PUMA Sensitizes Lung Cancer Cells to ChemotherapeuticAgents and IrradiationJian Yu,1,3 Wen Yue,2,3 BinWu,2,3 and Lin Zhang2,3

Abstract Purpose: Lung cancer, the leading cause of cancer mortality worldwide, is often diagnosedat late stages and responds poorly to conventional therapies, including chemotherapy and irradi-ation. A great majority of lung tumors are defective in the p53 pathway, which plays an importantrole in regulating apoptotic response to anticancer agents. PUMA was recently identified as anessential mediator of DNA damage ^ induced and p53-dependent apoptosis. In this study, weinvestigated whether the regulation of PUMA by anticancer agents is abrogated in lung cancercells and whether PUMA expression suppresses growth of lung cancer cells and/or sensitizeslung cancer cells to chemotherapeutic agents and irradiation through induction of apoptosis.Experimental Designs: The expression of PUMA was examined in lung cancer cells withdifferent p53 status treated with chemotherapeutic agents. An adenovirus expressing PUMA(Ad-PUMA), alone or in combination with chemotherapeutic agents or g-irradiation, was used totreat lung cancer cells. The growth inhibitory and apoptotic effects of PUMA in vitro and in vivowere examined. The mechanisms of PUMA-mediated growth suppression and apoptosis wereinvestigated through analysis of caspase activation and release of mitochondrial apoptogenicproteins.The cytotoxicities of PUMA on cancer and normal/nontransformed cells were compared.The efficacy of PUMA and p53 in suppressing the growth of lung cancer cells was also compared.Results: We showed that the induction of PUMA by chemotherapeutic agents is abolished inp53-deficient lung cancer cells. PUMA expression resulted in potent growth suppression of lungcancer cells and suppressed xenograft tumor growth in vivo through induction of apoptosis.Low dose of Ad-PUMA significantly sensitized lung cancer cells to chemotherapeutic agentsand g-irradiation through induction of apoptosis.The effects of PUMA are mediated by enhancedcaspase activation and release of cytochrome c and apoptosis-inducing factor into the cytosol.Furthermore, PUMA seems to be selectively toxic to cancer cells and more efficient than p53 insuppressing lung cancer cell growth.Conclusions: Our findings indicate that PUMA is an important modulator of therapeuticresponses of lung cancer cells and is potentially useful as a sensitizer in lung cancer therapy.

Lung cancer is the leading cause of cancer mortality worldwide(1). The current treatment options for patients with advanceddisease are limited to chemotherapy and radiation therapy,which unfortunately produce a low rate of response and virtuallyno cure. The 5-year survival rate, currently at 14%, has only been

marginally improved in the last two decades despite thedevelopment of new therapeutic agents and improved patientcare. Novel therapeutic interventions are critical for improve-ment of the survival and prognosis of lung cancer patients.Cancer development is a multistage process which involves a

number of genetic and epigenetic changes in the genescontrolling cell survival, cell death, cell-cell communication,cell-microenvironment interactions, and angiogenesis (2, 3). Ithas become increasingly clear that abnormalities in cell deathpathways play an important role in tumorigenesis and thedevelopment of resistance to chemotherapy and radiationtherapy (4, 5). Most of the agents used in cancer therapydirectly or indirectly damage DNA and induce apoptosis.Defects in the apoptotic machinery can lead to multidrugresistance (6). Therapeutic manipulation of the apoptoticpathways has become an attractive avenue to improve theclinical response of lung cancer patients (7–9).Defective p53 pathway is one of the most common

signatures of human cancer (2). p53 mutations occur inf50% of non–small-cell lung cancers and >70% of small-celllung cancers (10). A major physiologic function of p53 is to killdamaged or stressed cells through induction of apoptosis (11).p53 induces apoptosis by transactivation of its downstream

Cancer Therapy: Preclinical

Authors’ Affiliations: Departments of 1Pathology and 2Pharmacology, and3University of Pittsburgh Cancer Institute, University of Pittsburgh School ofMedicine, Pittsburgh, PennsylvaniaReceived 11/9/05; revised 1/17/06; accepted 2/27/06.Grant support: The Flight Attendant Medical Research Institute, the Alliance forCancer GeneTherapy, the Hillman Foundation (J. Yu), NIH grant CA106348, theEdward Mallinckrodt Jr. Foundation, the Elsa U. Pardee Foundation, the GeneralMotors Cancer Research Foundation, the V Foundation for Cancer Research,University of Pittsburgh Cancer Institute Lung Specialized Program of ResearchExcellence career development award, and the Outstanding Overseas YoungScholarAward from the Chinese Natural Science Foundation (L. Zhang).The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.Requests for reprints: JianYu, University of Pittsburgh Cancer Institute, HillmanCancer Center Research Pavilion, Suite 2.26H, 5117 Centre Avenue, Pittsburgh, PA15213. Phone: 412-623-7786; Fax: 412-623-7778; E-mail: yuj2@upmc.edu.

F2006 American Association for Cancer Research.doi:10.1158/1078-0432.CCR-05-2429

www.aacrjournals.orgClin Cancer Res 2006;12(9) May1, 2006 2928

Cancer Research. on November 28, 2020. © 2006 American Association forclincancerres.aacrjournals.org Downloaded from

apoptotic regulators. p53 mutations in cancer cells almostinvariably abolish this activity, indicating that the apoptoticfunction of p53 is critical for its tumor suppressor activity (11).Restoration of the p53 pathway by activating p53 itself or p53downstream targets has been explored to improve efficacy ofanticancer therapies (12).PUMA was recently identified as a BH3-only Bcl-2 family

