Mutation Research, 244 (1990) 303-308 Elsevier
303
MUTLET 0379
Mutagenicity of commercial hair dyes and detection of 2,7-diaminophenazine Tetsushi W a t a n a b e , Teruhisa H i r a y a m a and Shozo Fukui Kyoto Pharmaceutical University, 5 Nakauchi-cho, Misasagi, Yamashina-ku, Kyoto 607 (Japan) (Accepted 19March 1990)
Keywords: Hair dyes; Salmonella/microsome test; Blue-rayon extraction; 2,7-Diaminophenazine; Phenylenediamine
Summary Four commercial oxidative-type hair dye formulations, A, B, C, and D, were treated with hydrogen peroxide (H202) to simulate normal conditions of use, and the oxidized hair dyes were tested for their mutagenicity in Salmonella typhimurium TA98 in the presence of a mammalian metabolic activation system ($9 mix). Most of them did not show obvious mutagenicity in the range of 1-25/A/plate and all exhibited bactericidal activity at 10 #I/plate. In order to evaluate the mutagenicity of hair dyes both before and after H202 oxidation, rayon linked to a copper-phthalocyanine derivative (blue rayon) was used as an adsorbent for the elimination of interfering bactericidal compounds. Adsorbed compounds on blue rayon were eluted with ammoniacal methanol and eluents were subjected to the Ames test. The mutagenicity of the blue-rayon extracts in TA98 with $9 mix was increased by H202 oxidation. The blue-rayon extracts obtained from oxidized A and B were potent mutagens and reverted 334 and 999 colonies/10/~1 of original substance, respectively. In addition, 88 and 249 ng of 2,7-diaminophenazine, which was extremely mutagenic in TA98 with $9 mix, were detected in the extracts of 40 ml of the hair dye formulations A and B, respectively. The mutagenicity in oxidized hair dye formulations was successfully detected by use of blue-rayon extraction. 2,7-Diaminophenazine was only formed in the hair dye formulations containing m-phenylenediamine by H202 oxidation. Therefore, attention needs to be paid to the use of m-phenylenediamine as a hair dye component, not only for its own toxicity but also for that of its oxidation products.
The mutagenicity of hair dye ingredients was first evaluated by Ames et al. (1975a) using the Salmonella/microsome test. Later some hair dye ingredients have been shown to be mutagenic in Correspondence: Dr. T. Watanabe, Kyoto Pharmaceutical University, 5 Nakauchi-cho, Misasagi, Yamashina-ku, Kyoto 607 (Japan).
yeast (Mayer and Goin, 1980) and in insects (Blijleven, 1977), to induce chromosomal damage in cultured Chinese hamster cells (Kirkland and Venitt, 1976) and to be carcinogenic in rodents (Reznik and Ward, 1979). p-Phenylenediamine (PD) is the most common aromatic amine used in the preparation of permanent-type or oxidative-type hair dyes and the
0165-7992/90/$ 03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
304
mutagenicity of p - P D has been examined by several investigators (Crebelli et al., 1981; Burnett et al., 1982; Nishi et al., 1982; Rojanapo et al., 1986). After oxidation with HaO2, p - P D became mutagenic to TA98 in the presence of $9 mix, and this mutagenic compound has been suggested to be the Bandrowski's base (Ames et al., 1975a). In our previous papers (Watanabe et al., 1987, 1989a, in press), we reported the mutagenicity of 13 kinds of o-, m- and p - P D derivatives, both before and after H202 treatment, in S. typhimurium TA98 with or without $9 mix. The mutagenicity of mPD, which is used as a hair dye component in Japan, was markedly enhanced by H202 oxidation and 2,7-diaminophenazine was detected as the major mutagenic oxidative compound of m-PD. In this paper, we investigated the detection of the mutagenicity of commercial hair dye formulations and the existence of aminophenazine in the oxidized mixtures. Materials and methods
