional hazard analysis in both patientcohorts (Supplementary Table S1). Indeed tumor stage and graderemained strong predictors of poor clinical outcomes and

Choudhary et al.

Mol Cancer Ther; 15(1) January 2016 Molecular Cancer Therapeutics50




subsumed the prognostic effect of polyploidy. We conclude thatpolyploid breast cancers represent a high-risk subtype. Prognos-tically, this risk is captured in clinically observable variables,suggesting that polyploidy may promote advanced tumor gradeand stage. Yet, the finding that tumors with high ploidy haveworse clinical outcomes provides a strong clinical rationale forimproving treatment, and a biologic basis for specific anticancertherapy.

Physiologic effects of polyploidyThere is considerable evidence that polyploidy alters cell phys-

iology in diverse biologic systems (24, 25). To characterize theeffects in human epithelial cells, we obtained two diploid humanepithelial cell lines, RPE1 andMCF10a. RPE1 is an hTERT immor-talized retinal pigment epithelial cell line, and MCF10a is aspontaneously immortalized breast epithelial cell line (26). To

obtain genetically matched high-ploidy cells, we synchronizedcells in mitosis with brief nocodazole treatment, then chemicallyinterrupted cytokinesis using blebbistatin and generatedsubclones. As reported previously, many resulting binucleatedcells spontaneously undergo cytofission to resolve to a normaldiploid type (16). However, a fraction of the subclones recoveredare near tetraploid (4N). These near-tetraploid cells can undergofurther alterations in chromosome number due to centrosomeamplification and increased permissiveness of aneuploidy(10, 12). Because such genome evolution could complicatelater analyses, we selected lines that maintained a near-tetraploidchromosome number through >10 passages. As reported previ-ously (11, 27, 28), we found our polyploid cells to be enlargedcompared with their diploid counterparts (Fig. 2A). Flowcytometry and chromosome analysis confirmed tetraploidDNA content in the 4N cells (Fig. 2B and C, and Supplementary

11592694623

DLBCL

Colorectal

Lung

Pancreas

Ovary

RCC

Breast 14

19

27

31

34

23

9

Chromosomes

>3N, %3N

A

CEP4 CEP10 CEP17 DAPI

54321

HR+

HER2+

TNBC

Ploidy

21

21

5

Diploid Polyploid

>3N, %

B

Time (years)

PolyploidN = 158 (36 events)

NonpolyploidN = 937 (114 events)

840

1

0

1

0

Time (years)12840

TMA cohort, P = 0.009



DC

FE

Time (years)840

1

0

Time (years)840

1

0 TMA cohort, P = 0.009

12

Chromosome 17 cohort, P = 0.049Chromosome 17 cohort, P = 0.006

Rec

urre

nce-

free

sur

viva

l

Ove

rall

surv

ival

Rec

urre

nce-

free

sur

viva

l

Ove

rall

surv

ival





Figure 1.Polyploidy is a feature ofmany cancers. A,chromosome number in multiple cancertypes fromMitelman database. RCC, renalcell carcinoma; DLBCL, diffuse largeB-celllymphoma. B, top, representative three-color centromeric FISH images frombreast cancer tissue microarray (TMA)analysis of 354 distinct breast cancers.A total of 6 chromosomes were sampledby centromeric probes (4, 10, 17 on onesection and 3, 7, 9 on a separate section).Bottom, mean ploidy status by6-chromosome sampling of 354 breastcancers by subtype. C–F, Kaplan–Meiersurvival curves for polyploid versusnonpolyploid cancers. C, recurrence-freesurvival for TMA cohort where polyploidtumors are defined as those that havemean � 3 centromere signals per cellamong 6 centrosomes counted. D, overallsurvival for TMA cohort. E, relapse-freesurvival in a single-chromosome FISHcohort where polyploid tumors arethose defined as those that havemean� 3 centromere 17 per cell. F, overallsurvival for single-chromosome FISHcohort.






Fig. S2). Similarly, the tetraploid cells nearly double the proteincontent of diploids, consistent with expected gene dosageeffects (Fig. 2D and E). Although loss of p53 is permissive ofcontinued cycling after tetraploidization (29), our 4N cellsretained the ability to express and activate p53 (describedbelow). Cells arrested at the G1–S boundary commonly showedtwo pericentrin foci, suggesting supernumerary centrosomes(Fig. 2F). These findings demonstrate significant alterations incell physiology occur with tetraploidization, consistent withprior reports.

Given the association of high ploidy with high-risk breastcancer subtypes, we hypothesized that polyploidy might confera proliferative advantage. However, both tetraploid cell linesproliferated more slowly than their diploid counterparts,although the difference is modest for MCF10a (Fig. 2G). Theslowed proliferation of polyploid cells could arise from nonop-timal regulation of gene expression. If so, polyploid cancersmightadapt to recover rapid proliferation and polyploidy would notcorrelate with proliferation in human cancers. To test this, we

determined correlation of the ploidy of the human breast cancerswith Ki67, a standard marker of proliferation rate. High ploidydoes not correlate with decreased proliferation (Fig. 2H). Thus,high tumor ploidy does not necessarily slow proliferation, likelybecause of additional genetic adaptations that control prolifera-tion independent of ploidy.