protein and an essential mediator of p53-dependent andp53-independent apoptosis (13). PUMA is induced by DNA-damaging agents in a p53-dependent fashion and by non-genotoxic agents independent of p53 (14–16). Expression ofPUMA rapidly kills a variety of human cancer cells (14, 15).PUMA is localized in the mitochondria and induces apo-ptosis by activating caspases through mitochondrial dysfunc-tion (14, 15, 17). PUMA functions through other Bcl-2family members, such as Bax, Bcl-2, and Bcl-XL. Deletion ofPUMA in human cancer cells attenuated apoptotic responseto p53, DNA-damaging agents, and hypoxia (17). PUMA-knockout mice recapitulated several major apoptotic defi-ciencies observed in p53-knockout mice (18, 19). However,the role of PUMA in therapeutic responses in lung cancercells remains unclear.In this study, we examined the regulation of PUMA by

commonly used anticancer agents in lung cancer cells. Wefound that p53 mutations abolish the induction of PUMAby these agents. We also showed that PUMA is a potent andselective inducer of apoptosis in lung cancer cells and thatexpression of PUMA enhances the therapeutic responses oflung cancer cells to chemotherapeutic agents and g-irradiation.

Materials and Methods

Cell culture and drug treatment. The cell lines used in the studywere from American Type Culture Collection (Manassas, VA), exceptfor the lung cancer cell lines 273T and 201T, which were from theUniversity of Pittsburgh Cancer Institute lung cancer program, and thehuman primary fibroblast cell line WI-38, which was obtained fromthe Coriell Institute (Camden, NJ). All cell lines were maintained at37jC and 5% CO2. Cell culture media included RPMI 1640(Mediatech, Herndon, VA) for all lung cancer cell lines, McCoy’s 5A(Invitrogen, Carlsbad, CA) for colorectal cancer cell lines HCT116 andDLD1, DMEM (Mediatech) for 293 and 911 cells, and EMEM(Invitrogen) for WI-38. The cell culture media were supplementedwith 10% fetal bovine serum (HyClone, Logan, UT), 100 units/mLpenicillin, and 100 Ag/mL streptomycin (Invitrogen), except forEMEM, which was supplemented with 20% fetal bovine serum.Transfection was done with Lipofectamine 2000 (Invitrogen) follow-ing the instructions of the manufacturer. Reporter assays were carriedout as previously described (14).The anticancer drugs used in the study, including Adriamycin,

oxaliplatin, cisplatin, 5-fluorouracil (5-FU), etoposide, and Taxol, werepurchased from Sigma (St. Louis, MO). Caspase inhibitors, includingthe pan-caspase inhibitor Z-VAD-fmk, caspase-3 inhibitor Z-DEVD-fmk, caspase-8 inhibitor Z-IETD-fmk, and caspase-9 inhibitor Z-LEHD-fmk, were purchased from R&D Systems (Minneapolis, MN).All drugs and inhibitors were dissolved in DMSO and diluted toappropriate concentrations with cell culture media. Some cells wereexposed to g-irradiation at 8 Gy. For combination treatments, cellswere infected with adenoviruses for 16 to 18 hours before drugtreatment or g-irradiation.Constructs and adenoviruses. The expression constructs used in

the study included those expressing wild-type and different forms ofmutant p53, including V143A, R175H, R249S, and R273H (20), andconstructs expressing NH2-terminal green fluorescent protein (GFP)–

tagged proteins, including wild-type PUMA, PUMA lacking the BH3domain (DBH3), or the mitochondrial targeting sequence of PUMA(C43; ref. 17). For reporter assays, the previously described PUMAluciferase reporter constructs Frag 1 and Frag 2 were cotransfected withthe h-galactosidase reporter pCMVh (Promega; ref. 14).The recombinant adenoviruses Ad-PUMA, Ad-DBH3, and Ad-p53

were constructed using the Ad-Easy system as previously described(17, 21). High-titer viruses were produced in 293 cells and purified byCsCl2 gradient ultracentrifugation (22).Apoptosis and growth assays. After treatment, attached and floating

cells were harvested and analyzed for apoptosis by nuclear staining withpropidium iodide (23). A minimum of 300 cells were analyzed for each

treatment. The propidium iodide–stained cells were analyzed by flow

cytometry to determine the fraction of sub-G1 cells. Apoptosis wasalso analyzed by staining cells with Annexin V-Alexa 594 (Molecular

Probes, Carlsbad, CA), counterstaining with 4V,6-diamidino-2-phenyl-

indole, followed by flow cytometry. Cell growth was measured by

Fig. 1. p53-dependent activation of PUMA by chemotherapeutics in lung cancercells. A , eight lung cancer cell lines with different p53 status were treated withAdriamycin (0.2 Ag/mL) for 24 hours and analyzed for the expression of PUMA,p21, and p53 byWestern blotting. a-Tubulin was used as a loading control.B, reverse transcription-PCR was done for PUMA (25 and 35 cycles) and thecontrol GAPDH (20 cycles) using the cDNA prepared from cells with or withoutAdriamycin treatment. MW, molecular weight marker; Bl, no-template PCR control.C, PUMA luciferase reporters were cotransfected with wild-type (Wt) or mutantp53, along with the control pCMVh plasmid, into H1299 cells. Frag 1 (f500 bp)and Frag 2 (f200 bp) are previously described reporters with and without thep53-binding sites, respectively (14).The relative luciferase activities normalized tothe h-galactosidase activity were determined 48 hours following transfection.

Chemosensitization and Radiosensitization by PUMA

www.aacrjournals.org Clin Cancer Res 2006;12(9) May1, 20062929

Cancer Research. on November 28, 2020. © 2006 American Association forclincancerres.aacrjournals.org Downloaded from

3-(4,5-dimethyl-thiazol-2yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfo-

phenyl)-2H-tetrazolium (MTS) assay in 96-well plates (2,500 cells perwell) using the CellTiter 96 AQueous One Solution (Promega, Madison,

WI) following the instructions of the manufacturer. A490 nm was

measured using a Victor III (Perkin-Elmer/Wallace) plate reader. Eachexperiment was done in triplicate and repeated at least twice.

