[CANCER RESEARCH57. 267-271. January 15. 19971

ABSTRACT

Sulindac sulfoxide, a commonly prescribed anti-inflammatory drug,has cancer chemopreventive activity. During its metabolism, the inactiveprodrug sulindac sulfoxide undergoes either reduction to the active antiinflammatory metabolite sulindac sulfide or irreversible oxidation to sulindac sulfone, which lacks prostaglandin synthetase inhibitory activity.Interestingly, sulindac sulfone has been reported to have cancer chemopreventive activity. The objective of the experiments reported here was toinvestigate the chemopreventive activity of sulindac sulfone against mammary carcinogenesis and to study its mechanism. Rats were injected witheither 12.5 or 37.5 mg of 1-methyl-1-nitrosourea(MNU)/kg body weight at50 days of age. Sulindac sulfone was incorporated into a purified diet at aconcentration of either 0.03 or 0.06% (w/w) and fed to rats beginning 7days after the injection of MNU. Sulindac sulfoxide at a level of 0.06%(wlw) was fed as a reference for comparison. Thirty rats were assigned toeach dietary group treated with the high dose of MNU, and 44 rats wereassigned to each dietary group treated with the low dose of MNU. Thesulfone reduced cancer incidence and the number of cancers per ratirrespective ofthe dose of MNU injected, and its chemopreventive activitywas comparable to that of sulindac sulfoxide. Cancer latency was alsoprolonged significantly by sulindac sulfone; the effect was particularly

notable at the low dose ofcarcinogen, at which the prolongation of latencywas >8 weeks. The sulfone inhibited the occurrence of mammary carcinomas that were classified as having either a wild-type or a mutant codon12 In the Ha-tar gene however, the inhibitory effect was greater againstcarcinomas with a mutant Ha-ms genotype. Using a mammary glandorgan culture transformation assay, it was observed that sulindac sulfonealso inhibited the formation of 7,12.dimethylbenz(a)anthracene.mducedhyperplastic alveolar nodules and that the inhibitory activity of the sal

fone was comparable to that of the sulfoxide. These data indicate that theobserved effect of the sulfone on mammary carcinogenesis in vivo is likelyto be due to a tissue-specific effect rather than to other systemic effects.The findings that both the prodrug and the sulfone inhibited carcinogenesis in vivo and nodule formation in organ culture and that the sulfonelacks inhibitory activity on prostaglandin synthesis suggest a mecha

nism(s) of chemoprevention that is independent of the prostaglandinpathway. A candidate mechanism for the apparent clonal selection prossure exerted by the sulfone against mammary carcinogenesis Is apoptosis.To test this hypothesis, MCF-7 cells were exposed to a range of concentrations of sulindac sulfone and sulfoxide. Both compounds inhibited cellgrowth and induced apoptosis in the absence of necrosis. Collectively,these data support a specific chemopreventive effect of sulindac sulfoneagainst mammary carcinogenesis and Indicate that this compound mayhave a selective effect against carcinogenesis Involving alterations In thesignal transduction cascadeof which Ha-ras Is a component.EvidenceIsconsistent with the Involvement of apoptosis in the cancer-inhibitoryactivity observed.

Received 8/19/96; accepted 11/15/96.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

I This work was supported by Cell Pathways and Department of Veterans Affairs Merit

Review Grant 131.2 To whom requests for reprints should be addressed, at Division of Laboratory

Research, AMC Cancer Research Center. 1600 Pierce St., Lakewood, CO 80214. Phone:(303) 239-3464; Fax: (303) 239-3443.

INTRODUCTION

Sulindac sulfoxide and other NSAIDs3 have been reported to inhibit tumor formation in several models of experimentally inducedcancer (1—3).The cancer-preventive activity has generally been attributed to the inhibition of prostaglandin synthesis. Sulindac sulfoxide is a prodrug that is either reversibly reduced to sulindac sulfide,which has potent antiprostaglandin synthetase activity due to itsability to inhibit cyclooxygenase I and II, or is irreversibly oxidized tothe sulfone, which has been considered an inactive metabolite that iseliminated from the body (4, 5). Nonetheless, in several publicationsit has been reported that feeding the sulfone metabolite of sulindac,which has been shown not to affect prostaglandin synthetase activity,inhibited chemically induced mammary and colon carcinogenesis (6,7). This finding is significant relative to the use of NSAIDs for cancerchemoprevention because one factor limiting their use is gastrointestinal toxicity, an effect attributed to the inhibition of prostaglandinsynthesis by NSAIDs (8, 9). The possibility that cancer preventionmay be dissociable from prostaglandin synthetase inhibitory activityimplies that this aspect of toxicity may be avoidable. This issue gainsincreased importance in view of a recent clinical report that use ofNSAIDs was associated with a reduction in human breast cancer risk(odds ratio, 0.66; 95% confidence interval, 0.52—0.83)and that breastcancer risk declined with increasing NSAID exposure (10).

