Mutation Research, 242 (1990) 209-217

209

Elsevier MUTGEN 01598

Study on mutagenicity and antimutagenicity of BHT and its derivatives in a bacterial assay Yoshihiko Yoshida Osaka Prefectural Institute of Public Health, Nakamichi-l, Higashinari-ku, Osaka 537 (Japan) (Received 14 August 1989) (Revision received 7 May 1990) (Accepted 15 May 1990)

Keywords: Antimutagenicity; Salmonella typhymurium; Escherichia coli; Reverse mutation assay; Butylated hydroxytoluene and derivatives

Summary The mutagenicity of butylated hydroxytoluene (BHT) and its derivatives was investigated by the Ames method using Salmonella typhimurium TA98 and TA100 with or without $9 mix. The compounds were not mutagenic in either indicator strain at concentrations ranging from 50 to 330 /xg/plate (SQ: 3,5,3',5'tetra-tert-butylstilbenequinone; VI-III: unidentified), 500 #g/plate (BE: 3,5,3",5'-tetra-tert-4,4'-dihydroxy1,2-diphenylethylene; VI: 2,6-di-tert-butyl-4-methyl-4-tert-butylperoxy-2,5-cyclohexadienone; VI-I: unidentified; VI-II: 3-acetyl-2,5-di-tert-cyclopenta-2,4-dienone) and 1000 #g/plate (BHT). The antimutagenic effects of BHT and its derivatives on mutagenesis by chemical agents were investigated in Salmonella typhimurium TA98 and TA100 and Escherichia coli WP-2 hcr-. VI-II suppressed the mutagenesis induced in TA98 and TA100 by 2-(2-furyl)-3-(5-nitro-2-furyl) acrylamide (AF-2) and that induced in WP-2 hcr- by 4-nitroquinoline-l-oxide (4NQO) without decreasing cell viability. In WP-2 hcr-, the mutagenesis induced by AF-2 and ethyl methanesulfonate was also suppressed significantly. Mutations induced by methyl methanesulfonate were slightly inhibited. However, VI-II had no effect on the mutagenesis induced by N-methyl-N'-nitro-N-nitrosoguanidine.

Butylated hydroxytoluene (BHT) is a component of commercially available antioxidants, such as butylated hydroxyanisole (BHA), so that BHT has been utilized in foods, plastics, rubber and petroleum for preservation. Recently, the use of BHT and similar compounds has tended to de-

Correspondence: Dr. Y. Yoshida, Osaka Prefectural Institute of Public Health, Nakamichi-1, Higashinari-ku, Osaka 537 (Japan).

crease, because synthetic chemicals used as food additives have been refused by consumers, and natural substances have been favored. Moreover, BHA is suspected of being carcinogenic. On the other hand, BHT has been said to be anticarcinogenic and a suppressor of lipid peroxide, which, according to epidemiological evidence, causes cancer in humans. So BHT, as an antioxidant, is available to protect humans from cancer. We studied BHT and its derivatives to obtain some data about the mutagenicity and antimutagenicity of these substances.

0165-1218/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)

210 Materials and methods

Chemicals BHT was purchased from Katayama Chemical Co. (Osaka). Dimethyl sulfoxide (DMSO), 2-(2furyl)-3-(5-nitro-2-furyl) acrylamide (AF-2), 4nitroquinoline-l-oxide (4NQO), and methyl methanesulfonate (MMS) were from Tokyo Kasei Kogyo Co. (Tokyo). Ethyl methanesulfonate (EMS) was from Kanto Chemical Co. (Tokyo). Benzo[a]pyrene (B[a]P) was from ICN K & K Laboratories Inc. (New York) and N-methyl-N'nitro-N-nitrosoguanidine ( M N N G ) from Aldrich Chemical Co. (Milwaukee). Rat liver $9, flN A D H , fl-NADPH, and G-6-P were from Oriental Yeast Co. (Tokyo). Other chemicals were supplied by Wako Pure Chemical Industries (Osaka). Silica Gel 60 and Silica Gel 60 F254, both 2 mm thick, TLC plates were purchased from E. Merck (F.R.G.). Co(II) acetylacetone, used in this study, was prepared by the method of Charles and Pawlikowski (1958): 16 g CoC12 was dissolved with 125 ml water and added to 50 ml methyl alcohol with 20.5 g acetylacetone. This mixture was heated briefly on a hot plate and cooled to room temperature, and then placed in a refrigerator for several hours to crystallize. The solid was filtered off a Buchner funnel, washed with water and dried in a vacuum desiccator. Preparation of B H T derivatives All the B H T derivatives used here have been described in the literature. However, in some instances it was found to be desirable to modify the literature method of preparation or replace it by a more convenient procedure. BE and S Q were prepared according to Cook et al. (1955): a solution of 2 g BHT in 20 ml benzene was oxidized with 17 g KaFe(CN) 6 and 3 g K O H in 30 ml water. After 3 h stirring at 60 ° C, the layer was separated, and the volume of benzene was reduced. The benzene solution was applied on 2 mm thick Silica Gel 60 F254 TLC plates, and developed in CC14. The separated red and ultraviolet absorptive bands were stripped off the T L C plate individually, and extracted by acetone, and then the acetone was removed. The red-band substances were recrystallized from ethyl alcohol