Chemicals Four different commercial oxidative-type hair dye formulations, A, B, C and D, were purchased from local stores. The components of each hair dye formulation are shown in Table 1.2-Acetylaminofluorene (AAF), 4-nitroquinoline N-oxide (4NQO) and Alumina Activated 200 for column chromatography were purchased from Nakalai tesque, Kyoto (Japan). 30% hydrogen peroxide (H202), polychlorobiphenyl, penta(PCB, Kanechlor-500), and spectrophotometric-grade dimethyl sulfoxide (DMSO) were purchased from Wako Pure Chemicals, Osaka (Japan). 'Blue rayon' was purchased from Funakoshi Chemicals, Tokyo (Japan). 2,7-Diaminophenazine was prepared according to the method described in our previous paper (Watanabe et al., 1989b). Mutagenicity test The hair dye formulations (1 ml) were mixed with 6% H2Oz (1 ml) and stirred for 30 min at room temperature, following which treated hair dyes were diluted with sterile water (48 ml) and
TABLE 1 COMPONENTS OF THE TESTED HAIR DYE FORMULATIONS Sample
Components
A
2,5-diaminotoluene, m-aminophenol, mphenylenediamine, propylene glycol, ammonium thioglycollate,p-hydroxybenzoate,dibutyl hydroxytoluene,edetoate p-phenylenediamine, m-phenylenediamine, m-aminophenol, resorcinol, propylene glycol, thioglycolic acid, p-hydroxybenzoate,edetoate, polyoxyethylenelaurylethersulfate, cetanol, lanoline, benzoic acid, benzoate 2,5-diaminotoluene,p-aminophenol, m-aminophenol, o-aminophenol,p-amino-o-cresol, resorcinol, propylene glycol, ammonium thioglycollate, p-hydroxybenzoate,dibutyl hydroxytoluene, edetoate p-phenylenediamine,p-aminophenol, picramic acid, p-methylaminophenol,resorcinol, polyethyleneglycol, cetanol
neutralized by the addition of hydrochloric acid. Each hair dye solution plus 0.50 g of blue rayon was gently shaken for 10 min at room temperature. The rayon was taken out, and this treatment was repeated once again adding a new batch of blue rayon. After washing with water (3 times with 50 ml) and removal of moisture, the blue rayon was treated with 50 ml of methanol-28%0 NH4OH (50:1, v/v). The elution was repeated, and the combined eluate was evaporated to dryness under reduced pressure. Each residue was dissolved in 2 ml of DMSO and diluted with DMSO. 0.1 ml of each dilution, equivalent to 3 or l 0 / d of hair dye, was subjected to the mutagenicity test. Mutagenicity tests were performed essentially by the Ames method (Ames et al., 1975b; Maron and Ames, 1983) with the suspension assay modified by Yahagi 0975). The tester strain Salmonella typhimurium TA98 was checked routinely to confirm its genetic features for optimal response to known chemicals as follows: AAF (5/zg/plate), 4NQO (0.5 /~g/plate). The assay was performed in the absence
305
or presence of $9 mix to simulate mammalian metabolism. Every sample was tested on at least 3 separate occasions and every dose level was assayed in triplicate.
Metabolic activation system The postmitochondrial fraction, $9, was prepared from the livers of male Sprague-Dawley rats induced with PCB. The homogenate (3 ml of 0.15 M KC1/g liver) was centrifuged as described by Ames et al. (1975b). The $9 mix was prepared according to Yahagi (1975). I ml of $9 mix contained 8 /~mole MgClz, 33 /~mole KC1, 5 /zmole G6P, 4 #mole N A D P H , 0.5 units G6P, 150/zmole sodium phosphate buffer (pH 7.4) water and 50 ~tl of $9.
fraction corresponding to the retention time of authentic 2,7-diaminophenazine was collected and evaporated. The residue was dissolved in 2 ml of 20°7o CH3CN and 20 t~l of the solution was subjected to high-performance liquid chromatography (HPLC) using an LC-6A apparatus (Shimadzu Seisakusho, Kyoto) with a STR ODS-H column (5 #m particle size, 150 mm × 4 mm i.d., Shimadzu Techno Research, Inc., Kyoto). The material was eluted with 20°70 CH3CN at a flow rate of 0.7 ml/min by detecting the fluorescence with excitation wavelength at 490 nm and emission wavelength at 580 nm. Results and discussion
Determination of m-PD in hair dye formulations 100 mg of hair dye formulation was made up to 100 ml with ethyl acetate (AcOEt). 50 #1 of heptafluorobutyric anhydride was mixed with 2 ml of the AcOEt solution and the mixture was heated at 50°C for 30 rain. The reaction mixture was cooled to room temperature and subjected to gas chromatography-mass fragmentography (GC-MF). GCMF conditions were 2 m x 3 mm i.d. glass column packed with 3°70 silicone OV-330 on chromosorb W (AW-DMCS), injector temperature 230°C, oven temperature 160°C, helium gas flow 40 ml/min, ionization voltage 70 eV. The molecular ion (M ÷, m/z 500) was detected by a quadrupole mass spectrometer (Shimadzu QP-1000).