Ploidy-specific antiproliferative compoundsPrevious work identified specific treatments that can have

differential effects on specific polyploid versus diploid cell types(4, 24). To testwhether existing drugsmight have this property,weevaluated detailed response curves for several anticancer agents.We first evaluated standard anticancer agents including doxoru-bicin, paclitaxel, and etoposide, but these lacked a selective effecton tetraploid cells (Supplementary Fig. S3A–S3C). Previous worksuggested that polyploid yeasts aremore dependent than diploidsonCdc5, the yeast homolog of humanPlk1 (25); however, we didnot observe a selective effect of BI-2536, an inhibitor of Plk1, onpolyploid human cells (Supplementary Fig. S3D). Chloroquine,

Human breast cancer

Doubling timeDoubling time

420

1

10

2N 44.6 ± 4.2 h4N 53.9 ± 4.4 h

MCF10a

4200.1

1

10

100

4N2N 19.0 ± 0.6 h

26.4 ± 1.0 h

RPE1

Rel

ativ

e pr

olife

ratio

n

Ki6

7

A

2N

4N

RPE1 MCF10a RPE1 karyotypesCB

2N

4N

2N 4N2N4N2N4N2N4N52.51.25

RPE1

7.5

Prot

ein

conc

entra

tion

(ng/

cell)

0

0.5

1

1.5

MCF10aRPE1

2N4N

× 10 cells4

2N 4N

D FE

543210

20

40

60

80

100

PloidyDaysDays

HG

Cel

l cou

ntDNA content

RPE1

MCF10a

2N

2N

4N

4N

Actin DAPI

Actin DAPI Pericentrin98% 68%

Figure 2.Matched pairs of diploid–polyploidcells were generated from epithelialcell types. A, representative images ofdiploid (2N) and polyploid (4N) cellsderived from RPE1 and MCF10a celllines. Scale bar, 10 mm. B, flowcytometry demonstrating double DNAcontent of tetraploid cells after stainingwith propidium iodide. C, karyotype ofdiploid- and polyploid-matched RPE1cells. D and E, protein levels are higherin tetraploid cells relative to diploid asdetermined by Coomassie stainingof extracts (D) and Bradford Assay(E). F, polyploid RPE1 cells havesupernumerary centrosomes. Cellswere synchronized using aphidicolinand stained. Pericentrin foci aremarked with arrows and percentage ofcells that harbor the indicated numberof centrosomes is indicated (n > 40). G,4N cells proliferate more slowly than2N cells. Cell number was quantified inproliferating 2N and 4N cells anddoubling time was calculated. H,high ploidy in human breast cancersamples does not correlate stronglywith the Ki67 proliferation marker(R2 ¼ 0.04).

Choudhary et al.





17-AAG are aneuploidy-selective compounds (5), so we hypoth-esized that these might also operate selectively on polyploidtumors. Yet chloroquine and 17-AAG did not show polyploid-selective effects in our matched pair of diploid and tetraploidRPE1 cells (Supplementary Fig. S3E and S3F). These agents maybe selective for cells with near-diploid aneuploidy, generatingan unbalanced gene dose not found in our tetraploid cell linemodels. Thus, existing or potentially predicted anticancer agentslack polyploid-specific anticancer effects.

We sought to identify anticancer agents specific for high cellploidy, but not necessarily specific to the near-tetraploid state ofour model cell types, because many high-ploidy tumors havediverse aneuploid karyotypes. Thus, we designed a two-stepscreen that inherently addresses the diversity of high-ploidykaryotypes found in human cancer. To do this, we obtainedchemical sensitivity data from the NCI-60 database (30). The

data consist of 50% growth-inhibitory concentrations (GI50) of45,342 chemicals across approximately 60 human cancer celllines. For each chemical, we calculated the correlation of drugsensitivity (GI50) against ploidy and plotted distribution of thisscore, r, against the interline-variance in GI50 (Fig. 3A). A highinterline variance (vertical axis in Fig. 3A) indicates that otherfactors in addition to polyploidy contribute to drug selectivity(e.g., p53withDPBQ, describedbelow).Ofnote, the vastmajorityof chemicals had little differential effect for polyploid versusnonpolyploid cells. Those chemicals that lack selectivity includedoxorubicin, etoposide, and paclitaxel; of standard chemothera-py drugs, only vincristine is found to have weak correlation withincreased sensitivity of polyploid cells (Table 1). Nevertheless, asmall proportion of compounds elicit sensitivity that correlateshighlywith ploidy on a diverse set of human cancer cell types, andare worthy of further analysis.

50–5Polyploid selectivity, ρ

Inte

rline

varia

nce

High-ploidyselective

compounds

Low-ploidyselective

compounds

A

∆AUC

4N-selective compound∆AUC > 0

2N-selective compound∆AUC < 0

100100

1

4N2N

100100

1

4N 2N

–50

∆AUCRPE

∆AUCMCF10

50–50

50

8-AzaguanineDPBQ

B2N

0

0.4

1.0

2.5

6.3

(DPB

Q, m

mol

/L)

4NC

ED

(8-Azaguanine, mmol/L)101 100

2N4NR

elat

ive

coun

t

1.0

0.5

0

Rel

ativ

e co

unt

1.0

0.5

0

(DPBQ, mmol/L)0.1 1 10

2N4N

NSC7436

Figure 3.Discovery of DPBQ and 8-azaguanineas polyploid-specific disruptors ofproliferation. A, correlation with drugsensitivity and cell ploidy from NCI-60data of over 45,342 chemicals acrossapproximately 60 cell lines. Polyploidselectivity, r, is plotted against interlinevariance for each chemical. Red circlesindicate chemicals obtained forsecondary screens. B, secondary screento test selectivity of hits in matchedpairs of diploid and polyploid RPE1 andMCF10a cells. Red circles indicate hitsthat are polyploid selective in bothMCF10a and RPE1 cell types. C, DPBQ 8-day proliferation assay in RPE1 cells.Crystal violet stainingmarks viable cellsremaining. D and E, quantification ofproliferation assay as in C. Plottingabsorbance for 8-azaguanine (D) andDPBQ (E) in RPE1. n ¼ 3; SD shown.