Western blotting and antibodies. Western blotting was done aspreviously described (24). The antibodies used for Western blotting

included rabbit polyclonal antibodies against PUMA (17), caspase 9

(Cell Signaling Technology, Danvers, MA), caspase 3 (Stressgen, SanDiego, CA), and HA (Santa Cruz Biotechnology, Santa Cruz, CA);

monoclonal antibodies against a-tubulin (Oncogene Sciences, SanDiego, CA), p21 (Oncogene Sciences), cytochrome c (BD Biosciences,

San Jose, CA), cytochrome oxidase subunit IV (Molecular Probes),

caspase 8 (Cell Signaling Technology), and p53 (DO1, Santa CruzBiotechnology); and a goat antibody against apoptosis-inducing factor

(Santa Cruz Biotechnology).Cellular fractionation. Floating and attached cells were harvested

from two 75-cm2 (T75) flasks by centrifugation, resuspended inhomogenization buffer [0.25 mol/L sucrose, 10 mmol/L HEPES(pH 7.4), and 1 mmol/L EGTA], and subjected to 40 strokes ofhomogenization on ice in a 2-mL Dounce homogenizer. Thehomogenates were centrifuged at 1,000 � g at 4jC for 10 minutesto pellet nuclei and unbroken cells. The supernatant was subsequentlycentrifuged at 10,000 � g at 4jC for 15 minutes to obtain cytosolic(supernatant) and mitochondrial (pellet) fractions.Reverse transcription-PCR. Total RNA was isolated using the

RNAgents Total RNA Isolations System (Promega) according to theinstructions of the manufacturer. First-strand cDNA was synthesizedusing Superscript II reverse transcriptase (Invitrogen). Reversetranscription-PCR was done to amplify PUMA using the cycleconditions previously described (14). The primers used to amplifyPUMA included 5-tcctcagccctcgctctcgc-3Vand 5V-ccgatgctgagtccatcagc-3V.The primers for amplifying the control glyceraldehyde-3-phosphatedehydrogenase (GAPDH) were 5V-ctcagacaccatggggaaggtga-3V and 5V-atgatcttgaggctgttgtcata-3V.Xenograft tumors and tissue staining. All animal experiments were

approved by the Institutional Animal Care and Use Committee at

the University of Pittsburgh. Xenograft tumors were established by

s.c. injection of 4 � 106 A549 or H1299 cells into both flanks of

5- to 6-week-old female athymic nude mice (Harlan, Indianapolis,IN). Tumor treatment was initiated by injecting each tumor

(f50-100 mm3) with Ad-PUMA or Ad-DBH3 at 5 � 108 plaque-

forming units in 100 AL of PBS. Each treatment was repeated

twice. Tumor growth was monitored thrice a week by calipers to

calculate tumor volumes according to the formula (length �width2) / 2.Terminal deoxyribonucleotidyl transferase–mediated dUTP nick end

labeling staining on frozen sections was done using recombinantterminal transferase (Roche, Indianapolis, IN) and dUTP-Alexa 594(Molecular Probes) according to the instructions of the manufacturersand counterstained by 4V,6-diamidino-2-phenylindole. PUMA expres-sion in frozen sections was determined by fluorescence microscopicanalysis of GFP. Frozen sections were also analyzed by H&E staining. Allimages were acquired with a Nikon TS800 fluorescence microscopeusing SPOT camera imaging software.Statistical analysis. Statistical analysis was done using GraphPad

Prism IV software. P values were calculated by Student’s t test or two-way ANOVA. The mean F SD were displayed in the figures.

Results

PUMA Induction by chemotherapeutic agents is abolished inp53-deficient lung cancer cells. PUMA is normally expressed atlow levels in human tissues but can be induced by p53 or DNA-damaging agents (14, 15). To investigate the regulation of

PUMA by anticancer agents in lung cancer cells, eight lungcancer cell lines with different p53 status were treated withAdriamycin (0.2 Ag/mL), a DNA-damaging agent producingdouble-stranded DNA breaks and a chemotherapeutic drugcommonly used for treating lung cancer (25). The expression ofPUMA and p21, which mediate p53-dependent apoptosis andcell-cycle arrest, respectively, were analyzed (11). Both PUMAand p21 were found to be induced by Adriamycin in the wild-type p53 cell lines A549 and 128.88T, but not in the p53-mutant cell lines DMS53, 201T, 273T, and H1752 and thep53-null cell lines H1299 and Calu 1 (Fig. 1A). PUMA was alsofound to be induced only in the wild-type p53 lung cancer cellsby another chemotherapeutic agent, 5-FU (50 Ag/mL), andg-irradiation (8 Gy; data not shown).To determine whether the deficiency in PUMA induction

is at the mRNA level, semiquantitative reverse transcription-PCR was used to examine PUMA in cells with and withoutAdriamycin treatment. PUMA transcripts were found to beincreased in the wild-type p53 cells but not in the p53-mutant or p53-null cells (Fig. 1B). To test whether p53activates the PUMA promoter in lung cancer cells, PUMAreporter constructs, along with wild-type p53 and severaltumor-derived p53 mutants (10), were cotransfected intoH1299 cells. The reporter containing the p53 responsiveelements was activated by wild-type p53 but not by p53mutants (Fig. 1C).These results indicate that the induction of PUMA by

chemotherapeutic agents in lung cancer cells is mediated byp53 and that this induction is abolished in p53-deficient lungcancer cells.PUMA profoundly suppresses growth of lung cancer cells

through induction of apoptosis. The proapoptotic function ofPUMA and lack of PUMA induction in p53-deficient lungcancer cells prompted us to investigate whether PUMAsuppresses lung cancer cell growth. Six lung cancer cellslines (A549, Caul 1, 128.88T, H1299, H1752, and DMS53)were infected with an adenovirus expressing PUMA (Ad-PUMA) or a control adenovirus (Ad-DBH3) expressing PUMAlacking the BH3 domain, which is essential for its proapop-totic function (14). For all six cell lines, at least 80% of thecells were infected by Ad-PUMA and Ad-DBH3 as indicatedby the GFP signal (data not shown). After infection for 48hours, cells were analyzed by 3-(4,5-dimethyl-thiazol-2yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazoliumassay. Ad-PUMA was found to cause profound growthsuppression in all the cell lines whereas Ad-DBH3 hadvirtually no effect on cell growth compared with theuntreated cells (Fig. 2A).The cells infected by Ad-PUMA, but not those infected by Ad-

