Ther
SynCanInhi
Kedar
Abst
Intro
Breand twomemajorgene chumabasal-typesnomachemoestrogHER2/characpithelietiologcers re
AuthorMedicin
Note: SOnline
CorresMedicin650-49
doi: 10
©2010
Cance7970
Published OnlineFirst August 26, 2010; DOI: 10.1158/0008-5472.CAN-09-4521
Canceresearch
apeutics, Targets, and Chemical Biology
ergistic Chemosensitivity of Triple-Negative Breastcer Cell Lines to Poly(ADP-Ribose) Polymerase
R
bition, Gemcitabine, and Cisplatin
Hastak, Elizabeth Alli, and James M. Ford
ractThe
receptcers shBRCAinhibitstudy,canceritor PJcells. Tsuggesexcisioa concseeme
y and agmain unk
s' Affiliatione, Stanford
upplementa(http://cance
ponding Aue, 269 Cam8-6689; Fax
.1158/0008-
American A
r Res; 70(
Downlo
basal-like subtype of breast cancer is characterized by a triple-negative (TN) phenotype (estrogenor, progesterone receptor, and human epidermal growth factor receptor-2/neu negative). TN breast can-are similar gene expression profiles and DNA repair deficiencies with BRCA1-associated breast cancers.1-mutant cells exhibit sensitivity to gemcitabine, cisplatin, and poly(ADP-ribose) polymerase (PARP)ion; therefore, we hypothesized that TN cancer cells may also exhibit sensitivity to these drugs. In thiswe report that TN breast cancer cells are more sensitive to these drugs compared with non-TN breastcells. Moreover, combination treatments indicated that PARP inhibition by the small-molecule inhib-34 or siRNA knockdown synergized with gemcitabine and cisplatin in TN cells but not in luminal cancerN cells exhibited reduced repair of UV-induced cyclobutane pyrimidine dimers after PARP inhibition,ting that the synergistic effect of PJ34 and gemcitabine or cisplatin reflected inefficient nucleotiden repair. Mechanistic investigations revealed that in TN cells, PJ34 reduced the levels of ΔNp63α withurrent increase in p73 and its downstream target p21. Thus, the sensitivity to combination treatmentd to be mediated by sustained DNA damage and inefficient DNA repair triggering p63/p73–mediated
apoptosis. Our results suggest a novel therapeutic strategy to treat women with TN breast cancer, an aggres-sive disease that presently lacks effective treatment options. Cancer Res; 70(20); 7970–80. ©2010 AACR.
AstheraprepairIndeedshownplatinetopowe hato gemanalogpropeGemcwith oor traare nosubtyp
duction
ast cancer is the most common cause of malignancyhe second most common cause of cancer death inn (1). This heterogeneous disease is composed of fivebiological subtypes, which are based on microarraylassifications: luminal A, luminal B, normal breast-like,n epidermal growth factor receptor 2 (HER2), andlike breast cancers (1, 2). Whereas many of these sub-can be treated with much success, the basal-like carci-s are associated with high rates of relapse followingtherapy (3, 4). Basal-like breast tumors are largelyen receptor (ER), progesterone receptor (PR), andneu negative [triple negative (TN)] and express genesteristic of basal epithelial and normal breast myoe-al cells (5–7). However, the genes responsible for the
gressive phenotype of basal-like breast can-nown.
Theinvolvdoubltion (Hbase eRecenshownmeraswhenthis scin DNinhibi
: Division of Oncology, Stanford University School of, California
ry data for this article are available at Cancer Researchrres.aacrjournals.org/).
thor: James M. Ford, Stanford University School ofpus Drive, CCSR #1115, Stanford, CA 94305. Phone:: 650-725-1420; E-mail: [email protected].
5472.CAN-09-4521
ssociation for Cancer Research.
20) October 15, 2010
Research. on June 26, 202cancerres.aacrjournals.org aded from
many cancer chemotherapeutic drugs and radiationy cause DNA damage, tumor cells defective in DNApathways are predicted to be sensitive to their effects., cell lines deficient in BRCA1 (and BRCA2) have beento be sensitive to the DNA cross-linking agents cis-and mitomycin C (8, 9), the topoisomerase inhibitorside (10), and oxidative DNA damage (11). Recently,ve shown that BRCA1-deficient cells are sensitivecitabine (2′,2′-difluoro-2′-deoxycytidine; ref. 12), anue of cytosine arabinoside that exhibits anticancerrties due to potent inhibition of DNA synthesis (13).itabine is often used either alone or in combinationther drugs such as taxanes, vinorelbine, carboplatin,stuzumab in metastatic breast cancer. However, therereports of the efficacy of gemcitabine in the basal-likee of breast cancers.BRCA1 breast cancer susceptibility gene is known to beed in a number of DNA repair pathways, including DNAe-strand break repair through homologous recombina-R; ref. 14), nucleotide excision repair (NER; ref. 15), andxcision repair (BER) of oxidative DNA damage (11).tly, BRCA1- and BRCA2-deficient cells have also beento be sensitive to inhibitors of poly(ADP-ribose) poly-e (PARP; refs. 16, 17), an enzyme involved in BER, which,inhibited, leads to DNA strand breaks and cell death. Inenario, BRCA-mutant tumor cells with primary defects
A repair are particularly sensitive to small-moleculetors of BER, such as PARP inhibitors.0. © 2010 American Association for Cancer
Recmors bin herstrongdefectexhibiPARPThe
aminePARPBRCAcells aand ciwith bluminand geof sen
Mate
Cell liAll
MDAMobtainand mT47D1640 wtion inimmubarcodpressiwere ocatedmetrixHER2with 1shRNAsilenctific).mycininhibitlene bmicals
Cell vCell
assay,PJ34,was adzan crcrystawas mmentspendefrom aFor coat equgemci
ment,platedblue a
DrugFor
MDAMand trthe coin at t72 houBiosoflogram(21).used aCalcuthe isoCI areCI = 1was rdata a
γH2ABT4
HCC1slides10 μm4 houincludalone.hyde a(H-92;Ser139with Tat roo(Invitrtively.diamitempetion oeach gthreeThe rgroupchangcontrorandoscopefromImageAPO o
ApopApo
Annexin cha
Chemotherapy for Triple-Negative Breast Cancer
www.a
Published OnlineFirst August 26, 2010; DOI: 10.1158/0008-5472.CAN-09-4521
ent studies indicate that sporadic basal-like or TN tu-ear a striking resemblance to breast tumors that ariseeditary BRCA1 mutation carriers. These similaritiesly suggest that sporadic basal-like tumors might bears in BRCA1-mediated DNA repair pathways (11) andt similar sensitivities to DNA-damaging agents andinhibitors.refore, using a panel of breast cancer cell lines, we ex-d the cytotoxic effects of gemcitabine, cisplatin, and ainhibitor alone and in combination. We show that, like1-mutant cells, basal-like TN breast cancer (TNBC)re sensitive to the PARP inhibitor PJ34, gemcitabine,splatin. We further show that PJ34 acts synergisticallyoth gemcitabine and cisplatin in TNBC cells, but not inal breast cancer cell lines. Moreover, we show that PJ34mcitabine disrupt NER, suggesting a novel mechanismsitivity to these drugs in TNBC cells.