Although this analysis identifies compounds with sensitivitythat correlates with cell ploidy, it does not demonstrate thatploidy per se controls selectivity because it does not excludeunknown confounding variables. To determine if the responseto these compounds is intrinsic to polyploidy, we validated themin matched diploid–polyploid human cells. We procured avail-able chemicals with the highest ploidy-specific growth inhibitionscores (red circles in Fig. 3A) and established concentration–response curves on the matched 2N and polyploid (4N) RPE1and MCF10a cells. We measured the difference in the areasunder the curve between 2N and 4N lines (Fig. 3B, left) andplotted results for each chemical. As shown, few chemicalshave confirmed polyploid selectivity in both paired cell lines(Fig. 3B, right), suggesting that the correlation with ploidy formost chemicals is due to confounding factors. However, a smallnumber of compounds had confirmed selectivity in both poly-ploid RPE1 and MCF10a lines.

One intriguing hit is 8-azaguanine, an antimetabolite that isactivated by the purine salvage enzyme hypoxanthine phospho-ribosyltransferase 1 (HPRT1), encoded on the X-chromosome. 8-Azaguanine can be used to select cells that lack HPRT1, which arethen resistant. Our data here suggest that polyploid cells, whichhave additional X chromosomes, have enhanced sensitivity to 8-azaguanine possibly due to elevated expression of HPRT1; excessX chromosomes are not inactivated in polyploid cells (31).Second, we identified NSC7436 (4-amino-N-(4-tert-butylphe-nyl)benzenesulfonamide), a sulfonamide compound whichwas not further characterized here. A third hit discovered isDPBQ.To quantify themagnitude of the 8-azaguanine andDPBQ effects,we performed detailed concentration–response curves in 8-dayproliferation assays onmatched diploid-polyploid cells (Fig. 3C–E). This confirmed significant selectivity for polyploid celltypes. We conclude that these are bona fide polyploid-selectivecompounds.

DPBQ does not have a knownmechanism of action, so we firsttested the hypothesis that it may operate similarly to existingcancer therapeutics. To identify potential matches, we used thePrediction of Activity Spectra for Substances (PASS) score, whichis available for all compounds in the NCI-60 database (32). PASSestimates the probability that a given compound has one of the565 biologic activities based on known activities of a learning setof approximately 35,000 compounds. We obtained a PASS scoreof 0.8 (range, 0–1) for DPBQ as a topoisomerase inhibitor. Wewere initially puzzledby thisfindingbecause other topoisomeraseinhibitors lacked selectivity in our in silico screen and bothdoxorubicin and etoposide failed to exhibit any differential effectin diploid and tetraploid RPE1 in separate assays (SupplementaryFig. S2). Nevertheless, we directly evaluated DPBQ activity in atopoisomerase II assay, and foundnoactivity (Supplementary Fig.S4A).Moreover, we observed that the planar aromatic structure of

DPBQresemblesDNA intercalators, butwedidnot detect bindinga direct assay by circular dichroism (Supplementary Fig. S4B). Weconclude that DPBQ mechanism appears distinct from DNAbinding or inhibition of topoisomerase II.

Mechanism of DPBQ actionPreliminary data suggested that DPBQ caused cancer cell death

rather than inhibition of proliferation. To evaluate the cell bio-logic effects of DPBQ, we evaluated mechanisms of death byAnnexin and 7-AAD staining to detect apoptotic/necrotic cellpopulations (Fig. 4AandB). These results demonstrate thatDPBQelicits apoptosis and cell death and is selective for effects in 4Ncells. The tumor suppressor p53 is a central mediator of apoptosisfrom chemically induced stress (33). We therefore reasoned thatDPBQ may elicit p53 activation to produce the observed apo-ptosis. Indeed, DPBQ elicits expression and phosphorylation ofp53 and this effect is specific to tetraploid cells (Fig. 4C). Addi-tionally, this is bona fide activation of p53 transcriptional activityas it results in expression of p21, a downstream effector. Incontrast, doxorubicin causes activation of p53 in both diploidand tetraploid cells, consistent with the lack of cell line–specificselectivity. To test if p53 mediates the antiproliferative effect ofDPBQ in polyploid cells, we knocked down p53 and reanalyzedantiproliferative effects. Indeed, knockdown restores prolifera-tion of tetraploid cells in the presence of DPBQ (Fig. 4D). Weconclude that DPBQ elicits 4N-selective apoptosis mediatedby p53.

If p53 is indeed a mediator of DPBQ effect on polyploid cells,then we would anticipate that cell lines with high ploidy andintact p53 would be most sensitive to DPBQ. The p53 status isknown for most cell lines in the NCI-60 panel (34). Consideringthose for which p53 is known to be wild type, we find thatsensitivity to DPBQ is highest in cells that have both intact p53and high ploidy (Fig. 4E). In contrast, there is no correlation ofdrug sensitivity with p53 status or ploidy for other agents includ-ing 8-azaguanine and the agents in Supplementary Fig. S2. Weconclude that DPBQ activates p53 and induces apoptosis in apolyploid-selective manner.

