A program of cancer prevention, which aimed at identifying and minimizing human exposure to hazardous chemicals, requires the development of a rapid inexpensive screening assay as a complement to expensive long-term animal tests. Among thousands of hazardous chemicals to which humans are exposed, very few have been tested for genotoxicity. This report presents 255 chemicals (mutagens, non-mutagens, etc.) of a wide variety of chemical types which have been tested for genotoxicity, both in the Salmonella microsome test (Ames test) and in 8-hydroxyguanine (8-OH-Gua) assay, in order to detect carcinogens. Of the 255 chemicals tested, 83 chemicals (32.5%) were mutagenic in S. Typhimurium YG1041, YG1042, YG3003 and/or YG7108 strain, and 21 chemicals (8.2%) formed 8-OH-Gua in the rat hepatocytes oxidized DNA. These results may demonstrate the utility and limitation of both the Ames test and 8-OH-Gua assay in detecting chemicals likely to be toxic
- 25 - Genotoxicity of 255 chemicals in the Salmonella microsome test (Ames test) and 8-hydroxyguanine (8-OH-Gua) assay for the detection of carcinogens Nobuyuki Sera*, Yoshito Tanaka*, Hiroko Tsukatani*, Nobuhiro Shimizu*, Shigeji Kitamori* and Hideo Utsumi** *Fukuoka Institute of Health and Environmental Sciences, 39 Dazaifu Fukuoka 818-0135, Japan **Graduate School of Pharmaceutical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan Abstract A program of cancer prevention, which aimed at identifying and minimizing human exposure to hazardous chemicals, requires the development of a rapid inexpensive screening assay as a complement to expensive long-term animal tests. Among thousands of hazardous chemicals to which humans are exposed, very few have been tested for genotoxicity. This report presents 255 chemicals (mutagens, non-mutagens, etc.) of a wide variety of chemical types which have been tested for genotoxicity, both in the Salmonella microsome test (Ames test) and in 8-hydroxyguanine (8-OH-Gua) assay, in order to detect carcinogens. Of the 255 chemicals tested, 83 chemicals (32.5%) were mutagenic in S. Typhimurium YG1041, YG1042, YG3003 and/or YG7108 strain, and 21 chemicals (8.2%) formed 8-OH-Gua in the rat hepatocytes oxidized DNA. These results may demonstrate the utility and limitation of both the Ames test and 8-OH-Gua assay in detecting chemicals likely to be toxic. Keywords Mutagenicity, Ames test, S. Typhimurium YG1042, oxidative DNA damage, 8-OH-Gua Introduction Genotoxicity of carcinogens is considered one of the most important contributors to cancer and a marker of the cancer risk of various populations. There are a number of short-term genotoxicity tests such as the Ames test, chromosomal aberration test and micronuclei test, etc. The Ames test is widely used as a simple rapid in vitro method for detecting the mutagenicity of a variety of genotoxic chemicals. The Ames test, which is also used for predicting possible carcinogenicity, has been sufficiently developed and seriously validated for widespread use. There is a high correlation between mutagenicity in the Ames test and carcinogenicity in the animal tests (Sugimura 1986). 8-OH-Gua is one of the most popular markers for the evaluation of oxidative DNA damage and oxidative stress. Exposure to a large number of genotoxic chemicals including - irradiation, 4-nitroquinoline-1-oxide and asbestos causes formation of 8-OH-Gua in their respective target organs (Floyd 1990). Herein we report the genotoxicity of 255 chemicals performed by two kinds of experiments; the Ames test using new strains and 8-OH-Gua assay using rat hepatocytes. We suggest that the Ames test using YG1041 and/or YG1042 strain is quite useful for the efficient detection of genotoxic chemicals in the environment. Materials and methods 255 chemicals of a wide variety of chemical types were purchased from Wako