Cancer Letters. 59 (1991) 89-94

Eisevier Scientific Publishers Ireland Ltd

a9

Effect of allixin, a phytoalexin produced by garlic, on mutagenesis, DNA-binding and metabolism of aflatoxin B1

T. Yamasaki”, R.W. Teelb and B.H.S. Laua

*Department of Microbiology and bDepartment of Physiology, School of Medicine, Loma

Linda University, Loma Linda, California 92350 (U.S.A.)

(Received 9 May 1991)

(Accepted 14 May 1991)

Summary

Allixin, a phytoalexin isolated from garlic, was examined for its effects on aflatoxin BI(AFBI)-induced mutagenesis using Salmonella typhimurium TAlOO as the bacteri- al tester strain and rat liver S9 fraction as the metabolic acrioation system. The effects of allixin on the binding of [3H]AFB1 to calrf thymus DNA and on the formation of merabolites of t3H]AFB1 were also determin- ed. Allixin showed a dose-related inhibition of Histidine+ reoertants induced by AFk31. Allixin at 75 pg/ml inhibited [3.F_rlAFB, binding to calf thymus DNA and reduced formation of AFBI-DNA adducts. In addition, allixin ex- hibited a concentration-dependent inhibition of the formation of organosoluble metabolites and the glutathione conjugates of [3H]AFB1. The data indicate that the effect of allixin on AFBI-induced mutagenesis and binding of metabolires to DNA may be mediated through an inhibition of microsomal P-450 enzymes. Allixin may thus be useful in the chemopreven- tion of cancer.

Correspondence to: B.H.S. Lau, Department of Microbiology,

School of Medicine, Loma Linda University, Loma Linda, CA

92350, U.S.A.

Keywords: allixin; garlic; phytoalexin; mutagenesis; aflatoxin B1

Introduction

Plants produce bioactive chemicals to pro- tect themselves from pathogens and insect pests. These chemicals, so-called ‘phytoalex- ins’ (phyto = plant, alexin = to ward off), are major weapons for plant defense. Well known examples are rishitin (produced by potatoes, tomatoes and tobacco), glyceollin (produced by soybean) and pisatin (produced by pea) [ 11. Phytoalexins have also been described as ‘stress compounds’ because their synthesis is induced by exposure of a plant to certain kinds of stress, such as contact with bacteria, viruses, fungi, ultraviolet rays or heavy metal salts [5].

Recently, Kodera et al. [8] isolated a new phytoalexin, a stress compound synthesized by garlic, and named it allixin. This is a phenolic compound (Fig. l), shown to be an effec- tive inhibitor of phospholipid metabolism stimulated in vitro by the tumor promoter, 12-0-tetradecanoylphorbol-13-acetate (TPA) [17]. Using a mouse skin tumor model with dimethylbenz[a]anthracene (DMBA) as the in- itiator, treatment with allixin was shown to significantly reduce the promoting effects of TPA on tumorigenesis [17].

We hypothesized that allixin has certain an-

0304-3835/91/$03.50 0 1991 El sevier Scientific Publishers Ireland Ltd

Published and Printed in Ireland

90

H&O

H,C CH,CH,CH,CH,CH,

Fig. 1. Structure of allixin (3-hydroxy-5-methoxy-6-

methyl-2-pentyl-4H-pyran-4-one).

timutagenic effects against naturally-occurring mutagens because it is (1) produced by garlic, which has been shown to be anticarcinogenic [9], (2) a phenolic compound and plant phenolics have been shown to inhibit car- cinogenesis as well as mutagenesis [16], and [3] a phytoalexin, which is produced by plants to protect themselves from many stresses in- cluding mycotoxins. We have studied aflatoxin Br (AFBi), a potent mycotoxin, mutagen and carcinogen, using in vitro models [22,23]. AFBr requires metabolic activation to manifest its mutagenicity and carcinogenicity [ 18,201. In this paper, we report the effects of allixin on rat liver S-9 mediated mutagenesis, DNA- binding and metabolism of AFBr.