To further elucidate mechanism of polyploid-selective p53activation, we tested how 6-hour DPBQ alters gene expression.We initially matched the conditions of the connectivity mapin the hopes of identifying a matching pharmacophore (35).Although we identified DPBQ-matching pharmacophores,these did not match the polyploid-selective effect of DPBQ.These results suggest that this agent operates by a uniquemechanism. To narrow down specific mechanisms, wereviewed the effects of DPBQ on gene expression (Fig. 5A) andperformed gene set enrichment analysis against standardizedhallmark profiles (22). As expected, the enrichment profilematched the canonical effect of p53 activation, but we alsoidentified the hypoxia hallmark, which could explain mecha-nism of p53 activation (Fig. 5B). Hypoxia is known to induceoxidative stress and can have a reciprocal effect on p53activation (36). We therefore evaluated whether oxidativestress can have polyploid-selective effects. In contrast to othermethods of activating p53, we found that oxidative stressfrom H2O2 and is modestly polyploid-selective in inducingantiproliferative effects (Fig. 5C) and induces polyploid-selec-tive p53 activation (Fig. 5D). Thus, DPBQ activates p53 in apolyploid-selective manner at least partially through oxidativestress.

Table 1. Correlation of sensitivity with ploidy for standard cancer therapies

Drug r

Vincristine 1.7Chloroquine 0.7Doxorubicin 0.4Etoposide 0.1Paclitaxel �0.417-AAG �0.6Methotrexate �0.85-FU �1.1Gemcitabine �1.9

Choudhary et al.





Structure–function analysisDPBQ has a planar oxidized tricyclic structure with benzyl

substituents. To evaluate which components of the structureare critical for its polyploid-specific effects, we procured relatedcompounds that were readily available (Fig. 6A). First, we testedtwo related compounds that are missing the complete ben-zo[g]quinoxaline-5,10-dione structure, DQD and QXD (Fig. 6Band C). However, these lacked not only selectivity but alsopotency, as demonstrated by the high concentrations requiredto elicit antiproliferative effects. Similarly, potency and selectivitywere lost in BQD (Fig. 6D), which lacks both the phenyl sub-stituents and a heterocyclic nitrogen. In contrast, selectivity andpotency were maintained in NQD (Fig. 6E), in which the corebenzo[g]quinoxaline-5,10 dione structure was retained. Impor-tantly, this suggests that the phenyl groups are dispensable andthat modifications of the nitro- group could be made to improvepotency, selectivity, or pharmacologic properties. Thus, thisbenzo-quinoxaline-dione structure represents a core structure fordevelopment of polyploid-selective therapeutics.

DiscussionCancers have diverse karyotypes with highly divergent chro-

mosome numbers ranging from 33 to 133 in common cancers(23). The presence of supernumerary chromosome sets alterscell physiology in replication, cell size, mitosis, gene expres-sion, and tolerance for mutation (1, 11, 12, 24, 25, 27, 28).Previous work demonstrates that human tetraploid cells haveelevated sensitivity to certain drugs, resistance to others, andaltered regulation of apoptosis (4, 24). Although tetraploid-specific drugs could be useful for prevention, cancer cells oftenhave additional mutations, chromosomal aberrations, or epi-genetic changes that result in deviation from a purely tetra-ploid karyotype. In support of this idea, no human cancer cellsof the common types shown in Fig. 1A have strictly tetraploidchromosome content, as all have one or more chromosomegains/losses, creating a high-ploidy aneuploid state. Moreover,high ploidy does not correlate with low proliferation inhuman breast cancer, although laboratory-derived near-

2N 4N

p53

Actin

P(S15) p53Actin

p21

Actin

DoxDPBQ

+–– – –+––– + +–

CP

loid

y

NCI-60 cell lines

wt p53 ++++++++++++

4N

2N

DPBQ8-Azaguanine

PaclitaxelDoxorubicin

Etoposide17-AAG

Chloroquine

KeyResistant (high GI50)NeutralSensitive (low GI50)

D

Annexin PE7-

AA

D

A StaurosporineDMSO DPBQ

2N

4N

B

1.192

1.74.8

12

138.3

1.890

4.13.9

8.479

9.13.2

1.091

3.14.9

7.681

7.83.3

67

Untreate

dDPBQ

Stauro

sporin

e0

10

20

30

% A

popt

osis

2N4N *

Prol

ifera

tion

1

0

siCtrl

siCtrl+DPBQsiTP53

siTP53+DPBQ

DoxorubicinsiCtrl

siTP53

DPBQ

++

++

E

*

Figure 4.Mechanism of DPBQ. A and B, DPBQelicits polyploid-specific apoptosis. A,apoptosis by representative Annexinassay. B, averaged apoptosis (early andlate) for n ¼ 3 assays, SD shown. � , P <0.05 by t test. C, 1 mmol/L DPBQ elicits4N-specific p53 induction andactivation; dox, doxorubicin. D, p53 isrequired for the DPBQ effect. 4N RPE1cells were transfected with siRNAagainst p53 (siTP53) or control (siCtrl)and then exposed to DPBQ or vehicle.DPBQ restrained proliferation onlywhen p53 was present (red). Right, blotdemonstrating suppression of phospho(S15)-p53 with knockdown. � , P < 0.05by the t test. E, among NCI-60 lines,DPBQ has its strongest effects againstpolyploid cell lines that express wild-type p53.






tetraploid lines do proliferate more slowly. These observationssuggest that human cancers with high-ploidy aneuploid stateshave additional adaptations that need to be accounted for indiscovery of polyploid-selective cancer therapies.

