Liver subcellular fractions from rats treated by organosulfur compounds from Allium modulate mutagen activation

Mutation Research 466 Ž2000. 17–26 www.elsevier.comrlocatergentox Community address: www.elsevier.comrlocatermutres

Liver subcellular fractions from rats treated by organosulfur compounds from Allium modulate mutagen activation Denis Guyonnet, Christine Belloir, Marc Suschetet, Marie-Helene ´ ` Siess, Anne-Marie Le Bon ) Institut National de la Recherche Agronomique, Unite´ de Toxicologie Nutritionnelle, BV 1540, 17 rue Sully, 21034 Dijon Cedex, France Received 21 September 1999; received in revised form 7 December 1999; accepted 14 December 1999

Abstract The effects of in vivo administration of naturally occurring organosulfur compounds ŽOSCs. from Allium species were studied on the activation of several mutagens. Male SPF Wistar rats were given p.o. one of either diallyl sulfide ŽDAS., diallyl disulfide ŽDADS., dipropyl sulfide ŽDPS. or dipropyl disulfide ŽDPDS. during 4 consecutive days and the ability of hepatic S9 and microsomes from treated rats to activate benzow axpyrene ŽBaP., cyclophosphamide ŽCP., dimethylnitrosamine ŽDMN., N-nitrosopiperidine Ž N-PiP. and 2-amino-1-methyl-6-phenylimidazow4,5-b xpyridine ŽPhIP. was determined in the Ames test. Administration of DAS, DPS and DPDS resulted in a significant increase of the activation of BaP, CP, N-PiP and PhIP mediated by S9 and microsomes while DADS treatment only increased the mutagenicity of PhIP. In contrast, S9 from DADS-treated rats significantly inhibited the mutagenicity of N-PiP and BaP. DAS, DADS and DPS strongly inhibited DMN mutagenicity while DPDS enhanced it. To understand the mechanisms underlying these effects, the modifications of the activities of specific isozymes of CYP involved in the activation of these mutagens were studied. DAS, DPS and DPDS strongly enhanced pentoxyresorufin O-dealkylase ŽPROD. activity related to CYP2B and slightly increased ethoxyresorufin O-deethylase ŽEROD. and methoxyresorufin O-demethylase ŽMROD. activities related to CYP1A family. DADS exerted the same effects than other OSCs but to a lesser extent. p-Nitrophenol hydroxylase ŽPNPH. activity related to CYP2E1 was inhibited by DAS and DADS, whereas DPDS significantly increased this activity. Hence, the effects of OSCs on the mutagenicity of several genotoxic compounds are mediated by modification Ženhancement or inhibition. of specific CYP involved in their activation. q 2000 Elsevier Science B.V. All rights reserved. Keywords: Organosulfur compounds; Modulation of mutagenicity; Benzow axpyrene; Cyclophosphamide; Nitrosamines; PhIP

AbbreÕiations: BaP, benzow axpyrene; CP, cyclophosphamide; CYP, cytochrome P450; DADS, diallyl disulfide; DAS, diallyl sulfide; DMN, dimethylnitrosamine; DPDS, dipropyl disulfide; DPS, dipropyl sulfide; EH, epoxide hydrolase; EROD, ethoxyresorufin O-deethylase; GST, glutathione S-transferase; MROD, methoxyresorufin O-demethylase; MSF, microsomal fraction; N-PiP, N-nitrosopiperidine; OSCs, organosulfur compounds; PhIP, 2-amino-1-methyl-6-phenylimidazow4,5-b xpyridine; PROD, pentoxyresorufin O-dealkylase; PNPH, p-nitrophenol hydroxylase; UGT, UDP-glucuronosyltransferase ) Corresponding author. Tel.: q33-3-80-69-32-15; fax: q33-3-80-69-32-25; e-mail: [email protected] 1383-5718r00r$ - see front matter q 2000 Elsevier Science B.V. All rights reserved. PII: S 1 3 8 3 - 5 7 1 8 Ž 9 9 . 0 0 2 3 4 - X

18

D. Guyonnet et al.r Mutation Research 466 (2000) 17–26

1. Introduction Recent epidemiologic studies have shown that high consumption of garlic and other Allium vegetables is associated with reduced stomach and colon cancer risk in humans w1–5x. In addition, experimental studies conducted with naturally occurring organosulfur compounds ŽOSCs. from garlic and onion have demonstrated the ability of these compounds to reduce chemical carcinogenesis in different animal models w6–9x. Thus, it is becoming increasingly evident that Allium vegetables and related OSCs are potential human anticarcinogens. Most experimental studies have focused on allyl sulfides, mainly derived from garlic, whereas much less is known about alkyl sulfides, which are more specific to other Allium species such as onion. The precise mechanisms through which OSCs exert their anticarcinogenic effects have not been clarified but it seems that they can act at all stages of chemical carcinogenesis and especially at the initiation stage. One conceivable mechanism by which chemopreventive agents, such as OSCs, antagonize the effects of carcinogenic chemicals during the initiation phase of tumorigenesis is the modulation of the enzyme systems involved in their bioactivation to reactive intermediates. OSCs could exert their anticarcinogenic action according to distinct mechanisms, such as the inhibition of phase I enzymes andror enhancement of phase II detoxication enzymes. It was demonstrated that several allyl sulfides, such as diallyl sulfide ŽDAS., were efficient inhibitors of CYP2E1 w10,11x and, therefore, could block the activation of nitrosamines and other molecules activated by this CYP. OSCs are also inducers of phase II enzymes Ži.e., glutathione S-transferase ŽGST., epoxide hydrolase ŽEH., DT-diaphorase, UDPglucuronosyltransferase ŽUGT.., which contribute to enhanced deactivation and excretion of reactive metabolites of carcinogens w6,12–15x. However, OSCs are able to induce several other isoforms of CYP belonging to CYP1A and 2B subfamilies that are involved in the activation of different carcinogens. OSCs are moderate inducers of CYP1A subfamily, whereas they are strong inducers of CYP2B w14,16,17x. Thus, the induction of these CYPs could