Materials and Methods

Chemicals Allixin was kindly supplied by Wakunaga

Pharmaceutical Co. Ltd., Osaka, Japan, and dissolved in dimethylsulfoxide for each assay. Unlabeled AFBr and calf thymus DNA were purchased from Sigma Chemical Co., St. Louis, MO. [3H]AFBI (28 Ci/mmol) was ob- tained from Moravek Biochemicals Inc., Brea, CA. Aroclor 1254 (polychlorinated biphenyl) was purchased from Foxboro Co., North Ha- ven, CT. Bactoagar was obtained from Difco Laboratories, Detroit, MI, and nutrient broth (Oxoid #2) came from Oxoid Co., Bas- ingstoke, Hants, U.K.

Preparation of liver S9 Male Sprague-Dawley rats weighing

180-200 g (Harlan Laboratories, Indianapo- lis, IN) were given a single intraperitoneal in- jection of Aroclor 1254 (500 (mg/kg), 5 days before sacrifice. Standard Purina laboratory chow and water were provided ad libitum. Following removal of livers, S9 was prepared [14] and the protein concentration was deter- mined [ll]. Aliquots of S9 (31 mg/ml) were stored at - 80°C and used as needed.

Mutagenesis assay Mutagenesis assays were performed using

the plate incorporation method of Maron and Ames [14]. Salmonella typhimurium TAlOO (a gift from Dr. Bruce Ames, University of California, Berkeley) was used as a tester strain to evaluate the effects of allixin on AFBr-induced mutagenesis. An aliquot of 0.1 ml of an overnight culture of TA-100 in nutrient broth, 0.78 mg of S9, allixin (25- 150 pg) and AFBr (0.05 pg) were mixed with 0.5 ml of co-factor buffer. After the addition of 2 ml of top agar the contents were mixed and then poured onto minimal glucose agar plates. Histidine+ revertants were scored after 48-h incubation at 37OC and data were ex- pressed as mean revertant colonies/plate after subtraction of spontaneous revertants. The tests were performed in triplicate.

[3H]AFB1 metabolite binding to DNA and for- mation of [31-flAFB,-DNA adducts

[3H]AFB1 (500 pmol, 14 j&i), S9 (1.55 mg) and allixin (75 @g/ml) in 1 ml of co-factor buffer were incubated with 0.5 mg purified calf thymus DNA at 37OC for 1 h. DNA was ex- tracted [13], dried under Ns and redissolved in 10 mM Tris buffer containing 0.1 mM EDTA. Aliquots were counted in a Beckman LS5801 liquid scintillation counter (Beckman Instrument Co., Berkeley, CA). DNA was quantitated by the diphenylamine assay [3], and AFBr binding was expressed as pmol/mg DNA.

To determine AFBr-DNA adducts, DNA samples were acid hydrolyzed [12] and ex- tracted into l-butanol. The butanol layer was dried under N2 and redissolved in 1 ml of

91

10% ethanol in 20 mM triethylammonium for- mate (TEAF). AFBi-DNA adducts were separated by reverse-phase high performance liquid chromatography (HPLC) on a Zorbax ODS column (4.6 mm x 250 mm, DuPont In- struments Co., Wilmington, DE), using a linear gradient of 10% to 18% ethanol in 20 mM TEAF for 60 min at a flow rate of 1 ml/min [7]. The fractions were collected on an ISCO Retriever II fraction collector fitted with a flow- interrupter valve (Instrument Specialties Co., Lincoln, NE). The radioactivity was determin- ed by liquid scintillation spectroscopy and analyzed with a Beckman Datagraph Program in the DU Data Leader software package on an IBM PS/2 model 50 computer. Data were ex- pressed as pmol/mg DNA. The tests were per- formed in duplicate.