Polyploidy is found in a significant fraction of high-risk breastcancers. Although it comprises only 10% to 14% of breast cancer,

this represents 20 to 28,000 cases per year. Moreover, we find thatpolyploid breast cancers have worse clinical outcomes thannonpolyploid cancers, indicating a need for effective and spe-cific therapies for these patients. This study associates poly-ploidy with poor clinical outcomes, but other evidence suggeststhat polyploidy causes cancer. For example, polyploidy is

A

–DPBQ

+DPBQ

CTG

FA

TF3

HS

PA

6R

GS

2M

AFF

GE

ME

GR

3R

GC

32H

SP

A6

AN

KR

D1

PM

AIP

1E

GR

4A

RE

GN

R4A

3IL

8H

BE

GF

EG

R2

BIR

C3

FOS

L1D

DIT

3LI

FA

RL1

4M

T1F

NR

4A2

CY

R61

GD

F15

PH

LDA

1E

GR

1G

AB

AR

AP

L1D

US

P1

DN

AJB

4G

AD

D45

AP

P1R

15A

IL18

HM

OX

1S

CG

5JA

G1

CH

AC

1R

GS

20E

MP

1M

DM

2IR

S2

TIP

AR

PC

AM

TA1

CG

AC

LDN

1C

XC

R4

PTG

ER

4G

AD

D45

BFO

SC

15O

RF2

8K

CN

V1

BC

L11B

STO

N1

PIT

X2

MP

PE

D2

PLC

H1

NC

AM

2S

OX

2H

IST1

H2B

GG

PR

109B

INA

DL

LOC

1507

59FA

M5C

VP

S13

BFL

J235

56S

LC4A

8C

OL4

A6

TRIB

2E

ZD2

TTC

30A

C11

OR

F32

EG

LN3

NB

EA

PTC

H1

PIK

3R3

US

TK

IAA

1005

ZC3H

AV

1D

RD

1S

NE

1G

YP

AH

OX

C6

GR

AM

D1C

CB

FA2T

3K

IAA

1107

AD

AM

TSL3

CD

H18

PE

X11

AM

ELS

1G

PR

30C

1OR

F25

DM

XL1

FLJ2

2655

ZNF2

48N

OTC

H3

SG

CG

ER

C2

TRE

RF1

DA

CH

1

B

D

C

1010.1

0.0

0.5

1.0

(Nutlin, µµmol/L)

Rel

ativ

e co

unt

2N4N

1001010.1

0.0

0.5

1.0

(CaCl2, µmol/L)

Rel

ativ

e co

unt

2N4N

1,00010010

0.0

0.5

1.0

(H2O2, µmol/L)

Rel

ativ

e co

unt

2N4N

CtrlDox H2O2

4N2N4N2N4N2N

p-p53

Actin

Figure 5.Evaluation of DPBQ mechanism by gene expression analysis. A, top gene expression altered by DPBQ is shown, both upregulated (red) and downregulated(blue) in three technical replicates. B, gene set enrichment analysis showing strong enrichment of hallmark pathways for p53 and hypoxia. C, concentration–response curves for nutlin, CaCl2, and H2O2 on diploid and tetraploid cells. D, expression of phospho-p53 (top) and actin after treatment withdoxorubicin, control, or H2O2.

Choudhary et al.





common in early stages of malignant transformation of theesophagus (37) and of the cervix (38). Certain hereditarycancer syndromes also increase the rate of cytokinesis failureand concomitant tetraploid tumors; these include hereditarymutations in BRCA2 (39, 40) and APC (41–43). Once cells aretetraploid, supernumerary centrosomes can promote abnormalchromosome segregation leading to gains and losses of chro-mosomes (2, 30). Polyploid cells have a greater propensity foroncogenic transformation than diploid cells in zebrafish (14)and in p53�/� mice (13). Together, these observations supportthe idea that genome doubling is an early event in oncogenictransformation in some cancers, and suggests that high ploidymay also produce aggressive cancer phenotypes.

DPBQ is a novel lead compound discovered to exhibitselective lethality for polyploid cells. The screen used is inno-vative in that the primary screen accounts for the genetic andphysiologic diversity of cancer, thereby reducing the likelihoodof finding a hit restricted to a specific genetic background or celltype. Moreover, the secondary screen distinguishes polyploidy-specific effects by testing in matched diploid–polyploid cell linepairs. Of note, many of the chemical hits in the primary screendo not have specificity for polyploid cells, demonstrating thatother factors sometimes underlie the correlation seen with drugsensitivity and polyploidy for certain chemicals. Thus, this two-step screen provides validation that is both intrinsic (polyploid-specific) and extrinsic (diverse cancer cell types and geneticbackgrounds).