In the experiments reported in this study, the effect of two dietaryconcentrations of sulindac sulfone on the occurrence of mammarycarcinogenesis induced by a low or high dose of MNU was evaluated.The hypothesis that the sulfone exerts a selective inhibitory effect onthe clonal expansion of MNU-initiated cells harboring a mutated rasgene was also tested. Such a selective effect has been attributed toanother NSAID, piroxicam, in an azoxymethane-induced colon carcinogenesis model (1 1). In chemical carcinogenesis models, specificras mutations are induced and are believed to be involved in the earlystages of tumor development (12—16).Depending on the dose ofcarcinogen administered, a proportion of the mammary adenocarcinomas induced by MNU in the rat contain a G—*Atransition mutationin codon I2 of the Ha-ras gene (17, 18). Therefore, selective inhibition of the clonal expansion of subpopulations of initiated cellsidentified by this marker is expected to result in an altered proportionof tumors harboring the mutation. This hypothesis was evaluated bydetermining the proportion of induced tumors in each treatment groupthat had the codon 12 mutation in the Ha-ras gene. The hypothesisthat sulindac sulfone exerts specific effects on the mammary glandindependent of systemic drug metabolism was investigated by determining the effect of sulindac sulfoxide and sulfone on the formationof DMBA-induced hyperplastic alveolar nodules in a mammary glandorgan culture transformation assay. This procedure is widely used inthe evaluation of agents for chemopreventive activity and identifiesagents effective in inhibiting mammary carcinogenesis in the ratmodel that was used (19). Furthermore, the effect of these metaboliteson the growth of a mammary carcinoma cell line also was studied todetermine whether these metabolites affect cell growth by inducing

3 The abbreviations used are: NSAID, nonsteroidal anti-inflammatory drug; MNU,1-methyl- 1-nitrosourea; DMBA, 7. 12-dimethylbenz(a)anthracene.

267

Sulfone Metabolite of Sulindac Inhibits Mammary Carcinogenesis1

Henry J. Thompson,2 Cheng Jiang, Junxuan Lu, Rajendra G. Mehta, Gary A. Piazza, Nancy S. Paranka,Rifat Pamukcu, and Dennis J. Ahnen

AMC Cancer Research Center, Lakewood, Colorado 80214 fH. J. T., C. J., J. LI; Cell Pathways, Aurora, Colorado 80012 [0. A. P., N. S. P., R. P.]; Department of SurgicalOncology. University of Illinois, Chicago, Illinois 60612 (R. G. MI; and Department of Veterans Affairs Medical Center, Denver. Colorado 80220 (D. J. A.J

on March 5, 2020. © 1997 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

SULINDACSULFONEINHIBITSMAMMARYCARCINOGENESIS

apoptosis. Apoptosis is a physiological cell death pathway involved inthe regulation of tissue size and has been reported to be dysregulatedduring carcinogenesis (20, 21). NSAIDs have recently been reportedto induce apoptosis in colon carcinoma cells (22), but effects onepithelial cells derived from the mammary gland have not beenreported. The rationale for assessing the ability of sulindac metabolites to induce apoptosis also took into account reports that alteredregulation of the ras signal transduction pathway can induce apoptosis(23) and that azomethane-initiated colon cells harboring mutated rasgenes were selectively inhibited by the NSAID piroxicam (1 1). Asreviewed in Ref. 24, an increase in the rate of apoptosis in a cellpopulation could inhibit that population of cells from developing.

MATERIALS AND METHODS

Drug Synthesis. The sulfone metabolite [cis-5-fluoro-2-methyl-l(p-methylsulfonylbenzylidene)-3-indenylacetic acid] of sulindac sulfoxide (purchased

from Therapicon, Milan, Italy) was prepared as described in Ref. 22. Chemicalanalyses by high-performance liquid chromatography, nuclear magnetic resonance,infrared,andelementalanalysisdemonstratedthelot of sulfoneusedinthis study to be 97.8% sulindac sulfone, 1.4% sulindac sulfoxide, and 0.8%

sulindac sulfone epoxide.

Animal Studies. Female Sprague Dawley rats were obtained from TaconicFarms, Germantown, NY, at 21 days of age, housed three per cage, andmaintained in an environmentally controlled room at 22°Cwith 50% relativehumidity and a 12-h-light, 12-h-dark cycle. At 50 days of age, rats wereinjected i.p. with either 12.5 or 37.5 mg of MNUIkg of body weight (25).