Analysis of [3H]AFB1 mefabolifes [3H]AFB1 (51.9 pmol, 1.4 &i), S9 (1.55

mg) and various concentrations of allixin (25, 50 and 75 pg) in 1 ml of co-factor buffer were incubated at 37OC in a shaking waterbath for 25 min. The samples were extracted twice with chloroform/ethyl acetate (1:l) [lo] and the pooled organosoluble metabolites were dried under Nz. After redissolving, the metabolites were separated isocratically by HPLC on a Zorbax ODS column in 5% tetrahydrofuran, 20% methanol, 20% acetonitrile, and 55% water over 23 min at a flow rate of 1 ml/min. Water-soluble fractions were lyophilized, redissolved in 50% methanol and passed through a Cl8 Sep-Pak column (Water Associates, Milford, MA). Samples were then analyzed isocratically by HPLC in 25% ethanol in 20 mM TEAF (pH 3). The radioac- tivity in the collected fractions was determined by liquid scintillation spectroscopy and analyz- ed with a Beckman Datagraph Program as described above. Data were expressed as pmol/mg S9 protein. Organosoluble metabolites and glutathione conjugates of AFBi were identified from authentic stan- dards. The tests were performed in duplicate.

Statistical analysis The data were subjected to one-way analysis

600

500

400

300

200

100

l *

w z I I I

0 50 100 150

Con~~twion of Allixh (pg/plate)

Fts- 2. Effects of allixin on S9-mediated mutagenesis of

aflatoxin B1 in S. typhimurium TAlOO. Histidine+ rever- tants, scored after 48 h incubation, were expressed as

means f S.E.M. from triplicate plates. Asterisks indicate significant difference from control (P < 0.05).

of variance (ANOVA) followed by Tukey’s test for honestly significant difference (HSD). The analysis was performed with Statgraphics soft- ware version 3.0 (STSC, Inc., Rockville, MD).

Results

Figure 2 shows the effects of allixin on AFB,-induced mutagenesis in S. typhimurium strain TAlOO. Allixin exhibited a concentra- tion-dependent inhibition of AFBr-induced mutagenesis with 66% inhibition at a concen- tration of 150 pg/plate. All the inhibitory con- centrations of allixin were non-toxic to the bacteria.

The results of [3H]AFBI metabolite binding to calf thymus DNA and formation of AFBi-DNA adducts are shown in Table I. Allixin at 75 pg/ml significantly reduced the binding of AFBi metabolites to DNA. AFBi-DNA adducts were analyzed by HPLC and four peaks were identified. Allixin ex- hibited a significant inhibition of the formation of AFBi-N7-FAPyr(minor), AFBi-N7-FAPyr

(major), AFBi-diol and AFB,-N7-Gua adducts.

Figure 3 illustrates the effects of allixin on the formation of organosoluble metabolites of [3H]AFBI. The metabolized AFBi was

92

Table 1. Effect of allixin on [3H]AFB, binding to calf thymus DNA and on the formation of [3H]AFB,-DNA adducts.

Control Allixin (no allixin) (75 pg/ml)

AFB, binding to DNA” 106.48 =t 1.45 70.72 f 0.34’

AFB,-DNA adductsb AFB,-N7-FAPyr

(minor) 2.86 f 0.13 0.31 f 0.09’ AFB,-N7-FAPyr

(major) 1.28 f 0.04 0.87 f 0.04’ AFB1-diol 1.76 f 0.02 1.14 f 0.08’ AFB1-N7-Gua 52.80 f 1.20 38.51 f 1.41’

Values are means + S.E.M. from duplicate samples, ex- pressed as pmol/mg DNA.