rials and Methods
nes and reagentscell lines were used within 6 months of purchase.B468, hs578t, MCF7, BT549, and BT474 cell lines wereed from the American Type Culture Collection (ATCC)aintained in DMEM with 10% fetal bovine serum (FBS).and HCC1806 cells (ATCC) were maintained in RPMIith 10% FBS. ATCC provides molecular authentica-support of their collection through their genomics,
nology, and proteomic cores, as described, using DNAing and species identification, quantitative gene ex-on, and transcriptome analyses (18). SUM149PT cellsbtained from Asterand Plc, where they were authenti-through gene expression data through the use of Affy-GeneChips and biomolecular markers, such as ER andstatus. They were maintained in Ham's F-12 medium0% FBS. HCC1806 cells were transfected with themir retroviral vector in pSM2 against nonspecific
ing control, PARP1 or PARP2 (Thermo Fisher Scien-Stable clones were maintained with 1 μg/mL puro-and confirmed by Western blotting. The PARP
or PJ34 was purchased from EMD Biosciences. Methy-lue, MTT, and cisplatin were purchased from Sigma Che-. Gemcitabine was a gift from Eli Lilly Pharmaceuticals.
iability and colony formation assayviability was measured by an MTT assay. For the MTTcells were seeded in 96-well plates and treated withgemcitabine, and cisplatin as indicated. MTT reagentded to the cells, which was reduced to purple forma-ystals by the mitochondria of living cells. The reducedls were solubilized with DMSO, and the absorbanceeasured at 570 nm by spectrophotometry. All experi-were done in triplicate and also repeated three inde-nt times, and data were plotted as mean ± SD. Datarepresentative experiment are shown in the figures.
lony formation assay, HCC1806 and MCF7 were plated
al density and treated with 10 μmol/L PJ34, 0.6 nmol/Ltabine, or 4 μmol/L cisplatin for 72 hours. After treat-PJ34,24 hou
acrjournals.org
Research. on June 26, 202cancerres.aacrjournals.org Downloaded from
cells were counted and either 500 or 1,000 cells were. After 15 days, the cells were stained with methylenend individual colonies were counted.
combination studiescombination studies, BT474, MCF7, HCC1806, andB468 cells were seeded in triplicate in 96-well plateseated with PJ34, gemcitabine, or cisplatin alone or withmbination of PJ34 and gemcitabine or PJ34 and cisplat-he indicated doses. MTT assays were performed afterrs of treatment. The data were plotted using Calcusyn-t software (19, 20). Combination index (CI) and isobo-s were plotted using the CI equation of Chou-Talalay
A nonconstant ratio drug combination design wasnd normalized isobolograms were constructed usingsyn-Biosoft software. The experimental protocol andbologram plotted along with the equation for calculatingshown in Supplementary Fig. S1. CI < 1 was synergistic,was additive, and CI > 1 was antagonistic. Study
epeated three independent times and representativere shown.
X and Rad51 staining74, MDAMB468, HCC1806, HCC1806shP1, and806shP2 cells were plated overnight in chambered(1,500 cells per chamber). Cells were treated withol/L PJ34 or 5 nmol/L gemcitabine for either 1 tors for Rad51 staining or 24 hours for γH2AX. Controlsed primary alone, isotype control, and secondaryAfter treatment, cells were fixed in 4% paraformalde-nd stained overnight with primary antibody for Rad511:200 dilution; Santa Cruz Biotechnology) or γH2AX-(1:500 dilution; Cell Signaling). Cells were washedBS/bovine serum albumin and incubated for 1 hourm temperature with either Alexa 488 or Alexa 594ogen) secondary antibody for Rad51 or γH2AX, respec-Cells were fixed in Prolong gold antifade with 4′,6-dino-2-phenylindole (Invitrogen) and cured at roomrature for 24 hours before visualizing. For quantifica-f Rad51 foci and γH2AX foci, at least 100 cells fromroup were visually scored. Cells showing more thanfoci were counted as positive for γH2AX or Rad51.atio of Rad51 foci in the control versus the treateds was represented as fold change. For γH2AX, the folde was the ratio of γH2AX-positive cells in HCC1806l cells versus shPARP1 and shPARP2. Images fromm fields were taken using a Nikon Eclipse E 800 micro-with an attached camera and using Spot SoftwareDiagnostic Instruments (Diagnostic Instruments).s were taken with a 40× lens (40×/1.0 DIC H Planil immersion lens).
tosis detectionptosis was detected by staining HCC1806 cells forin V and cleaved caspase-3. Cells were plated overnightmbered slides followed by treatment with 10 μmol/L
0.6 nmol/L gemcitabine, or 4 μmol/L cisplatin forrs. For Annexin V-FITC, the standard manufacturer'sCancer Res; 70(20) October 15, 2010 7971
0. © 2010 American Association for Cancer
protocstaine(Asp17
NucleMD
and Hplatesdiatedtreateas meknockmini kidine d
suredtriplicamineTDM-Dr. ToUnivewere arametsuredindepe
Weste
Figure4.8 nmcells ascolonyin triplic
Hastak et al.
