(CANCERRESEARCH58, 1723—1729.April 15. 19981

therapy-induced apoptosis (16, 17). There is also a good correlationbetween a tumor's p53 functional status and its response to somechemotherapeutic agents (i8—20). In the past few years, p53 functionhas been reported to be redox-regulatable in vitro through its cysteineresidues (21—23).In the present report, we extend the previous observations on the redox regulation of p53 protein to the cellular level.We demonstrate that several sulfur-containing antioxidants selectivelyinduce p53-dependent apoptosis in transformed cells. In contrast,antioxidants whose action is limited to scavenging radicals do notseem to have this activity.

MATERIALS AND METHODS

CellLines. PrimaryMEF cells(passage,<5) andE1A/Ha-ras-transformedMEFs were generously provided by Dr. S. Lowe (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY) (18). The 308 papilloma cell line, which hasa mutant Ha-ras allele (24) and wild-typep53 (25), was the gift of Dr. S. Yuspa(National Cancer Institute, Bethesda, MD). The (10)1 MEF cell line was thegift of Dr. A. Levine (Princeton University, Princeton, NJ) (26). BALB/c 3T3A3i cells were obtained from the American Type Culture Collection (Bethesda, MD). The 308 papilioma cells were maintained in 0.05 mMCa24 Eagle'sminimal essential medium with 10% fetal bovine serum, and all other cells

were grown in DMEM with 10% heat-inactivated fetal bovine serum. Cellculture confluence was maintained below 80%.

Chemicals and Cell Treatments. All chemicals were obtained fromSigma, except for Trolox, which was purchased from Aldrich. These compounds were freshly dissolved in medium and adjusted to a neutral pH, ifnecessary, or (for vitamin E acetate and BRA) first dissolved in ethanol andadded to the medium. Cell viability was determined by trypan blue exclusionas a measure of cell death independent of any growth suppression by p53.

Apoptosis Assays. For fixed cells, apoptosis-associated DNA strand breakswere visualized by fluorescent in situ end labeling as described previously(27). For isolated DNA, DNA fragmentation analysis was performed (18).

For flow cytometry analysis, approximately 106cells/sample were washedwith ice-cold PBS and fixed in 95% ethanol. Cells were then resuspended in1 mg/nil RNase (Sigma) for 30 mm at 37°Cand stained with 0.05 mg/mipropidium iodide (Sigma) for 1 h on ice. Flow cytometric analysis was

performed with a fluorescence-activated cell-sorting Vantage flow cytometer(Becton Dickinson). Cells were excited at 488 nm, and the emission wasdetected through a 630/22-nm band pass filter. A minimum of 10,000 cellswere analyzed for each sample. Cell cycle analysis was performed usingModfit 5.2 software (Verity Software House). Cells were considered to be inapoptosis ifthey exhibited sub-G1 DNA fluorescence and a forward angle lightscatter the same as or slightly lower than that ofcells in G@(28). Cellular debriswas gated out using the electronic threshold.

Northern and Western Blot Analysis. Northern and Western blot analyses were performed as described previously (25).

Analysis of p53 PrOteIn Synthesis. Cultures of 308 cells were treated inthe absence or presence of 20 mi@tNAC. At 4.5 h posttreatment, cells wereincubated with methionine-free medium containing 2% dialyzed and chelexedfetal bovine serum for 0.5 h. At 5 h posttreatment, biosynthetic labeling wasinitiated by adding 200 @Ciof [35S]methionineper milliliter of methioninefree medium. The labeling was terminated at 5, 10, or 15 mm. Throughout theexperiment, 20 mMNAC was included in the group ofNAC-treated cells. Cellswere then washed twice with 10 ml of ice-cold PBS, scraped, and pelleted bycentrifugation at 1500rpm at 4°Cfor 5 mm. The supernatant was removed, andthe cell pellet was lysed in ice-cold cell lysis buffer [0.5% Triton X-lOO,300mM NaCl, 50 mM Tris-HC1 (pH 7.4), 10 pg/mi leupeptin, and 0.1 mMphenylmethylsulfonyl fluoride]. Aliquots of cell lysate containing equal

1723

Antioxidant Action via p53-mediated Apoptosis1

Ming Liu, Jifi C. Pelling, Jingfang Ju, Edward Chu, and Douglas E. Brash2

Departments of l'herapeutic Radiology [M. L, D. E. B.J and Genetics ID. E. B.J, Yak University School of Medicine, New Haven, Connecticut 06520; Department of Pathologyand Laboratory Medicine, University of Kansas Medical Center, Kansas City, Kansas 66160 Ii. C. P.1; and Departments of Medicine and Pharmacology. Yale Cancer Center,YaleUniversitySchoolofMedicineand VAConnecticutHealthcareSystem,NewHaven,Connecticut06520Ii. J., E.C.]

ABSTRACT

The biological meets ofantloxidants are often considered in terms of their@edsonoxygenorlipidradicals.However,antio@ddantscanalsoexerttheir@ctsthrough altering the cellular redox potentiaL Herein, we report that

suIbJrh.alntaining antioxidants audi as N-acetylcysteine and dimercaptopropsnol induced apoptosis in several trai@ormed cell lines and transformedprinal7 nsltures but not in normal celLs.In contrast, chain-breaking antioxidants such as vitamin E lacked this activity. An increased glutathione levelw—nst required for apoptosls however, all apOptOsiS-indUcingantioxidantselevated the total cellular thiol levels.Antioxidant-induced apoptosis requiredthe p53 tumor suppressor gene@N-Acetylcysteine elevated p53 expression

@rameriptionallyby increasing the rate of p53 mRNA translation ratherthan by altering the protein stability. The p53 induction occurred in normalcalk. These observations indicate a redox sensor for p53 inductIon in vivo,with additional transformation-speelfic information being required for apopto@s.Manipulating p53-dependent apoptosLswith nontoxic antioxidants mayhave a direct dinical app&ation.