“Calf thymus DNA, rat liver S-9, t3H]AFB, and aitixin were incubated as described in Materials and Methods. Radioactivity was determined by liquid scintillation spec-

troscopy.

bAfter extraction, DNA samples were hydrolyzed and analyzed by HPLC. Four peaks were detected and quan- titated to each [3H]AFB,-DNA adduct. *Significant difference from control (P < 0.05).

n Control (no ailixin) 15 t n Allixin 25 pg/ml

u H Allixin 50 m/ml .- s q Allixin 75 pgglml l

El ul ‘0

3

$5 0 3 a a

Total Metabolhed Unmetabolhed Am Am

Ownosoluble Metabolites

Fig. 3. Effects of allixin on the formation of organosolu- ble metabolites of AFB,. Values are means f S.E.M. from duplicate samples. After incubation of allixin and 13H]AFB, with S9, metabolites were extracted and

analyzed by HPLC. Asterisks indicate significant dif- ference from control (P < 0.05).

n Control (no allixin)

1 Allixin 25 wrnl

n 1.5 .I S

Allixin 75 wrnl

k 3 1.0

3 3 0.5 3 a a

0.0 Glntathione Coqjugates

Fig. 4. Effects of allixin on the formation of

AFB,-glutathione conjugates. Water-soluble fractions were passed through a Cl8 Sep-Pak column and analyz-

ed by HPLC. Values are means * S.E.M. from duplicate

samples. Asterisks indicate significant difference from control (P < 0.05).

significantly decreased by allixin at 25-75 pg/ml; this was accompanied by significant in- creases of the unmetabolized AFBr. The ef- fects of different concentrations of ailixin on the formation of glutathione conjugates of [3H]AFBr are shown in Fig. 4. All concentra- tions of allixin tested significantly inhibited the formation of AFBr glutathione conjugates.

Discussion

Garlic (Allium satiuum) has been shown to inhibit tumor growth and to prevent chemical carcinogenesis [9]. Plant phenolics, such as flavonoids, have also been found to inhibit the actions of chemical carcinogens [16]. It is an- ticipated that many more plants will be discovered to play a role in cancer prevention [6]. Wargovich et al. [24] reported that diallyl sulfide (DAS), an oil-soluble organosulfur compound of garlic, inhibited the nuclear aber- ration of mouse colonic epithelial cells induced by 1,2_dimethylhydrazine (DMH) . Four organosulfur compounds of garlic were shown to inhibit benzoIa]pyrene-induced neoplasia of the forestomach in mice [19]. We have recent- ly reported that an aged garlic extract, ajoene,

93

and DAS suppressed AFB1-induced

mutagenesis in Salmonella typhimurium strain TAlOO [22].

In the present study, we demonstrated that allixin, a phenolic compound isolated from stress-exposed garlic, inhibited the number of histidine+ revertants induced by AFBl in TAlOO (Fig. 2) in incubations with rat liver S9 fraction. In addition, allixin decreased the binding of AFBl metabolites to calf thymus DNA and the formation of AFB1-DNA ad- ducts (Table I).

Cytochrome P-450 mediated mixed- function oxidases (MFO), localized in microsomes, metabolize AFBl to reactive spe- cies. AFBl &g-oxide is reported to bind to the N7 atom of guanine [4,7]. Brady et al. [2] reported that DAS was a competitive inhibitor of N-dimethylnitrosamine demethylase. Allixin inhibited the metabolism of AFBl in our studies as shown by the increase in unmetabolized AFBl and by the decrease in organosoluble metabolite formation (Fig. 3). In addition, allixin decreased the glutathione con- jugates of [3H]AFB1 (Fig. 4). The conjugation of xenobiotic metabolites with glutathione is mediated by glutathione S-transferase (GST) [7]. Sumiyoshi et al. [21] correlated the inhibi- tion of DMH-induced colon carcinogenesis in vivo with increased GST activity by S-ally1 cys- teine which is a water-soluble compound of garlic. Our results of inhibition of AFB1- glutathione conjugates by allixin may reflect the inhibition of cytochrome P-450 activity resulting in reduced amount of the substrate available for conjugation with glutathione. AFBl &g-oxide is reported to be the substrate for glutathione conjugation and forms adducts with DNA [4,7]. The inhibition of the forma- tion of AFBl &g-oxide by allixin could ac- count for the inhibition of AFB1-induced mutagenesis in TAlOO, AFBl binding to calf thymus DNA and AFB1-DNA adducts forma- tion Plants synthesize and store phytoalex- ins in response to intruders such as microorganisms. Phytoalexins have been shown to be toxic in animals [l]; however, it is possible that these chemicals may have useful