Followingcarcinogentreatment,ratswererandomizedintoone of fourdietarygroups at each dose of carcinogen. Forty-four rats were assigned to eachdietary group that received the low dose of carcinogen, and 30 rats wereassigned to each dietary group that received the high dose of MNU. The

sulindac sulfoxide or sulfone was administered in a purified high-fat (24.6%,w/w) diet formulated with corn oil (diet composition given in Ref. 26)beginning 7 days after the injection of MNU. Thus, the effects of the sulfoneand sulfoxide were limited to the postinitiation events in mammary carcinogenesis. Sulindac sulfone was incorporated into the diet at 0.03 or 0.06%(w/w); sulindac sulfoxide served as a positive control and was incorporated at0.06% (w/w). A purified diet formulation was used to eliminate the variabilityin diet composition associated with the use of rodent chow diets. Furthermore,

a high-fat diet formulation more closely models the typical Western dietconsumed by individuals to whom this intervention may someday be applied.High-performance liquid chromatography analyses showed the sulfone to bestable in this diet for a period in excess of 30 days.

Rats were weighed weekly and palpated twice each week for detectablemammary tumors. Rats injected with the high dose of MNU were killed at 24weeks postcarcinogen. Animals injected with the low dose of MNU were killedat 37 weeks postcarcinogen. At necropsy, animals were euthanized withgaseous CO2 and then skinned and their pelts examined under incandescentlight. All tumors and suspicious areas were excised and processed for histopathological classification (27).

Ha-ras Mutation Assay. The mutationalstatus of Ha-ms codon 12 (OGAGAA) in mammaryadenocarcinomaswas assessedby a modifiedPCR-generatedrestricted fragment length polymorphism method on paraffin-embedded tissue

blocks(28).Briefly,DNA extractsfromtheparaffm-embeddedtumorswereusedas templates for PCR amplificationof the rat codon 12 region. The downstreamPCR primer was designed with two mismatchesso that a codon 12 GGA-GAAmutationwould introducea Xmnl site in the PCR product.Upon digestionof theproducts with Xmnl and separationby electrophoresison a 6% polyactylamidegel, a 96-bp band was generatedif the mutation was present,whereasthe normalPCR product was not sensitive to this enzyme and remained as a I 18-bp band. Toincrease detection sensitivity, PCR products were labeled by incorporation of

[a-32P]dCTPand detected by autoradiography.Drug Preparation for Organ Culture and Cell Culture Experiments.

Stock solutions of sulindac sulfoxide or sulfone were made at l000X concentrations in DMSO and diluted with medium to the final working concentrations. The final concentration of DMSO for all treatments was maintained at0.1%. All drug solutions were prepared fresh on the day of testing.

Organ Culture Experiments. Virgin female BALB/c mice, 3—4weeks of

age, were obtained from Charles River, Wilmington, MA. Upon arrival themice were treated daily with subcutaneous injections of 1 p@g of estradiol

l7@ + I mg progesterone for 9 days. This treatment is a prerequisite inasmuch

as animals not pretreated with steroids fail to respond to hormones in vitro(29). The entire culture procedure has been described in detail (30). Briefly, the

animals were killed by cervical dislocation and the thoracic pair of mammaryglands was dissected and spread out on a silk raft. These tissue preparationswere incubated for 10 days in Waymouth MB752/l medium (5 glands/S ml ofmedium/dish). The medium was supplemented with 2 nmi glutamine, antibi

otics (penicillin and streptomycin, 100 units of each/ml medium), and growth

promoting hormones, 5 @gof insulin, S @gof prolactin, 1 @gof aldosterone,and 1 i.@gof hydrocortisone per ml of medium. The carcinogen DMBA (2

@g/ml) was added to the medium for 24 h between days 3 and 4. For the

present study, DMBA was dissolved in DMSO at a final concentration of 4mg/mi, and SO @lwere added to 100 ml of medium, resulting in 2 j@g/mlfinalconcentration. The control dishes contained DMSO as vehicle. On day 4,DMBA was removedfromthe mediumby rinsingthe glandsin freshmediumand transferring them to new dishes containing fresh medium without DMBA.After 10days of incubation, the glands were maintained for another 14days inthe medium containing only insulin (S @g/ml).During the entire culture period,the glands were maintained at 37C in a 95% 02 and 5% CO2 environment.

Sulindac sulfoxide and sulindac sulfone were added to the medium on theday the culture was initiated, and drug treatment was continued throughout theexperiment. Carcinogen-treated glands without the test agent served as a

positive control. At the end of the experiment, glands were fixed and stainedfor microscopic evaluation and lesion identification as described in Ref. 30.

Cell Culture Experiments. MCF-7 cells were obtained from AmericanType Culture Collection (Rockville, MD) and were grown in RPM! (Celox,Hopkins, MN) supplemented with 5% FCS (Gemini Bioproducts, Inc., Calabases, CA), and 2 mt@tglutamine, 100 units/mi penicillin, 100 units/mi streptomycin, and 0.25 @g/m1amphotericin (Life Technologies, Inc., Grand Island,NY). Cultures were maintained in a humidified atmosphere of 95% air and 5%CO2at 37°C.Thecultureswerepassagedatpreconfluentdensitieswiththeuseof a solution of 0.05% trypsin and 0.53 mr@iEDTA (Celox). Cells were platedat 1 X 106 cells/25-cm2 flask.