lead to an increased bioactivation of certain carcinogens. It was, therefore, pertinent to investigate the effects of in vivo administration of individual OSCs on the mutagenicity of known environmental carcinogens that require metabolic activation by different CYPs. Moreover, most studies on antimutagenic effects of OSCs have been conducted in vitro whereby OSCs are directly introduced in the incubation system w18–21x. It is important not to restrict antimutagenicity studies to in vitro systems because in vitro mechanisms are not always relevant in vivo w22x. The present study was undertaken to evaluate the incidence of the in vivo modulation of phase I enzymes by several individual OSCs on carcinogen activation in the Ames test. Four OSCs from Allium species were selected for this study: DAS, diallyl disulfide ŽDADS., dipropyl sulfide ŽDPS. and dipropyl disulfide ŽDPDS.. These compounds have been shown to significantly modify drug-metabolizing enzymes in rat liver w14,16x. We evaluated their effects on the genotoxicity of mutagens that need to be activated by CYPs. Therefore, the effects of hepatic subcellular fractions Žmicrosomes and S9 fraction. from OSCs-treated rats were evaluated on the mutagenicity of Ži. a food heterocyclic amine, 2-amino-1-methyl-6-phenylimidazow4,5-b xpyridine ŽPhIP., which is activated by CYP1A2 w23x; Žii. a polycyclic aromatic hydrocarbon, benzow axpyrene ŽBaP., which is activated by CYP1A1; w24x and Žiii. two nitroso-compounds, N-nitrosopiperidine Ž N-PiP. and dimethylnitrosamine ŽDMN., which are, respectively, activated by CYP2B w25x and CYP2E1 w26x. These mutagens are major environmental carcinogens that either are generated during cooking or occur in tobacco smoke or are food contaminants. They could, therefore, be consumed together with OSCs. We also investigated the effect of OSCs on the activation of cyclophosphamide ŽCP., a widely used anticancer prodrug in humans that is mainly activated by CYP2C11 and 2B w27,28x. The results of our study showed that OSCs were able to modulate the genotoxicity of these carcinogens either by increasing or decreasing their activation. This is related to the influence of OSC treatments on the CYP isoenzymes activities.

D. Guyonnet et al.r Mutation Research 466 (2000) 17–26

2. Material and methods

2.1. Chemicals DAS Žpurity 97%., DADS Žpurity 80%, remainder other allyl sulfides., DPS Žpurity 97%. and DPDS Žpurity 98%. were obtained from Aldrich Chemical ŽStrasbourg, France.. They were used without further purification. N-PiP, BaP, DMN and CP were purchased from Sigma-Aldrich Chimie ŽSaint-Quentin Fallavier, France.. PhiP was obtained from Toronto Research Chemicals ŽToronto, Ontario, Canada.. Other chemicals were of the highest quality available.

2.2. Animals and treatments Male SPF Wistar rats, 5 weeks old, from IffaCredo ŽL’Arbresle, Lyon, France. were housed in individual stainless steel cages and maintained at 218C, with constant humidity and a 12-h light–dark cycle. During the experiment, they were fed ad libitum a semi-liquid purified diet as previously described w29x. After 2 weeks of feeding, five groups, each with eight rats, were given one of either corn oil Žvehicle control., DAS Ž1 mmolrkg., DADS Ž1 mmolrkg., DPS Ž1 mmolrkg. or DPDS Ž1 mmolr kg. daily by gavage for 4 consecutive days. In a preliminary study, a dose of 1 mmolrkg was shown to modulate significantly CYP isoenzyme activities Žunpublished data..