modulatory functions as well. Nishino et al. [17] reported that allixin acted as an inhibitor of phospholipid metabolism induced by the tumor promoter, TPA. The effects of allixin we report may be in part due to modulation of phospholipids in microsomal membranes which play a role in the metabolism of xenobiotics [ 181. Another possible explanation for the effect of allixin we report may relate to a competitive antagonism caused by cytochrome P-450 metabolism of allixin. Mat- thews et al. [15] found that pisatin, a phytoalexin from pea plants (Pisum satiuum), was detoxified by fungal cytochrome P-450 demethylation reactions. Varying the concen- trations of the mutagen/carcinogen to that of allixin would be informative in this regard. The effects of allixin on AFB1-induced mutagenesis in S. typhimurium strain TAlOO and on the metabolism of AFBl were concen- tration dependent (Figs. 2-4).

Wattenberg [25] has classified the inhibitors of carcinogenesis into three categories based on the mode of action: [l] metabolic inhibitors prevent the formation of carcinogens from precursor compounds, [2] blocking agents in- hibit carcinogens from reaching or reacting with critical target sites and [3] suppressing agents inhibit the process of neoplastic manifestation. Phenolic compounds have previously been shown to act at all three levels [25]. The present data indicate that allixin acts as a metabolic inhibitor and a blocking agent as the effects of allixin on AFB1-induced mutagenesis and on AFBl binding to DNA correlated with inhibition of S9-mediated me- tabolism. The data thus suggest that allixin may play a role in cancer prevention warrant- ing further investigations into the possible anti- carcinogenic properties of phytoalexins.

Acknowledgements

This work was supported by Chan Shun In- ternational Foundation, Burlingame, Califor- nia, U.S.A. and by Wakunaga Pharmaceutical Co., Ltd, Osaka, Japan. We thank Padma P. Tadi, Brian Y .Y. Wong and Kiok Lim for assis- tance in this study.

94

References

1

2

3

4

5

6

7

8

9

10

11

12

Beier, R.C. (1990) Natural pesticides and bioactive com-

ponents in foods. Rev. Environ. Contam. Toxicol., 113,

47-137.

Brady, J.F., Li, D., Ishizaki, H. and Yang, C.S. (1988) Ef-

fect of diallyl sulfide on rat liver microsomal nitrosamlne

metabolism and other monooxygenase activities. Cancer

Res., 48, 5937-5940.

Burton, K. (1956) A study of the conditions and mecha-

nism of diphenylamine reaction for the calorimetric estima-

tion of the deoxyribonucleic acid. Biochem. J., 62,

315-323.

Essigmann, J.M., Croy, R.G., Nadzan, A.M., Busby, Jr.,

W.F., Reinhold, V.N., Buchi, G. and Wogan, G.N.

(1977) Structural identification of the major DNA adduct

formed by aflatoxin B, in uitro. Proc. Natl. Acad. Sci.

USA, 74, 1870-1874.

Grisebach, H. and Ebel, J. (1978) Phytoalexins, chemical

defense substances of higher plants? Angew. Chem. Int.

Ed. Engl.. 17. 635-647.

Hocman, G. (1989) Prevention of cancer: Vegetables and

plants. Comp. Biochem. Physiol., 93, 201-212.

Kensler, T.W., Egner, P.A., Davidson, N.E., Roebuck,

B.D., Pikul, A. and Groopman, J.D. (1986) Modulation of

aflatoxin metabolism, aflatoxin-N7-guanine formation, and

hepatic tumorigenesis in rats fed ethoxyquin: role of induc-

tion of glutathione S-transferases. Cancer Res., 46,

3924-3931.