We have studied the mutagenicity of 4 hair dye formulations, A, B, C and D, which are used in Japan. In a preliminary experiment, the tests were carried out after a brief oxidation to simulate normal conditions of use. The hair dye formulations were mixed with 6o70 H202 (1:1, v/v) and allowed to stand for 30 min at room temperature. For the mutagenesis assay, the oxidized hair dyes were used at 1, 2.5, 10 and 25/zl/plate (Table 2). Oxidized A
Determination of 2, 7-diaminophenazine in 1-120:treated hair dyes
Sample
40 ml of hair dye formulation was mixed with 6o70 H2Oz (1:1, v/v) for 30 min at room temperature, and the oxidized mixture was diluted with 80 ml of water. The dilution was treated with 2 batches of blue rayon (2 g each) as described above. The compounds adsorbed on blue rayon were eluted 3 times with methanol-28% NH4OH (50:1, 200 ml each). The eluate was evaporated under reduced pressure and the residue was subjected to alumina column chromatography eluting with CHCI3, CHCI3-MeOH (19:1) and MeOH. The eluate was separated into several fractions, and the
TABLE 2 MUTAGENICITY OF O X I D I Z E D COMMERCIAL HAIR DYES IN S. typhimurium TA98 WITH OR W I T H O U T $9 MIX
A B C D
DMSO 4NQO AAF
Dose (#l/plate)
His + rev./plate - $9 mix
+ $9 mix
1 2.5 1 2.5 1 2.5 1 2.5
34 KL 22 KL 20 KI 36 KL
145 320 66 44 81 106 59 67
16 327
55
0.5/zg/plate 5/~g/plate
KL, killing effect.
736
306
induced 320 revertants at 2.5/~l/plate with $9 mix, but the other hair dyes did not show obvious mutagenicity at that dose and all hair dyes exhibited bactericidal activity at 10/A/plate. Hayatsu et al. (1983) have reported that cotton bearing covalently bound trisulfo-copper-phthalocyanine (blue cotton) can adsorb multicyclic mutagenic compounds in crude samples and the adsorbed mutagens can be recovered by elution with ammoniacal methanol. In addition, they could detect the mutagenicity in smokers' urine, cooked beef and river water with this method. Recently rayon, instead of cotton, bearing covalently linked trisulfo-copper-phthalocyanine residues were developed (blue rayon). For the elimination of interfering bactericidal compounds
TABLE 3 M U T A G E N I C I T Y OF BLUE-RAYON EXTRACTS FROM C O M M E R C I A L HAIR DYES WITH OR W I T H O U T H202 TREATMENT 1N S. typhimurium TA98 WITH OR W I T H O U T $9 MIX Sample
Dose (/~1 equivalent of hair dye/plate)
His + rev./plate - $9 mix
+ $9 mix
Not treated with 11202 A B
C D
3 10 3 10 3 10 3 10
NT 39 NT 19 NT 15 NT 17
45 138 24 45 27 37 30 29
NT 21 NT 22 NT 19 NT 20
110 334 327 999 34 94 41 59
Treated with 02112 A B
C D
DMSO 4NQO AAF
3 10 3 10 3 10 3 10
0.5 #g/plate 5 #g/plate
NT, not tested.
16 327
22 364
from oxidized hair dyes and the evaluation of the mutagenicity of oxidized hair dyes, blue rayon was used as an adsorbent. Table 3 shows the mutagenicity of blue-rayon extracts obtained from the 4 hair dye formulations in Salmonella typhimurium TA98 with or without $9 mix. The oxidative conditions of hair dyes and the extraction procedure with blue rayon have been described in Materials and methods. Of the 4 hair dye formulations tested, only A exhibited obvious mutagenicity without H 2 0 2 treatment. The mutagenic potency of all hair dye formulations tested in this study was enhanced by H202 treatment and their activity was observed in the presence of $9 mix. A marked enhancement of mutagenicity was observed for A and B, and bluerayon extracts obtained from oxidized A and B induced 334 and 999 revertants per l0/A of original substance, respectively. m-Phenylenediamine (PD), which was used as a coupler in the hair dye formulations, was determined as the heptafluorobutyryl derivative by GC-MF. The conditions have been described in Materials and methods. Hair dye formulations A and B contained 8 and 2.5 mg of m-PD per gram of hair dye, respectively. In our previous papers (Watanabe et al., 1989a,b), we reported that the mutagenicity of mPD was markedly increased by H202 oxidation, and 2,7-diaminophenazine was identified as the major mutagenic oxidative compound of m-PD. Since A and B contained m-PD as a hair dye component and their mutagenic potency was markedly increased by the oxidation, the formation of 2,7-diaminophenazine in the oxidative process was expected. We presumed that the yield of 2,7-diaminophenazine in oxidized hair dyes was small because of the low m-PD content in hair dye formulations. The blue-rayon extracts were subjected to further treatment with alumina column chromatography to remove contaminants. In a preliminary experiment, recoveries of 2,7-diaminophenazine from a 3°7o H 2 0 2 solution by blue-rayon and alumina column treatment were 85.0 and 90.7%, respectively, and those are satisfactory levels. In order to confirm the existence of 2,7-diamino-