and the ultraviolet absorptive white substances from acetic acid. The former is SQ, m.p. 314315°C (reported 3 1 4 - 3 1 5 ° C ) and the latter is BE, m.p. 169-170 o C (reported 169-170 ° C). Vl and Vl-II were prepared according to Lerchovh and Pospisil (1977). To 3 g B H T in 20 ml benzene in a water bath at 80 ° C, 4 ml of 90% tert-butylhydroperoxide/benzene solution was added with stirring, then 1 ml 2% benzene solution of Co(II) acetylacetone was added. After 1 and 4 h, a 1-ml aliquot of this catalyst was added again. The reaction mixture, strongly frothing, was stirred for 8 h at 80 o C. The mixture was filtered, and the benzene and the transformation products of tertbutylhydroperoxide were removed by evaporation in vacuum at 60-80 ° C. The residue thus obtained was dissolved in methyl alcohol and repeatedly extracted with 15 ml of n-heptane. The n-heptane fractions were analyzed by TLC in the solvent systems n-heptane : ether (8 : 1) and CC14. According to TLC analysis, the same component fractions were combined, then the combined fractions were concentrated in a rotary evaporator; after that the residues thus obtained were dissolved in hot methyl alcohol, and left to crystallize in the cold. These recrystallizations were run 3 times. The obtained yellowish crystal (VI), m.p. 86.4°C (methyl alcohol) (reported 87 ° C), was identified using reported references from UV, IR, and FDMS spectra (m.w. 308). This crystal VI was exposed for 1 month to sunshine filtered through a window glass, and the phototransformation was followed by T L C analysis (n-heptane : ether, 8 : 1). This chromatographic separation was repeated several times. Preparative T L C (silica gel with a UV indicator, 254 nm) yielded VI-II as the main product of transformation, in addition to trace amounts of 2 other compounds, Vl-I and VI-Ill. VI-II was purified by crystallization from methyl alcohol with 10% water. This formed orange needles, m.p. 65 ° C (reported 64.5-65.5 ° C), and was identified from reported reference UV, IR and FD-MS spectra (m.w. 234). BHT and its derivatives are given as TLC chromatograms in Fig. 1. Methods Mutagenic activity assays in the Ames test. The mutagenicity in strains TA98 and TA100 of S.

211

005

0

°

©

@

0

Sample

Solvent

BHT

BE

0

SQ Vl V I - I

0

O

V I - I I VI-III

C CI 4

I

I

BHT

BE

I

SQ

I

I

Vl

VI-I

Hep : E t h e r / 8

I

I

Vl-II Vl-ll[

:1

Fig. 1. Thin-layer chromatogram of BHT and its derivatives.

typhimurium was investigated with the preincubation plate incorporation method (modified Ames test). The $9 mix contained 10% rat liver $9. AF-2 and B[a]P were used as positive controls. The mutagenic potency is expressed as the number of revertants in the corresponding amount of compounds. Reverse mutation assays. The compounds were examined with the mutant frequency test using S. typhimurium TA98 and TA100 and E. coli WP-2 h c r - ( K a t a et al.). 5 ml of fresh nutrient broth was inoculated with 0.5 ml of the bacterial overnight culture and incubated with shaking for 3 h at 37°C. The bacteria in log phase (about 2 × 109/ml) were washed twice with M / 1 5 phosphate buffer (pH 7.2). 0.5 ml of the mutagen dissolved in distilled water or D M S O was added to the bacterial suspension, and the bacteria were treated for 1 h at 37 ° C. After the treatment, the cells were washed twice with phosphate buffer by centrifugation and were resuspended in phosphate buffer. In using S. typhimurium TA98 and TA100, these assays were carried out in the same way as