In addition to DPBQ, we identified 8-azaguinine as anadditional compound with polyploid-specific effects. 8-Aza-guanine is an antimetabolite that interferes with nucleotidesynthesis after it is activated by HPRT1. Similar compoundssuch as mercaptopurine and 6-thioguanine are used to treatacute lymphocytic leukemia (44). Cells with inactive HPRT1are resistant to the effects to these antimetabolites, but here-

tofore it has not been demonstrated that polyploid tumorsmight have enhanced sensitivity to these agents. The increasedsensitivity of polyploid tumors here suggests a novel biomarkerfor sensitivity to existing anticancer drugs. In addition, poly-ploidy would decrease the likelihood of secondary resistancethrough HPRT1 loss of function as human polyploid cells havemultiple activated X chromosomes (31).

The finding of enhanced sensitivity of tetraploid cells toH2O2 is consistent with prior observations that tumorigenicpolyploid cells have enhanced reactive oxidative species (ROS)signaling (45). However, this does not completely account forthe mechanism by which DPBQ elicits p53 activation. First, wefind no selectivity with other ROS activators, such as doxoru-bicin. Second, the preferential effect of DPBQ on polyploidversus diploid cells is larger than that of H2O2. These findingssuggest that DPBQ either elicits a particular type of oxidativestress or also operates in additional ways that have a strongselectivity to polyploid cells. It will be important to elucidatethe details of this mechanism.

Despite the strength of the screen, there are some limitationsof the findings. The effects of DPBQ appear to be limitedto high-ploidy cell types with intact p53, and a significantfraction of cancers have lost the function of this tumor sup-pressor. However, there is good reason to anticipate that somepolyploid cancers will retain p53. Loss of p53 is not strictlyrequired for development of polyploid cancers, and can beovercome by mitogenic growth signals or loss of hippo path-way (46). Indeed, cell-cycle progression and mitotic entry hasbeen observed in p53-wild-type cells after cytokinesis failure asreported by others (17, 19). Finally, we identify high-ploidyNCI-60 cell lines reported to have wild-type p53. A substantialfraction of high-risk breast cancer subtypes appear to have wild-type p53, although it remains unclear how this overlaps withthe polyploid fraction. Thus, it is plausible that a significant

AO

O

N

N

O

O

N

N

O

O

N

N

O

O

N

O

O

N

N

NO2

DPBQNSC629713

DQDNSC128830

QXDNSC101615

BQDNSC281974

NQDNSC362342

Potency(IC504N, μmol/L)

Selectivity(IC502N/IC504N)

6.1 1.3

6.6 1.1

0.2 2.9

10.3 1.1

0.4 6.0

0.0

0.5

1.0

101

Rel

ativ

e co

unt

2N4N

0.0

[DQD, mmol/L] [QXD, mmol/L]

[NQD, mmol/L][BQD, mmol/L]

0.5

1.0

10010

Rel

ativ

e co

unt 2N

4N

0.0

0.5

1.0

1010.1

Rel

ativ

e co

unt 2N

4N

0.0

0.5

1.0

10

Rel

ativ

e co

unt

2N4N

B C

D E

Figure 6.Structure–function analysis of DPBQ. A, DPBQ-like molecules were tested for potency and specificity for in paired diploid–polyploid RPE1 cells. B–E, concentrationresponse curves for DPBQ-like molecules for 8-day proliferation assays, as in Fig. 3C–E. n ¼ 2; SD shown.






fraction of breast and other cancers may have the characteristicsrequired for sensitivity to DPBQ-like drugs.

We anticipate that the therapeutic window for DPBQ (6-fold-selectivity) is sufficient for an effective therapeutic, and might beenhanced by pharmacologic modification.

We conclude that polyploidy correlates with, and may cause,high-stage and high-grade breast cancer with worse clinical out-comes. This finding provides insight into an underlying cause ofsome high-risk breast cancers. Moreover, we have discoveredcompounds that exploit the underlying differences of polyploidversus diploid cell types, and may serve as a lead compound fornovel cancer therapy. Given that high-ploidy aneuploid statesexist in a significant numbers of cancers, targeting polyploidycould be an effective therapeutic strategy.

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

Authors' ContributionsConception and design: A. Choudhary, B. Zachek, L.M. Zasadil, B.A. Weaver,M.E. BurkardDevelopment of methodology: A. Choudhary, B. Zachek, R.F. Lera, C. Kanugh,M.E. BurkardAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): A. Choudhary, R.F. Lera, L.M. Zasadil, A. Lasek,H. Kim, J.J. Laffin, J.M. Harter, K.B. Wisinski, B.A. Weaver, M.E. BurkardAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): A. Choudhary, B. Zachek, R.F. Lera, L.M. Zasadil, R.A.Denu, S. Saha, M.E. Burkard

Writing, review, and/or revision of the manuscript: A. Choudhary, B. Zachek,R.F. Lera, L.M. Zasadil, A. Lasek, R.A. Denu, H. Kim, K.B. Wisinski, B.A. Weaver,M.E. BurkardAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): A. Choudhary, B. Zachek, A. Lasek, C. Kanugh,M.E. BurkardStudy supervision: M.E. Burkard

AcknowledgmentsThe authors acknowledge the UW Biophysics Instrumentation Facility (NSF

BIR-9512577 and NIH S10 RR13790), and the University of Wisconsin Bio-technology Center–Gene Expression Center for equipment and expertise. Theauthors acknowledge Amy Fothergill, Ross Fedenia, John Feltenberger, MelissaMartowicz, and Robert Millholland for contributions of clinical data, reagents,and preliminary analyses.