307
phenazine in oxidized hair dyes, 40 ml of hair dye formulations, which is the amount most commonly used for hair dyeing, was oxidized, diluted and treated with blue rayon. The blue-rayon extracts were column-chromatographed on alumina and the fractions corresponding to 2,7-diaminophenazine were subjected to HPLC on an ODS column. As shown in Fig. 1, a peak was detected at a retention time of 3.9 min, which coincided with that of authentic 2,7-diaminophenazine, on each chromatogram of hair dye extracts A and B by monitoring fluorescence. In addition, the maximal excitation and emission wavelengths of the peaks were the same as those of the authentic sample. The yields of 2,7-diaminophenazine detected on the ODS column were calculated, with the aid of the standard curve of authentic 2,7-diaminophenazine, to be 88 ng (A) and 249 ng (B) in 40 ml of hair dye formulation. In non-oxidized hair dye formulations A and B, 2,7-diaminophenazine could not be detected under similar conditions described in Materials and methods (data not shown). In the model system, m-PD (5 mg/ml in water) was treated with 6°7o H 2 0 2 (1:1, v/v) in the same manner as the hair dye formulations used in this study. 16.2 #g of 2,7-diaminophenazine was then formed from 0.2 g of m-PD, which amount was almost comparable to that of m-PD in 40 ml of hair
dye formulation, after H 2 0 2 treatment for 30 min. Thus the yield of 2,7-diaminophenazine in the model system was more than 50 times larger than that in the oxidized hair dyes. For the reason above, it was considered that in hair dye formulations m-PD was a minor component, and most m-PD was consumed by binding with other abundant hair dye components such as p-phenylenediamine derivatives, mono- or polyphenol, aminophenol, etc. Corbett (1969) reported that oxidation of mixture of p-aminophenol and m-PD gave 2,4-diamino-4'-hydroxyldiphenylamine. In addition, he reported that m-PD reacted with p-PD under oxidative conditions, and it gave 2,4,4'-triaminodiphenylamine (Corbett, 1969, 1973). In the present study, the mutagenicity in hair dye formulations was successfully detected with the use of blue-rayon extraction. 2,7-Diaminophenazine, which was extremely mutagenic and reverted 730 colonies/ng in TA98 with $9 mix, was apparently detected in the blue-rayon extracts obtained from oxidized A and B. Since m-PD was capable of forming 2,7-diaminophenazine in the oxidation process of hair dye formulations, the use of m-PD as a hair dye component should receive attention for the toxicity of not only itself but also its oxidation products.
A
C
D
c
DAPz o ,:j
!
DAPz
I 0
I 4
I 8
I 12
I 16
I 0
I 4
I 8
I 12
Retention
I 0 time
I 4
I 8
I 12
I 16
I
I
I
I
0
4
8
12
(rain)
Fig. 1. HPLC chromatograms of blue-rayon extracts obtained from oxidized hair dye A (A), B (B), C (C) and D (D) by detecting fluorescence. DAPz, 2,7-diaminophenazine.
308
Acknowledgements The authors are grateful to Prof. B.N. Ames, University of California, Berkeley, CA (U.S.A.) who kindly supplied Salmonella typhimurium TA98. We also wish to thank Mr. T. Nishigaki and Mr. Y. Tada for their excellent technical assistance in this investigation.
References Ames, B.N., H.O. Kammen and E. Yamasaki (1975a) Hair dyes are mutagenic: identification of a variety of mutagenic ingredients, Proc. Natl. Acad. Sci. (U.S.A.), 72, 2433-2437. Ames, B.N., J. McCann and E. Yamasaki (1975b) Methods for detecting carcinogens and mutagens with the Salmonella/ mammalian-microsome mutagenicity test, Mutation Res., 31, 347-364. Blijleven, W.G.H. (1977) Mutagenicity of four hair dyes in Drosophila melanogaster, Mutation Res., 48, 181-185. Burnett, C., C. Fuchs, J. Corbett and J. Menkart (1982) The effect of dimethyl sulfoxide on the mutagenicity of the hair dye p-phenylenediamine, Mutation Res., 103, 1-4. Corbett, J.F. (1969) Benzoquinone imines. Part VI. Mechanism and kinetics of the reaction ofp-benzoquinone di-imines with m-phenylenediamines, J. Chem. Soc. (B), 827-835. Corben, J.F. (1973) Role of m-difunctional benzene derivatives in oxidative hair dyeing, I. Reaction with p-diamines, J. Soc. Cosmet. Chem., 24, 103-134. Crebelli, R., L. Conti, A. Carere and R. Zito (1981) Mutagenicity of commercial p-phenylenediamine and of an oxidation mixture of p-phenylenediamine and resorcinol in Salmonella typhimurium TA98, Food Cosmet. Toxicol., 19, 79-84. Hayatsu, H., T. Oka, A. Wakata, Y. Ohara, T. Hayatsu, H. Kobayashi and S. Arimoto (1983) Adsorption of mutagens to