2.3. Preparation of hepatic subcellular fractions Twenty-four hours after the last treatment, the animals were killed by cervical dislocation following 16 h of fasting. Livers were removed and pooled. Liver homogenate S9 fraction was prepared according to Maron and Ames w30x. Liver microsomes were prepared as previously described w29x. All steps of the preparation of hepatic subcellular fractions were carried out at 0–48C using cold, sterile solutions and glassware. To minimize bacterial contamination, liver microsomes were filtered through an Acrodisc w dis-

19

posable 25-mm filter assembly ŽGelman Sciences, Ann Arbor, MI.. Sterility of hepatic preparations was checked before their use. Liver S9 homogenates and microsomes were stored in aliquots of 1 ml at y808C. 2.4. Enzyme assays The protein levels of microsomes were measured by the method of Bradford w31x, adapted for automatic measurement using a Cobas Fara II centrifugal analyzer ŽRoche Instruments, Basel, Switzerland.. The protein levels of S9 were measured by the same method but were done manually. Cytochrome P450 content and enzyme assays Žpentoxyresorufin Odealkylase ŽPROD., ethoxyresorufin O-deethylase ŽEROD . and methoxyresorufin O-demethylase ŽMROD.. were performed as previously described w29x. p-Nitrophenol hydroxylase ŽPNPH. activity was measured with p-nitrophenol as substrate according to the method of Tassaneeyakul et al. w32x. 2.5. Mutagenicity assays The Ames test was performed with Salmonella typhimurium strains TA98 or TA 100 according to Maron and Ames w30x with minor modifications. S. typhimurium strains were provided by Dr. B. Ames ŽDepartment of Biochemistry, University of California, Berkeley, USA.. A bacterial suspension obtained after an overnight culture of 12-h was used in the Ames test. The effects of treatments on the mutagenicity of mutagens were determined by a liquid preincubation method using liver S9 homogenates or microsomes as the activation system. The mutagenicity assay was carried out with a 10% Žvrv. activation system for BaP, CP and PhIP, a 20% Žvrv. activation system for N-PiP and a 30% Žvrv. activation system for DMN. The volume of subcellular fractions was adjusted in order to introduce the same protein level per plate for each treatment. Each mutagen was preincubated at 378C for 60 min with the bacteria and subcellular fractions in the presence of a NADPH-generating system. When microsomes were used, the activation system was supplemented with glucose-6-phosphate dehydrogenase Ž1 unitrplate.. The mixtures were diluted with soft

20

D. Guyonnet et al.r Mutation Research 466 (2000) 17–26

to the control group. p F 0.05 was chosen as indicating significance. Calculations were made with the SAS system ŽCary, NC..

3. Results Hepatic S9 and microsomes from rats treated with DAS, DPS and DPDS significantly increased the activation of CP to mutagens when compared with

Fig. 1. Metabolic activation of CP by hepatic preparations from control and OSCs-treated rats. The mutagenicity assay was carried out using the S. typhimurium strain TA100 and a 10% Žvrv. activation system. The activation system was supplemented with glucose-6-phosphate dehydrogenase when isolated microsomes were used. The spontaneous reversion rates were 142"6 and 136"7 for S9 and microsomes, respectively. Results are presented as mean"SEM of two series of triplicates. The mutagenicity assay with microsomes was done with smaller levels of CP per plate because in this situation, CP mutagenicity reached a plateau starting from 500 mgrplate Ždata not shown.. U Significantly different from the corresponding control ŽDunnett’s test, pF 0.05..

agar, plated onto minimal glucose agar plates and further incubated for 48 h at 378C to allow the development of histidine revertant colonies ŽHisq. . The number of Hisq revertants was counted on two repetitions of triplicate plates for each dose of mutagen. 2.6. Statistical analysis Data were compared by analysis of variance followed by Dunnett’s test to compare the treated groups

Fig. 2. Metabolic activation of N-PiP by hepatic preparations from control and OSCs-treated rats. The mutagenicity assay was carried out using the S. typhimurium strain TA100 and a 20% Žvrv. activation system. The activation system was supplemented with glucose-6-phosphate dehydrogenase when isolated microsomes were used. The spontaneous reversion rates were 149"7 and 134"7 for S9 and microsomes, respectively. Results are presented as mean"SEM of two series of triplicates. U Significantly different from the corresponding control ŽDunnett’s test, pF 0.05..

D. Guyonnet et al.r Mutation Research 466 (2000) 17–26

21

control ŽFig. 1.. S9 and microsomes from DADStreated rats slightly increased the mutagenicity of CP but this effect was not significant. Like the results with CP, S9 and microsomes from DAS-, DPS- and DPDS-treated rats significantly increased N-PiP mutagenicity ŽFig. 2.. By contrast, S9 and microsomes from DADS-treated rats were less effective than control preparations in converting N-PiP to mutagens. S9-mediated mutagenicity of BaP was significantly reduced by DADS when compared with con-

Fig. 4. Metabolic activation of PhIP by hepatic preparations from control and OSCs-treated rats. The mutagenicity assay was carried out using the S. typhimurium strain TA98 and a 10% Žvrv. activation system. The activation system was supplemented with glucose-6-phosphate dehydrogenase when isolated microsomes were used. The spontaneous reversion rates were 40"4 and 24"2 for S9 and microsomes, respectively. Results are presented as mean"SEM of two series of triplicates. U Significantly different from the corresponding control ŽDunnett’s test, pF 0.05..