Cance7972
Published OnlineFirst August 26, 2010; DOI: 10.1158/0008-5472.CAN-09-4521
ol was followed (Sigma). For caspase-3, cells wered with a 1:250 dilution of cleaved caspase-3 antibody5, Cell Signaling) and visualized as mentioned above.
otide excision repair assayAMB468, HCC1806, HCC1806 control, HCC1806shP1,CC1806shP2 cells were grown overnight in six-well(in triplicate). Cells were rinsed with PBS and then irra-with 20 J/m2 UVC. MDAMB468 and HCC1806 cells wered with 10 μmol/L PJ34 or 5 nmol/L gemcitabine, where-dium without drugs was added to PARP1- and PARP2-down cells. Genomic DNA was extracted (QIAamp DNA
it, Qiagen) at 0 to 24 hours. Repair of cyclobutane pyrim- Totformation assay after 15 d. The numbers of colonies are shown; insets, represenate. Columns, average cell viability in log scale; bars, SD.
r Res; 70(20) October 15, 2010
Research. on June 26, 202cancerres.aacrjournals.org Downloaded from
using an ELISA. Briefly, genomic DNAwas distributed inate onto microtiter plates precoated with 0.003% prot-sulfate. DNA lesions were detected with either 1:5,0002 (for CPDs) or 1:5,000 64M-2 (for 6-4PPs; a gift fromshio Mori, Radioisotope Research Center, Nara Medicalrsity School ofMedicine, Nara, Japan; ref. 22). The signalsmplified and subsequently developed with 3,5,3′,5′-tet-hylbenzidine (Sigma Chemicals). Absorbance was mea-at 450 nm. Each experiment was repeated threendent times, and representative data are shown.
rn blot analyses
al cellular protein was isolated by lysing the cells inimers (CPD) and 6-4 photoproducts (6-4PPs) was mea- modified radioimmunoprecipitation assay buffer. Proteins
1. Cell viability of breast cancer cells after PJ34, gemcitabine, or cisplatin treatment. A, cells were treated with 0.01 to 62.5 μmol/L PJ34 (i), 0.15 tool/L gemcitabine (ii), or 0.256 to 10 μmol/L cisplatin (iii) for 72 h, and cell viability was determined by an MTT assay. B, apoptosis in HCC1806shown by cleaved caspase-3 and Annexin V staining (inset, bright-field image). C, HCC1806 and MCF7 cells were stained with methylene blue for
tative figures. Experiments were done three independent times
Cancer Research
0. © 2010 American Association for Cancer
(25–50PAGEmembwith aBiotecthe blogeJ (Nthe tre
Resu
SensitgemciTo
BRCAtreatmHCC1cell li
(BT47PJ34 (platinity wasignifiin treaSupplTNBCmatiohighlyminalsis inAnnexto BRCto PARthe ot
Table 1. Synergistic effect of PJ34 with either gemcitabine or cisplatin in breast cancer cell lines
A B
Gemcitabine(μmol/L)
PJ34(μmol/L)
CI Effect PJ34(μmol/L)
Cisplatin(μmol/L)
Cl Effect
(i) HCC1806 (i) HCC18060.0016 1 0.88 Synergism 1.875 10 1.09 Additive0.008 1 0.95 Synergism 3.75 10 0.85 Synergism0.04 1 0.97 Synergism 7.5 10 0.83 Synergism0.2 1 0.98 Synergism 15 10 0.646 Synergism1 1 0.98 Synergism
(ii) MDAMB468 (ii) MDAMB4680.0016 1 0.74 Synergism 1.875 10 0.83 Synergism0.008 1 0.48 Synergism 3.75 10 0.76 Synergism0.04 1 0.25 Synergism 7.5 10 0.82 Synergism0.2 1 0.72 Synergism 15 10 0.91 Synergism1 1 0.90 Synergism
(iii) BT474 (iii) BT4740.0016 1 1.34 Antagonism 1.875 10 74 Antagonism0.008 1 3.51 Antagonism 3.75 10 1.31 Antagonism0.04 1 1.33 Antagonism 7.5 10 1.53 Antagonism0.2 1 18.30 Antagonism 15 10 1.54 Antagonism1 1 50.22 Antagonism
(iv) MCF7 (iv) MCF70.0016 1 1.29 Antagonism 1.875 10 1.71 Antagonism0.008 1 2.93 Antagonism 3.75 10 1.28 Antagonism0.04 1 7.79 Antagonism 7.5 10 1.20 Antagonism0.2 1 0.74 Synergism 15 10 0.89 Synergism1 1 1.19 Antagonism
NOTE: HCC1806, MDAMB468, MCF7, and BT474 cells were treated with PJ34, gemcitabine, or cisplatin alone or a combination ofthe drugs at the indicated concentrations for 72 h, and an MTT assay was performed. CI was calculated by Calcusyn-Biosoftsoftware. (A) CI values of PJ34 in combination with gemcitabine; (B) CI values of PJ34 in combination with cisplatin.
Chemotherapy for Triple-Negative Breast Cancer
www.a
Published OnlineFirst August 26, 2010; DOI: 10.1158/0008-5472.CAN-09-4521
μg/lane) were separated by electrophoresis (10% SDS-) and electroblotted onto polyvinylidene difluorideranes (GE Healthcare). The membranes were probedntibodies against PARP, p63, p73, p21, actin (Santa Cruzhnology), or MCM (BD Biosciences). Protein levels fromts were evaluated using the gel analysis software Ima-IH), and the ratio of protein levels in the control versusated groups was represented as fold change.
lts
ivity of basal-like breast cancer cells to PJ34,tabine, and cisplatintest our hypothesis that PARP inhibitors that target1-pathway dysfunction might also be efficacious in theent of TNBCs, a panel of sporadic TNBC cells (BT549,
806, and MDAMB468) along with a BRCA1-mutant TNne (SUM149PT) and luminal breast cancer cell linesluminHER2/
acrjournals.org
Research. on June 26, 202cancerres.aacrjournals.org Downloaded from
4, MCF7, and T47D) were tested for their sensitivity to0–62.5 μmol/L), gemcitabine (0–4.8 nmol/L), and cis-(0–10 μmol/L). After 72 hours of treatment, cell viabil-s measured by an MTT assay. All TNBC cell lines werecantly more sensitive to PJ34, gemcitabine, and cisplat-tment than the luminal breast cancer cell lines (Fig. 1A,ementary Table S1). We confirmed the sensitivity ofcells to PJ34, gemcitabine, and cisplatin by colony for-n assay, and as shown in Fig. 1C, HCC1806 cells weresensitive to all the three drugs compared with the lu-MCF7 cells. Furthermore, all the drugs induced apopto-TNBC cells as evidenced by caspase-3 cleavage andin V staining (Fig. 1B). Therefore, we found that, similarA1-deficient cells, TNBC cells were selectively sensitiveP inhibition, gemcitabine, and cisplatin compared withher breast subtypes. Moreover, the resistance of the
al cell lines does not depend on HER2 status, as BT474 isneu positive whereas MCF7 is HER2/neu negative.Cancer Res; 70(20) October 15, 2010 7973
0. © 2010 American Association for Cancer
SynerOur
to PJ3Howeacts ingemciand MnationPJ34 a15 μmwas us(Suppand Cand ancoefficand aleffectAs s
and McombiluminexhibiSimilaeffectsi andBT474We
PJ34 aand cowhilecell linantagonot sh
PARPand cMan
PARP2Theresitizinated sthe TNPARPblottincontroeach)8 μmoMTT ashownsensitiknockbine cknock(Fig. 2
DNA dPJ34Bec
repairtigateby stasites oTo
TN MPJ34 o
Hastak et al.