INTRODUCTION

Antioxidants have a wide range of biochemical activities. Theseinclude inhibiting the generation of reactive oxygen species, directlyor indirectly scavenging free radicals, and altering the intracellularredox potential (1). Some antioxidants have been used as inhibitors ofapoptosis, because apoptosis was at first thought to be mediated byoxidative stress (2). However, it is clear that reactive oxygen speciesare not always required to induce apoptosis (3, 4). Furthermore,antioxidants have been shown to trigger apoptosis in smooth musclecells independent of oxidative reactions (5).

Pro-oxidant states have been considered to be contributing factorsfor tumorigenesis (6). Correspondingly, an increasing body of cvidance indicates that antioxidants have anticancer activities. Antioxidants can inhibit tumor initiation, tumor promotion, and cell transformation (7, 8). For example, the antioxidant NAC3 has antimutagemcand chemopreventive activities in a variety of organs, such as thelung, liver, skin, and colon (9—13).NAC inhibits the transformation ofcultured rat tracheal epithelial cells by the carcinogen benzo(a)pyrene,as well (7).

The tumor suppressor protein p53 is also known to play an important role in inhibiting tumorigenesis. This transcription factor is in

volved in cell cycle arrest and apoptosis after DNA damage (14, 15).Manipulating p53-mediated pathways has thus been a major focus forcancer therapy (14, 15). For example, restoring the expression ofwild-type p53 renders cells more sensitive to spontaneous or chemo

Received 11/18/97; accepted 2/18/98.Thecostsof publicationof thisarticleweredefrayedin partby thepaymentof page

charges. This article must therefore be hereby marked advertisement in accordance with18U.S.C.Section1734solelyto indicatethis fact.

1 Supported by NIH Grants CA55737 (to D. E. B.), CA40847 (to J. C. P.), and

CA16359 (to E. C.) and the Yale Cancer Center Flow Cytometry Shared Resource, andby the LeslieH. WarnerPostdoctoralFellowshipin CancerResearch(to M. L.).

2 To whom requests for reprints should be addressed, at Department of Therapeutic

Radiology,YaleUniversitySchoolofMedicine,NewHaven.CT06520-8040.Fax:(203)785-6309; E-mail: donglas.brash@yaIe.edu.

3 1@abbreviafions used are: NAC, N-acetylcysteine; BHA, butylated hydroxyanisole;BSO, buthionine suifoximine; DMP, 2,3-dimercaptopropanol; GSH, glutathione; MEF,mouseembryofibroblast;OTC,L-2-oxo-4-thiazolidinecarboxylate;tMEF,transformedmouse ensbiyo fibroblast; PDTC, pyrrolidinedithiocarbamate; C/EBP@3,CAAT/enhancerbinding protein @.

on July 26, 2021. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

,

.‘ 4

p53-MEDIATED APOVrOSIS BY ANTIOXIDANTS

amounts of protein (30 j.@g)were subjected to immunoprecipitation analysiswith anti-p53 antibody PAbl22 (25) and protein A-agarose (Life Technologies, Inc.). The immunoprecipitated proteins were resolved on a 10% SDSPAGE gel. The levels of synthesized p53 proteinwere then determinedbydensitometric scanning using a Hewlett-Packard ScanJet 4P scanner and NIHImage 1.59 analysis software.

Analysis of p53 Protein Half-Life. Cultures of 308 cells were treated inthe absence or presence of 20 msi NAC. After a 3.5-h incubation, cells wereincubated with methionine-free medium containing 2% dialyzed and chelexedfetal bovine serum for 0.5 h. The cells were then labeled by adding 100 @Ciof [35S]methionine per milliliter of methionine-free medium for 1 h. At 5 h

posttreatment, cells were washed with PBS and incubated with a chase mediumcontaining a 2-fold excess of unlabeled methionine (45 gg/ml) and cysteine(72 @g/ml)for 0, 20, or 40 mm. Throughout the experiment, 20 mM NAC was

included in the group of NAC-treated cells. Aiiquots of each sample lysatewere subjected to immunoprecipitation analysis as described in the measurement of the p53 protein synthesis rate.

Measurement of GSH and Total Thiols. Cells (4 X 106)were harvestedfrom each sample. The GSH-400 kit (R&D Systems) was used following themanufacturer's instructions.

RESULTS

p53-dependent Apoptosis by NAC. Treatment of the murine papilloma 308 cell line with the chemopreventive agent NAC led todose-dependent cell death (Fig. 1A). Death was apoptotic, with cellsshowing in situ end labeling of DNA strand breaks after 24 h oftreatment with 20 m@iNAC, but not at 6 h (data not shown), as wellas morphological changes such as cell shrinkage and nuclear conden

sation (Fig. 1B). Morphological changes were minimal in cells treated

with doses of NAC associated with high cell viability (data notshown).