cotton bearing covalently bound trisulfo-copper-phthalocyanine, Mutation Res., 119, 233-238. Kirkland, D.J., and S. Venitt (1976) Cytotoxicity of hair colourant constituents: chromosome damage induced by two nitrophenylenediamines in cultured Chinese hamster cells, Mutation Res., 40, 47. Maron, D.M., and B.N. Ames (1983) Revised methods for the Salmonella mutagenicity test, Mutation Res., 113, 173-215. Mayer, V.W., and C.J. Goin (1980) Induction of mitotic recombination by certain hair-dye chemicals in Saccharomyces cerevisiae, Mutation Res., 78,243-252. Nishi, K., and H. Nishioka (1982) Light induces mutagenicity of hair-dye p-phenylenediamine, Mutation Res., 104, 347-350. Reznik, G., and J.M. Ward (1979) Carcinogenicity of the hairdye component 2-nitro-p-phenylenediamine: induction of eosinophilic hepatocellular neoplasms in female B6C3F1 mice, Food Cosmet. Toxicol., 17, 493-500. Rojanapo, W., P. Kupradinun, A. Tepsuwan, S. Chutimatae and M. Tanyakaset (1986) Carcinogenicity of an oxidation product of p-phenylendiamine, Carcinogenesis, 7, 1997 -2002. Watanabe, T., M. Ono, T. Hirayama and S. Fukui (1987) 2,7-Diamino-3,8-dimethylphenazine as the major mutagenic product from the reaction of 2,4-diaminotoluene with hydrogen peroxide, Mutation Res., 190, 113-117. Watanabe, T., T. Hirayama and S. Fukui (1989a) Phenazine derivatives as the mutagenic reaction product from o- or mphenylenediamine derivatives with hydrogen peroxide, Mutation Res., 227, 135-145. Watanabe, T., Y. Hanasaki, T. Hirayama and S. Fukui (1989b) Mutagenicity of nitro- and amino-substituted phenazines in Salmonella typhimurium, Mutation Res., 225, 75-82. Watanabe, T., T. Hirayama and S. Fukui (1990) The mutagenic modulating effect of p-phenylenediamine on the oxidation of o- or m-phenylenediamine with hydrogen peroxide in Salmonella test, Mutation Res., in press. Yahagi, T. (1975) Screening methods using microbes for the environmental carcinogen, Tanpakushitsu Kakusan Koso (Protein, Nucleic Acid and Enzyme), 20, 1178-1189. Communicated by H.S. Rosenkranz
303
MUTLET 0379
Mutagenicity of commercial hair dyes and detection of 2,7-diaminophenazine Tetsushi W a t a n a b e , Teruhisa H i r a y a m a and Shozo Fukui Kyoto Pharmaceutical University, 5 Nakauchi-cho, Misasagi, Yamashina-ku, Kyoto 607 (Japan) (Accepted 19March 1990)
Keywords: Hair dyes; Salmonella/microsome test; Blue-rayon extraction; 2,7-Diaminophenazine; Phenylenediamine
Summary Four commercial oxidative-type hair dye formulations, A, B, C, and D, were treated with hydrogen peroxide (H202) to simulate normal conditions of use, and the oxidized hair dyes were tested for their mutagenicity in Salmonella typhimurium TA98 in the presence of a mammalian metabolic activation system ($9 mix). Most of them did not show obvious mutagenicity in the range of 1-25/A/plate and all exhibited bactericidal activity at 10 #I/plate. In order to evaluate the mutagenicity of hair dyes both before and after H202 oxidation, rayon linked to a copper-phthalocyanine derivative (blue rayon) was used as an adsorbent for the elimination of interfering bactericidal compounds. Adsorbed compounds on blue rayon were eluted with ammoniacal methanol and eluents were subjected to the Ames test. The mutagenicity of the blue-rayon extracts in TA98 with $9 mix was increased by H202 oxidation. The blue-rayon extracts obtained from oxidized A and B were potent mutagens and reverted 334 and 999 colonies/10/~1 of original substance, respectively. In addition, 88 and 249 ng of 2,7-diaminophenazine, which was extremely mutagenic in TA98 with $9 mix, were detected in the extracts of 40 ml of the hair dye formulations A and B, respectively. The mutagenicity in oxidized hair dye formulations was successfully detected by use of blue-rayon extraction. 2,7-Diaminophenazine was only formed in the hair dye formulations containing m-phenylenediamine by H202 oxidation. Therefore, attention needs to be paid to the use of m-phenylenediamine as a hair dye component, not only for its own toxicity but also for that of its oxidation products.