Apoptosis and Necrosis. Drug effects on apoptosisand necrosisof culturedMCF-7 cells were determined morphologicallyby fluorescentmicroscopyafterlabeling with acridme orange and ethidium bromide as described by Duke andCohen (31). Floating and attached cells were collected by trypsinization andwashed three times in PBS. One-mi aliquots of 1 X 106 cells were centhfuged

(300 X g). The pellet was gently resuspendedin 25 @.dof media and 1 @.dof dyemixturecontaining 100 @tg/miacridineorange and 100 @tg/mlethidiumbromidepreparedin PBS.Ten @lofmixture wereplacedon a microscopeslideandcoveredwith a 22-mm coverslipand examinedunder X 40 dry objectiveswith the use ofepillumination and filter combination. An observer blinded to the identity oftreatments scored at least 100 cells/sample. Live cells were determined by theexclusionofethidium bromidestain.Live and dead apoptoticcellswere identifiedby nuclear condensation of chromatin stained by the acridine orange and theethidiumbromide,respectively.Necroticcells were identifiedby uniformlabelingof the cell with ethidium bromide.

Statistical Analyses. Statistical differences in cancer incidence were determined by x2 analysis (32). Tumor count data were analyzed by factorialANOVA following square root transformation (32). Differences among groupsin the proportion of tumors with mutant or wild-type ras status were evaluatedby the Fisher exact test (32). Differences in cancer latency were assessed by alife table procedure (33).

RESULTS

Mammary Carcinogenesis Experiments

Low Dose of MNU. The effect of feeding sulindac sulfone onmammary carcinogenesis induced by the low dose of MNU is shownin Table 1. In comparison to animals that received no drug, both thesulfone and the sulfoxide reduced cancer incidence and reduced thetotal number of carcinomas detected. The inhibitory activity of bothcompounds was statistically significant at P < 0.05 for all comparisons with the untreated group; however, the effects of the three drug

268

on March 5, 2020. © 1997 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Table 1 Effect of sulindac sulfone and sulfoxide on mammary carcinogenesis induced by a low dose ofMNUTreatment―

No. of rats― Mammary carcinomas, incidence (%)C Mammary carcinomas, total no.'@ Final body weight,gCControl

44/43 29.6 (bOY 16 (lOOf 399 ±8 (100/0.03% sulfone 44/42 11.6 (39)8 9 (56)@ 386 ±6(97)@0.06%

sulfone 44/43 16.3 (55)g 8 (50)@ 354 ±8 (89)@0.06% sulfoxide 44/44 9.1 (3l)@ 6 (38)R 367 ±8(92)ga

Allratswereinjectedwith12.5mgMNU/kgbodyweightat50daysofageasdescribedin“MaterialsandMethods.―Thestudywasterminated37weekspost-carcinogentreatment.b Number of rats randomized to an experimental group/number of rats surviving to the end of theexperiment.C

Incidence of mammary carcinomas found at the end of the experiment. The number in parentheses is the percent of the response observed in the control group.

Toul numberof mammarycarcinomasfoundat theendof the experiment.The numberin parenthesesis the percentof the responseobservedin thecontrolgroup.eBody weight of the animals at the end of the study. The number in parentheses is the percent of the response observed in the control group. Values are means ± SE.

f.g Values with different superscripts are statistically different (P < 0.05).

30.00

@‘ 25.00

EE@@ 20.00.QIv@I15.00

@ 1E

Table2 Effectof sulindacsulfoneand sulfoxideon theHa-rascodon12mutationalstatusof mammarycarcinomasinducedbyhigh-doseMNUHa-Ras

statusof carcinomas, Ha-Rasstatusof carcinomas,Treatment― No. of animals― Total no. of carcinomasc mutant codon 12 number― wild-type codon 12number'@Control

30/30 126 (lOOf 60 (100) 66 (100)0.03% sulfone 30/30 103 (82)e 54 (90) 49 (74)0.06% sulfone 30/30 68 (54f 25 (42) 43 (65)0.06% sulfoxide 30/30 82 (65/ 34 (57) 48(74)a

All rats were injected with 50 mg of MNUIkg of body weight at 50 days of age as described in “Materialsand Methods.―The study was terminated 24 weeks post-carcinogentreatment.

b Number of rats randomized to an experimental group/number of rats surviving to the end of the experiment.

C Total number of mammary carcinomas found at the end of the experiment. The number in parentheses is the percent of the response observed in the control group.

d The effect of treatment on the proportion of the total number of carcinomas within a dietary group that had mutant or wild-type status was evaluated by contingency table analyses.Number in parentheses is the percent of the response observed in the control group. The high dose of both the sulfone and the sulfoxide altered the proportion of induced tumors withthe mutant versus wild-type genotype (P < 0.05). Regression analyses of the control and sulfone-treated groups indicated that the effect of sulfone on carcinomas harboring mutantgenes was approximately 66% greater than its effect on carcinomas harboring the wild-type genes.

e,fValues with different superscripts are statistically different (P < 0.05).