In view of the fact that 308 cells contain a mutant Ha-ras allele andwild-type p53 (24, 25), we sought a matched pair of normal andtransformed cells for comparison. We compared normal primary MEFcells to a matched line of MEF cells transformed by Ha-ran plus E1A(18). As shown in Fig. 2, A and B, the transformed fibroblasts (tMEFp53@'@)weresensitiveto NAC-inducedapoptosis,buttheirnormalcounterparts (MEF p53@'@) were strikingly resistant. In contrast, bothtransformed and normal primary cells from p53' null mice weredeficient in apoptosis induced by NAC (Fig. 14). A specificity ofapoptosis toward transformed cells has been observed previously withchemotherapeutic agents and with hypoxia (18, 29). The specificity ofapoptosis toward transformed cells was not due to the level of p53induction alone, because p53 was induced in their normal counterparts(MEF p53@'@) without causing apoptosis (Fig. 2, A and C). An

additional transformation-related signal is evidently also required forapoptosis. Apoptosis in response to transformation or other heritableabnormalities has been observed in other systems (i8, 29—32).

p53 Induction by NAC via Increased p53 Translation RateS Wenext investigated the molecular mediator of antioxidant-induced apoptosis. The tumor suppressor protein p53 is required for the inductionof apoptosis in response to DNA-damaging agents such as y orUV-irradiation (18, 27) and after hypoxia (29). As shown in Fig. 3A,treatment of 308 cells with NAC resulted in a dose-dependent 5—10-fold increase of p53 protein levels within 3—8h. Northern blotanalysis revealed no major difference in the steady-state level of p53mRNA between control and NAC-treated cells (Fig. 3A), indicating

80

60

A

.0

>—0—— 1 n*A MC

A 5n*ANAC

—0——@ mM MC

Fig. 1. Induction of apoptosis by NAC in 308papilloma cells. A, cell viability measured by trypanblue exclusion; results from three independent cxperiments (mean ± SD). B, apoptosis-associatedDNA strand breaks in cells exposed to 20 mt.i NACfor 24 h, visualized by fluorescent in situ end labeling. Both photographs have the same magnification.In NAC-treated cells, apoptotic nuclei are indicatedby arrows, and other cells have a condensed morphology. Cytoplasmic background results from RNAstaining.

NAC

1724

TreatmentPeriod(h)

B no treatment

on July 26, 2021. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

p53-MEDIATED APOPTOSIS BY ANTIOXIDANTS

A

B

—D--MEFp5311+U

—0— tMEF p53@@

—6-—tMEFp53@

>

024Fig. 2. p53-dependentapoptosisby NACis selecLiveto transformed cells. A, viability of normal (MEF)and transformed (tMEF) cells measured by trypan blueexclusion after treatment with NAC for 24 h. 0, MEFp53+/+; ,MEF p53@; 0, tMEF p53'@; & LMEF

p53@. B,analysisof DNAfragmentationfromtMEFcellsexposedto NAC.C, immunohistochemicalstainlug of p53 protein in p53@'@ MEF cells after 5 h ofexposure to 20 mMNAC.

C

@‘?k@@

NAC

.@@

a

1725

NAC (mM)

p53@'@tMEF: p53W+

h:O 2448

1kb

no treatment

1@@

I'

on July 26, 2021. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

p53-MEDIATED APOPTOSIS BY ANTIOXIDANTS

ATime(h): 0 3 5 8 11

p53 @.@i@r@==::::::::i Time (mm)

NAC Notreatment

5 10 15 5 10 15

p53@ @- @. —@ .@

—fr— NAC

—0--•No treatment

NAC (mM):

p53

z0

25000

15000

10000

5000

/@ NAC—ci--- No treatment

Fig. 3. NAC induces p53 protein through an increased p53 translation rate in 308 cells. A, time courseof p53 protein induction in total cell lysates by 20 mt.iNAC (top panel), dose-dependent p53 protein induction by NAC after 5 h of treatment (middle panel), andNorthern blot of total RNA from cells untreated orexposed to 20 m@iNAC for 5 h (bottom panel). B,synthesis of p53 protein in cells at 5 h posttreatmentwith 20 mMNAC. Arbitrary units of p53 protein banddensity on the autoradiogram were plotted. C, p53protein half-life in cells at 5 h posm-eatment with 20mM NAC. For B and C, data represent one of twosimilar experiments.

5 10 15

35 S-Met labeling time (mm)

Notreatment0 20 40

20

NAC (mM): 0 20

C

0

NAC

ChaseTmme(mln) 0 20 40p53@

Ca00.1

0.

a0

0.1

10

@pp

110 2'O 3'O dO

Chase Time (mmn)

that p53 induction was controlled at the posttranscriptional level.NAC also induced p53 in the murine fibroblast cell line BALB/c 3T3A31 (data not shown).

Most p53-inducing agents damage DNA (15). Some of theseagents increase p53 posttranscriptionally (33—35).In some cases,the induction has been shown to be due to increased p53 proteinstability (34—37). NAC, in contrast, does not induce DNA damage(38—40). To determine the precise molecular mechanism(s) for theinduction of p53 in response to NAC treatment (Fig. 3A), wedirectly measured both the biosynthetic rate of p53 protein and thep53 protein half-life in 308 cells. As shown in Fig. 3, B and C, thebiosynthetic rate of p53 protein was elevated by nearly 5-fold afterNAC treatment. In contrast, the half-life of p53 protein was notaltered in the presence of NAC. These results indicate that enhanced translation of p53 mRNA, and not increased protein stability, accounts for the induction of p53 protein after NAC exposure.