DBH3, underwent massive apoptosis revealed by nuclearstaining, cell cycle analysis, and Annexin V staining (Fig. 2Band C). Activation of caspases 3, 8, and 9 was detected in all celllines after Ad-PUMA infection and correlated with growthsuppression (Fig. 2A and D). Interestingly, DMS53 cells weremost sensitive to Ad-PUMA infection, showing signs of lateapoptosis 48 hours after infection, including PUMA degrada-tion and fully activated caspases (Fig. 2A and D; ref. 14). Toexamine whether PUMA-induced apoptosis in lung cancer cellsis mediated through mitochondrial pathway (26, 27), cytosolicfractions were isolated from H1299 cells infected with Ad-PUMA and Ad-DBH3 for 48 hours, and then analyzed by

Cancer Therapy: Preclinical

www.aacrjournals.orgClin Cancer Res 2006;12(9) May1, 2006 2930

Cancer Research. on November 28, 2020. © 2006 American Association forclincancerres.aacrjournals.org Downloaded from

Western blotting. Cytochrome c and apoptosis-inducing factorwere found to be released into the cytosol in thecells infected with Ad-PUMA but not in those infected withAd-DBH3 (Fig. 2E).These data show that PUMA is a potent inducer of growth

suppression and apoptosis in lung caner cells and that PUMApromotes the release of mitochondrial apoptogenic proteinsand caspase activation to induce apoptosis in lung cancercells.PUMA sensitizes lung cancer cells to chemotherapeutic agents

and g-irradiation. Abnormalities of apoptosis regulation havebeen shown to contribute to the development of resistance tochemotherapy and radiation therapy in cancer cells (4–6). Theimportant role of PUMA in DNA damage–induced and p53-dependent apoptosis suggests that elevated PUMA expressionmay restore sensitivity of cancer cells to anticancer agents. Totest this hypothesis, A549 lung cancer cells were treated withlow dose of Ad-PUMA [10 multiplicity of infection (MOI)],alone or in combination with g-irradiation or chemotherapeu-tic agents, including Taxol, 5-FU, oxaliplatin, cisplatin, etopo-side, and Adriamycin. PUMA was found to significantlyenhance the growth inhibitory effects of these chemotherapeu-tic drugs and g-irradiation, with the synergy most pronouncedwhen combined with DNA-damaging agents, including Adria-mycin, cisplatin, etoposide, and g-irradiation (Fig. 3A). For

example, as much as 8-fold increase in growth suppression wasachieved when Ad-PUMA was combined with Adriamycin. Wealso determined the IC50s of several chemotherapeutic agentsin A549 cells with or without Ad-PUMA and found thatAd-PUMA significantly lowered the IC50s of these agents by3-fold (Adriamycin) to over 10-fold (Taxol and 5-FU; Fig. 3B;Table 1).We then determined whether PUMA expression sensitizes

lung cancer cells to the anticancer agents through induction ofapoptosis. A549 cells are resistant to apoptosis induced byAdriamycin (up to 2 Ag/mL) and g-irradiation (up to 8 Gy; data

Table 1. IC50 of the chemotherapeutics in A549 cellswith or without Ad-PUMA

Drug IC50

�Ad-PUMA +Ad-PUMA

Taxol 74.0 nmol/L 5.0 nmol/LCisplatin 110.7 Amol/L 18.5 Amol/L5-FU 77.8 Ag/mL 7.5 Ag/mLEtoposide 48 Amol/L 6 Amol/LAdriamycin 0.76 Ag/mL 0.22 Ag/mL

Fig. 2. PUMA suppresses the growth of lung cancer cells through induction of apoptosis. A , cell growth was measured by 3-(4,5-dimethyl-thiazol-2yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium assay at 48 hours following adenoviral infection (100 MOI forA549 and 128.88T; 25 MOI for others) asdescribed in Materials and Methods.The growth of untreated cells was defined as 100%. B, after the indicated treatments for 48 hours, H1299 cells were visualized withphase-contrast (Phase) or fluorescence microscopy following nuclear staining with propidium iodide (PI). C, after the indicated treatments for 24 hours, H1299 cells werestained by propidium iodide orAnnexinV/4V,6-diamidino-2-phenylindole, and then analyzed by flow cytometry to determine the fraction of sub-G1cells and early apoptotic cells(R4), respectively. D, PUMA expression and caspase activation in lung cancer cell lines were analyzed byWestern blotting 48 hours after adenoviral infection. Arrows,active forms of caspases. E , release of cytochrome c and apoptosis-inducing factor (AIF) was examined in the cytosolic fractions of the cells following indicated treatmentsbyWestern blotting. a-Tubulin and cytochrome oxidase subunit IV (Cox IV), which are localized in the cytosol and mitochondria, respectively, were used as controls forfractionation.