Cance7974
Published OnlineFirst August 26, 2010; DOI: 10.1158/0008-5472.CAN-09-4521
gism between PJ34 and gemcitabine or cisplatinresults show that TNBC cells are selectively sensitive4, gemcitabine, and cisplatin when used individually.ver, to determine whether inhibition of PARP by PJ34a synergistic, additive, or antagonistic fashion with
tabine or cisplatin, we treated BT474, MCF7, HCC1806,DAMB468 cell lines with the agents alone or in combi-for 72 hours. Cells were treated either with 1 μmol/Lnd 0.0016 to 1 μmol/L gemcitabine or with 1.875 tool/L PJ34 and 10 μmol/L cisplatin. CalcuSyn softwareed to calculate the CI and plot normalized isobologramslementary Figs. S2 and S3; refs. 19, 21). CI < 1, CI = 1,I > 1 quantitatively indicate synergism, additivity,tagonism, respectively. We also calculated the linearient r value to estimate the accuracy of measurement,l our experiments had an r value >0.90 for the median-plot.hown in Table 1A (i and ii), TNBC cell lines HCC1806DAMB468 exhibited synergism for all the differentnations of PJ34 and gemcitabine doses, whereas theal BT474 and MCF7 cell lines (Table 1A, iii and iv)ted antagonistic effects for most dose combinations.rly, PJ34 and cisplatin had additive to synergisticin both HCC1806 and MDAMB468 cells (Table 1B,
ii), whereas antagonism was observed in the luminaland MCF7 cell lines (Table 1B, iii and iv).also treated the cells with varied concentrations ofnd kept the concentration of gemcitabine constant,nversely, we kept the concentration of PJ34 constantchanging the concentration of cisplatin. Again, TNBCes exhibited additive to synergistic effects, whereas
nism was observed in BT474 and MCF7 cell lines (dataown).Fig. 3Athe Ra
r Res; 70(20) October 15, 2010
Research. on June 26, 202cancerres.aacrjournals.org Downloaded from
knockdown sensitizes cells to gemcitabineisplatiny PARP inhibitors are known to inhibit PARP1 and, both of which are involved in DNA repair pathways.fore, to investigate the role of PARP inhibition in sen-g TNBC cells to gemcitabine and cisplatin, we gener-table clones of PARP1- and PARP2-knockdown cells inHCC1806 cell line. Decreased protein expression of
1 and PARP2 protein was confirmed by immuno-g (Fig. 2A). We then treated HCC1806 nonsilencingl and HCC1806 shPARP1 and shPARP2 clones (twowith either 0.15 to 4.8 nmol/L gemcitabine or 0.25 tol/L cisplatin for 72 hours, measured cell viability by anssay, and calculated IC50 values by Prism software. Asin Fig. 2B and C, both PARP1 and PARP2 knockdownzed HCC1806 cells to gemcitabine and cisplatin. PARP2down significantly (P < 0.05) sensitized cells to gemcita-ompared with HCC1806 (Fig. 2B). Conversely, PARP1down significantly (P < 0.05) sensitized cells to cisplatinC).
amage in basal-like breast cancer cell lines afterand gemcitabine treatmentause PARP and gemcitabine play a major role in DNAand inhibition of DNA synthesis, respectively, we inves-d the effect of PJ34 and gemcitabine on DNA damageining for Rad51 foci and γH2AX, which accumulate atf broken DNA.study Rad51 foci formation, the luminal BT474 andDAMB468 cell lines were treated with either 10 μmol/Lr 5 nmol/L gemcitabine for 1 to 4 hours. As shown in
, 4 hours of PJ34 or gemcitabine treatment increasedd51 foci in MDAMB468 cells. BT474 cells had a higherurePAr getmelysitrolARe trcencit
T ash. ICPrisgraper gCotri
s, SD; ★, P < 0.05.
FigandaftetreaanaconshPwercongemMT72byaseith(C).frombar
0. © 2010 American Associat
2. Cell viability of PARP1-RP2-knockdown cellsmcitabine and cisplatinnt. A, Western blots of PARP in HCC1806and shPARP1- andP2-knockdown cells. Cellseated with increasingtrations of eitherabine or cisplatin and ansay was performed after
50 values were determinedm and are representedhs for cells treated withemcitabine (B) or cisplatinlumns, average cell viabilityplicate experiments;
Cancer Research
ion for Cancer
basal ldid nothe nuaftermiddland Pparedin FigWe
with 1and sting trγH2AHCC1increa
PARP2numbcontroobserdamagof DNsis (dacells iblockDNA d
IneffiPJ34
FigureMDAMC, quan dle) trthree fo
Chemotherapy for Triple-Negative Breast Cancer
www.a
Published OnlineFirst August 26, 2010; DOI: 10.1158/0008-5472.CAN-09-4521
evel of Rad51 foci, but PJ34 and gemcitabine treatmentt further increase the number of foci. We next countedmber of Rad51 foci and observed a 2-fold increasePJ34 and gemcitabine treatment (Fig. 3C, top ande). Similar to PJ34 treatment, knockdown of PARP1ARP2 also increased the number of Rad51 foci com-with HCC1806 control cells, as seen and quantified. 3B and C (bottom).next treated BT474, MDAMB468, and HCC1806 cells0 μmol/L PJ34 or 5 nmol/L gemcitabine for 24 hoursained the cells for γH2AX. As shown in Fig. 4A, follow-eatment, there was no increase in the number ofX-positive cells in BT474 cell line. On the other hand,
tification of Rad51 foci after 1, 2, or 4 h of PJ34 (top) or gemcitabine (midci or more were counted as positive.