I... 0) ‘- ro in N. 0) @- ro in@ ,- (0@ (‘4 N (‘1 N ro t')

Weeks Po@ MNU

SULINDAC SULFONE INHIBITS MAMMARY CARCtNOGENESIS

treatments were not statistically different from one another. Detection ofcarcinomas was significantly delayed. (P < 0.05 for each control versustreatment comparison) as determined by life table analysis (Fig. 1). Thedelay in time to detection of a comparable incidence of carcinomasobserved in all treatment groups relative to untreated animals exceeded 8weeks. It is noteworthy that there were 44 animals assigned to each groupand that this study was carried out for 37 weeks following the administration ofcarcinogen, yet only four animals were removed from the studybefore its termination due to the presence of necrotizing tumors. Finalbody weights of drug treated animals were between 89 and 97% of theuntreated rats, and final body weights showed no correlation with differences among groups in cancer latency (Table 1 and Fig. 1). Ha-ras statuswas determined on all carcinomas that were detected, but only two orthree tumors in each group had the 12th codon mutation; this smallnumber was insufficient for statistical analyses.

High Dose of MNU. A preliminary report of the effect of sulindacsulfoxide and the sulfone on the carcinogenic response to the highdose of MNU was published in Ref. 6. In that paper, a dose-dependentreduction in tumor number and a dose-dependent prolongation ofcancer latency was reported. Sulindac sulfoxide produced an effect comparable to that ofthe sulfone. Table 2 shows the effects ofthe sulfone and

sulfoxide on the number of mammary carcinomas categorized by Ha-rascodon 12 mutational status (wild-type or mutant). All carcinomas thatoccurred in each group were analyzed. Thus, the number of carcinomasharboring the wild-type or mutant Ha-ras gene reported in the untreatedgroup provides an estimate of the number of each of these tumor typesthat would be expected in the treated groups ifthe treatment had no effecton the occurrence of that “pathogeneticallydefined type―of carcinoma.The data indicate that both the sulfone and the sulfoxide inhibited theoccurrence of approximately the same number of wild-type carcinomasin each treatment group. Both the sulfone and the sulfoxide treated groupshad a proportionately greater effect on carcinomas that had mutant versuswild-type Ha-ras genes (P < 0.05). As noted previously, these treatmentswere well tolerated by the animals, and effects of sulindac sulfone orsulfoxide on growth rate (4—10%)were unrelated to the degree ofinhibition observed (3).

Organ Culture Experiments

Experiments were performed to evaluate the effects of sulindacsulfone and sulfoxide on the development of DMBA-induced mammary lesions in organ culture. A total of five organ culture experi

Fig. I. The effect of treatment with sulindac sulfone or sulindac sulfoxide

on the induction of mammary carcinomas by a low dose (12.5 mg/kg ofbody weight) of MNU. The incidence of palpable mammary carcinomas asa functionof timepost-carcinogentreatmentis shown.Theeffectof sulfoneand sulfoxideon cancerlatencywerestatisticallysignificantby life tableanalyses (P < 0.05). •,control; @,sulfone, 0.03% w/w;@ , sulfone, 0.06%w/w; •,sulfoxide, 0.06% w/w.

[@-1-oH

@TJ

269

on March 5, 2020. © 1997 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

2 3 4 5 6

SULINDAC SULFONE INHIBITS MAMMARY CARCINOGENESIS

DISCUSSION

The results reported in Tables 1 and 2 and Figs. I and 2 confirm andexpand our preliminary report that sulindac sulfone, which has a minimaleffect on either COX I or COX H activity in the prostaglandin biosynthetic pathway, significantly inhibits mammary carcinogenesis inducedby either a low or a high dose of l-methyl-l-nitrosurea (6). The datapresented on body weights (Table 1) indicate that these treatments werewell tolerated. In fact, the levels of either sulindac sulfoxide or sulfonerequired for chemopreventive activity against mammary carcinogenesisin the rat were lower than other laboratories have reported for thechemoprevention of colon carcinogenesis in the same species (1, 2). Anadditional hypothesis that was evaluated in the carcinogenesis experimentwas formulated based on a report that piroxicam, another NSAID, inhibited the selective amplification of initiated cells harboring mutated rasgenes in the azoxymethane-induced colon carcinogenesis model (1 1). Asshown in Tables 1 and 2, sulindac sulfone and sulfoxide reduced carcinoma number; however, the data in Table 2 indicate that both compoundshad greater inhibitory activity against carcinomas harboring mutant Haras genes. These data indicate a selective inhibition of the clonal expansion of initiated cells harboring mutated Ha-ms. This observation suggests that an activated ras signal transduction pathway may be a target ofsulindac sulfone. Of interest in this regard is a recent review (34) in whichdysregulation of downstream components of the Ras signal transductionpathway was reported to play a role in human breast carcinogenesis.