Fig. 3. Metabolic activation of BaP by hepatic preparations from control and OSCs-treated rats. The mutagenicity assay was carried out using the S. typhimurium strain TA98 and a 10% Žvrv. activation system. The activation system was supplemented with glucose-6-phosphate dehydrogenase when isolated microsomes were used. The spontaneous reversion rates were 41"2 and 25"2 for S9 and microsomes, respectively. Results are presented as mean"SEM of two series of triplicates. U Significantly different from the corresponding control ŽDunnett’s test, pF 0.05..

trol ŽFig. 3.. Surprisingly, S9 from DAS-, DPS- and DPDS-treated rats significantly inhibited BaP mutagenicity at the dose of 2.5 mgrplate of BaP, whereas they increased the mutagenic response of BaP at higher doses. The strongest mutagenicity was observed with S9 from DPS-treated rats. When microsomes were used as the activation system, DADS treatment did not modify the mutagenesis of BaP when compared to control. By contrast, microsomes of DAS-, DPS- and DPDS-treated rats were more effective than microsomes of control rats in convert-

D. Guyonnet et al.r Mutation Research 466 (2000) 17–26

22

Fig. 5. Metabolic activation of DMN by hepatic preparations from control and OSCs-treated rats. The mutagenicity assay was carried out using the S. typhimurium strain TA100 and 30% Žvrv. activation system. The activation system was supplemented with glucose-6-phosphate dehydrogenase when isolated microsomes were used. The spontaneous reversion rates were 134"7 and 120"5 for S9 and microsomes, respectively. Results are presented as mean"SEM of two series of triplicates. U Significantly different from the corresponding control ŽDunnett’s test, pF 0.05..

ing BaP to mutagens, but no difference could be discerned among these three treatments. When PhIP was employed as mutagen, hepatic S9 and microsome preparations from OSCs-treated rats were significantly more effective than controls in converting PhIP to mutagen metabolites ŽFig. 4.. Among all treatments, DAS and DPS were the most potent inducers of PhIP mutagenesis. S9 from DAS and DADS-treated rats markedly decreased the mutagenicity of DMN as compared to S9 from control rats ŽFig. 5.. They significantly inhibited the mutagenicity induced by DMN by more than 60–70%. DPS also reduced DMN mutagenicity but this inhibitory effect was less than that of DAS or DADS. Conversely, S9 from DPDS-treated rats significantly increased DMN mutagenicity. Similar observations were made when microsomes were used as the activation system. The effects of treatments on total cytochrome P450 content and different monooxygenase activities in hepatic microsomes are shown in Table 1. Total hepatic CYP was increased by DPS Ž45%. and DPDS Ž25%., whereas it was reduced by DADS Ž31%.. DPS and DPDS induced the activities of the four monooxygenases. DPS and DPDS significantly enhanced PROD activity Žrespectively, =43 and =60. and, to a lesser extent, EROD activity Žrespectively, =2.9 and =2.2. and MROD activity Žrespectively, =1.8 and =1.4.. The increase of PNPH activity was only significant in DPDS-treated rats. In DAS-treated rats, a similar pattern of induction was observed since EROD and MROD activities were enhanced 1.7-fold and 1.4-fold, respectively. But DAS was the most potent inducer of PROD activity Ž=125..

Table 1 Effects of DAS, DADS, DPS and DPDS on CYP-dependent monooxygenase activities in hepatic microsomes Note: Rats were administered the corresponding OSC Ž1 mmolrkg. daily by gavage for 4 days. Control rats received corn oil alone. Values are means " SEM of three repetitions. Treatments

EROD ŽCYP1A1., pmol miny1 mgy1 protein

MROD ŽCYP1A2., pmol miny1 mgy1 protein

PROD ŽCYP2B1,2.,a pmol miny1 mgy1 protein

PNPH ŽCYP2E1., pmol miny1 mgy1 protein

Cytochrome P450, nmolrmg protein

Control DAS DADS DPS DPDS

57.6 " 3 95.7 " 2.4U 111.7 " 4.3U 165.5 " 9.8U 126.5 " 13.4U

19.7 " 0.9 31 " 0.7U 21.8 " 1.7 34.5 " 2.6U 28.2 " 0.9U

13.3 " 1.2 1668 " 26U 193 " 11U 569 " 41U 803 " 72U

1.058 " 0.05 0.864 " 0.01U 0.605 " 0.01U 1.134 " 0.01 1.237 " 0.02U

0.97 " 0.04 0.934 " 0.02 0.667 " 0.01U 1.409 " 0.05U 1.210 " 0.06U

U a

Significantly different from the corresponding control ŽDunnett’s test, p F 0.05.. Statistical analysis was carried out on log-transformed data.

D. Guyonnet et al.r Mutation Research 466 (2000) 17–26

Moreover, DAS significantly inhibited PNPH activity Ž25%.. The induction of CYP activities was less efficient in DADS-treated rats. EROD and PROD activities were significantly increased Žrespectively, =1.9 and =15., whereas MROD activity was not modified. On the other hand, DADS was the strongest inhibitor of PNPH activity Ž45%..