the top agar system of the Ames mutagenicity test, and the tested sample was diluted to the same degree as with mutagenic activity assays. When E. coli was used, top agar consisting of 0.1 ml of the sample at various concentrations, 0.5 ml of M / 1 0 phosphate buffer ( p H 7.4), 2 ml of 0.65% molten agar, and 0.1-ml portions of the bacterial suspension were well mixed and poured onto SEM (semi-enriched minimal) agar medium plates, which consisted of V o g e l - B o n n e r E medium (2 g citric acid monohydrate, 10 g K 2 H P O 4, 3.5 g N a ( N H 4 ) H P O 4 . 4 H 2 0 , 0.2 g MgSO 4 • 7 H 2 0 per liter) supplemented with 0.4% glucose, 1.5% agar, and 5% ( v / v ) liquid nutrient broth (8 g Difco nutrient broth and 5 g NaC1 per liter). To determine cell viability, the suspension of mutagenized cells was diluted t o 10 - 6 , and a 0.1-ml portion was poured in the same way as described above. These procedures are shown in Scheme 1. After incubation for 2 days at 37 o C, revertants and viable cells were counted. All assays were performed 2 - 4 times and all values reported are the averages of triplicate plates.

212 0.5 ml Bacterial overnight culture 5 ml fresh nutrient broth 3 h 37 o C incubation wash twice suspend in M / 1 5 P buffer (pH 7.2) 0.5 ml mutagen solution (in H 2 0 or DMSO) 1 h 37 o C treatment wash twice resuspend in M / 1 5 P buffer (pH 7.2) (Bacterial suspension) Bacterial cell treatment with mutagens 0.1-ml sample at various concentrations 0.5 ml M / 1 0 P buffer (pH 7.4) 2 ml soft agar 0.1-ml portions (Bacterial suspension) spread on M M or SEM plates cells were counted after 47-h 37 o C incubation

0.1-ml sample at various concentrations 0.5 ml M / 1 0 P buffer (pH 7.4) 2 ml soft agar 0.1-ml aliquots of 106-fold diluted (Bacterial suspension) spread on M M or SEM plates cells were counted after 48-h 37 o C incubation

For the detection of Trp + revertants

Determination of the cell viability Scheme 1. Reverse mutation assays.

To examine the susceptibility of Trp + revertant colonies to VI-II, 4NQO-induced revertant colonies on an SEM plate without VI-II were picked at random and purified on tryptophan-free minimal agar plates by single-colony isolation. These purified Trp ÷ colonies were then cultured in liquid minimal medium for 1 day. After 106-fold dilutions, 0.1 ml of this Trp + bacterial suspension was well mixed with 0.5 ml phosphate buffer, 2 ml 0.6% molten top agar and 0.1 ml VI-II at various concentrations, and then poured onto the SEM plate. These plates were incubated at 37 ° C for 2 days. The number of Trp + colonies was counted. Results and discussion

It is thought that B H T and its derivatives are phenolic antioxidants like BHA and act through peroxide radical binding from free radical propagation by lipids. Radical-bound antioxidants are oxidized by dehydrogenation at the hydroxyl or methyl base. In the case of methyl-base dehydration, 2 dehydrated molecular antioxidants combine, become stable, and then stop the propagation reaction. These chemicals correspond to BE and SQ. In the case of the hydroxyl base, one

antioxidant binds 2 radicals, resonates, and then becomes stable. VI is of this sort. Photo-oxidation of VI produces VI-I, VI-II and VI-III. S Q and VI-III are poorly soluble in water, therefore when these were mixed with phosphate buffer in the top agar system, after some time a sediment formed. For that reason, those compounds were used at lower concentrations than the others. In the mutagen test, the preincubation method according to the modified Ames test was used. The results are given in Table 1. Each of the sample groups was examined at the same time, but all samples were not done simultaneously, so there are differences between solvent controls or positive (mutagenic) controls of the sample groups. In all cases, the n u m b e r of revertant colonies from these samples was less than twice the number of solvent controls (0 g g / p l a t e ) , and dose-response relationships were not observed between the sample concentrations and the numbers of revertant colonies. For this reason, these compounds are presumed to be non-mutagenic. The effects of B H T and its derivatives on AF-2-treated S. typhimurium TA98 and TA100 revertant colonies are shown in Tables 2 and 3. In this experiment, the concentrations of the samples