Chemosensitization and Radiosensitization by PUMA

www.aacrjournals.org Clin Cancer Res 2006;12(9) May1, 20062931

Cancer Research. on November 28, 2020. © 2006 American Association forclincancerres.aacrjournals.org Downloaded from

not shown). Ad-PUMA (10 MOI) and Adriamycin (0.2 Ag/mL)alone did not induce significant apoptosis in A549 cells(Fig. 4A). However, almost 90% of cells underwent apoptosisafter the combination treatment for 72 hours (Fig. 4A).Similarly, a combination of g-irradiation (8 Gy) with Ad-PUMA also led to a markedly enhanced apoptotic responsein A549 cells (Fig. 4B). In contrast, the control Ad-DBH3 hadno effect when combined with Adriamycin or g-irradiation(Fig. 4A and B). Analysis of apoptosis using other approaches,including nuclear staining and cell cycle analysis, confirmedthese results (data not shown). Furthermore, the apoptosis wasaccompanied by enhanced activation of caspases 3, 8, and 9(Fig. 4C), as well as release of cytochrome c and apoptosis-

inducing factor into the cytosol (Fig. 4E). Pretreating the cellswith caspase inhibitors significantly decreased the apoptosisand growth inhibitory effects of the combination treatments(Fig. 4D and data not shown). Similarly, PUMA was found toenhance apoptosis induced by Adriamycin and g-irradiation in128.88T cells (data not shown).These data indicate that PUMA can synergize with different

chemotherapeutic agents and irradiation to trigger the release ofcytochrome c and apoptosis-inducing factor and caspaseactivation to initiate apoptosis in lung cancer cells.PUMA is selectively toxic to cancer cells and more potent than

p53 in growth suppression. Selectivity, or therapeutic index,is an important issue to consider for any therapeutic agent.

Fig. 3. PUMA sensitizes lung cancer cells to chemotherapeutics andg-irradiation. A549 cells were treated with chemotherapeutic agentsand g-irradiation, alone or in combination with Ad-PUMA orAd-DBH3(10 MOI), as described in Materials and Methods. Cell growthwas measured by 3-(4,5-dimethyl-thiazol-2yl)-5-(3-carboxyme-thoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium assay. A , ratios ofgrowth inhibition conferred by the combinations to that of drugsalone.Taxol, 5 nmol/L; 5-FU, 25 Ag/mL; Oxa, oxaliplatin, 25 Amol/L;Cis, cisplatin, 25 Amol/L; Etp, etoposide, 2 Amol/L; Adr, Adriamycin,0.2 Ag/mL; IR, g-irradiation, 8 Gy. B, effects of the indicatedchemotherapeutic drugs at different concentrations, with or withoutcombination with Ad-PUMA, on cell growth.

Cancer Therapy: Preclinical

www.aacrjournals.orgClin Cancer Res 2006;12(9) May1, 2006 2932

Cancer Research. on November 28, 2020. © 2006 American Association forclincancerres.aacrjournals.org Downloaded from

Thus, the toxicities of PUMA in several cancer and normal/nontransformed cell lines were evaluated. Cells were transfectedwith a vector expressing NH2-terminal GFP-fused wild-typePUMA, a control vector expressing PUMA lacking the BH3domain (DBH3), or a vector expressing the mitochondriallocalization sequence of PUMA (C43; ref. 17). Because cellskilled by PUMA do not express significant amount of GFP(examples shown in Fig. 5A), the toxicities of PUMA todifferent cell lines can be compared through analysis of GFP-positive cells after transfection. This assay is similar to thewidely used h-galactosidase cotransfection method in assessingcytotoxicity (28). The fractions of GFP-positive cells wereindistinguishable in all cell lines after transfection with thecontrol vectors expressing DBH3 or C43 (Fig. 5A and B).However, transfection with wild-type PUMA diminished GFP-positive cells in the cancer cell lines, including the lung cancercell lines A549, H1299, Calu 1, 128.88T, and 201T and thecolon cancer cell lines HCT116 and DLD1. In contrast, PUMAhas little effect on the normal/nontransformed cell lines,including the primary human fibroblast cell line WI-38, the

immortalized but nontransformed kidney epithelial cell line293, and the retinal epithelial cell line 911 (P < 0.0001;Fig. 5B). These results suggest that PUMA is selectively toxic tocancer cells although the underlying mechanism of thisselectivity remains unclear.Adenovirus-mediated transfer of p53 has been extensively

explored in cancer gene therapy (12). To test the potentialuse of PUMA in cancer gene therapy, Ad-PUMA and anadenovirus expressing p53 (Ad-p53) constructed in the samesystem were compared for their potency in suppressing thegrowth of lung cancer cells. Strikingly, Ad-PUMA was foundto be at least 5- to 10-fold more potent than Ad-p53 insuppressing the growth of H1299 and A549 cells (Fig. 5C).We observed much higher expression of PUMA but lowerexpression of p21 in the cells infected with Ad-PUMAcompared with those infected with Ad-p53 at the same titer(Fig. 5D). We also compared the chemosensitization effectsof Ad-p53 to those of Ad-PUMA and found that Ad-p53(10 MOI) resulted in little enhancement of these effects inA549 cells (data not shown). These results show that PUMA is

Fig. 4. The sensitizing effects of PUMA are mediated by enhanced apoptosis induction. A549 cells were subjected to the indicated treatments and analyzed for apoptosis asdescribed in Materials and Methods. Adenoviruses were used at 10 MOI. A and B, apoptosis was analyzed by nuclear fragmentation assays. Adriamycin, 0.2 Ag/mL;g-irradiation, 8 Gy. C, PUMA expression and caspase activation were analyzed byWestern blotting. Arrows, active forms of caspases.D, cells were pretreated with the caspaseinhibitors (20 Amol/L) for 4 hours and then subjected to the indicated treatments for 24 or 48 hours. Apoptosis was analyzed by nuclear fragmentation assay. E , release ofcytochrome c and apoptosis-inducing factor into the cytosol was analyzed byWestern blotting.