Kodera, Y., Matsuura, H., Yoshida, S., Sumida, T.,

Itakura, Y., Fuwa, T. and Nishino, H. (1989) Allixin, a

stress compound from garlic. Chem. Pharm. Bull., 37,

1656- 1658.

Lau, B.H.S., Tadi, P.P. and Tosk, J.M. (1990) Allium

satiuum (garlic) and cancer prevention. Nutr. Res.. 10,

937-948.

Lotlikar,.P.D., Raj, H.G., Bohm, L.S.. Ho, L.L., Jhee,

E.C., Tsuji, K. and Gopalan, P. (1989) A mechanism of

inhibition of aflatoxin B,-DNA binding in the liver by

phenobarbital pretreatment of rats. Cancer Res., 49,

951-957.

Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall,

R.J. (1951) Protein measurement with the folin ph&nol

reagent. J. Biol. Chem., 193, 265-275.

Mandal, S., Ahuja, A., Shivapurkar, N.M., Cheng, S.J.,

Groopman, J.D. and Stoner, G.D. (1987) Inhibition of

aflatoxin B, mutagenesis in Salmonella typhimurium and

13

14

15

16

17

18

19

20

21

22

23

24

25

DNA damage in cultured rat and human tracheobronchial

tissues by ellagic acid. Carcinogenesis, 8, 1651-1656.

Maniatis, T., Fritsh, E.F. and Sambrook. J. (Edit.) In:

Molecular cloning: A laboratory manual. p. 458, Cold

Spring Harbor Laboratory, 1982.

Maron, D.M. and Ames, B.N. (1983) Revised methods for

the Salmonella mutagenicity test. Mutat. Res., 113,

173-215.

Matthews, D.E. and Van Etten. H.D. (1983) Detoxifica-

tion of the phytoalexin pisatin by a fungal cytochrome

P-450. Arch. Biochem. Biophys., 224, 494-505

Newmark, H.L. (1987) Plant phenolics as inhibitors of

mutational and precarcinogenic events. Can. J. Physiol.

Pharmacol., 65, 461-466.

Nishino, H., Nishino. A., Takayasu, J., Iwashima, A.,

Itakura, Y., Kodera, Y., Matsuura, H. and Fuwa, T. (1990) Antitumor-promoting activity of allixin, a stress

compound produced by garlic. Cancer J., 3, 20-21.

Singer, B. and Grunberger, D. (Edit.) In: Molecular

biology of mutagens and carcinogens. pp. 97-188,

Plenum Press, New York, 1983.

Sparnins, V.L., Barany, G. and Wattenberg, L.W. (1988)

Effects of organosulfur compounds from garlic and onions

on benzo[a]pyrene-induced neoplasia and glutathione S-

transferase activity in the mouse. Carcinogenesis, 9.

131-134.

Stark, A.A. (1980) Mutagenicity and carcinogenicity of

mycotoxins: DNA binding as a possible mode of action.

Annu. Rev. Microbial., 34, 235-262.

Sumiyoshi, H. and Wargovich, M.J. (1990)

Chemoprevention of 1,2-dimethylhydrazine-induced col-

on cancer in mice by naturally occurring organosulfur com-

pounds. Cancer Res., 50, 5084-5087.

Tadi, P.P., Teel, R.W. and Lau, B.H.S. (1991)

Organosulfur compounds of garlic modulate mutagenesis,

metabolism, and DNA binding of aflatoxin B1. Nutr. Can-

cer, 15, 87-95.

Teel, R.W. and Strother. A. (1990) Glucose alters rat liver

S9-mediated mutagenesis, metabolism and DNA-binding

of aflatoxin B,. Cancer Lett., 54, 163- 169.

Wargovich, M.J. and Goldberg, M.T. (1985) Diallyl

sulfide: A naturally occurring thioether that inhibits

carcinogen-induced nuclear damage to colon epithelial

cells in uioo. Mutation Res., 143, 127-129.

Wattenberg, L.W. (1985) Chemoprevention of cancer.

Cancer Res., 45, 1-8.