The mutagenicity of hair dye ingredients was first evaluated by Ames et al. (1975a) using the Salmonella/microsome test. Later some hair dye ingredients have been shown to be mutagenic in Correspondence: Dr. T. Watanabe, Kyoto Pharmaceutical University, 5 Nakauchi-cho, Misasagi, Yamashina-ku, Kyoto 607 (Japan).
yeast (Mayer and Goin, 1980) and in insects (Blijleven, 1977), to induce chromosomal damage in cultured Chinese hamster cells (Kirkland and Venitt, 1976) and to be carcinogenic in rodents (Reznik and Ward, 1979). p-Phenylenediamine (PD) is the most common aromatic amine used in the preparation of permanent-type or oxidative-type hair dyes and the
0165-7992/90/$ 03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
304
mutagenicity of p - P D has been examined by several investigators (Crebelli et al., 1981; Burnett et al., 1982; Nishi et al., 1982; Rojanapo et al., 1986). After oxidation with HaO2, p - P D became mutagenic to TA98 in the presence of $9 mix, and this mutagenic compound has been suggested to be the Bandrowski's base (Ames et al., 1975a). In our previous papers (Watanabe et al., 1987, 1989a, in press), we reported the mutagenicity of 13 kinds of o-, m- and p - P D derivatives, both before and after H202 treatment, in S. typhimurium TA98 with or without $9 mix. The mutagenicity of mPD, which is used as a hair dye component in Japan, was markedly enhanced by H202 oxidation and 2,7-diaminophenazine was detected as the major mutagenic oxidative compound of m-PD. In this paper, we investigated the detection of the mutagenicity of commercial hair dye formulations and the existence of aminophenazine in the oxidized mixtures. Materials and methods
Chemicals Four different commercial oxidative-type hair dye formulations, A, B, C and D, were purchased from local stores. The components of each hair dye formulation are shown in Table 1.2-Acetylaminofluorene (AAF), 4-nitroquinoline N-oxide (4NQO) and Alumina Activated 200 for column chromatography were purchased from Nakalai tesque, Kyoto (Japan). 30% hydrogen peroxide (H202), polychlorobiphenyl, penta(PCB, Kanechlor-500), and spectrophotometric-grade dimethyl sulfoxide (DMSO) were purchased from Wako Pure Chemicals, Osaka (Japan). 'Blue rayon' was purchased from Funakoshi Chemicals, Tokyo (Japan). 2,7-Diaminophenazine was prepared according to the method described in our previous paper (Watanabe et al., 1989b). Mutagenicity test The hair dye formulations (1 ml) were mixed with 6% H2Oz (1 ml) and stirred for 30 min at room temperature, following which treated hair dyes were diluted with sterile water (48 ml) and
TABLE 1 COMPONENTS OF THE TESTED HAIR DYE FORMULATIONS Sample
Components
A
2,5-diaminotoluene, m-aminophenol, mphenylenediamine, propylene glycol, ammonium thioglycollate,p-hydroxybenzoate,dibutyl hydroxytoluene,edetoate p-phenylenediamine, m-phenylenediamine, m-aminophenol, resorcinol, propylene glycol, thioglycolic acid, p-hydroxybenzoate,edetoate, polyoxyethylenelaurylethersulfate, cetanol, lanoline, benzoic acid, benzoate 2,5-diaminotoluene,p-aminophenol, m-aminophenol, o-aminophenol,p-amino-o-cresol, resorcinol, propylene glycol, ammonium thioglycollate, p-hydroxybenzoate,dibutyl hydroxytoluene, edetoate p-phenylenediamine,p-aminophenol, picramic acid, p-methylaminophenol,resorcinol, polyethyleneglycol, cetanol
neutralized by the addition of hydrochloric acid. Each hair dye solution plus 0.50 g of blue rayon was gently shaken for 10 min at room temperature. The rayon was taken out, and this treatment was repeated once again adding a new batch of blue rayon. After washing with water (3 times with 50 ml) and removal of moisture, the blue rayon was treated with 50 ml of methanol-28%0 NH4OH (50:1, v/v). The elution was repeated, and the combined eluate was evaporated to dryness under reduced pressure. Each residue was dissolved in 2 ml of DMSO and diluted with DMSO. 0.1 ml of each dilution, equivalent to 3 or l 0 / d of hair dye, was subjected to the mutagenicity test. Mutagenicity tests were performed essentially by the Ames method (Ames et al., 1975b; Maron and Ames, 1983) with the suspension assay modified by Yahagi 0975). The tester strain Salmonella typhimurium TA98 was checked routinely to confirm its genetic features for optimal response to known chemicals as follows: AAF (5/zg/plate), 4NQO (0.5 /~g/plate). The assay was performed in the absence
305
or presence of $9 mix to simulate mammalian metabolism. Every sample was tested on at least 3 separate occasions and every dose level was assayed in triplicate.