Table 2 Summary of the results obtained in the mutagenicity assays Treatments

DAS DADS DPS

4. Discussion Several studies have shown that organosulfur compounds of Allium vegetables are able to alter the expression of xenobiotic-metabolizing enzymes ŽXME. w6,12,14,16,33,34x. Nevertheless, the exact incidence of these modifications, especially those of phase I enzymes, on carcinogen metabolism remains to be determined. The outcome with these modifications could result in a finely balanced, metabolic situation for which the equilibrium might easily be disturbed in favor of either a toxic response or a detoxication enhancement. Increased or decreased activity of specific CYP can be directly beneficial by decreasing metabolisation of some carcinogens. But such a protective effect would not, however, extend to other carcinogens. The aim of this study was to provide an overview of the repercussions of in vivo modifications of some CYP by OSCs on the activation of a wide range of mutagens, especially environmental carcinogens, in order to evaluate chemopreventive or noxious effects provided by OSCs treatment. This study shows that hepatic subcellular fractions from OSCs-treated rats were able to modify the activation of several promutagens ŽTable 2.. These promutagens are not mutagenic per se but need to be metabolized by CYP to induce genetic damage. Hence, the elevated or decreased mutagen activation by OSCs might be mediated by the induction or inhibition of specific CYPs that convert promutagens into active metabolites. To study the effects of OSCs, we used two different subcellular fractions: a postmitochondrial supernatant ŽS9. and a microsomal fraction. When using microsomes, the effects of OSCs are evaluated in relation to their capacity to increase or to inhibit specific CYPs. When using S9, the activities of cytosolic enzymes and the effects of

23

DPDS

S9 MSF S9 MSF S9 MSF S9 MSF

Mutagen CP

N-PiP

BaP

PhIP

DMN

qq qq 0 0 qq qq qq qq

qq q y y qq qq qq qq

q qq y 0 qq qq q qq

qq qq q q qq qq q q

yy yy yy yy y y q q

MSF s microsomal fraction; Žq. senhancement of activation when compared with control; Žy. sdecrease of activation when compared with control; 0 s no influence on the activation. The number of symbols is indicative of the intensity of the effect.

OSCs or their metabolites, which could be present in this fraction, are, in addition, taken into account. CP is principally activated in liver by a 4-hydroxylation reaction primarily catalyzed by CYP2C11 in control rats and by CYP2B1 in phenobarbital-induced rats w27x. The activation of N-PiP is mainly catalyzed by CYP2B w25x. Therefore, the induction of CYP2B activity by DAS, DPS and DPDS might be responsible for the increased activation of CP and N-PiP. The slight induction of CYP2B occurring in DADStreated rats would be inefficient for increasing the activation of CP. Conversely, an unexpected finding was the inhibition of N-PiP mutagenicity by the hepatic subcellular fraction from DADS-treated rats. This cannot be attributed to modification of CYP2B activity, which was slightly increased by DADS. Such an effect could be related to the inhibition of other CYP isoforms involved in the activation of N-PiP since DADS strongly reduced total cytochrome P450 content in the liver. The four OSCs slightly increased EROD and MROD activities, which are related to the CYP1A family. Since the CYP1A family is closely associated with the metabolic activation of chemical carcinogens such as heterocyclic amines and polycyclic aromatic hydrocarbons, it is conceivable that OSCs might enhance the activation of chemical carcinogens that rely on the same CYP for their activation. To test this hypothesis, we investigated the modulation of the food carcinogen, PhIP, and the environ-

24

D. Guyonnet et al.r Mutation Research 466 (2000) 17–26

mental pollutant, BaP. The initial activation step of PhIP via N-oxidation is catalysed predominantly by CYP1A2 w23x. The weak inductions of activity related to the CYP1A family were enough to significantly increase the activation of PhIP to mutagens. It is well-known that the activation of BaP is mainly catalysed by CYP1A1 in the rat w24,35x. We observed that DAS, DPS and DPDS equally affected the mutagenicity of BaP and PhIP in the Ames test. By contrast, S9 from DADS-treated rats inhibited the mutagenicity of BaP although this effect was not observed when microsomes were used as the activation system. Two mechanisms are likely to mediate the antimutagenic effect of DADS treatment on S9mediated mutagenicity of BaP. Metabolites of DADS might be present in the cytosol and scavenge the reactive metabolites of BaP. A second mechanism might involve the modulation of BaP metabolism by cytosolic enzymes. This latter effect cannot be due to the detoxication of 7b,8a-dihydroxy-9a ,10a-oxy7,8,9,10-tetrahydrobenzow axpyrene ŽBPDE., the most mutagenic metabolite of BaP, by GST since GST activity is extremely limited without addition of glutathion w36x. In addition to BPDE, quinones constitute another major class of BaP metabolites generated during its oxidative metabolism that are able to induce genetic damages w24,37x. DT-diaphorase can reduce the mutagenicity of BaP quinones w38,39x. In a previous study, we have shown that DADS was a potent inducer of DT-diaphorase in the rat liver w12x. Moreover, the addition of dicoumarol, an inhibitor of DT-diaphorase, diminishes the mutagenicity of BaP and abolishes the antimutagenic effect of DADS against BaP ŽD. Guyonnet, unpublished results.. This increased detoxification of BaP quinones through the rise of DT-diaphorase might explain the antimutagenic properties of S9 from DADS-treated rats. The mutagenicity of DMN, a nitrosamine selectively activated by CYP2E1, was strongly inhibited by DAS and DADS in accordance with inhibition of CYP2E1 activity. These results are in agreement with previous studies that have shown an inhibitory capacity of DAS and DADS on CYP2E1 w10,11,34x. For the first time, we have described the inhibition of DMN mutagenicity by DPS even if the antimutagenic ability of DPS was lower than DAS and DADS. The inhibition of DMN mutagenicity by DPS was not accompanied by an inhibition of PNPH