806 and MDAMB468 cell lines showed a significantse in the number of γH2AX-positive cells. PARP1- and
Recirrad
acrjournals.org
Research. on June 26, 202cancerres.aacrjournals.org Downloaded from
-knockdown cells also exhibited 2-fold increases in theer of γH2AX-positive cells compared with HCC1806l cells, as shown and quantified in Fig. 4B and C. Thesevations suggest that TNBC cells accumulate DNAe due to inhibition of PARP activity or due to inhibitionA synthesis by gemcitabine. Moreover, cell cycle analy-ta not shown) illustrates that PJ34 treatment arrestedn the G2-M phase and gemcitabine induced an S-phasein TNBC cells, consistent with cell cycle arrest afteramage.
cient repair of UV-induced DNA damage afterand gemcitabine treatment
eatment or PARP1 and PARP2 knockdown (bottom). Cells with
3. Increase in Rad51 foci after PARP inhibition or gemcitabine treatment in TNBC cells. A, representative images of Rad51 foci in BT474 andB468 cells treated with either 10 μmol/L PJ34 or 5 nmol/L gemcitabine for 4 h. B, HCC1806 control and PARP1- and PARP2-knockdown cells.
ent studies have shown that PARP is activated by UViation (23, 24); however, the role of PARP after
Cancer Res; 70(20) October 15, 2010 7975
0. © 2010 American Association for Cancer
UV-indshowetive inwild-tycells bUV-inin TNBPARP1As s
cells tin repPJ34 o(botto
rendechangthe syand ction orepair
RolecanceStu
breast
Figurecells trein HCCPARP2
Hastak et al.
Cance7976
Published OnlineFirst August 26, 2010; DOI: 10.1158/0008-5472.CAN-09-4521
uced DNA damage is not clear. Previously, our laboratoryd that human and mouse BRCA1-mutant cells are defec-global genomic repair (GGR) compared with BRCA1pe and luminal breast cancer cells (15). Because TNBCear resemblance to BRCA-mutant cells, we investigatedduced DNA damage repair through the GGR pathwayC cells treated with either PJ34 or gemcitabine and in- and PARP2-knockdown cells.hown in Fig. 5A and B, both MDAMB468 and HCC1806reated with either PJ34 or gemcitabine were efficientairing 6-4PPs (top). However, treatment of cells with
r gemcitabine completely inhibited the repair of CPDs induce1806 control and HCC1806 shPARP1- and shPARP2-knockdown cells. C, quan-knockdown cells.
r Res; 70(20) October 15, 2010
Research. on June 26, 202cancerres.aacrjournals.org Downloaded from
red the cells inefficient in repairing CPDs with noe in the ability to repair 6-4PPs (Fig. 5C). Therefore,nergistic effect observed in TNBC cells between PJ34isplatin or gemcitabine may be, in part, due to inhibi-f the GGR pathway along with defects in other DNApathways.
of p63 and p73 in sensitizing basal-like breastr cells to PJ34 and gemcitabinedies have suggested that 0% to 30% of invasive ductalcarcinomas express ΔNp63α protein (25–27). p73 can
apoptosis by p53-independent mechanisms, making itm). Similarly, knockdown of both PARP1 and PARP2 particularly important for therapeutics in basal-like breast
4. DNA damage in cells after PARP inhibition or gemcitabine treatment. A, representative images of γH2AX in BT474, MDAMB468, and HCC1806ated with 10 μmol/L PJ34 or 5 nmol/L gemcitabine for 24 h. Inset, magnified image showing distinct punctate staining. B, γH2AX-positive cells
tification of γH2AX foci in HCC1806 control and PARP1- and
Cancer Research
0. © 2010 American Association for Cancer
carcinfoundlines blishedan onand Peven mAs s
HCC1expresits prop73 pp73 inexpresp53 amaintpublisare ovsion oFig. 6CMCM7knockas shothe pr
p73 amay tand a
Discu
Thesuggeswith ito idemensapy odeveloand hcytotoexplorrepairin TNPre
lines amatchthis o
Figureeither 10 μmol/L PJ34 or 5 nmol/L gemcitabine for 1 to 24 h (A and B). Top, repair of 6-4PPs in MDAMB468, HCC1806, and HCC1806 shPARP1 andshPARP
Chemotherapy for Triple-Negative Breast Cancer
www.a
Published OnlineFirst August 26, 2010; DOI: 10.1158/0008-5472.CAN-09-4521
omas, which are most often null or mutant for p53. Wethat p63 is significantly overexpressed in TNBC celly analyzing gene expression data from a study pub-by Neve and colleagues (28). Because p63 can act ascogene (25), the correlation between p63 expressionJ34- and gemcitabine-mediated cell death becomesore important.hown in Fig. 6A, PJ34-treated hs568t, MDAMB486, and806 TN cells exhibit a more than 3-fold decrease in thesion of ΔNp63α. ΔNp63α binds to p73 and preventsapoptotic activity. We therefore analyzed the levels ofrotein and observed an increase in the expression ofPJ34-treated cells with a concurrent increase in thesion of p21 (Fig. 6B). A recent study showed thatnd p73, through p21, can repress minichromosomeenance (MCM) proteins (29). Because analysis of thehed microarray data (28) revealed that MCM proteinserexpressed in TNBC cells, we examined the expres-f MCM proteins after PJ34 treatment. As shown in, PJ34 treatment decreased the levels of MCM4 andin TNBC cells. We next treated PARP1- and PARP2-
down cells with gemcitabine for 24 or 48 hours, and
2. Bottom, repair of CPDs.
wn in Fig. 5D, gemcitabine treatment downregulatedotein level of ΔNp63α with a concurrent increase in
volvedstrand
acrjournals.org
Research. on June 26, 202cancerres.aacrjournals.org Downloaded from
nd p21. Thus, the results suggest that DNA damagerigger the p63/p73 pathway to induce cell cycle arrestpoptosis.
ssion
discovery of molecular subclasses of breast cancerts that treatments may be targeted more selectivelymproved outcomes. Currently, a major challenge isntify such targets and more effective therapeutic regi-for TNBCs that are not responsive to endocrine ther-r trastuzumab. BRCA1 mutation carriers commonlyp basal-like breast tumors with defects in DNA repairave been shown to have altered sensitivity to certainxic DNA-damaging agents. In the current study, weed whether agents known to selectively target DNA-–deficient BRCA1-mutant cells would also be effectiveBC cells.viously, we have established that BRCA1-mutant cellre more sensitive to cisplatin and gemcitabine thaned BRCA1 wild-type cells (12). Mechanisms to explainbservation include defects in DNA repair pathways in-
5. Decreased NER in TNBC cells after PARP inhibition or gemcitabine treatment. Cells were irradiated with 20 J/m2 UVC (A–C) and then treated with
in HR, NER, and resolution of the intra- and inter-DNA cross-links induced by cisplatin. In our current
Cancer Res; 70(20) October 15, 2010 7977
0. © 2010 American Association for Cancer
study,compathat Tbine, ctive tochemobreastPAR
mechavate tsuggesenceBRCAity wiTNBCcells (in treathat Tthan tsensitimutanBec
damagPARPagentsin breses anor cisp
ably,repairA re
quiredthe syinhibicationreplicthe stresultscombiinhibiproveingly,gemcitive tPARPby difbeingOur
distincell linside atroxacstudiean acc
Figure(A); p73PJ34 (C
Hastak et al.