Metabolic activation system The postmitochondrial fraction, $9, was prepared from the livers of male Sprague-Dawley rats induced with PCB. The homogenate (3 ml of 0.15 M KC1/g liver) was centrifuged as described by Ames et al. (1975b). The $9 mix was prepared according to Yahagi (1975). I ml of $9 mix contained 8 /~mole MgClz, 33 /~mole KC1, 5 /zmole G6P, 4 #mole N A D P H , 0.5 units G6P, 150/zmole sodium phosphate buffer (pH 7.4) water and 50 ~tl of $9.
fraction corresponding to the retention time of authentic 2,7-diaminophenazine was collected and evaporated. The residue was dissolved in 2 ml of 20°7o CH3CN and 20 t~l of the solution was subjected to high-performance liquid chromatography (HPLC) using an LC-6A apparatus (Shimadzu Seisakusho, Kyoto) with a STR ODS-H column (5 #m particle size, 150 mm × 4 mm i.d., Shimadzu Techno Research, Inc., Kyoto). The material was eluted with 20°70 CH3CN at a flow rate of 0.7 ml/min by detecting the fluorescence with excitation wavelength at 490 nm and emission wavelength at 580 nm. Results and discussion
Determination of m-PD in hair dye formulations 100 mg of hair dye formulation was made up to 100 ml with ethyl acetate (AcOEt). 50 #1 of heptafluorobutyric anhydride was mixed with 2 ml of the AcOEt solution and the mixture was heated at 50°C for 30 rain. The reaction mixture was cooled to room temperature and subjected to gas chromatography-mass fragmentography (GC-MF). GCMF conditions were 2 m x 3 mm i.d. glass column packed with 3°70 silicone OV-330 on chromosorb W (AW-DMCS), injector temperature 230°C, oven temperature 160°C, helium gas flow 40 ml/min, ionization voltage 70 eV. The molecular ion (M ÷, m/z 500) was detected by a quadrupole mass spectrometer (Shimadzu QP-1000).
We have studied the mutagenicity of 4 hair dye formulations, A, B, C and D, which are used in Japan. In a preliminary experiment, the tests were carried out after a brief oxidation to simulate normal conditions of use. The hair dye formulations were mixed with 6o70 H202 (1:1, v/v) and allowed to stand for 30 min at room temperature. For the mutagenesis assay, the oxidized hair dyes were used at 1, 2.5, 10 and 25/zl/plate (Table 2). Oxidized A
Determination of 2, 7-diaminophenazine in 1-120:treated hair dyes
Sample
40 ml of hair dye formulation was mixed with 6o70 H2Oz (1:1, v/v) for 30 min at room temperature, and the oxidized mixture was diluted with 80 ml of water. The dilution was treated with 2 batches of blue rayon (2 g each) as described above. The compounds adsorbed on blue rayon were eluted 3 times with methanol-28% NH4OH (50:1, 200 ml each). The eluate was evaporated under reduced pressure and the residue was subjected to alumina column chromatography eluting with CHCI3, CHCI3-MeOH (19:1) and MeOH. The eluate was separated into several fractions, and the
TABLE 2 MUTAGENICITY OF O X I D I Z E D COMMERCIAL HAIR DYES IN S. typhimurium TA98 WITH OR W I T H O U T $9 MIX
A B C D
DMSO 4NQO AAF
Dose (#l/plate)
His + rev./plate - $9 mix
+ $9 mix
1 2.5 1 2.5 1 2.5 1 2.5
34 KL 22 KL 20 KI 36 KL
145 320 66 44 81 106 59 67
16 327
55
0.5/zg/plate 5/~g/plate
KL, killing effect.
736
306
induced 320 revertants at 2.5/~l/plate with $9 mix, but the other hair dyes did not show obvious mutagenicity at that dose and all hair dyes exhibited bactericidal activity at 10/A/plate. Hayatsu et al. (1983) have reported that cotton bearing covalently bound trisulfo-copper-phthalocyanine (blue cotton) can adsorb multicyclic mutagenic compounds in crude samples and the adsorbed mutagens can be recovered by elution with ammoniacal methanol. In addition, they could detect the mutagenicity in smokers' urine, cooked beef and river water with this method. Recently rayon, instead of cotton, bearing covalently linked trisulfo-copper-phthalocyanine residues were developed (blue rayon). For the elimination of interfering bactericidal compounds
TABLE 3 M U T A G E N I C I T Y OF BLUE-RAYON EXTRACTS FROM C O M M E R C I A L HAIR DYES WITH OR W I T H O U T H202 TREATMENT 1N S. typhimurium TA98 WITH OR W I T H O U T $9 MIX Sample
Dose (/~1 equivalent of hair dye/plate)
His + rev./plate - $9 mix
+ $9 mix
Not treated with 11202 A B
C D
3 10 3 10 3 10 3 10
NT 39 NT 19 NT 15 NT 17
45 138 24 45 27 37 30 29
NT 21 NT 22 NT 19 NT 20
110 334 327 999 34 94 41 59
Treated with 02112 A B
C D
DMSO 4NQO AAF
3 10 3 10 3 10 3 10
0.5 #g/plate 5 #g/plate
NT, not tested.