Apoptosis by Other Sulfur-containing Antioxidants. BecauseNAC is well known to ameliorate oxidative stress (12, 41), we

investigated the capacity of other antioxidants (42—44)for transformarion-specific apoptosis. As demonstrated in Fig. 4, A—C,the sulfurcontaining reducing agents DMP and OTC also selectively inducedapoptosis in E1A/Ha-ras-transformed cells, but not in their normalcounterparts. Lipoic acid behaved similarly (data not shown). DMPwas active at doses as low as 50 p.M.DMP, OTC, and lipoic acid also

required p53 (Fig. 4A; data not shown). These agents, as well as NAC,

all induced apoptosis in the human p53@―@colorectal carcinoma cellline RKO (data not shown).

In contrast, the non-sulfur-containing antioxidants vitamin E acetate (tocopherol acetate), BHA, and the water-soluble analogue ofvitamin E, Trolox (4), had little effect on the cell viability of p53@tMEFs for at least 48 h (Fig. 5). The chosen antioxidant concentra

tions here were basically the highest soluble or noncytotoxic doses top53@'@ tMEFs. DNA analysis also confirmed that no apoptosis occurred in p53@'@ tMEF cells treated with these chain-breaking antioxidants (data not shown). Thus, the significant feature of thesesulfur-containing compounds seems to be their effect on the intracel

1726

0 1 5 20 C

@d:r@

on July 26, 2021. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

eo

so

40

p53-MEDIATED APOPTOSIS BY ANTIOXIDANTS

A

5

DMP OTC

1oo..@_@

@EFp53@

—•0-—MEFp53@1@

.—.-- MEF p&

—6-— tMEFP53@

\\@ @——•0—MEFp53@\ —..-\\ __0@_IMEFp5341+

@ __@.._ tMEFp53@

so

so

40

i4 4@8

Treatment Period (h)Fig. 4. p53-dependent apoptosis induced by 50

@AMDMP and 20 mai OTC in tMEFs. A, cell via

bility. B. DNA fragmentation analysis at 24 h posttreatimnt C, flow cytometry analysis at 48 h posttreatment. Box RI represents viable cells. Box R2showsapoptoticcells, definedas havingsub-G1DNA fluorescence (1 axis) and a forward anglelight scatter (X axis) to that of cells in G1. Alldata represent two to three independent experirecalL

>

I

DISCUSSION

Previous mechanisms proposed for the beneficial effects of antioxidants have involved the prevention of oxidative damage to cellularDNA or membrane lipids. Indeed, antioxidants do block apoptosis andcell cycle arrest when these are mediated by reactive oxygen speciesor oxidative stress (47, 48). The present study identifies a novelcellular mechanism for NAC and other sulfur-containing antioxidantsthat involves the induction of p53-mediated apoptosis selectively intransformed cells. The two mechanisms seem to use different properties of antioxidant compounds, radical scavenging activity and alteration of intracellular redox potential.

In this study, we demonstrated that BSO cannot block NAC-induced

0 NAC

D

Bj

0 EIOH

a viiEacetateUD

Fig@5. Sulfur-free antioxidants tocopherol acetate (200 pi@i),Trolox (200 psi), andBHA (100 ,LM)do not induce cell death in tMEF p53@@ cells. Antioxidants weredissolved in ethanol and then added to the medium. The final concentration of ethanol inthe medium is 1:2000 (vlv). At 48 h, cell viability was measured by trypan blue exclusion.All data represent two independent experiments. 0, ethanol; I, Trolox; @,BHA; @,vitamin E acetate.

lular redox potential rather than their effect on radical species. In fact,all of the apoptosis-inducing agents tested above elevate the cellularthiol level (Fig. 6; data not shown).

GSH4ndependence of NAC-induced Apoptosis. A major intracellular pathway of NAC metabolism is deacetylation to the thiolcysteine, the limiting amino acid precursor for synthesis of GSH (45).GSH, in turn, is the major cellular antioxidant (46). To test thepossibility that NAC acts by increasing the level of GSH, we pretreated and coincubated cells with BSO; this agent inhibits all GSHsynthesis by inactivating y-glutamylcysteine synthetase (46). As cxpected, Fig. 6 shows that BSO completely blocked the induction ofcellular GSH by NAC, whereas it only partially blocked the inductionof total thiols. However, BSO did not block NAC-induced apoptosis(Fig. 6), implying that NAC exerts its redox effect by a means otherthan by increasing GSH.

—C

o@

2C

(02

.@

>

Fig. 6. Analysis of GSH. total thiols, and viability in tMEF p53@ cells treated withBSO and/or NAC. Cells treated with both compounds were preincubated with mediumcontaining 20 @.LMBSO for 1 h and then treated with medium containing 20 @MBSO and20 mss NAC for 5 h (GSH and thiols) or for 24 h (viability).

1727

24 48

Treatment Period (h)

no treatment DMP

C@@@@@ @‘:@:@@@

@@@rTR18 @5

apoptosis-@@ L@@ R2@ 1@'R@, R2

FSC@H FSC.H

on July 26, 2021. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

p53-MEDIATED APOPTOSIS BY ANTIOXIDANTS

apoptosis, although it inhibits cellular GSH elevation by NAC (Fig. 4).This finding indicates that the present apoptosis differs from the BSOsensitive biphasic toxicity (Fenton reaction) of some antioxidants otherthan NAC (49). Furthermore, we found that penicillamine (50 p.M), apotent chelator of copper, had no effect on NAC-inducedapoptosis (datanot shown). Similarly, it has been reported that apoptosis in smoothmuscle cells by the antioxidant PDTC was not attributable to the Fentonreaction through autooxidation of PDTC (5).