Grant SupportThis work was supported by R01 GM097245, the Mary Kay Foundation,

American Cancer Society IRG-58-011-48, the Wisconsin Partnership NewInvestigator Program Award #2261, and Effcansah (M.E. Burkard), and R01CA140458 (B.A. Weaver). Additional support was provided by a training grantfrom the National Center for Advancing Translational Sciences UL1TR000427(M.E. Burkard), and Cancer Center Support P30 CA014520 supporting collab-orative resources for all investigators.

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 June 23, 2015; revised November 5, 2015; accepted November 9,2015; published OnlineFirst November 19, 2015.

References1. Storchova Z, Kuffer C. The consequences of tetraploidy and aneuploidy.

J Cell Sci 2008;121:3859–66.2. Storchova Z, Pellman D. From polyploidy to aneuploidy, genome insta-

bility and cancer. Nat Rev Mol Cell Biol 2004;5:45–54.3. Janssen A, Kops GJ, Medema RH. Elevating the frequency of chromosome

mis-segregation as a strategy to kill tumor cells. Proc Natl Acad Sci U S A2009;106:19108–13.

4. Lissa D, Senovilla L, Rello-Varona S, Vitale I, MichaudM, Pietrocola F, et al.Resveratrol and aspirin eliminate tetraploid cells for anticancer chemo-prevention. Proc Natl Acad Sci U S A 2014;111:3020–5.

5. Tang YC, Williams BR, Siegel JJ, Amon A. Identification of aneuploidy-selective antiproliferation compounds. Cell 2011;144:499–512.

6. Duelli DM, Hearn S, Myers MP, Lazebnik Y. A primate virus generatestransformed human cells by fusion. J Cell Biol 2005;171:493–503.

7. Edgar BA, Orr-Weaver TL. Endoreplication cell cycles: more for less. Cell2001;105:297–306.

8. Lacroix B, Maddox AS. Cytokinesis, ploidy and aneuploidy. J Pathol2011;226:338–51.

9. Lanni JS, Jacks T. Characterization of the p53-dependent postmitoticcheckpoint following spindle disruption. Mol Cell Biol 1998;18:1055–64.

10. Ganem NJ, Godinho SA, Pellman D. A mechanism linking extra centro-somes to chromosomal instability. Nature 2009;460:278–82.

11. Comai L. The advantages and disadvantages of being polyploid. Nat RevGenet 2005;6:836–46.

12. Dewhurst SM, McGranahan N, Burrell RA, Rowan AJ, Gronroos E, Endes-felder D, et al. Tolerance of whole-genome doubling propagates chromo-somal instability and accelerates cancer genome evolution. Cancer Discov2014;4:175–85.

13. Fujiwara T, Bandi M, Nitta M, Ivanova EV, Bronson RT, Pellman D.Cytokinesis failure generating tetraploids promotes tumorigenesis inp53-null cells. Nature 2005;437:1043–7.

14. Shepard JL, Amatruda JF, Finkelstein D, Ziai J, Finley KR, Stern HM, et al. Amutation in separase causes genome instability and increased susceptibil-ity to epithelial cancer. Genes Dev 2007;21:55–9.

15. Ganem NJ, Cornils H, Chiu SY, O'Rourke KP, Arnaud J, Yimlamai D, et al.Cytokinesis failure triggers hippo tumor suppressor pathway activation.Cell 2014;158:833–48.

16. Choudhary A, Lera RF, Martowicz ML, Oxendine K, Laffin JJ, WeaverBA, et al. Interphase cytofission maintains genomic integrity ofhuman cells after failed cytokinesis. Proc Natl Acad Sci U S A 2013;110:13026–31.

17. Uetake Y, Sluder G. Cell cycle progression after cleavage failure: mamma-lian somatic cells do not possess a "tetraploidy checkpoint". J Cell Biol2004;165:609–15.

18. Davoli T, de Lange T. The causes and consequences of polyploidy innormaldevelopment and cancer. Annu Rev Cell Dev Biol 2011;27:585–610.

19. Wong C, Stearns T. Mammalian cells lack checkpoints for tetraploidy,aberrant centrosome number, and cytokinesis failure. BMC Cell Biol2005;6:6.

20. Wang L, Zhao Z, Meyer MB, Saha S, YuM, Guo A, et al. CARM1methylateschromatin remodeling factor BAF155 to enhance tumor progression andmetastasis. Cancer Cell 2014;25:21–36.

21. Roschke AV, Tonon G, Gehlhaus KS, McTyre N, Bussey KJ, Lababidi S, et al.Karyotypic complexity of the NCI-60 drug-screening panel. Cancer Res2003;63:8634–47.

22. SubramanianA, TamayoP,Mootha VK,Mukherjee S, Ebert BL,GilletteMA,et al. Gene set enrichment analysis: a knowledge-based approach forinterpreting genome-wide expression profiles. Proc Natl Acad Sci U S A2005;102:15545–50.

23. Mitelman F, Johansson B, Mertens F. Mitelman database of chromosomeaberrations and gene fusions in cancer. 2012. Available from: http://cgap.nci.nih.gov/Chromosomes/Mitelman

24. Castedo M, Coquelle A, Vivet S, Vitale I, Kauffmann A, Dessen P,et al. Apoptosis regulation in tetraploid cancer cells. EMBO J 2006;25:2584–95.

25. Storchova Z, Breneman A, Cande J, Dunn J, Burbank K, O'Toole E, et al.Genome-wide genetic analysis of polyploidy in yeast. Nature 2006;443:541–7.