The organ culture model used in the experiments reported hasbeen used to evaluate the chemopreventive efficacy of many agents(19). Our observations indicate that both sulindac sulfoxide andsulfone inhibit DMBA-induced alveolar nodulogenesis in thismouse mammary gland organ culture transformation assay. Thesefindings in the organ culture system are also consistent with thosein the MNU model in that both sulindac sulfone and sulfoxide wereinhibitory and that in both models the inhibitory effect of thesulfone was dose-dependent. This finding provides strong evidencethat these sulindac metabolites exert a specific effect on the mammary gland and tumorigenesis therein that is apart from anysystemic effects that either compound may exert. This observation

80

a0 @4)C,)

.aI@aaasa@ 20

0

Sulindac sulfone __:. 0.01 0.1Sulindac sulfoxide

________1 10 100- uM. - - 10 uM

Concentration of agent added

Fig. 2. The effect of sulindac sulfone and sulindac sulfoxide on the occurrence ofmammary gland nodular lesions induced in mouse mammary gland organ culture bytreatment with DMBA. The inhibitory activity of the sulfone on nodule occurrence wassignificant by logistic regression analysis (P < 0.05).

ments were conducted. In a preliminary study, sulindac sulfone andsulfoxide resulted in a 40—50%inhibition of lesion occurrence at 10p.M. On the basis of these data, the sulfone was studied over a range

of concentrations. As shown in Fig. 2, the inhibitory activity was dosedependent, and inhibition was approximately 80% of the controlresponse at a concentration of 100 @.LM.

Cell Culture Experiments

A concentration-dependent reduction in cell number was observedin response to sulindac sulfone treatment (Fig. 3A), and the effect ofthe sulfone was somewhat greater than that of the sulfoxide at comparable concentrations (Fig. 3C). Accompanying the observed reductions in cell number was a proportionate increase in the percentage ofcells undergoing apoptosis induced by both compounds (Fig. 3, B andD). These compounds had no significant effect on necrosis irrespective of the concentration of drug treatment (Fig. 3, B and D).

-a.

200 400 600 84

3

2Eaa

c)

% 50

0

100@ apoptosts B

0 200 400 600 800

[Sulindac sulfone], uM

Fig. 3. The effect of sulindac sulfone and sulindac sulfoxide on cell number,apoptosis, and necrosis as assessed in MCF-7 cells. Results shown are from arepresentative experiment. Data points, mean; bars, SE. Experiments were doneat least three times to confirm that the same patterns of response were consistently observed.

0

200 400 600

[Sulindac sulfoxide], uM

270

on March 5, 2020. © 1997 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

SULINDACSULFONEINHIBfl'S MAMMARYCARCINOGENESIS

is counter to reports that the effects of NSAIDs on lung carcinogenesis involve systemically mediated effects on prostaglandins(35). These data support the observation that inhibition of prostaglandin biosynthesis, an activity attributed to sulindac sulfide,appears unnecessary for chemopreventive activity in the organculture assay or in the in vivo mammary carcinogenesis model.

It has been reported in cell culture systems and in in vivo colonpolyps that metabolites of sulindac inhibit cell growth by inducingapoptosis (7, 22). Because one mechanism by which selection againstparticular transformed cell populations might be affected is inductionof apoptosis and because, as noted above, sulindac sulfone and sulfoxide appear to preferentially inhibit the clonal expansion of MNUinitiated cells harboring mutated Ha-ras genes, we asked whetherthese metabolites of sulindac would induce apoptosis in a well-studiedmammary carcinoma cell line, MCF-7. As shown in Fig. 3, bothsulindac metabolites inhibited the growth of this cell line. It also wasnoted that each compound induced a comparable level of apoptoticcell death at the drug concentration that inhibited growth by 50%. Itis also noteworthy that the sulfone and the sulfoxide induced increasing levels of apoptosis in the absence of a concomitant increase ofnecrosis, which supports the hypothesis that sulfone acts by triggeringphysiological rather than pathological cell death.

In their totality, the data presented indicate that sulindac sulfone, whichlacks significant prostaglandin synthetase inhibitory activity (6), inhibitsMNU-induced mammary carcinogenesis in a dose-dependent manner.Carcinogenesis induced by a low or high dose of carcinogen was inhibiteti to a comparable degree. The observed effects of the sulfone and thesulfoxide on the occurrence of Ha-ras codon 12 mutant carcinomas areconsistent with the idea that clones of transformed cells with a specificpathogenetic characteristic were preferentially inhibited. Moreover, thedata indicate that the protective effect is exerted on the target tissue andis not dependent on systemic effects that NSAIDS may exert. Theseeffects were observed at concentrations of the sulfone that were welltolerated by the animal, and the cell culture data indicate that apoptosiscan be induced by the sulfone without necrosis. These data provideimportant leads about the mechanisms by which sulindac sulfone works,which merit further investigation.