The induction of p53 is often associated with the cellular response toDNA-damaging agents (14), and the trigger is believed to result fromstrand breaks in nuclear DNA (50). However, we do not expect thatstrand breaks are the mechanism of antioxidant action in the present case.

The 5-10-fold induction of p53 observed after exposure to NAC exceedsthat produced by 5—10Gy of X-ray (33, 51, 52). Alkaline elution andnucleoid sedimentation routinely detect DNA strand breaks produced bydoses 10-fold smaller (53, 54). However, alkaline elution studies haveindicated no DNA-damaging activity of NAC (38—40), nor are therequisite numbers of strand breaks seen by nucleoid gradients (55). Onthe contrasy, NAC blocks DNA strand breaks and mutagenesis by DNAdamaging agents (12, 38, 39). Our results also show that NAC inducesthe expression of p53 protein through enhanced p53 mRNA translationalefficiency. The increased protein stability that has been previously described after DNA strand breaks (34—37) is apparently not the mecha

nism here. In 308 cells, for example, UVB irradiation increases the p53

protein half-life by 7-fold (35).Three mechanisms for translational regulation of p53 expression

have been reported. In mouse cells, p53 expression seems to beautoregulated by a negative feedback loop involving p53 protein, the5 I untranslated region, and some 280 nucleotides of the coding region

of p53 mRNA (56). In contrast, it has been suggested that human p53mRNA translation is regulated by a RNA binding factor(s) that acts onthe 3' untranslated region of p53 mRNA (57). Lastly, p53 mRNAtranslation can be regulated by thymidylate synthase. This protein hasbeen demonstrated to bind to p53 mRNA, thereby decreasing p53protein synthesis in vitro (58). Thymidylate synthase has recentlybeen shown to regulate the p53 protein level in vivo, as well.4 Theactivity of thymidylate synthase, in turn, is redox regulatable (58).

The agents that induced p53 and apoptosis in this study increasedcellular thiol levels, directly or indirectly. Thus, p53 induction may bean important biological effect of increasing thiols or increasing thereducing potential. p53 function has itself been reported to be redoxregulatable through its cysteine residues (21—23).Antioxidants maytherefore also induce p53 DNA binding activity. p53 transactivationactivity can be activated by redox enzyme Ref-l (59). Redox regulation of other transcription factors, such as AP- 1 and nuclear factor KB,by NAC and other thiols is well established (reviewed in Ref. 60).

Sulfur-containing antioxidants might find cancer-related clinicaluses. p53-mediated apoptosis is an important determinant of a tumor'sresponse to chemotherapy (18, 19) and is a critical natural regulator oftumorigenesis (30, 3 1). Hypoxia, which creates a milieu similar to thatof antioxidants, induces p53-dependent apoptosis in tumors in vivo(29). Sulfur-containing antioxidants, including those studied here, arealready in clinical use for other purposes. NAC is used for acetaminophen poisoning and chronic bronchitis at a whole-body distributionof approximately 1.5 m@t(61). DMP, which was active here at 50 p.M,is used clinically for metal poisoning at a dose corresponding to awhole-body distribution of approximately 60 j.&M(62). Therefore,antioxidants may provide a practical nongenotoxic route to cancerprevention or therapy of the 50% of human cancers that retain wild

sulfur-containing chain-breakingantioxidants antloxidants

intracellular thiols C/EBP@

p53 transformation Rb' p21t ‘ I /I \ I /@ \@ fr/'

apoptosis

Fig. 7. Twin antioxidant pathways for apoptosis. The present data indicate thatsulfur-containing antioxidants alter intracellular thiol levels, elevate p53 protein, andinduce apoptosis in transformed cells (left). Because p53 rises even in normal cells (Fig.2C), apoptosis requires an additional transformation-related signal (right). Such signalsare known. For example, cells with aberrant cell cycles caused by viral or transgenicinactivation of Rb undergo apoptosis; this apoptosis requires p53 (30, 31). Evidently,apoptosis requires both a cell cycle abnormality signal and a detector. Sulfur-containingantioxidants augment the p53-dependent detector via redox regulation. Because p21regulates the cell cycle, the free radical pathway described by Chinery et aL (63) mayenter this model by creating or augmenting a cell cycle perturbation. Preexisting perturbations would result from transformation or chemotherapy. Like the detector pathway, thecell cycle abnormality signal is necessary but not sufficient for apoptosis. Overexpressingp21 via C/EBP@3does not lead to apoptosis, although blocking C/EBP@transcription withantisense transcripts blocks apoptosis (64). Green tea polyphenols may also provokeapoptosis via the non-p53 free radical pathway (65). Different antioxidants will elicit thetwo pathways to different degrees.

ACKNOWLEDGMENTS

We are very grateful to Drs. Scott Lowe (Cold Spring Harbor Laboratory,Cold Spring Harbor, NY) and A. Levine (Princeton University, Princeton, NJ)for providing cell lines and thank A. Jonason for technical assistance.4 Jingfang Ju and Edward Chu, unpublished data.