Pure Chemical Co. Ltd. (Osaka, Japan) and 10 complex mixtures such as condensates of river water were collected in Kyushu, Japan. The 255 chemicals tested were 65 pesticides (26 insecticides, 16 herbicides and 14 fumigants, etc.), 168 industrial chemicals (104 aromatic hydrocarbons, 60 allyl hydrocarbons, etc.) and others (natural toxins, etc.). These samples were dissolved in dimethylsulfoxide (DMSO) or sterilized distilled water at sufficient concentration to react by dose-response curves. Mutagenicity was assayed by the standard Ames test (Maron et al. 1983) with a modification of pre- incubation (37°C, 20min), in the presence or absence of a mammalian metabolic activation system (S9mix). The newly constructed Salmonella tester strains used were YG1041 (hisD3052/pKM101/pYG233) and YG1042 (hisG46/pKM101/pYG233) strain (nitroreductase and O-acetyltransferase-overproducing) sensitive - 26 - to nitroarenes and/or aromatic amines, YG3003 (hisD(G)8476/rfa/pAQ1/pKM101/mutM ST ::Km r ) strain (deficient in 8-hydroxyguanine DNA glycosylase) sensitive to oxidative chemicals, and YG7108 (hisG46/rfa/d uvrB/ d ada ST ::Km r / d ogt tST ::Cm r ) strain (deficient in O 6 -methylguanine DNA methyltransferase) sensitive to alkylating chemicals. These strains were kindly supplied by Dr. T. Nohmi of the National Institute of Health Sciences. S9 fraction of phenobarbital/5,6-benzoflabone-pretreated male Sprague-Dawley (SD) rat liver and human liver were purchased from Oriental Yeast Co. Ltd. (Tokyo, Japan) and Funakoshi Co. Ltd. (Tokyo, Japan), respectively. Fresh rat hepatocytes were isolated from SD rats (8 weeks) and they were plated on a collagen substratum of 12 well plates with Dulbecco’s Modified Eagle Media (Gibco Co. Ltd.) supplemented with fetal calf serum (5%), insulin (6.25g/ml), penicillin (50U/ml), streptomycin (50g/ml) and dexamethasone (1M) at a density of 3x10 3 cells/viable cells/dish (Delclos et al. 1987, Nakae et al. 1995). 8-OH-Gua level in rat hepatocytes DNA treated in an in vitro reaction with chemicals for 24hrs was measured by high performance liquid chromatograph with an electrochemical detector (HPLC-ECD), and/or by a competitive ELISA system using a monoclonal antibody specific for 8-OH-Gua. Results and discussion Nitroreductase (NR) and O-acetyltransferase (OAT) are enzymes involved in the intracellular metabolic activation of aromatic hydrocarbons, etc. in S. Typhimurium TA strains. YG1041 and YG1042 have 50-100 times respectively higher levels of both NR and OAT enzyme activity than their host strains TA98 and TA100 (Hagiwara et al. 1993). We compared the mutagenic activity (shown as the number of revertants per nanomolecule of chemicals, rev./nmol) of YG strains with that of the conventional TA strains as shown in Fig. 1. YG1041 showed about 40-1,800 times higher mutagenic activity toward 2-nitrofuorene, 1- nitropyrene and 1,8-dinitropyrene than did TA98. YG1042 showed a 20-70 times stronger mutagenic response to 2-nitronaphthalene, 2,4-dinitroluene and 2-aminoanthracene than TA100. We tested the 1 10 100 1000 10000 100000 1000000 10000000 Chemicals Fig. 1 Comparison of mutagenicity in Ames test using S .Typhimurium YG starins (1041, 1042) and their host TA strains (98, 100) TA strain YG strain 2-Nitrofluoranthene, 1-nitropyrene and 1,8-dinitropyrene are tested without S9mix, Glu-P-1 and Trp-P-2 are tested with S9mix using TA98 and YG1041. 