Chemosensitization and Radiosensitization by PUMA

www.aacrjournals.org Clin Cancer Res 2006;12(9) May1, 20062933

Cancer Research. on November 28, 2020. © 2006 American Association forclincancerres.aacrjournals.org Downloaded from

more potent than p53 in growth suppression and chemo-sensitization in lung cancer cells.PUMA suppresses tumor growth in vivo. To determine

whether PUMA confers antitumor activity in vivo , A549xenograft tumors (f50-100 mm3) were treated with threeinjections of Ad-PUMA and the control Ad-DBH3 at 5 � 108

plaque-forming units (Fig. 6A). To avoid potential systemiceffects of different viruses, Ad-PUMA and Ad-DBH3 wereinjected into separate tumors in the same animals. Ad-DBH3did not have any effect on tumor growth compared with PBSalone (data not shown), with tumors reaching eight times theinitial volumes in 24 days (Fig. 6A and B). In contrast, tumorssubjected to Ad-PUMA treatment grew much slower andreached less than twice the initial volume, with at least 80%growth suppression compared with those treated by Ad-DBH3(P < 0.01; Fig. 6A and B). Ad-PUMA was also used to treatH1299 xenograft tumors and found to suppress tumor growthby >80% (Fig. 6C). GFP fluorescence patterns indicated thattransgenes were highly expressed in the tumor cells 48 hoursfollowing the second injection (Fig. 6D). Analyzing tumorhistology by H&E staining and apoptosis in situ by terminaldeoxyribonucleotidyl transferase–mediated dUTP nick endlabeling staining revealed significant cell loss and large fractionsof apoptotic cells in the tumors treated with Ad-PUMA but notin those treated with Ad-DBH3 or PBS alone (Fig. 6D). Thesedata show that PUMA can effectively inhibit growth ofestablished tumors in vivo through induction of apoptosis.

Discussion

Defects in the p53 pathway, which often abolish p53-mediated apoptotic response to DNA damage, are thought toplay an important role in the development of drug resistancein cancer cells (29, 30). Although a number of proapoptoticproteins have been identified as p53 targets, only a few havebeen shown to play an important role in p53-dependentapoptosis in gene-targeting studies (13, 31). PUMA wasidentified as a p53 downstream gene that plays a critical rolein p53-dependent apoptosis. Targeted deletion of PUMA inhuman colorectal cancer cells resulted in resistance to apoptosisinduced by p53, the DNA-damaging agent Adriamycin, andhypoxia (17). Two recent studies using PUMA-knockout miceindicated that PUMA is an essential mediator of p53-dependentand p53-independent apoptosis in vivo (18, 19). It has alsobeen shown that the apoptotic mediators of p53, includingPUMA, the BH3-only proteins Bid and Noxa, and the deathreceptor DR5, can be induced by chemotherapeutics andirradiation in tissue-specific patterns, which correlated withradiosensitivity in these tissues (32). These observations, theprevalence of p53 mutations in lung cancer, as well asthe ineffectiveness of current lung cancer treatments, promptedus to investigate the role of PUMA in the therapeutic responsesof lung cancer cells.Our study showed that lung cancer cells with abnormali-

ties in p53 are deficient in the activation of PUMA by

Fig. 5. PUMA is selectively toxic to cancer cells and more potent than p53 in growth suppression.A , A549, H1299, and 911cells were transfected with the indicated plasmidsand analyzed by fluorescence microscopy. A549 and H1299 cells are shown at�200 magnification. 911cells are shown at�400 magnification.B, the fractions of GFP-positivecells were quantified after transfection and compared with those of GFP-C43 (100%). C, H1299 and A549 cells were infected with Ad-PUMA, Ad-p53, orAd-DBH3 at theindicated MOI for 48 hours. Cell growth was measured by 3-(4,5-dimethyl-thiazol-2yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium assay.The growthof the cells infected with Ad-DBH3 was defined as100%. D, p53, PUMA, and p21were analyzed byWestern blotting afterAd-p53 orAd-PUMA infection at the indicated MOIfor 24 hours.

Cancer Therapy: Preclinical

www.aacrjournals.orgClin Cancer Res 2006;12(9) May1, 2006 2934

Cancer Research. on November 28, 2020. © 2006 American Association forclincancerres.aacrjournals.org Downloaded from

chemotherapeutics and irradiation, which is likely due todeficiency in the activation of PUMA promoter by p53. Theseresults suggest that lack of PUMA induction may contribute tothe development of resistance to anticancer agents in lungcancer. Delivery of PUMA into lung cancer cells throughadenoviral gene transfer resulted in apoptosis and enhancedsensitivity to chemotherapeutic agents and g-irradiation,suggesting that adequate levels of PUMA are crucial fortriggering apoptotic responses to these agents. Interestingly,PUMA seems to be the most effective in enhancing growthsuppression and apoptosis when combined with DNA-damag-ing agents, such as Adriamycin, etoposide, cisplatin, and g-irradiation (Fig. 3A). This observation is consistent with thenotion that PUMA plays a critical role in DNA damage–induced apoptosis (17–19). PUMA also synergized with otherclasses of chemotherapeutic agents, including the microtubulepoison Taxol and antimetabolite 5-FU, at lesser but stillsignificant levels (Fig. 3A). These observations suggest thatPUMA is rather a broad-spectrum chemosensitizer and radio-sensitizer of lung cancer cells and warrants further evaluation.Restoration of the p53 pathway by introducing p53 itself or

p53 downstream targets into cancer cells has become anattractive approach in cancer gene therapy (12). Nonreplicatingadenoviruses expressing p53 have been shown to suppress cellgrowth and induce apoptosis in several types of cancer in vitroand in vivo (33, 34). Radiation therapy and p53 adenoviral