1728

Redox Pathway Radical Pathway

type p53. Even for p53@ tumors, use of antioxidants may eliminate

normal cells mutagemzed during chemotherapy.While this paper was under review, the antioxidants PDTC and a

water-soluble vitamin E analogue, Trolox, were reported to induceapoptosis by inducing p21 via CIEBPI3 (particularly in the presence ofchemotherapeutic drugs; Ref. 63). This process differs from the present mechanism in two respects: (a) It seems to be p53 independent,because apoptosis was not preceded by an increase in p53 protein.Although p53 independence cannot be determined from the p53@HCT-l5 cell line used (20, 63), our own unpublished results indicatethat PDTC is equally toxic in p53@ and p53@ transformed MEFs.(b) This non-p53 pathway is activated by a different set of antioxidants. Chain-breaking antioxidants such as vitamin E acetate andTrolox did not induce p53-dependent apoptosis of transformed MEFs(Fig. 5), although we have not investigated the high Trolox concentrations used in Ref. 63. Whereas the p53-dependent mechanismseems to reflect changes in the redox potential only, the non-p53apoptosis mechanism seems to involve radical species (63, 64). Apossible pathway relating the two mechanisms is shown in Fig. 7.

on July 26, 2021. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

p53-MEDIATED APOPTOSIS BY ANTIOXIDANTS

33. Kastan, B. K., Onyekwere, 0., Sidransky, D., Vogelstein, B., and Craig, R. W.Participation of p53 in the cellular response to DNA damage. Cancer Res., 51:6304—6311,1991.

34. Fritsche, M., Haessler, C., and Brandner, G. Induction of nuclear accumulation of thetumor-suppressorprotein p53 by DNA-damagingagents. Oncogene, 8: 307—318,1993.

35. Liu, M., Dhanwada, K. R., Bitt, D. F., Hecht, S., and Pelling, J. C. Increase in p53protein half-life in mouse keratinocytes following UV-B irradiation. Carcinogenesis(Lond.), 15: 1089—1092,1994.

36. Maltzman, W., and Czyzyk, L. UV irradiation stimulates levels of p53 cellular tumorantigen in nontransformed mouse cells. Mol. Cell. Biol., 4: 1689—1694,1984.

37. Price, B. D., and Calderwood, S. K. Increased sequence-specific p53-DNA bindingactivity after DNA damage is attenuated by phorbol ester. Oncogene, 8: 3055—3062,1993.

38. Yunis, A. A., Lim, L. 0., and Arimura, G. K. DNA damage induced by chloramphenicol and nitroso-chloramphenicol: protection by N-acetylcysteine. Respiration,50 (Suppl.): 50—55,1986.

39. Chan, J. Y. H., Stout, D. L., and Becker, F. F. Protective role of thiols in carcinogeninduced DNA damage in rat liver. Carcinogenesis (Load.), 7: 1621—1624,1986.

40. Solen, 0. Radioprotective effect of N-acetylcysteine in vitro using the induction ofDNA breaks as end-point. Int. J. Radiat. Biol., 64: 359—366,1993.

41. Aruoma, 0. I., Halliwell, B., Hoey, B. M., and Butler, J. The antioxidant action ofN-acetylcysteine: its reaction with hydrogen peroxide, hydroxyl radical, superoxideand hypochlorous acid. Free Radic. Biol. Med., 6: 593—597,1989.

42. Anderson, M. E., and Meister, A. Intracellular delivery of cysteine. Methods Enzymol., 143: 313—325,1987.

43. Ceconi, C., Curello, S., Cargnoni, A., Boffa, 0. M., and Ferrari, R. Antioxidantprotection against damage during cardiac ischemia and reperfusion: effect of dimercaptopropanol. Cardioscience, 1: 191—198,1990.

44. Packer, L., Wits, E. H., and Tritschler, H. J. a-Lipoic acid as a biological antioxidant.Free Radic. Biol. Med., 19: 227—250,1995.

45. Burgunder, J. M., Varriale, A., and Lauterburg, B. H. Effect of N-acetylcysteine onplasma cysteine and glutathione following paracetamol administration. Eur. J. Clin.Pharmacol., 36: 127—131,1989.

46. Dolphin, D., Poulson, R., and Avramovic, 0. (eds.) Glutathione: Chemical, Biochemical and Medical Aspects, Vol. 3. New York: John Wiley & Sons, 1989.

47. Lotem, J., Peled-Kamar, M., Groner, Y., and Sachs, L. Cellular oxidative stress andthe control of apoptosis by wild-type p53, cytotoxic compounds, and cytokines. Proc.NatI. Acad. Sci. USA, 93: 9166—9171,1996.

48. Irani, K., Xia, Y., Zweier, J. L, Scollot, S. J., Ocr, C. D., Fearon, E. R., Sundaresan, M.,Finkel,1., and Ooldschmidt-Clermont,P. J. Mitogenicsignalingmediatedby oxidants inras-transformedfibroblasts.Science (WashingtonDC), 275: 1649-1652, 1997.

49. Held, K. D., Sylvester, C. S., Hopcia, K. L., and Biaglow, J. E. Role of Fentonchemistry in thiol-induced toxicity and apoptosis. Radiat. Res., 145: 542—553,1996.

50. Nelson, W. 0., and Kastan, M. B. DNA strand breaks: the DNA template alterationsthat trigger p53-dependent DNA damage response pathways. Mol. Cell. Biol., 14:1815—1823,1994.

51. Lu, X., and Lane, D. P. Differential induction of transcriptionally active p53 following UV or ionizing radiation: defect in chromosome instability syndrome? Cell, 75:765—778,1993.

52. Khanna, K. K., and Lavin, M. F. Ionizing radiation and UV induction of p53 proteinby different pathways in ataxia-telangiectasia cells. Oncogene, 8: 3307—3312, 1993.

53. Lipetz, P. D., Brash, D. E., Joseph, L. B., Jeweu, H. D., Lisle, D. R., Lantry, L. E.,Hart, R. W., and Stephens, R. E. Determination of DNA superhelicity and extremelylow levels of DNA strand breaks in low numbers of nonradiolabeled cells byDNA-4',6-diamidino-2-phenylindole fluorescence in nucleoid gradients. Anal. Biochem.,121:339—348,1982.