Cance7978
Published OnlineFirst August 26, 2010; DOI: 10.1158/0008-5472.CAN-09-4521
we found that TNBC cell lines are sensitive to cisplatinred with luminal cells. We also report for the first timeNBC cell lines exhibit profound sensitivity to gemcita-ompared with the luminal types, which are not sensi-gemcitabine. These novel findings suggest a targetedtherapeutic approach to TN and BRCA1-deficientcancer.P1 and PARP2 are involved in various DNA repairnisms. They bind to the DNA damage sites and acti-hemselves by automodification. Recent studies haveted that decreasing PARP expression by RNA interfer-or by chemical inhibitors sensitizes BRCA1- and2-deficient cells to cell death through synthetic lethal-th their DNA repair defects (16, 17). Our finding thatcells share DNA repair defects with BRCA1-mutant11) suggests that PARP inhibitors may be effectiveting these tumors, as well. In fact, our results showNBC cells are more susceptible to PARP inhibitionhe luminal type of breast cancer cells. Moreover, theirvity to PARP inhibitors is similar to that of BRCA1-t cells.ause PARP plays a major role in the response to DNAe, we also wished to examine whether inhibitor ofacts synergistically with DNA-damaging cytotoxic. Therefore, to explore the effects of drug combinationsast cancer subtypes, we performed isobologram analy-
d found that the combination of PJ34 with gemcitabinelatin had a synergistic effect in TNBC cells. Remark-foci reand/o
and p21 in hs578t, MDAMB468, and HCC1806 cells treated with PJ34 (B); and M). D, expression of PARP, ΔNp63α, p73, and p21 in HCC1806 and HCC1806 sh
r Res; 70(20) October 15, 2010
Research. on June 26, 202cancerres.aacrjournals.org Downloaded from
however, the combination proved antagonistic in-proficient luminal breast cancer cell lines.cent study (30) showed that PARP1 and PARP2 are re-to reactivate replication at stalled DNA forks. Thus,nergism observed between gemcitabine and PARPtion in TNBC cells may be attributed to stalled repli-forks caused by incorporation of gemcitabine into
ating DNA and failure to reactivate replication atalled fork due to inhibition of PARP activity. Thesehave clinical significance and suggest that a regimenning a platinum agent with gemcitabine and a PARPtor may have unique efficacy in TNBC but may noteffective in other subtypes of breast cancers. Interest-knocking down PARP2 further sensitized TN cells totabine, whereas PARP1-knockdown cells were sensi-o cisplatin. It is therefore possible that PARP1 and2 responded preferentially to DNA damage causedferent DNA-damaging agents, and this observation isfurther investigated.work shows that H2AX is phosphorylated and forms
ct nuclear foci in response to gemcitabine in TNBCes, similar to the effect of other deoxycytidine nucleo-nalogues, such as 1-h-D-arabinofuranosylcytosine anditabine (31). Additionally, consistent with previouss (32, 33), we found that gemcitabine treatment causedumulation of Rad51 nuclear foci. It is not clear if these
sult from gemcitabine-induced stalled replication forksr gemcitabine-induced accumulation of cells in S phase.6. Modulation of p63, p73, p21, and MCMs in TNBC cells. Western blot analysis of ΔNp63α in MDAMB468 and hs578t cells treated with PJ34
CM4 and MCM7 in MDAMB468 and HCC1806 cells treated withPARP1 and shPARP2 cells treated with 0.6 nmol/L gemcitabine.Cancer Research
0. © 2010 American Association for Cancer
Howevby gemDNA dSim
also lesistenforma(34, 35defectin HRrepairAlth
portedmuchis stimour labtive incoupleof PARor byUVC-ireportto be isynergbe parpair isway oinvestUnd
treatmment.geneBRCAmembgroup
PJ34 oΔNp6p73 ain turnfor liccationcells (regulalines oto dowthe exexprescell deto BRgemcisponsseemsisms,p63/p73–mediated apoptotic signaling cascade. These datasugges
Disclosure of Potential Conflicts of Interest
No p
Grant
NIHand SusE. Alli).
Theof pageaccorda
Refe1. Pe
bre2. So
breimp
3. CapriCa
4. Lieapcan
5. YeingMe
6. Raan20
7. Nieclincar
8. Momitco
9. Bh
Chemotherapy for Triple-Negative Breast Cancer
www.a
Published OnlineFirst August 26, 2010; DOI: 10.1158/0008-5472.CAN-09-4521
er, it is possible that stalled replication forks causedcitabine trigger HR repair of chemotherapy-inducedamage.ilarly, inhibition of PARP either by drugs or by RNAid to an increase in γH2AX foci and Rad51 foci, con-t with the idea that loss of PARP might increase thetion of DNA strand breaks that are repaired by HR). Recently, we (11) have shown that TNBC cells areive in BER; moreover, these cells may also be defective. Thus, inhibition of PARP along with defective DNAmechanisms may lead to synthetic lethality.ough UV-induced activation of PARP has been re-, the possible role of PARP in NER has not receivedattention. Studies have shown that repair of 8-oxoGulated by XPC (36) and CSB (37, 38), and becauseoratory has shown that BRCA1-mutant cells are defec-the GGR pathway of NER, but not in transcription-d repair (15) and BER (11), we investigated the roleP in GGR. We showed that PARP inhibition chemicallyRNAi decreased the capacity of TNBC cells to removenduced CPD lesions, a result consistent with previouss of PARP playing a role in GGR (39, 40). NER is knownnvolved in platinum-DNA adduct repair. Therefore, theism observed with PARP inhibition and cisplatin maytly due to inhibition of NER. However, whether the re-facilitated by the transcription-coupled repair path-f NER is not established by this study, and furtherigation into the mechanism is under way.erstanding the molecular mechanism behind drugent is critical to predict the clinical efficacy of treat-Meta-analyses of published microarray data (28) forexpression changes common to the basal-like and1-mutated cell lines identified p63 and MCM family
ers to be significantly overexpressed (P < 0.05) in both Receomycin-C resistance, and chromosome stability is restored withrrection of a Brca1 mutation. Cancer Res 2001;61:4842–50.attacharyya A, Ear US, Koller BH, Weichselbaum RR, Bishop DK.