activity related to CYP2E1. In contrast, DPDS strongly increased DMN mutagenicity which is related to an induction of CYP2E1 activity. The inhibition of DMN mutagenicity by DAS and DADS was in agreement with other results by showing, on one hand, anticarcinogenic effects of these two OSCs against chemical carcinogenesis induced by DMN in rats w7x and, on the other hand, in vivo antigenotoxic effects against DMN w40x. Inhibition of CYP2E1, which is responsible for the activation of numerous carcinogenic chemicals, is believed to be a major mechanism by which DAS would exert its chemopreventive effect w7–9,20,41x. The antimutagenic properties of DADS against BaP were supported by another study that showed that DADS inhibits BaP-induced bone marrow micronuclei formation in mice w42x. Conversely, the anticarcinogenic effects of DAS against BaP-induced neoplasia of forestomach and lung in mice w6x were not in agreement with our results. Likewise, DADS had chemopreventive potential against PhIP-induced mammary carcinogenesis in rats w43x, whereas it induced PhIP mutagenicity in our study. In addition, DAS, which strongly enhanced CP mutagenicity in the Ames test, was capable of blocking nuclear aberration induction by CP in urinary bladder in mice w44x. Some phase II enzymes ŽGST, UGT., involved in the metabolism of these genotoxic agents, are likely inactive under our experimental conditions. Therefore, the lack of activity of phase II enzymes could explain the differences observed between our results and those of others. Among the four OSCs tested, DADS seems to be the most interesting chemopreventive agent in view of its antimutagenic effects against DMN-, N-PiPand BaP mutagenicity, even considering that DADS increases the mutagenicity of PhIP. The antimutagenic potential of DAS and DPS was demonstrated against DMN but, at the same time, they enhanced the activation of BaP, CP, N-PiP and PhIP by virtue of their induction of CYP activities related to the activation of these carcinogens. The least interesting OSCs tested would be DPDS, which enhances the mutagenicity of all mutagens. Nevertheless, OSCs were also able to induce phase II enzymes ŽGST, UGT, EH, DT-diaphorase. involved in the detoxification of carcinogen. Further studies are in progress to evaluate the repercussions of the induction of

D. Guyonnet et al.r Mutation Research 466 (2000) 17–26

phase II enzymes by OSCs on carcinogen metabolism. These studies will determine if this action can tip the scales in favor of detoxification pathways.

w8x

w9x

5. CAS registry number Benzow axpyrene ŽBaP. Cyclophosphamide ŽCP. Dimethylnitrosamine ŽDMN. N-nitrosopiperidine Ž N-PiP. 2-Amino-1-methyl-6-phenylimidazo w4,5-b xpyridine ŽPhIP. Diallyl sulfide ŽDAS. Diallyl disulfide ŽDADS. Dipropyl sulfide ŽDPS. Dipropyl disulfide ŽDPDS.

50-32-8 6055-19-2 62-75-9 100-75-4 105650-23-5

w10x

w11x

592-88-1 2179-57-9 111-47-7 629-19-6

w12x

Acknowledgements

w13x

This work was supported by funds from the Conseil Regional de Bourgogne. ´

w14x

References w15x w1x E. Dorant, P.A. Van den Brandt, R.A. Goldbohm, F. Sturmans, Consumption of onions and a reduced risk of stomach carcinoma, Gastroenterology 110 Ž1996. 12–20. w2x E. Dorant, P.A. Van den Brandt, R.A. Goldbohm, R.J. Hermus, F. Sturmans, Garlic and its significance for the prevention of cancer in humans: a critical view, Br. J. Cancer 67 Ž1993. 424–429. w3x K.A. Steinmetz, L.H. Kushi, R.M. Bostick, A.R. Folsom, J.D. Potter, Vegetables, fruit, and colon cancer in the Iowa Women’s Health Study, Am. J. Epidemiol. 139 Ž1994. 1–15. w4x E. Buiatti, D. Palli, A. Decarli, D. Amadori, C. Avellini, S. Bianchi, R. Biserni, F. Cipriani, P. Cocco, A. Giacosa, E. Marubini, R. Puntoni, C. Vindigni, J. Fraumeni, W. Blot, A case–control study of gastric cancer and diet in Italy, Int. J. Cancer 44 Ž1989. 611–616. w5x F. Levi, C. Pasche, C. La Vecchia, F. Lucchini, S. Franceschi, Food groups and colorectal cancer risk, Br. J. Cancer 79 Ž1999. 1283–1287. w6x V.L. Sparnins, G. Barany, L.W. Wattenberg, Effects of organosulfur compounds from garlic and onions on benzow axpyrene-induced neoplasia and glutathione S-transferase activity in the mouse, Carcinogenesis 9 Ž1988. 131– 134. w7x D. Haber-Mignard, M. Suschetet, R. Berges, ` P. Astorg, M.H. Siess, Inhibition of aflatoxin B1- and N-nitrosodiethyl-