gene transfer were shown to have synergistic growth-suppres-sive effects in a variety of cancer cells, including those of lung,colorectal, ovarian, and head and neck cancer (35–38).Clinical trials testing combinations of p53 gene replacementwith chemotherapy have yielded some encouraging results inlung cancer patients (33, 34). Depending on its expressionlevel, p53 can induce cell cycle arrest or apoptosis in lungcancer cells (39). In our study, Ad-PUMA seems to be morepotent than Ad-p53 at the same titer in growth suppression,apoptosis induction, and chemosensitization (Fig. 5C and datanot shown). This can be explained by the much higher levels ofPUMA but lower levels of p21 induced by Ad-PUMA comparedwith Ad-p53 (Fig. 5D), as high level of p21 is known to inhibitapoptosis (17). These observations suggest that PUMA might bea more effective radiosensitizer and chemosensitizer for lungtumors compared with p53. However, this possibility needs tobe further evaluated in lung cancer cells and other in vivo tumormodels.A recent study showed that PUMA can regulate diverse

apoptotic pathways by antagonizing all known antiapoptoticBcl-2 family members, including Bcl-2, Bcl-XL, Bcl-w, Mcl-1,and A1, which are frequently overexpressed in cancer cells,suggesting that PUMA may be useful for targeting a variety ofapoptotic defects in cancer cells (40). PUMA was also foundto be required for apoptosis induced by oncogenes c-myc andE1A (18, 19). In addition, insulin-like growth factor I and

Fig. 6. PUMA suppresses the growth of xenograft lung tumors. A , the growth curve of A549 tumors (n = 5 per group) subjected to Ad-PUMA and Ad-DBH3 treatmentsas described in Materials and Methods. Arrows, treatments were done on days 0, 2, and 4, respectively. B, top, a representative A549 tumor-bearing mouse before(Pre-treatment) and 22 days after the indicated treatment (Post-treatment). Bottom, tumors from different treatment groups at the end of the experiment. C, growth curve ofH1299 tumors (n = 5 per group) subjected to Ad-PUMA and Ad-DBH3 treatments. Arrows, treatments were done on days 0, 3, and 6, respectively. D, frozen sections ofthe H1299 tumors 48 hours following the second injection were analyzed by fluorescence microscopy to determine the efficiency of adenoviral infection (GFP, �200).Apoptosis was analyzed by terminal deoxyribonucleotidyl transferase ^ mediated dUTP nick end labeling (TUNEL) staining (red, �200) with 4V,6-diamidino-2-phenylindolecounterstaining (blue).Tumor histology was analyzed by H&E staining (�400).

Chemosensitization and Radiosensitization by PUMA

www.aacrjournals.org Clin Cancer Res 2006;12(9) May1, 20062935

Cancer Research. on November 28, 2020. © 2006 American Association forclincancerres.aacrjournals.org Downloaded from

epidermal growth factor signaling pathways are frequentlyderegulated in cancer and can suppress PUMA expression inserum-starved tumor cells (16, 18, 19). Because tumor cellgrowth often relies on overexpression of oncogenes orantiapoptotic proteins (3), these observations may explainwhy tumor cells are much more sensitive to PUMA than normalcells (Fig. 5B).

Acknowledgments

We thank other members of our laboratories for helpful discussions and com-ments, Dr. Jill M. Siegfried at the University of Pittsburgh for providing lung cancercell lines, Dr. BertVogelstein at the Howard Hughes Medical Institute and the SidneyKimmel Comprehensive Cancer Center at Johns Hopkins for providing the p53expression constructs, and Dr. Pamela A. Hershberger for careful reading and criticalcomments.

References1. Jemal A, Murray T, Ward E, et al. Cancer statistics,

2005. CA CancerJ Clin 2005;55:10 ^ 30.2. Vogelstein B, Kinzler KW. Cancer genes and the

pathways they control. Nat Med 2004;10:789 ^ 99.3. Hanahan D, Weinberg RA. The hallmarks of cancer.

Cell 2000;100:57 ^ 70.4. Johnstone RW, Ruefli AA, Lowe SW. Apoptosis: a

link between cancer genetics and chemotherapy. Cell2002;108:153 ^64.

5. Yu J, Zhang L. Apoptosis in human cancer cells. CurrOpin Oncol 2004;16:19 ^ 24.

6. Coultas L, Strasser A. The molecular control ofDNA damage-induced cell death. Apoptosis 2000;5:491 ^ 507.

7. Reed JC. Apoptosis-based therapies. Nat Rev DrugDiscov 2002;1:111 ^21.

8. DrevsJ, Medinger M, Schmidt-Gersbach C,Weber R,Unger C. Receptor tyrosine kinases: the main targetsfor new anticancer therapy. Curr DrugTargets 2003;4:113 ^ 21.

9. Langer CJ. Emerging role of epidermal growth factorreceptor inhibition in therapy for advanced malignan-cy: focus on NSCLC. Int J Radiat Oncol Biol Phys2004;58:991 ^ 1002.

10. Hollstein M, Sidransky D,Vogelstein B, Harris CC.p53 mutations in human cancers. Science 1991;253:49 ^ 53.

11. Vogelstein B, Lane D, Levine AJ. Surfing the p53network. Nature 2000;408:307 ^ 10.

12. Lane DP, Lain S.Therapeutic exploitation of the p53pathway.Trends Mol Med 2002;8:S38 ^ 42.

13. Yu J, Zhang L. No PUMA, no death: implicationsfor p53-dependent apoptosis. Cancer Cell 2003;4:248 ^9.

14.YuJ, Zhang L, Hwang PM, Kinzler KW,Vogelstein B.PUMA induces the rapid apoptosis of colorectalcancer cells. Mol Cell 2001;7:673 ^82.

15. Nakano K,Vousden KH. PUMA, a novelproapoptoticgene, is inducedby p53. Mol Cell 2001;7:683 ^94.

16. Han J, Flemington C, Houghton AB, et al. Expres-sion of bbc3, a pro-apoptotic BH3-only gene, is regu-

lated by diverse cell death and survival signals. ProcNatl Acad Sci U S A 2001;98:11318 ^ 23.