213 TABLE 1 MUTAGENIC EFFECTS OF BHT AND ITS DERIVATIVES IN Sample

Dose (/~g/plate)

S. typhimurium (colonies/plate)

TA98 - $9

+ $9

TA100 - $9

+ $9

0 100 330 1000

24 22 19 24

30 21 19 23

131 145 142 132

110 119 103 112

0

50 165 500

17 15 18 19

30 22 25 20

131 141 150 140

110 121 157 138

SQ

0 33 110 330

17 21 18 16

30 23 30 16

168 128 115 131

110 108 118 119

Vl

0 50 165 500

24 20 19 13

30 30 21 21

168 159 134 127

110 100 96 124

vI-I

0 50 165 500

24 20 24 14

30 18 17 16

147 171 164 115

110 132 118 117

Vl-ll

0 50 165 500

28 26 27 29

26 20 20 17

126 118 139 141

147 145 137 127

0 33 110 333

24 22 25 22

30 23 33 28

147 157 160 117

110 101 97 128

BHT

BE

j vI-Ill

Positive(negative)controls

AF-2 (0.1 ~g/plate)

B[a]P (5 ~g/plate)

AF-2 (0.01 ~g/plate)

B[a]P (5 ~g/plate)

480 (28) 529(24)

220 (30) 335 (26)

959 (131) 829 (168) 1039(147) 968 (126)

566 (110) 808 (147)

w e r e o f t h e s a m e level. I n T a b l e 2, S. typhimurium T A 9 8 w a s u s e d as t h e t e s t e r o r g a n i s m , a n d d e creases in the number of colonies were observed w i t h B H T , B E , V I a n d V I - I I . I n T a b l e 3, S. typhimurium T A 1 0 0 w a s u s e d , a n d t h e s a m e d e creases were observed with BHT, BE, VI and VI-II. Table 4 shows the effects of VI-II on muta-

tion induced

b y A F - 2 o r o t h e r m u t a g e n s i n S. 4NQO- or MNNG-induced mutagenesis was more significantly decreased by VI-II than AF-2-induced mutagenesis. VI-II at 200 # g / p l a t e c a u s e d a b o u t 90% r e d u c t i o n i n t h e n u m ber of revertants induced by 4NQO or MNNG. In this way, the numbers of His + revertant colonies

typhimurium.

214

in S. typhimurium TA98 and TA100 were estimated according to the mutagenic or bio-antimutagenic activity against B H T and its derivatives. In addition to the above examination, using E. coli WP-2 h c r - and SEM plates, we could investigate at the same time the number of mutant colonies in relation to the need for amino acids, while the range of sample concentrations, not decreased by the number of viable cells, was adopted to estimate the antimutagenic effects on mutants induced by the same mutagens. First of all, the effects of VI-II on mutants induced by some mutagens were tested (Fig. 2). Bacteria pretreated

TABLE 3 ANTIMUTAGENIC EFFECT OF BHT AND ITS DERIVATIVES ON AF-2 (0.1 #g/ml) IN S. typhimurium TA100 Sample

Dose (~g/plate)

Colonies/plate

BHT

0 100 330 1000

1121 1087 633 538

BE

0 50 165 500

1307 1286 1205 993

SQ

0 33 110 330

1326 1253 1254 1192

Vl

0 50 165 500

1326 1153 1162 996

VI-I

0 50 165 500

1013 1001 961 943

Vl-II

0 50 100 200 300 400 500

901 630 527 357 181 102 76

VI-III

0 33 110 333

979 989 942 947

TABLE 2 ANTIMUTAGENIC EFFECTS OF BHT AND ITS DERIVATIVES ON AF-2 (0.1 ttg/ml) IN S. typhimurium TA98 Sample

Dose (/~g/plate)

Colonies/plate

BHT

0 100 330 1000

56 44 37 40

BE

0 50 165 500

56 56 50 54

SQ

0 33 110 330

56 49 52 56

VI

0 50 165 500

51 45 36 31

VI-!