Thnuthe23
10. TrePoetoCa
11. Alloxsen20
12. Shcisem
13. Hu2′,61
14. Zhdo
15. Hafac32
16. Fa
acrjournals.org
Research. on June 26, 202cancerres.aacrjournals.org Downloaded from
r gemcitabine resulted in a decreased expression of3α with a concurrent increase in the expression ofnd the cyclin-dependent kinase inhibitor p21, which,, repressed the MCM proteins (41). MCMs are requiredensing of origins, providing a signal for initiating repli-in S phase, and are frequently overexpressed in cancer42–44). Our data showed that MCM4/MCM7 are down-ted in TNBC cells treated with PJ34. Based on thesef evidence, we hypothesize that PJ34 treatment leadsnregulation of ΔNp63α with a concurrent increase inpression of p73 and p21, which, in turn, decreases thesion of MCM4/MCM7, leading to cell cycle arrest andath. Overall, we found that human TNBC cells, similarCA1-deficient cell lines, were more sensitive to PJ34,tabine, and cisplatin and exhibited synergistic re-es to combinations of these agents. This sensitivityto be dependent on inefficient DNA repair mechan-causing sustained DNA damage that may trigger the
t novel options for targeted treatment of TNBCs.
otential conflicts of interest were disclosed.
Support
grant R01 CA108794, the Breast Cancer Research Foundation (J.M. Ford),an G. Komen for the Cure Postdoctoral Fellowships (K. Hastak and
costs of publication of this article were defrayed in part by the paymentcharges. This article must therefore be hereby marked advertisement innce with 18 U.S.C. Section 1734 solely to indicate this fact.
ived 12/21/2009; revised 07/12/2010; accepted 07/27/2010; published
s. Our study showed that tre atment of TNBC cells with OnlineFirst 08/26/2010.rencesrou CM, Sorlie T, Eisen MB, et al. Molecular portraits of humanast tumours. Nature 2000;406:747–52.rlie T, Perou CM, Tibshirani R, et al. Gene expression patterns ofast carcinomas distinguish tumor subclasses with clinicallications. Proc Natl Acad Sci U S A 2001;98:10869–74.rey LA, Dees EC, Sawyer L, et al. The triple negative paradox:mary tumor chemosensitivity of breast cancer subtypes. Clinncer Res 2007;13:2329–34.dtke C, Mazouni C, Hess KR, et al. Response to neoadjuvant ther-y and long-term survival in patients with triple-negative breastcer. J Clin Oncol 2008;26:1275–81.hiely F, Moyano JV, Evans JR, Nielsen TO, Cryns VL. Deconstruct-the molecular portrait of basal-like breast cancer. Trends Mold 2006;12:537–44.kha EA, Tan DS, Foulkes WD, et al. Are triple-negative tumoursd basal-like breast cancer synonymous? Breast Cancer Res07;9:404, author reply 5.lsen TO, Hsu FD, Jensen K, et al. Immunohistochemical andical characterization of the basal-like subtype of invasive breastcinoma. Clin Cancer Res 2004;10:5367–74.ynahan ME, Cui TY, Jasin M. Homology-directed DNA repair,
e breast cancer susceptibility gene BRCA1 is required for sub-clear assembly of Rad51 and survival following treatment withDNA cross-linking agent cisplatin. J Biol Chem 2000;275:
899–903.szezamsky AD, Kachnic LA, Feng Z, Zhang J, Tokadjian C,well SN. BRCA1- and BRCA2-deficient cells are sensitive toposide-induced DNA double-strand breaks via topoisomerase II.ncer Res 2007;67:7078–81.i E, Sharma VB, Sunderesakumar P, Ford JM. Defective repair ofidative DNA damage in triple-negative breast cancer conferssitivity to inhibition of poly(ADP-ribose) polymerase. Cancer Res09;69:3589–96.arma VB, Hartman AR, Cowan K, Ford JM. Enhanced sensitivity toplatin and gemcitabine in DNA repair deficient Brca1 null mousebryonic fibroblasts. Proc Am Assoc Cancer Res 2005;46:4390.ang P, Chubb S, Hertel LW, Grindey GB, Plunkett W. Action of2′-difluorodeoxycytidine on DNA synthesis. Cancer Res 1991;51:10–7.ang J, Powell SN. The role of the BRCA1 tumor suppressor in DNAuble-strand break repair. Mol Cancer Res 2005;3:531–9.rtman AR, Ford JM. BRCA1 induces DNA damage recognition
tors and enhances nucleotide excision repair. Nat Genet 2002;:180–4.rmer H, McCabe N, Lord CJ, et al. Targeting the DNA repair defectCancer Res; 70(20) October 15, 2010 7979
0. © 2010 American Association for Cancer
in917
17. BrydeNa
18. AT19. Ch
immPro
20. ChtionorJ 1
21. ChshiAd
22. MoSimeitsamPh
23. Chpoforker
24. VoShribSc
25. RibZuinvof45
26. Waand
27. Lep6bio20
28. NelineCe
29. Scp7On
30. Br
starec
31. Ewgeup12
32. PabyDN20
33. WaKatio55
34. Scpona
35. Yadoco20
36. D'Epro20
37. Sugege
38. Tutalgro
39. Floerapro
40. GhDetivhu
41. Defam
42. BloDN
43. BloNa
44. Ts
Hastak et al.