The data presented in this study illustrate an emerging concept forcancer prevention and control that may be applied to carcinogenesis atmany organ sites. Namely, with the expanding ability to identify thepathogenetic characteristics of clones of cells that progress throughthe carcinogenic process, it will be possible to target specific pathogenetic cascades to prevent and control the development of cancer. Bydetermining an individual's genetic predisposition for a specific cancer, it may be possible to assign that individual to a treatment regimethat specifically blocks the processes of clonal expansion and selection ofpopulations ofcells with those pathogenetic characteristics thatare associated with that individual's cancer risk. From such an approach, it is likely that highly specific and effective methods for theprevention and control of cancer will be developed.

ACKNOWLEDGMENTS

We thank Dr. Meenakshi Singh for her diagnosis of mammary tumors andKim Rothhammer, John McGinley, and Stephanie Briggs for their excellenttechnical assistance.

REFERENCES

1. Rao, C. V., Rivenson, A., Simi, B., Zang, E., Kelloff, G., Steele, V., and Reddy, B. S.Chemoprevention of colon carcinogenesis by sulindac, a nonsteroidal anti-inflammatory agent. Cancer Res., 55: 1464—1472, 1995.

2. Pepin, P., Bouchard, L., Nicole, P., and Castonguay, A. Effects of sulindac andoltipraz on the tumorigemcity of 4-.(methylnitrosamino)l-(3-pyridyl)-1-butanone inAIJ mouse lung. Carcinogenesis (Land.), 13: 341—348,1992.

3. Caner, C. A., Ip, M. M., and Ip, C. A comparison of the effects of the prostaglandinsynthesis inhibitors indomethacin and carprofen on 7,12-dimethylbenz[ajanthraceneinduced mammary tumorigenesis in rats fed different amounts of essential fatty acid.Carcinogenesis (Land.), 10: 1369—1374, 1989.

4. Duggan, D. E., Hooke, K. F., Risley, E. A., Shen, T. Y., and Arman, C. 0.Identification of the biologically active form of sulindac. J. Pharmacol. Exp. Ther.,201: 8—13,1977.

5. Shen, T. Y., Ham, E. A., Cirillo, V. J., and Zanetti, M. Prostaglandin synthetaseinhibitors (H. J. Robinson and 3. R. Vane [eds]) Raven Press, NY, 1974.

6. Thompson, H. J., Briggs, S., Paranka, N. S., Piazza, 0. A., Brendel, K., Gross, P. H.,Sperl, 0. J., Pamukcu, R., and Ahnen, D. J. Inhibition of mammary carcinogenesis inrats by sulfone metabolite of sulindac. J. NatI. Cancer Inst., 87: 1259—1260, 1995.

7. Alberta, D. S., Hixson, L., Ahnen, D., Bogert, C., Einspahr, J., Paranka, N., Brendel,K., Gross, P. H., Pamukcu, R., and Burt, R. W. Do NSAIDS exert their colon cancerchemoprevention activities through the inhibition of mucosal prostaglandin synthetate? J. Cell Biochem. Suppl., 22: 18—23,1995.

8. Roberts,A.Effectsofprostaglandinsonthestomachandintestine.Prostaglandins,6:523—532,1974.

9. Davies,N.M.Toxicityofnonsteroidalanti-inflammatorydrugsin thelargeintestine.Dis.ColonRectum,38: 1311—1321,1995.

10. Harris, R. E., Namboodiri, K. K., and Farrar, W. B. Nonsteroidal anti-inflammatorydrugs and breast cancer. Epidemiology, 7: 203—205,1996.

11. Singh, J., Kulkarni, N., Kelloff, 0., and Reddy, B. S. Modulation of azoxymethaneinduced mutational activation of ras protooncogenes by chemopreventive agents incoloncarcinogenesis.Carcinogenesis(Lond.),15: 1317—1323,1994.

12. Barbacid, M. ras genes. Annu. Rev. Biochem., 56: 779-827, 1987.13. Bos, i. L. ras oncogenes in human cancer: a review. Cancer Res., 49: 4682—4689,

1989.14. Sukumar, A., Nolario, V., Martin-Zanca, D., and Barbacid. M. Induction of mammary

carcinoma in rats by nitroso-methylurea involves malignant activation of H-ras-llocus by single point mutations. Nature (Land.), 306: 658—661, 1983.

15. Zarbl, H., Sukumar, S., Arthur, A. V., Martin-Zanca, D., and Barbacid, M. Directmutagenesis of Ha-ras-l oncogenes by N-nitroso-N-methylreas during initiation ofmammary carcinogenesis in rats. Nature (Lond.), 315: 382—385,1985.