0 50 165 500

56 55 45 47

VI-II

0 50 165 50O

54 34 16 6

VI-III

0 33 110 333

53 52 49 49

w i t h A F - 2 (0.1 / t g / m l ) , 4 N Q O (1.2 m g / m l ) , M N N G

(2 / ~ g / m l ) , M M S

( 2 / ~ g / m l ) , E M S (4 m g / m l ) ,

or untreated (for spontaneous mutagenesis), and

the various concentrations of VI-ll with top agar were poured onto the SEM plates. Mutants induced by AF-2 were depressed about 805% on the SEM plate containing 500 /~g VI-II. 4NQO- and EMS-induced mutations were inhibited more: about 90% reduction was observed on the plate with 500 /~g VI-II. About 70% decrease in the number of revertants induced by MMS was observed on the SEM plate containing 500 /~g VI-II.

215

C

AF-2 ( 0 . 1 # g / m L )

DO0

M M S (1.2 mg

4NQO (l#g/mL)

:3 pO

40C

---x .....

x. . . . . . .

mL)

- ;.EX

200

~ 10(:

100 0

3

60(?

300

~0 40C

200

i

I 2OOI -

10(3

0 0

"E

100

200

300

400

t~, 0 (

500

I 100

I 200

I 300

I II 400 500

0

!

[ 100

I 200

I 300

b

I II 400 500

O

/

._~

F MNNG(2#g/mL)

:

EMS ( 4 m g / m L )

o

O

>~

~ x

v

soo-

.~

c

: "

-- .

.

-*x

/

-~3oo

-- 200

;K +D-

400~--

-,

.

-x

.

- 200

.

.- 1C

- - -,x - 1OO

--)

-0

.

.

.

--

Spontaneous mutagenesis

- 100

> I

I 0

I

300

200

100

0 ] 0

I 100

I J I II 200300400500

0 Concentration

I 100 of

I I I II 200300400500 V ! - [I

ol 0

I 100

I I I 20030040%0

I

(#g //plate )

Fig. 2. Effects of VI-II on mutation induction in E. coil WP-2 h e r - pretreated with AF-2 (0.1 # g / m l ) , 4NQO (1/zg/ml), MMS (1.2 mg/ml), M N N G (2/sg/ml), or EMS (4 m g / m l ) for 60 rain at 37 o C and spontaneous mutagenesis.

216 TABLE 4 A N T I M U T A G E N I C EFFECTS OF VI-ll ON VARIOUS M U T A G E N S IN S. typhimurium TA98

TA100

AF-2 ( ~ g / m l )

0

0.1

0.3

0.6

AF-2 ( ~ g / m l )

0

0.07

0.13

0.2

Vl-ll(~g/plate) 0 100

29 20

56 31

139 5

118 7

VI-II(~g/plate) 0 100

143 108

684 207

1155 458

1436 487

AF-2 (0.1)

4NQO (2)

MNNG (2)

901 630 527 357

1677 446 295 81

1039 361 266 105

Mutagens(~g/ml)

Vl-ll(~g/plate) 0 50 100 200

On the contrary, VI-II was not effective against M N N G mutagenesis. VI-I! at a concentration of 100 /~g/plate resuited in a significant reduction in the numbers of revertants induced by various concentrations of 4 N Q O (Table 5). The effects of B H T and its derivatives on 4 N Q O mutagenesis at 2 # g / m l and the cell viability are shown in Table 6. VI-II at concentrations increasing from 0 to 500/~g on SEM agar plates reduced the number of Trp + revertant colonies, whereas no decrease was found in cell viability. About 88% reduction in the number of revertants induced by 4 N Q O was observed on the SEM plate containing 500 /~g VI-II. In order to examine whether VI-II would inhibit the expression of Trp +, a reconstruction experiment was carried out (Table 7). The Trp ÷ colonies grew well even in the presence of VI-II. These results suggest that VI-II may activate the error-free recombinant repair by mod-

TABLE 6 A N T I M U T A G E N I C EFFECTS OF BHT A N D ITS DE RIVATIVES ON 4 N Q O (2/.tg/ml) IN E. coli WP-2 h c r Sample

Dose (#g/plate)