Cance7980
Published OnlineFirst August 26, 2010; DOI: 10.1158/0008-5472.CAN-09-4521
BRCA mutant cells as a therapeutic strategy. Nature 2005;434:–21.ant HE, Schultz N, Thomas HD, et al. Specific killing of BRCA2-ficient tumours with inhibitors of poly(ADP-ribose) polymerase.ture 2005;434:913–7.CC Bulletin. Maintaining high standards in cell culture; 2010.ou TC, Stepkowski SM, Kahan BD. Computerized quantitation ofunosuppressive synergy for clinical protocol design. Transplantc 1994;26:3043–5.ou TC, Talalay P. Generalized equations for the analysis of inhibi-s of Michaelis-Menten and higher-order kinetic systems with twomore mutually exclusive and nonexclusive inhibitors. FEBS981;115:207–16.ou TC, Talalay P. Quantitative analysis of dose-effect relation-ps: the combined effects of multiple drugs or enzyme inhibitors.v Enzyme Regul 1984;22:27–55.ri T, Nakane M, Hattori T, Matsunaga T, Ihara M, Nikaido O.ultaneous establishment of monoclonal antibodies specific for
her cyclobutane pyrimidine dimer or (6-4)photoproduct from thee mouse immunized with ultraviolet-irradiated DNA. Photochem
otobiol 1991;54:225–32.ang H, Sander CS, Muller CS, Elsner P, Thiele JJ. Detection ofly(ADP-ribose) by immunocytochemistry: a sensitive new methodthe early identification of UVB- and H2O2-induced apoptosis inatinocytes. Biol Chem 2002;383:703–8.denicharov MD, Ghodgaonkar MM, Halappanavar SS, Shah RG,ah GM. Mechanism of early biphasic activation of poly(ADP-ose) polymerase-1 in response to ultraviolet B radiation. J Celli 2005;118:589–99.eiro-Silva A, Ramalho LN, Garcia SB, Brandao DF, Chahud F,coloto S. p63 correlates with both BRCA1 and cytokeratin 5 inasive breast carcinomas: further evidence for the pathogenesisthe basal phenotype of breast cancer. Histopathology 2005;47:8–66.ng X, Mori I, Tang W, et al. p63 expression in normal, hyperplasticmalignant breast tissues. Breast Cancer 2002;9:216–9.
ong CO, Vidnovic N, DeYoung MP, Sgroi D, Ellisen LW. The3/p73 network mediates chemosensitivity to cisplatin in alogically defined subset of primary breast cancers. J Clin Invest07;117:1370–80.ve RM, Chin K, Fridlyand J, et al. A collection of breast cancer cells for the study of functionally distinct cancer subtypes. Cancerll 2006;10:515–27.ian MJ, Carchman EH, Mohanraj L, et al. Wild-type p53 and
3 negatively regulate expression of proliferation related genes.cogene 2008;27:2583–93.yant HE, Petermann E, Schultz N, et al. PARP is activated atDNCd20
r Res; 70(20) October 15, 2010
Research. on June 26, 202cancerres.aacrjournals.org Downloaded from
lled forks to mediate Mre11-dependent replication restart andombination. EMBO J 2009;28:2601–15.ald B, Sampath D, Plunkett W. H2AX phosphorylation marksmcitabine-induced stalled replication forks and their collapseon S-phase checkpoint abrogation. Mol Cancer Ther 2007;6:39–48.rsels LA, Morgan MA, Tanska DM, et al. Gemcitabine sensitizationcheckpoint kinase 1 inhibition correlates with inhibition of a Rad51A damage response in pancreatic cancer cells. Mol Cancer Ther09;8:45–54.chters FM, van Putten JW, Maring JG, Zdzienicka MZ, Groen HJ,mpinga HH. Selective targeting of homologous DNA recombina-n repair by gemcitabine. Int J Radiat Oncol Biol Phys 2003;57:3–62.hultz N, Lopez E, Saleh-Gohari N, Helleday T. Poly(ADP-ribose)lymerase (PARP-1) has a controlling role in homologous recombi-tion. Nucleic Acids Res 2003;31:4959–64.ng YG, Cortes U, Patnaik S, Jasin M, Wang ZQ. Ablation of PARP-1es not interfere with the repair of DNA double-strand breaks, butmpromises the reactivation of stalled replication forks. Oncogene04;23:3872–82.rrico M, Parlanti E, Teson M, et al. New functions of XPC in thetection of human skin cells from oxidative damage. EMBO J06;25:4305–15.nesen M, Stevnsner T, Brosh RM, Jr., Dianov GL, Bohr VA. Globalnome repair of 8-oxoG in hamster cells requires a functional CSBne product. Oncogene 2002;21:3571–8.o J, Chen C, Zeng X, Christiansen M, Bohr VA. Functional cross-k between hOgg1 and the helicase domain of Cockayne syndromeup B protein. DNA Repair (Amst) 2002;1:913–27.hr C, Burkle A, Radicella JP, Epe B. Poly(ADP-ribosyl)ation accel-tes DNA repair in a pathway dependent on Cockayne syndrome Btein. Nucleic Acids Res 2003;31:5332–7.odgaonkar MM, Zacal N, Kassam S, Rainbow AJ, Shah GM.pletion of poly(ADP-ribose) polymerase-1 reduces host cell reac-ation of a UV-damaged adenovirus-encoded reporter gene inman dermal fibroblasts. DNA Repair (Amst) 2008;7:617–32.Laurenzi V,MelinoG. Evolution of functionswithin the p53/p63/p73ily. Ann N Y Acad Sci 2000;926:90–100.w JJ, Laskey RA. A role for the nuclear envelope in controllingA replication within the cell cycle. Nature 1988;332:546–8.w JJ, Dutta A. Preventing re-replication of chromosomal DNA.t Rev Mol Cell Biol 2005;6:476–86.uyama T, Tada S, Watanabe S, Seki M, Enomoto T. Licensing forA replication requires a strict sequential assembly of Cdc6 and
t1 onto chromatin in Xenopus egg extracts. Nucleic Acids Res05;33:765–75.Cancer Research
0. © 2010 American Association for Cancer
2010;70:7970-7980. Published OnlineFirst August 26, 2010.Cancer Res Kedar Hastak, Elizabeth Alli and James M. Ford Gemcitabine, and CisplatinCell Lines to Poly(ADP-Ribose) Polymerase Inhibition, Synergistic Chemosensitivity of Triple-Negative Breast Cancer
Updated version
10.1158/0008-5472.CAN-09-4521doi:
Access the most recent version of this article at:
Material
Supplementary
http://cancerres.aacrjournals.org/content/suppl/2010/08/26/0008-5472.CAN-09-4521.DC1
Access the most recent supplemental material at:
Cited articles
http://cancerres.aacrjournals.org/content/70/20/7970.full#ref-list-1
This article cites 43 articles, 13 of which you can access for free at:
Citing articles
http://cancerres.aacrjournals.org/content/70/20/7970.full#related-urls
This article has been cited by 20 HighWire-hosted articles. Access the articles at:
E-mail alerts related to this article or journal.Sign up to receive free email-alerts
Subscriptions
Reprints and
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Permissions
Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)
.http://cancerres.aacrjournals.org/content/70/20/7970To request permission to re-use all or part of this article, use this link
Research. on June 26, 2020. © 2010 American Association for Cancercancerres.aacrjournals.org Downloaded from
Published OnlineFirst August 26, 2010; DOI: 10.1158/0008-5472.CAN-09-4521