2-Nitronaphthalene, 2,4-dinitrotoluen and 4-nitroquinoline N -oxide (4-NQO) are tested without S9mix, 2-aminoanthracene and benzo[a]pyrene are tested with S9mix using TA100 and YG1042. - 27 - mutagenicity of 255 chemicals and listed the top twenty chemicals as shown in Table1. It turned out that YG1041 and/or YG1042 were more sensitive to the mutagenicity of chemicals classified as nitroarenes, aromatic amines and heterocyclic amines. In addition, free radicals may be the most important genotoxic agents contributing to age-related diseases including cancer, and produce various kinds of oxidative DNA damage; nearly 100 products (Floyd 1990). Among them, 8-OH-Gua is a major mutation-prone (G:C to T:A transversion) DNA base-modified product. Alkylating chemicals, as well as free radicals, are regarded as endogenous genotoxic chemicals which might be involved in colon cancer because it is suggested that colon cells may be frequently exposed to them (Sugimura 1986). We have chosen two other sensitive strains (Suzuki et al. 1997, Yamada et al. 1995) in order to assess the mutagenicity of 255 chemicals. YG3003 was used to aid the identification of a variety of mutagenic free radicals including quinones, nitro and chloro derivatives, while YG7108 was used to aid the identification of mutagenic alkylating chemicals such as N- methyl-N’-nitro-N-nitrosoguanidine (MNNG). Of the 255 chemicals tested, 8 chemicals (3.1%) were mutagenic in YG3003 and 12 chemicals (4.7%) were mutagenic in YG7108 presented in Table 2. In order to predict possible carcinogenicity to humans, the mutagenicity of 10 chemicals, which are well recognized mutagens requiring a metabolic activation system (S9mix), were compared using rat liver S9 fraction and human liver S9 fraction. The mutagenicity by each S9mix makes a small difference and is almost equivalent as shown in Fig. 2. In particular, relatively higher mutagenicity was observed for 2- aminoanthracene, MeIQx and 3-methylcholathrene not only by rat liver S9 fraction but also by human liver S9 fraction. These results suggest that most mutagens could be activated to the ultimate forms in the presence of a mammalian metabolic activation system, such as human liver S9 fraction as well as rat liver S9 fraction. This result suggests that human liver S9 fraction in the Ames test is useful for evaluation of genotoxicity to human. Reactive oxygen species and free radicals, such as superoxide anion radical (O 2 - ), hydrogen peroxide (H 2 O 2 ), hydroxy radical ( . OH) and singlet oxygen ( 1 O 2 ), have been well correlated with the genotoxicity of chemicals. 8-OH-Gua is one of the most critical lesions generated from deoxyguanine (dG) in oxidized DNA which was observed after reaction of DNA, protein and lipids both in vitro and in vivo with genotoxic chemicals that generate free radicals. 8-OH-Gua is related to the initiation, promotion and progression of many diseases Table 1 The top 20 chemicals of mutagenicity in S . Typhimurium YG1042 with or without S9mix YG1042, +S9mix YG1042, -S9mix Rank- Chemicals Mutagenicity Rank- Chemicals Mutagenicity ing (rev./nmol) ing (rev./nmol) 1 Aflatoxin B1 408,000 1 1,8-Dinitropyrene 2,240,000 2 MeIQx 116,000 2 1,6-Dinitropyrene 1,870,000 3 PhIP 82,700 3 1-Nitropyrene 203,000 4 Trp-P-2 58,800 4 NIP 56,000 5 3-Methylcholanthrene 47,000 5 4-Nitroquinoline-N-oxide 51,100 6 NIP 33,600 6 Malathion 34,000 7 2-Aminoanthracene 26,600 7 MEP 20,500 8 Benzo[a]pyrene 17,500 8 Maneb 18,300 9 MEP 16,100 9 Aniline 17,200 10 Benzo[e]pyrene 15,700 10 N,N-Dimethylformamide 15,600 11 3-Nitrofluoranthene 12,600 11 Methomyl 14,400 12 Benzo[b]fluoranthene 9,720 12 Trifluralin 12,700 13 EPN 9,230 13 1,4-Dioxane 12,100 14 Trifluralin 9,080 14 Benzophenone 11,800 15 4-Chloronitrobenzene 6,910 15 EPN 9,920 16 Bifenox 6,910 16 Pyrene 9,180 17 1,2,5,6-Dibenzanthracene 6,180 17 2-Aminoanthracene 7,020 18 Pyrene 4,280 18 Captans 6,120 19 2-Phenylene diamine 4,250 19 Bifenox 4,870 20 1,2-Dibromo-3-chloropropane 3,910 20 N-Nitrosodiphenylamine 4,070 - 28 - such as cancer, tissue injury and inflammation (Floyd 1990). 