Trio ÷ revertants/plate

Viable cells ( × 106)/plate

BHT

0 100 330 1000

696 636 622 616

15 17 24 21

BE

0 50 165 500

696 708 700 712

15 20 17 17

SQ

0 33 110 330

696 780 720 694

15 13 20 21

VI

0 50 165 500

696 638 728 694

15 21 18 18

VI-I

0 50 165 500

696 636 732 660

15 15 25 21

VI-II

0 50 165 500

905 560 281 112

26 31 30 30

VI-III

0 33 110 333

696 666 668 578

15 28 21 29

TABLE 5 A N T I M U T A G E N I C EFFECTS OF V|-II ON 4NQO IN E. coil WP-2 h c r VI-II

4NQO ( ~ g / m l )

0

1

2

4

Trp + revertants/plate Viable cells ( x 106)/plate Trp + revertants/plate Viable cells ( x 106)/plate

19 142 16 143

653 170 192 190

897 53 329 56

301 3 130 4

0,g/ plate) 0 100

217

References

TABLE 7 SENSITIVITIES OF INDUCED Trp + CLONES OF E. coli WP-2 hcr- TO VI-II UNDER RECONSTRUCTED CONDITIONS Vi-II (~tg/plate)

Trp + revertants ( )< 106)/plate

Survival (%)

0 100 200 300 400 500

184 213 195 203 197 185

100.0 115.8 105.9 110.3 107.1 100.5

ifying some catalytic properties of the recA protein.

Conclusions In mutagenic tests, BHT and its derivatives have shown nonmutagenicity on the preincubational modified Ames assay. One of the BHT derivatives, VI-II (3-acetyl-2,5-di-tert-cyclopenta2,4-dienone), decreased the number of mutagenesis revertants induced by AF-2, 4NQO and M N N G in S. typhimurium TA100. It also exhibited bio-antimutagenic activity against mutagenesis in E. coli WP-2 hcr- by AF-2, 4NQO, EMS or MMS, but had no effect on MNNG-induced mutagenesis.

Acknowledgements The author wishes to thank Mr. Hirotaka Obana of this laboratory for helpful discussions and assistance. A part of this report was presented at the annual meeting of the Japanese Food Hygienic Society on May 18, 1988, in Yokohama.

Charles, R.G., and M.A. Pawlikowski (1958) Comparative heat stabilities of some metal acetylacetone chelates, J. Phys. Chem., 62, 440-444. Cook, C.D., N.G. Nash and H.R. Flanagan (1955) Oxidation of hindered phenols. Ill. The rearrangement of the 2,6-ditert-butyl-4-methyl phenoxy radical. J. Am. Chem. Soc., 77, 1783-1785. Hara, Y. (1988) Tennenbutu tyu no Koototuzeninshi no Kensaku to Dootei (Screening and determination of antimutagenesis on natural substances), Farumashia, 24, 265270. Ito, N., S. Fukushima, A. Hagiwara, M. Shibata and T. Ogiso (1988) Carcinogenicity of butylated hydroxyanisole in F344 rats, J. Natl. Cancer Inst., 70, 343-352. Lerchov~, J., and J. Pospisil (1977) Antioxidants and stabilizers. LVIII. Reaction of hydroperoxides modeling oxidized polyolefins with 2,6-di-tert-butyl-4-methyl phenol. Eur. Polymer J., 13, 975-979. Lerchovh, J., L. Kotul~, K.J. Rotshov~ and J. Pospisil (1976) Antioxidants and stabilizers. LXI. Photochemical behavior of transformation products of the phenolic antioxidant 4-tert-butylperoxy-2,5-cyclohexadienone. J. Polymer Sci. Symp. 57, 229-235. Matsushita, S. (1987) Problems of lipid oxidation in foods, Yukagaku, 36, 3-9. Ohta, T., T. Watanabe, M. Moriya and Y. Shirasu (1983) Antimutagenic effects of cinnamaldehyde on chemical mutagenesis in Escherichia coli, Mutation Res., 107, 219227. Sawatari, K., T. Nishino and Y. Tabata (1976) Handbook of Antioxidants, pp. 25-31. Shimoi, K., Y. Nakamura, I. Tomita and T. Kada (1985) Bio-antimutagenic effects of tannic acid on UV and chemically induced mutagenesis in Escherichia coli B/r, Mutation Res., 149, 17-23. Terao, T. (1988) Synthetic antioxidants. Nippon Nogei Kagaku Kalshi, 62, 174-177. Yahagi, T. (1975) Screening methods using microbes for environmental carcinogens, Tanpaku Kakusan Kooso, 20, 1178-1189.