17. Yu J, Wang Z, Kinzler KW, Vogelstein B, Zhang L.PUMA mediates the apoptotic response to p53 in co-lorectal cancer cells. Proc Natl Acad Sci U S A 2003;100:1931 ^ 36.

18.VillungerA, Michalak EM, Coultas L, et al. p53- anddrug-induced apoptotic responses mediated by BH3-only proteins Puma and Noxa. Science 2003;302:1036 ^ 8.

19. JeffersJR, Parganas E, LeeY, et al. Puma is an essen-tial mediator of p53-dependent and -independentapoptotic pathways. Cancer Cell 2003;4:321 ^ 8.

20. Baker SJ, Markowitz S, Fearon ER, Willson JK,Vogelstein B. Suppression of human colorectal carci-noma cell growth by wild-type p53. Science 1990;249:912 ^ 5.

21. Yu J, Zhang L, Hwang PM, et al. Identification andclassification of p53-regulated genes. Proc Natl AcadSci U S A1999;96:14517 ^ 22.

22. HeTC, Zhou S, da Costa LT, et al. A simplified sys-tem for generating recombinant adenoviruses. ProcNatl Acad Sci U S A1998;95:2509 ^ 14.

23. Kohli M, Yu J, Seaman C, et al. SMAC/Diablo-dependent apoptosis induced by nonsteroidal antiin-flammatory drugs (NSAIDs) in colon cancer cells.ProcNatl Acad Sci U S A 2004;101:16897 ^ 902.

24. Yu J,Tiwari S, Steiner P, Zhang L. Differential apo-ptotic response to the proteasome inhibitor bortezo-mib [Velcade(TM), PS-341] in Bax-deficient andp21-deficient colon cancer cells. Cancer Biol Ther2003;2:694 ^ 9.

25. Sandler AB. Chemotherapy for small cell lung can-cer. Semin Oncol 2003;30:9 ^ 25.

26. Wang X. The expanding role of mitochondria inapoptosis. Genes Dev 2001;15:2922 ^ 33.

27. Cory S, Huang DC, Adams JM. The Bcl-2 family:roles in cell survival and oncogenesis. Oncogene2003;22:8590 ^ 607.

28. Graeber TG, Osmanian C, Jacks T, et al. Hypoxia-mediated selection of cells with diminished apoptoticpotential in solid tumours. Nature 1996;379:88 ^91.

29. Lowe SW, Ruley HE, Jacks T, Housman DE. p53-dependent apoptosis modulates the cytotoxicity ofanticancer agents. Cell1993;74:957 ^ 67.

30. Bunz F, Hwang PM, Torrance C, et al. Disrup-tion of p53 in human cancer cells alters the re-sponses to therapeutic agents. J Clin Invest 1999;104:263 ^ 9.

31. Yu J, Zhang L. The transcriptional targets of p53 inapoptosis control. Biochem Biophys Res Commun2005;331:851 ^ 8.

32. Fei P, Bernhard EJ, El-Deiry WS. Tissue-specificinduction of p53 targets in vivo. Cancer Res 2002;62:7316 ^ 27.

33. Roth JA, Grammer SF. Tumor suppressor genetherapy. Methods Mol Biol 2003;223:577 ^ 98.

34. Roth JA, Grammer SF. Gene replacement therapyfor non-small cell lung cancer: a review. HematolOncol Clin North Am 2004;18:215 ^ 29.

35. Gjerset RA,Turla ST, Sobol RE, et al. Use of wild-type p53 to achieve complete treatment sensitizationof tumor cells expressing endogenous mutant p53.Mol Carcinog 1995;14:275 ^ 85.

36. Spitz FR, Nguyen D, Skibber JM, et al. Adenoviral-mediated wild-type p53 gene expression sensitizescolorectal cancer cells to ionizing radiation. ClinCancer Res 1996;2:1665 ^ 71.

37. Gallardo D, Drazan KE, McBride WH. Adenovirus-based transfer of wild-type p53 gene increases ovar-ian tumor radiosensitivity. Cancer Res 1996;56:4891 ^ 3.

38. Xu L, Pirollo KF, Chang EH. Transferrin-liposome-mediated p53 sensitization of squamous cell carcino-ma of the head and neck to radiation in vitro. HumGeneTher1997;8:467 ^ 75.

39. Chen X, Ko LJ, Jayaraman L, Prives C. p53 levels,functional domains, and DNA damage determine theextent of the apoptotic response of tumor cells. GenesDev 1996;10:2438 ^ 51.

40. Chen L,Willis SN,Wei A, et al. Differential targetingof prosurvival Bcl-2 proteins by their BH3-onlyligands allows complementary apoptotic function.Mol Cell 2005;17:393^403.

Cancer Therapy: Preclinical

www.aacrjournals.orgClin Cancer Res 2006;12(9) May1, 2006 2936

Cancer Research. on November 28, 2020. © 2006 American Association forclincancerres.aacrjournals.org Downloaded from

2006;12:2928-2936. Clin Cancer Res   Jian Yu, Wen Yue, Bin Wu, et al.   Agents and IrradiationPUMA Sensitizes Lung Cancer Cells to Chemotherapeutic

  Updated version

  http://clincancerres.aacrjournals.org/content/12/9/2928

Access the most recent version of this article at:

   

   

  Cited articles

  http://clincancerres.aacrjournals.org/content/12/9/2928.full#ref-list-1

This article cites 40 articles, 13 of which you can access for free at:

  Citing articles

  http://clincancerres.aacrjournals.org/content/12/9/2928.full#related-urls

This article has been cited by 11 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

  .pubs@aacr.orgDepartment at

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

  Permissions

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

.http://clincancerres.aacrjournals.org/content/12/9/2928To request permission to re-use all or part of this article, use this link

Cancer Research. on November 28, 2020. © 2006 American Association forclincancerres.aacrjournals.org Downloaded from