16. Kumar,R.,Sukumar,S.,andBarbacid,M.Activationof ras oncogenesprecedingtheonset of neoplasia. Science (Washington DC), 248: 1101—1104, 1990.

17. Zhang, R., Haag, J. D., and Gould, M. N. Reduction in the frequency of activated rasoncogenesinratmammarycarcinomaswithincreasingN-methyl-N-nitrosoureadosesor increasing prolactin levels. Cancer Res., 50: 4286—4290, 1990.

18. Schneider, B. L., and Bowden, 0. T. Selective pressures and ras activation incarcinogenesis. Mol. Carcinog., 6: 1—4,1992.

19. Mehta, R. 0., and Moon, R. C. Characterization of effective chemopreventive agentsin vitro using an initiation-promotion protocol. Anticancer Res., I 1: 593—597,1991.

20. Kerr, J. F. R., Wyllie, A. H., and Currie, A. R. Apoptosis: a basic biologicalphenomenon with wide ranging implications in tissue kinetics. Br. J. Cancer, 26:239—257,1972.

21. Williams, 0. T. Programmed cell death: apoptosis and oncogenesis. Cell, 65: 1097—1098, 1991.

22. Piazza, 0. A.. Kulchak-Rahm, A. L., Knstzsch, M., Sperl, 0., Paranka, N. S., Gross,P. H., Brendel,K., Burt,R. W., Alberta,D. S., Pamukcu,R., and Ahnen,D. J.Antineoplastic drugs sulindac sulfide and sulfone inhibit cell growth by inducingapoptosis. Cancer Res., 55: 3110—3116, 1995.

23. Chen, C. Y., and Faller, D. V. Direction of p21 ras-generated signals towards cellgrowth or apoptosis is determined by protein kinase C and Bcl-2. Oncogene, 11:1487—1498,1995.

24. Thompson, H. J., Strange, R., and Schedin, P. J. Apoptosis in the genesis andprevention of cancer. Cancer Epidemiology, Biomarkers & Prey., 1: 597—602,1992.

25. Thompson, H. J., and Adlakha, H. Dose-responsive induction of mammary glandcarcinomas by the intraperitoneal injection of l-methyl-1-nitrosourea. Cancer Res.,51: 3411—3415,1991.

26. Thompson,H.3.,Ronan,A.M.,Ritacco,K.A.,andTagliaferro,A.R.Effectof typeand amount of dietary fat on the enhancement of rat mammary tumorigenesis byexercise. Cancer Res., 49: 1904—1908, 1989.

27. Young, S., and Hallowes, R. C. Tumors of the mammary gland. In: V. S. Tunisov(ed), Pathology of Tumours in Laboratory Animals, Vol. 1, pp. 31—73.IARCScientific Publication. Lyon, France, 1973.

28. Lu, J.. Jiang, C., Fontaine, S., and Thompson, H. J. ras may mediate mammary cancerpromotion by high fat. Nutr. Cancer, 23: 283—290,1995.

29. Mehta, R. 0., and Baneijee, M. R. Hormonal regulation of macromolecular biosynthesis during lobulo-alveolar development of mouse mammary gland in organ culture.Acta Endocrinol., 80: 501—511,1975.

30. Mehta. R. 0., Cemy, W. L., Mandigan, M. J., and Moon, R. C. Modification of mousemammary gland organ culture. J. Tissue Cul. Methods, 8: 27—35,1984.

31. Duke, R. C., and Cohen, J. J. Morphological and biochemical assays of apoptosis. In:J. E. Coligan and A. M. Kruisbeak (eds.), Current Protocols in Immunology, pp.3.17.1—3.17.16.New York: John Wiley & Sons, 1992.

32. Sokal, R. R., and Rohlf, F. J. Biometry: The Principles and Practice of Statistics inBiological Research, pp. 150, 384. San Francisco, CA: W. H. Freeman and Co., 1969.

33. Peto, R. Guidelines on the analysis of tumor rates and death rates in experimentalanimals. Br. J. Cancer, 29: 101—105,1974.

34. Clark, 0. J., and Der, C. Aberrant function of the ras signal transduction pathway inhuman breast cancer. Breast Cancer Res. Treat., 35: 133—144,1995.

35. Rioux, N., and Castonguay, A. Modulation of lung tumorigenesis and immuneresponse by nonsteroidal anti-inflammatory drugs in NJ mice. Proc. Am. Assoc.CancerRca.,37:284,1996.

271

on March 5, 2020. © 1997 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1997;57:267-271. Cancer Res   Henry J. Thompson, Cheng Jiang, Junxuan Lu, et al.   CarcinogenesisSulfone Metabolite of Sulindac Inhibits Mammary

  Updated version

  http://cancerres.aacrjournals.org/content/57/2/267

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/57/2/267To request permission to re-use all or part of this article, use this link

on March 5, 2020. © 1997 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from