8-OH-Gua has been commonly detected and identified by one of the already-existing methods; HPLC-ECD method (Helbock er al. 1998, Shigenaga et al. 1994). Despite its popularity, there have been doubts regarding its accuracy (possibility of artefactual production during sample preparation processes), expensive running cost and discrepancy in 8- OH-Gua level detected using different methods (gas chromatograph-mass spectrometry, 32 P postlabeling method, etc.). Therefore, it is important to develop a simple rapid in vitro method for evaluating the generation of reactive oxygen species from a variety of genotoxic chemicals. We developed a competitive ELISA method (Musarrat et al. 1994) and analyzed 8-OH-Gua level (expressed as the ratio of 8-OH-Gua per YG1041 (Rat S9) YG1041 (Human S9) YG1042 (Rat S9) YG1042 (Human S9) 0 50000 100000 150000 200000 250000 300000 350000 400000 450000 Chemicals Fig. 2 Comparison of mutagenicity of chemicals by rat and human liver S9mix Table 2 Mutagenicity in Ames test using S .Typhimurium YG strains (YG3003, YG7108) and their host strains (TA1535, TA102) Chemicals Muta genicity (rev./nmol) TA1535 TA102 YG3003 YG7108 Oxidative chemicals Meth ylene blue + visible light 0.51 (1.0) 8.5 (1.7) 26 (5.1) NT Neutral red + visible li ght 45 (1.0) 89 (2.0) 86 (1.9) NT 4-Nitro quinoline-N -oxide 22 (1.0) 128 (5.8) 132 (6.0) NT 2-Nitrofluorene 0.19 (1.0) 0.89 (4.7) 1.1 (5.8) NT H ydrogen peroxide ND 0.077 (1.0) 0.16 (2.1) NT Phenazine methosulfate ND ND ND NT Alk ylating chemicals MNNG 148 (1.0) NT NT 912 ( 6.1) (N -methyl-N '-nitro-N -nitrosoguanidine) ENNG ND NT NT 10,521 (N -ethyl-N '-nitro-N -nitrosoguanidine) PNNG 63 (1.0) NT NT 2,410 (38.3) (N -propyl-N '-nitro-N -nitrosoguanidine) BNNG 32 (1.0) NT NT 2,106 (65.8) (N -butyl-N '-nitro-N -nitrosoguanidine) - 29 - 10 5 dG per nanomolecule (nmol) of chemicals) induced by in vitro incubation of rat hepatocytes with 255 chemicals. The results suggest that 8-OH-Gua level was relatively correlated with mutagenicity for YG3003, as shown in Fig. 3. In particular, 8-OH-Gua level by incubation of rat hepatocytes with 4-NQO (0.86–2.8 8- OH-Gua/10 5 dG/nmol) was drastically increased in a dose-dependent curve using log transformation at the dose of 0.001-0.1mM (HPLC-ECD method) and 0.01-10mM (ELISA method). Fig. 3 Relationship between mutagenicity in YG3003 and 8-OH-Gua level in rat hepatocytes induced by chemicals 0 1 2 3 0.1 110100 1000 Mutagenicity (rev./nmol) Chemicals tested are Aflatoxin B1, 2-aminoanthracene, 2-aminoanthraquinone, 3-nitrofluoranthene, 1-nitropyrene, 4-nitroquinoline N-oxide, Cumene, dibenzyl ether, diethyl benzene mixture, formaldehyde, MEP, methoxychlor, paraquat, pentachlorophenol, potassium dichromate, simetryne, thiobencarb, TPN, 1,2,4-trichlorobenzene, 2,4,5-trichlorophenol and triphenyltin (IV) chloride. 4-NQO Fig. 4 Summary of 255 chemicals for mutagenicity in Ames test and oxidative DNA damage in 8-OH-Gua assay Ames test (83 chemicals) 8-OH-Gua assay (21 chemicals) Not detected (151 chemicals ) YG1041 (21 chemicals) YG1042 (51 chemicals) YG3003 (8 chemicals) YG7108 (12 chemicals) Fig. 3 Relationship between mutagenicity in YG3003 and 8-OH-Gua level in rat hepatocytes induced by chemicals 0 1 2 3 0.1 110100 1000 Mutagenicity (rev./nmol) Chemicals tested are Aflatoxin B1, 2-aminoanthracene, 2-aminoanthraquinone, 3-nitrofluoranthene, 1-nitropyrene, 4-nitroquinoline N-oxide, Cumene, dibenzyl ether, diethyl benzene mixture, formaldehyde, MEP, methoxychlor, paraquat, pentachlorophenol, potassium dichromate, simetryne, thiobencarb, TPN, 1,2,4-trichlorobenzene, 2,4,5-trichlorophenol and triphenyltin (IV) chloride. 