w16x

w17x

w18x

w19x

w20x

25

amine-induced liver preneoplastic foci in rats fed naturally occurring allyl sulfides, Nutr. Cancer 25 Ž1996. 61–70. M.J. Wargovich, Diallyl sulfide, a flavor component of garlic Ž Allium satiÕum., inhibits dimethylhydrazine-induced colon cancer, Carcinogenesis 8 Ž1987. 487–489. J.Y. Hong, Z.Y. Wang, T.J. Smith, S. Zhou, S. Shi, J. Pan, C.S. Yang, Inhibitory effects of diallyl sulfide on the metabolism and tumorigenicity of the tobacco-specific carcinogen 4-Žmethylnitrosamino.-1-Ž3-pyridyl.-1-butanone ŽNNK. in ArJ mouse lung, Carcinogenesis 13 Ž1992. 901– 904. J.F. Brady, D.C. Li, H. Ishizaki, C.S. Yang, Effect of diallyl sulfide on rat liver microsomal nitrosamine metabolism and other monooxygenase activities, Cancer Res. 48 Ž1988. 5937–5940. M.K. Kwak, S.G. Kim, J.Y. Kwak, R.F. Novak, N.D. Kim, Inhibition of cytochrome P4502E1 expression by organosulfur compounds allyl sulfide, allyl mercaptan and allyl methylsulfide in rats, Biochem. Pharmacol. 47 Ž1994. 531–539. D. Guyonnet, M.H. Siess, A.M. Le Bon, M. Suschetet, Modulation of phase II enzymes by organosulfur compounds from Allium vegetables in rat tissues, Toxicol. Appl. Pharmacol. 154 Ž1999. 50–58. B.S. Reddy, C.V. Rao, A. Rivenson, G. Kelloff, Chemoprevention of colon carcinogenesis by organosulfur compounds, Cancer Res. 53 Ž1993. 3493–3498. D. Haber, M.H. Siess, M.C. Canivenc-Lavier, A.M. Le Bon, M. Suschetet, Differential effects of dietary diallyl sulfide and diallyl disulfide on rat intestinal and hepatic drugmetabolizing enzymes, J. Toxicol. Environ. Health 44 Ž1995. 423–434. S.V. Singh, S.S. Pan, S.K. Srivastava, H. Xia, X. Hu, H.A. Zaren, J.L. Orchard, Differential induction of NADŽP.H: quinone oxidoreductase by anti-carcinogenic organosulfides from garlic, Biochem. Biophys. Res. Commun. 244 Ž1998. 917–920. M.H. Siess, A.M. Le Bon, M.C. Canivenc-Lavier, M. Suschetet, Modification of hepatic drug-metabolizing enzymes in rats treated with alkyl sulfides, Cancer Lett. 120 Ž1997. 195–201. J. Pan, J.Y. Hong, B.L. Ma, S.M. Ning, S.R. Paranawithana, C.S. Yang, Transcriptional activation of cytochrome P450 2B1r2 genes in rat liver by diallyl sulfide, a compound derived from garlic, Arch. Biochem. Biophys. 302 Ž1993. 337–342. R.W. Teel, Effect of phytochemicals on the mutagenicity of the tobacco-specific nitrosamine 4-Žmethylnitrosamino.-1Ž3-pyridyl.-1-butanone ŽNNK. in Salmonella typhimurium strain TA1535, Phytother. Res. 7 Ž1993. 248–251. P.P. Tadi, R.W. Teel, B.H. Lau, Organosulfur compounds of garlic modulate mutagenesis, metabolism, and DNA binding of aflatoxin B1, Nutr. Cancer 15 Ž1991. 87–95. Y.J. Surh, R.C. Lee, K.K. Park, S.T. Mayne, A. Liem, J.A. Miller, Chemoprotective effects of capsaicin and diallyl sulfide against mutagenesis or tumorigenesis by vinyl carbamate and N-nitrosodimethylamine, Carcinogenesis 16 Ž1995. 2467–2471.