Choudhary et al.




26. Soule HD, Maloney TM, Wolman SR, Peterson WD Jr., Brenz R, McGrathCM, et al. Isolation and characterization of a spontaneously immor-talized human breast epithelial cell line, MCF-10. Cancer Res1990;50:6075–86.

27. Epstein CJ. Cell size, nuclear content, and the development of polyploidyin the Mammalian liver. Proc Natl Acad Sci U S A 1967;57:327–34.

28. Galitski T, Saldanha AJ, Styles CA, Lander ES, Fink GR. Ploidy regulation ofgene expression. Science 1999;285:251–4.

29. Senovilla L, Vitale I, Galluzzi L, Vivet S, Joza N, Younes AB, et al. p53represses the polyploidization of primary mammary epithelial cells byactivating apoptosis. Cell Cycle 2009;8:1380–5.

30. Shoemaker RH. TheNCI60 human tumour cell line anticancer drug screen.Nat Rev Cancer 2006;6:813–23.

31. Monkhorst K, Jonkers I, Rentmeester E, Grosveld F, Gribnau J. X inactiva-tion counting and choice is a stochastic process: evidence for involvementof an X-linked activator. Cell 2008;132:410–21.

32. Anzali S, Barnickel G, Cezanne B, Krug M, Filimonov D, Poroikov V.Discriminating between drugs and nondrugs by prediction of activityspectra for substances (PASS). J Med Chem 2001;44:2432–7.

33. Lowe SW, Ruley HE, Jacks T, Housman DE. p53-dependent apo-ptosis modulates the cytotoxicity of anticancer agents. Cell 1993;74:957–67.

34. BerglindH, Pawitan Y, Kato S, IshiokaC, Soussi T. Analysis of p53mutationstatus in human cancer cell lines: a paradigm for cell line cross-contam-ination. Cancer Biol Ther 2008;7:699–708.

35. Lamb J, Crawford ED, Peck D, Modell JW, Blat IC, Wrobel MJ, et al. TheConnectivity Map: using gene-expression signatures to connect smallmolecules, genes, and disease. Science 2006;313:1929–35.

36. Sermeus A, Michiels C. Reciprocal influence of the p53 and the hypoxicpathways. Cell Death Dis 2011;2:e164.

37. Barrett MT, Pritchard D, Palanca-Wessels C, Anderson J, Reid BJ, Rabino-vitch PS. Molecular phenotype of spontaneously arising 4N (G2-tetra-ploid) intermediates of neoplastic progression in Barrett's esophagus.Cancer Res 2003;63:4211–7.

38. Olaharski AJ, Sotelo R, Solorza-Luna G, Gonsebatt ME, Guzman P, MoharA, et al. Tetraploidy and chromosomal instability are early events duringcervical carcinogenesis. Carcinogenesis 2006;27:337–43.

39. Daniels MJ, Wang Y, Lee M, Venkitaraman AR. Abnormal cytokinesis incells deficient in the breast cancer susceptibility protein BRCA2. Science2004;306:876–9.

40. Jonsdottir AB, Stefansson OA, Bjornsson J, Jonasson JG, OgmundsdottirHM, Eyfjord JE. Tetraploidy in BRCA2 breast tumours. Eur J Cancer2012;48:305–10.

41. Caldwell CM, Green RA, Kaplan KB. APC mutations lead to cytokineticfailures in vitro and tetraploid genotypes in Min mice. J Cell Biol2007;178:1109–20.

42. Danes BS. Increased tetraploidy: cell-specific for the Gardner gene in thecultured cell. Cancer 1976;38:1983–8.

43. DikovskayaD, SchiffmannD,Newton IP, Oakley A, Kroboth K, SansomO,et al. Loss of APC induces polyploidy as a result of a combination of defectsin mitosis and apoptosis. J Cell Biol 2007;176:183–95.

44. Adamson PC, Poplack DG, Balis FM. The cytotoxicity of thioguanine vsmercaptopurine in acute lymphoblastic leukemia. Leuk Res 1994;18:805–10.

45. Roh M, van der Meer R, Abdulkadir SA. Tumorigenic polyploid cellscontain elevated ROS and ARE selectively targeted by antioxidant treat-ment. J Cell Physiol 2012;227:801–12.

46. Godinho SA, Picone R, Burute M, Dagher R, Su Y, Leung CT, et al.Oncogene-like induction of cellular invasion from centrosome amplifica-tion. Nature 2014;510:167–71.






2016;15:48-59. Published OnlineFirst November 19, 2015.Mol Cancer Ther Alka Choudhary, Brittany Zachek, Robert F. Lera, et al. High-Ploidy Breast CancerIdentification of Selective Lead Compounds for Treatment of

Updated version

10.1158/1535-7163.MCT-15-0527doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2015/11/19/1535-7163.MCT-15-0527.DC1

Access the most recent supplemental material at:

Cited articles

http://mct.aacrjournals.org/content/15/1/48.full#ref-list-1

This article cites 45 articles, 20 of which you can access for free at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://mct.aacrjournals.org/content/15/1/48To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-15-0527

http://mct.aacrjournals.org/content/suppl/2015/11/19/1535-7163.MCT-15-0527.DC1

http://mct.aacrjournals.org/content/15/1/48.full#ref-list-1

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

mailto:[email protected]

http://mct.aacrjournals.org/content/15/1/48


identiﬁcation of selective lead compounds for...

Documents