4-NQO - 30 - Conclusions Of 255 chemicals tested, 83 chemicals (32.5%) were mutagenic in the Ames test using YG1041, YG1042, YG3003 and/or YG7108 strain, and 21 chemicals (8.2%) formed 8-OH-Gua in the rat hepatocytes oxidized DNA as shown in Fig. 4. These sensitive strains could detect a variety of mutagens that were not detected in the standard tester strains. They are suitable for mutagenicity of aromatic hydrocarbons, allyl hydrocarbons and insecticides, but not suitable for that of herbicides, fumigants, heavy metals, inorganic chemicals and natural toxins, etc. These results may suggest that the utility and limitations of both the Ames test and 8-OH- Gua assay in detecting chemicals likely to be environmental mutagens. We recommend the Ames test using these sensitive strains (especially YG1041 and/or YG1042) in combination with the standard tester strains (TA98 and TA100 series), in order to detect the mutagenicity of environmental chemicals efficiently, in extracts of river water, exhaust gases and urban atmosphere. References Delclos, K.B., Miller, D.W., Lay, J.O.Jr., Casciano, D.A., Walker, R.P., Fu, P.P. and Kadluber, F.F. (1987). Identification of C8-modified deoxyinosine and N2- and C8-modified deoxyguanosine as major products of the in vivo reaction of N-hydroxy-6-aminochrysene with DNA and the formation of the adducts in isolated rat hepatocytes treated with 6-nitrochrysene, Carcinogenesis, 8, 1703-1709. Floyd, R.A. (1990) The role of 8-hdroxyguanine in carcinogenesis, Carcinogenesis, 11, 1447-1450. Hagiwara, Y., Watanabe, M., Oda, Y., Sofuni, T. and Nohmi, T. (1993). Specificity and sensitivity of Salmonella typhimurium YG1041 and YG1042 strains possessing elevated levels of both nitroreductase and acetyltransferase, Mutat. Res., 261, 171-180. Helbock, H.J., Beckman, K.B., Shigenaga, M.K., Walter, P.B., Woodall, A.A., Yeo, H.C. and Ames, B.N. (1998). DNA oxidation matters: the HPLC-electrochemical detection assay of 8-oxo- deoxyguanosine and 8-oxo-guanine, Proc. Natl. Acad. Sci. USA, 6, 288-293. Maron, D.M. and Ames, B.N. (1983). Revised methods for the Salmonella mutagenicity test. Mutat. Res., 113, 173-215. Musarrat, J. and Wani, A.A. (1994). Quantitative immunoanalysis of promutagenic 8-hydroxy-2’- deoxyguanosine in oxidized DNA. Carcinogenesis, 15, 2037-2043. Nakae, D., Misumoto, Y., Kobayashi, E., Noguchi, O. and Konishi, Y. (1995). Improved genomic/nucleic DNA extraction for 8-hydroxydeoxyguanosine analysis of small amounts of rat liver tissue, Cancer Lett., 97, 233-239. Shigenaga, M.K., Aboujaoude, E.N., Cheon, Q. and Ames, B.N. (1994). Assays of oxidative DNA damage biomarkers 8-oxo-2-deoxyguanosine and 8-oxoguanine in nuclear DNA and biological fluides by high-performance liquid chromatography with electrochemical detection. Methods Enzymol., 234, 16-33. Sugimura, T., (1986) Studies on environmental chemical carcinogens in Japan, Science, 233, 312-318. Suzuki, M., Matsui, K., Yamada, M., Kasai, H., Sofuni, T. and Nohmi, T. (1997). Construction of mutants of Salmonella typhimurium deficient in 8-hydroxyguanine DNA glycosylase and their sensitivities to oxidative mutagens and nitro compounds, Mutat. Res., 393, 233-246. Yamada, M., Sedgwick, B., Sofuni, T. and Nohni T. (1995). Construction and characterization of mutants of Salmonella typhimurium deficient in DNA repair of O 6 -methylguanine, J. Bacteriology, 177(6), 1511-1519. Acknowledgements This research was supported by the research project ‘The research on the development of the total evaluation technique for the hazardous impacts by the chemical substances towards human and ecology (the project leader is Prof. Hideo Utsumi, Graduate School of Pharmaceutical Sciences, Kyushu University)’. This research project was supported by Fundamental Research Fund for the Environmental Future from Environmental Agency, Government of Japan. . 9,230 13 1, 4-Dioxane 12 ,10 0 14 Trifluralin 9,080 14 Benzophenone 11 ,800 15 4-Chloronitrobenzene 6, 910 15 EPN 9,920 16 Bifenox 6, 910 16 Pyrene 9 ,18 0 17 1, 2,5,6-Dibenzanthracene. Benzo[e]pyrene 15 ,700 10 N,N-Dimethylformamide 15 ,600 11 3-Nitrofluoranthene 12 ,600 11 Methomyl 14 ,400 12 Benzo[b]fluoranthene 9,720 12 Trifluralin 12 ,700 13 EPN