26

D. Guyonnet et al.r Mutation Research 466 (2000) 17–26

w21x M.E. Dion, M. Agler, J.A. Milner, S-allyl cysteine inhibits nitrosomorpholine formation and bioactivation, Nutr. Cancer 28 Ž1997. 1–6. w22x A. Bu-Abbas, M.N. Clifford, R. Walker, C. Ioannides, Selective induction of rat hepatic CYP1 and CYP4 proteins and of peroxisomal proliferation by green tea, Carcinogenesis 15 Ž1994. 2575–2579. w23x H. Wallin, A. Mikalsen, F.P. Guengerich, M. IngelmanSundberg, K.E. Solberg, O.J. Rossland, J. Alexander, Differential rates of metabolic activation and detoxication of the food mutagen 2-amino-1-methyl-6-phenylimidazow4,5b xpyridine by different cytochrome P450 enzymes, Carcinogenesis 11 Ž1990. 489–492. w24x H.V. Gelboin, Benzowalphaxpyrene metabolism, activation and carcinogenesis: role and regulation of mixed function oxidases and related enzymes, Physiol. Rev. 60 Ž1980. 1107– 1166. w25x A.D. Ayrton, J.N. Smith, C. Ioannides, Bioactivation of N-nitrosopiperidine to mutagens: role of hepatic cytochrome P450 proteins and contribution of cytosolic fraction, Carcinogenesis 8 Ž1987. 1691–1695. w26x C.S. Yang, J.S. Yoo, H. Ishizaki, J.Y. Hong, Cytochrome P450IIE1: roles in nitrosamine metabolism and mechanisms of regulation, Drug Metab. Rev. 22 Ž1990. 147–159. w27x L. Clarke, D.J. Waxman, Oxidative metabolism of cyclophosphamide: identification of the hepatic monooxygenase catalysts of drug activation, Cancer Res. 49 Ž1989. 2344– 2350. w28x L.J. Yu, P. Drewes, K. Gustafsson, E.G. Brain, J.E. Hecht, D.J. Waxman, In vivo modulation of alternative pathways of P450-catalyzed cyclophosphamide metabolism: impact on pharmacokinetics and antitumor activity, J. Pharmacol. Exp. Ther. 288 Ž1999. 928–937. w29x D. Haber, M.H. Siess, I. de Waziers, P. Beaune, M. Suschetet, Modification of hepatic drug-metabolizing enzymes in rat fed naturally occurring allyl sulphides, Xenobiotica 24 Ž1994. 169–182. w30x D.M. Maron, B.N. Ames, Revised methods for the Salmonella mutagenicity test, Mutat. Res. 113 Ž1983. 173–215. w31x M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72 Ž1976. 248– 254. w32x W. Tassaneeyakul, M.E. Veronese, D.J. Birkett, F.J. Gonzalez, J.O. Miners, Validation of 4-nitrophenol as an in vivo substrate probe for human liver CYP2E1 using cDNA expression and microsomal kinetic techniques, Biochem. Pharmacol. 46 Ž1993. 1975–1981.

w33x V.A. Gudi, S.V. Singh, Effect of diallyl sulfide, a naturally occurring anti-carcinogen, on glutathione-dependent detoxification enzymes of female CD-1 mouse tissues, Biochem. Pharmacol. 42 Ž1991. 1261–1265. w34x J.F. Brady, M.H. Wang, J.Y. Hong, F. Xiao, Y. Li, J.S. Yoo, S.M. Ning, M.J. Lee, J.M. Fukuto, J.M. Gapac, C.S. Yang, Modulation of rat hepatic microsomal monooxygenase enzymes and cytotoxicity by diallyl sulfide, Toxicol. Appl. Pharmacol. 108 Ž1991. 342–354. w35x A.H. Conney, R.L. Chang, D.M. Jerina, S.J. Wei, Studies on the metabolism of benzow axpyrene and dose-dependent differences in the mutagenic profile of its ultimate carcinogenic metabolite, Drug Metab. Rev. 26 Ž1994. 125–163. w36x H. Greim, W. Goggelmann, K.H. Summer, T. Wolff, Mutagenicity testing with Salmonella microsome test, Arch. Toxicol. 46 Ž1980. 31–40. w37x P.J. O’Brien, Molecular mechanisms of quinone cytotoxicity, Chem. Biol. Interact. 80 Ž1991. 1–41. w38x P. Joseph, A.K. Jaiswal, NADŽP.H:quinone oxidoreductase 1 reduces the mutagenicity of DNA caused by NADPH: P450 reductase-activated metabolites of benzow axpyrene quinones, Br. J. Cancer 77 Ž1998. 709–719. w39x P. Joseph, A.K. Jaiswal, NADŽP.H:quinone oxidoreductase 1 ŽDT diaphorase. specifically prevents the formation of benzow axpyrene quinone–DNA adducts generated by cytochrome P4501A1 and P450 reductase, Proc. Natl. Acad. Sci. U.S.A. 91 Ž1994. 8413–8417. w40x A.M. Le Bon, C. Roy, C. Dupont, M. Suschetet, In vivo antigenotoxic effects of dietary allyl sulfides in the rat, Cancer Lett. 114 Ž1997. 131–134. w41x L.W. Wattenberg, V.L. Sparnins, G. Barany, Inhibition of N-nitrosodiethylamine carcinogenesis in mice by naturally occurring organosulfur compounds and monoterpenes, Cancer Res. 49 Ž1989. 2689–2692. w42x H.S. Marks, J.L. Anderson, G.S. Stoewsand, Inhibition of benzow axpyrene-induced bone marrow micronuclei formation by diallyl thioethers in mice, J. Toxicol. Environ. Health 37 Ž1992. 1–9. w43x N. Suzui, S. Sugie, K.M. Rahman, M. Ohnishi, N. Yoshimi, K. Wakabayashi, H. Mori, Inhibitory effects of diallyl disulfide or aspirin on 2-amino-1-methyl-6-phenylimidazow4,5b xpyridine-induced mammary carcinogenesis in rats, Jpn. J. Cancer Res. 88 Ž1997. 705–711. w44x M.T. Goldberg, P.D. Josephy, Studies on the mechanism of action of diallyl sulfide, an inhibitor of the genotoxic effects of cyclophosphamide, Can. J. Physiol. Pharmacol. 65 Ž1987. 467–471.