54. Larsen, K. H., Brash, D., Cleaver, J. E., Hart, R. W., Maher, V. M., Painter, R. B., andSega, G. A. DNA repair assays as tests for environmental mutagens. Mutat. Res., 98:287—318,1982.

55. Renzing, J., Hansen, S., and Lane, D. P. Oxidative stress is involved in the UVactivation of p53. J. Cell Sci., 109: 1105—1112, 1996.

56. Mosner, J., Mummenbrauer, T., Bauer, C., Sczakiel, G., Grosse, F., and Deppert, W.Negative feedback regulation of wild-type p53 biosynthesis. EMBO J., 14: 4442—4449, 1995.

57. Fu, L., Minden, M. D., and Benchimol, S. Translational regulation of human p53 geneexpression. EMBO J., 15: 4392—4401,1996.

58. Chu, E., and Allegra, C. J. The role of thymidylate synthase as an RNA-bindingprotein. Bioessays, 18: 191—198,1996.

59. Jayaraman, L., Murthy, K. 0. K., Thu. C., Curran, T., Xanthoudakis, S., and Prives,C. Identification of redox/repair protein Ref-l as a potent activator of p53. GenesDcv., 11: 558—570,1997.

60. Sun, Y., and Oberley, L. W. Redox regulation of transcriptional activators. FreeRadic. Biol. Med., 21: 335-348, 1996.

61. Holdiness, M. R. Clinical pharmacokinetics of N-acetylcysteine. Clin. Pharmacokinet., 20: 123—134,1991.

62. Weiner, B. (ed.) PDR Generics, 2nd ed. Montvale, NJ: Medical Economics Data,1996.

63. Chinery, R. C., Brockman, J. A., Peeler, M. 0., Chyr, Y., Beauchamp, R. D., andCoffey, R. J. Antioxidants enhance the cytotoxicity of chemotherapeutic agents incolorectal cancer a p53-independent induction of@ via C/EBP@. Nat.Med.,3: 1233—1241,1997.

64. Kastan, M. B. To oxidize or not to oxidize? Nat. Med., 3: 1192—1193, 1997.65. Ahmad, N.. Feyes, D. K., Nieminen, A., Agarwal, R., and Mukhtar, H. Green tea

constituent epigallocatechin-3-gallate and induction of apoptosis and cell cycle arrestin human carcinoma cells. J. NatI. Cancer Inst., 89: 1881—1886,1997.

1729

REFERENCES

1. Miquel, J., Quintanilha, A. T., and Weber, H. CRC Handbook of Free Radicals andAntioxidants in Biomedicine, Vol. II. Boca Raton, FL: CRC Press, 1989.

2. Hockenbery,D. M., Oltaval, 1 N., Yin, X. M., Miuiman, C. L, and Korsmeyer,S. J.Bcl-2 functions in an antioxidant pathway to prevent apoptosis. Cell, 75: 241-251, 1993.

3. Shimizu,S., Eguchi,Y., Kosaka,H., Kamlike,W.,Matsuda,H.,andTsujimoto,Y.Prevention of hypoxia-induced cell death by Bcl-2 and Bcl-xL. Nature (Load.), 374:811—813,1995.

4. Jacobson,M. D., and Raff, M. C. Programmed cell death and Bcl-2 protection in verylowoxygen.Nature(Lond.),374: 814—816,1995.

5. Tsai, J.. lain, M., Hsieh,C., Lee, W., Yoshizum,M., Patterson,C., Perrella,M.,Cooke, C., Wang, H., Hater, E., Schlegel, R., and Lee, M. Induction of apoptosis bypyrrolidinedithiocarbamate and N-acetylcysteine in vascular smooth muscle cells.J. Biol.Chem.,271: 3667—3670,1996.

6. Cenmi,P.A.Prooxidantstatesandtumorpromotion.Science(WashingtonDC).227:375—381,1985.

7. Steele, V. E., Kelloff, G. J., Wilkinson, B. P., and Arnold, J. T. Inhibition oftransformation in cultured rat tracheal epithelial cells by potential chemopreventiveagents. Cancer Res., 50: 2068—2074,1990.

8. O'Brien,P.Antioxidantsandcancecmolecularmechanisms.In: D.Armstrong(ed),Free Radicals in Diagnostic Medicine, pp. 215-239. New York: Plenum Press, 1994.

9. floi@,S.D.,Astengo,M.,Serra,D.,andBennicelli,C. Inhibitionofurethane-inducedlung tumors in mice by dietary N-acetylcysteine. Cancer Lett., 32: 235-241, 1986.

10. Rostein,3.B., and Slaga,T. J. Anticarcinogenesismechanisms,as evaluatedin themultistage mouse skin model. Mutat. Res., 202: 421—427,1988.

11. Pereira,M.A.,andKhouiy,M.D.Preventionbychemopreventiveagentsofazoxymethsac-induced foci of aberrant ciypts in rat colon. Cancer Len.. 61: 27—33,1991.

12. flora, S. D., Izzotti, A., D'Agostino, F., Balansky, R., and Cesarone, C. F. Chemopreventivepropertiesof N-acetylcysteineand other thiols.In: L. Wattenberg,M.Lipkin, C. W. Boone, and G. J. Kelloff(eds.), CancerChemoprevention, pp. 183—194.Boos Raton, FL: CRC Press, 1992.

13. Izzotti,A.,D'Agostini,F.,Bagnasco,M.,Scatolini,L.,Rovida,A.,Balansky,R. M.,Cesarone, C. F., and Dc Flora, S. Chemoprevention of carcinogen-DNA adducts andchronic degenerative diseases. Cancer Res., 54 (Suppl.): 1994s—l998s,1994.

14. Ko, L. I., and Prives, C. p53: puzzle and paradigm. Genes Dcv., 10: 1054—1072,1996.

15. Levine,A.J. p53,thecellulargatekeeperforgrowthanddivision.Cell,88:323-331,1997.

16. Fujiwara, T., Grimm, E. A., Mukhopadhyay,T., Thang, W., Owen-Schaub,L B., andRoth, 3. A. Inductionof chemosensitivityin hwnan lung cancer cells in vivoby adenovirus-mediatedtransfer of the wild-typep53 gene. Cancer Rca.,54: 2287—2291,1994.

17. Uu, T. J., El-Naggar,A. K., McDonnell,T. J., Stack, K. D., Wang, M., Taylor,D. L., andclayman, 0. L Apoptosisinductionmediatedby wild-typep53 adenoviralgene transferin squamous cell carcinomaof the head and neck. Cancer Rca., 55: 3117—3122,1995.

18. Lowe,S. W.,Ruley,H. E.,Jacks,1., andHousman,D. E. p53-dependentapoptosismodulates the cytotoxicity of anticancer agents. Cell, 74: 957—967,1993.

19. Lowe, S. W., Bodis, S., McClatchey, A., Reminton, L., Ruley, H. E., Fisher, D. E.,Housman, D. E., and Jacks, T. p53 status and the efficacy of cancer therapy in vivo.Science (Washington DC), 266: 807—810,1994.

20. O'Connor.P. M.,Jackman,J., Bae,I., Myers,T. 0., Fan,S., Mutoh,M.,Scudiero,D. A., Monks,A., Sausville,E. A.,Weinstein,J. N., Friend,S., Fornace,A. 3.,andKobti, K. W. Characterization of the p53 tumor suppressor pathway in cell lines of theNational Cancer lnstltute anticancer drug screen and correlations with the growthinhibitory potency of 123 anticancer agents. Cancer Res., 57: 4285—4300,1997.

21. Hainaut,P., and Milner.J. Redoxmodulationof p53 conformationand sequencespecific DNA binding in vitro. Cancer Rca., 53: 4469—4473, 1993.

22. Hupp,T. R., Meek,D. W., Midgley,C. A., and Lane, D. P. Activationofthe crypticDNAbinding function of mutant forms of p53. Nucleic Acids Rca.,21: 3167—3174,1993.

23. Rainwater.R., Parks, D., Anderson,M. E., Tegtmeyer,P., and Mann, K. Role of cysteineresidues in regulationof p53 function.Mol. Cell. BioL, 15: 3892-3903, 1995.

24. Strickland,J. E.,Greenhalgh,D. A., Koceva-Chyla,A.,Hennings,H.,Restrepo,C.,Balaschak, M., and Yuspa, S. H. Development of murine epidermal cell lines whichcontain an activated ras@@oncogene and form papillomas in skin grafts on athymicnude mouse hosts. Cancer Res., 48: 165—169,1988.

25. Liu, M., and Pelling, J. C. UV-B/A irradiation of mouse keratinocytes results inp53-mediated WAF1/CIP1 expression. Oncogene, JO: 1955—1960,1995.

26. Harney,D.M.,andLevine,A.J. p53alterationisa commoneventinthespontaneoushnmottalization of primary BALB/c mmmc embryo fibroblasts. Genes Dcv., 5:2375—2385,1991.

27. Ziegler, A., Jonason, A. S., Leffell, D. J., Simon, J. A., Sharma, H. W., Kimmelman,3., Remington, L, Jacks, T., and Brash, D. E. Sunburn and p53 in the onset of skincancer. Nature (Land.), 372: 773—776,1994.

28. Darzynkiewicz,Z.,Bnmo,S.,DelBino,0., Gorczyca,W.,Hotz,M.A.,Lassota,P.,andTraganos, F. Features ofapoptotic cells measured by flow cytometry. Cytometry,13: 795—808,1992.

29. Grasber, T. G., Osmanian, C., Jacks, J., Housman, D. E., Koch, C. J., Lowe, S. W.,and Giaccia, A. J. Hypoxia-mediated selection of cells with diminished apoptoticpotential in solid tumours. Nature (Lond.), 379: 88—91,1996.

30. Symonds, H., Krall, L, Remington, L., Saenz-Robles, M., Lowe, S., Jacks, T., andDyke, T V. p53-dependent apoptosis suppresses tumor growth and progression invivo. Cell, 78: 703—711,1994.

31. Morgenbesser, S. D., Williams, B. 0., Jacks, T., and DePinhO,R. A. p53-dependent81,01*0518produced by Rb-deficiency in the developing mouse lens. Nature (Lond.),371: 72—74,1994.

32. Brash,D. E. Cellularproofreading.Nat.Med.,2: 525—526,1996.

on July 26, 2021. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1998;58:1723-1729. Cancer Res   Ming Liu, Jill C. Pelling, Jingfang Ju, et al.   Antioxidant Action via p53-mediated Apoptosis

  Updated version

  http://cancerres.aacrjournals.org/content/58/8/1723

Access the most recent version of this article 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. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/58/8/1723To request permission to re-use all or part of this article, use this link

on July 26, 2021. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from