16 327
22 364
from oxidized hair dyes and the evaluation of the mutagenicity of oxidized hair dyes, blue rayon was used as an adsorbent. Table 3 shows the mutagenicity of blue-rayon extracts obtained from the 4 hair dye formulations in Salmonella typhimurium TA98 with or without $9 mix. The oxidative conditions of hair dyes and the extraction procedure with blue rayon have been described in Materials and methods. Of the 4 hair dye formulations tested, only A exhibited obvious mutagenicity without H 2 0 2 treatment. The mutagenic potency of all hair dye formulations tested in this study was enhanced by H202 treatment and their activity was observed in the presence of $9 mix. A marked enhancement of mutagenicity was observed for A and B, and bluerayon extracts obtained from oxidized A and B induced 334 and 999 revertants per l0/A of original substance, respectively. m-Phenylenediamine (PD), which was used as a coupler in the hair dye formulations, was determined as the heptafluorobutyryl derivative by GC-MF. The conditions have been described in Materials and methods. Hair dye formulations A and B contained 8 and 2.5 mg of m-PD per gram of hair dye, respectively. In our previous papers (Watanabe et al., 1989a,b), we reported that the mutagenicity of mPD was markedly increased by H202 oxidation, and 2,7-diaminophenazine was identified as the major mutagenic oxidative compound of m-PD. Since A and B contained m-PD as a hair dye component and their mutagenic potency was markedly increased by the oxidation, the formation of 2,7-diaminophenazine in the oxidative process was expected. We presumed that the yield of 2,7-diaminophenazine in oxidized hair dyes was small because of the low m-PD content in hair dye formulations. The blue-rayon extracts were subjected to further treatment with alumina column chromatography to remove contaminants. In a preliminary experiment, recoveries of 2,7-diaminophenazine from a 3°7o H 2 0 2 solution by blue-rayon and alumina column treatment were 85.0 and 90.7%, respectively, and those are satisfactory levels. In order to confirm the existence of 2,7-diamino-
307
phenazine in oxidized hair dyes, 40 ml of hair dye formulations, which is the amount most commonly used for hair dyeing, was oxidized, diluted and treated with blue rayon. The blue-rayon extracts were column-chromatographed on alumina and the fractions corresponding to 2,7-diaminophenazine were subjected to HPLC on an ODS column. As shown in Fig. 1, a peak was detected at a retention time of 3.9 min, which coincided with that of authentic 2,7-diaminophenazine, on each chromatogram of hair dye extracts A and B by monitoring fluorescence. In addition, the maximal excitation and emission wavelengths of the peaks were the same as those of the authentic sample. The yields of 2,7-diaminophenazine detected on the ODS column were calculated, with the aid of the standard curve of authentic 2,7-diaminophenazine, to be 88 ng (A) and 249 ng (B) in 40 ml of hair dye formulation. In non-oxidized hair dye formulations A and B, 2,7-diaminophenazine could not be detected under similar conditions described in Materials and methods (data not shown). In the model system, m-PD (5 mg/ml in water) was treated with 6°7o H 2 0 2 (1:1, v/v) in the same manner as the hair dye formulations used in this study. 16.2 #g of 2,7-diaminophenazine was then formed from 0.2 g of m-PD, which amount was almost comparable to that of m-PD in 40 ml of hair
dye formulation, after H 2 0 2 treatment for 30 min. Thus the yield of 2,7-diaminophenazine in the model system was more than 50 times larger than that in the oxidized hair dyes. For the reason above, it was considered that in hair dye formulations m-PD was a minor component, and most m-PD was consumed by binding with other abundant hair dye components such as p-phenylenediamine derivatives, mono- or polyphenol, aminophenol, etc. Corbett (1969) reported that oxidation of mixture of p-aminophenol and m-PD gave 2,4-diamino-4'-hydroxyldiphenylamine. In addition, he reported that m-PD reacted with p-PD under oxidative conditions, and it gave 2,4,4'-triaminodiphenylamine (Corbett, 1969, 1973). In the present study, the mutagenicity in hair dye formulations was successfully detected with the use of blue-rayon extraction. 2,7-Diaminophenazine, which was extremely mutagenic and reverted 730 colonies/ng in TA98 with $9 mix, was apparently detected in the blue-rayon extracts obtained from oxidized A and B. Since m-PD was capable of forming 2,7-diaminophenazine in the oxidation process of hair dye formulations, the use of m-PD as a hair dye component should receive attention for the toxicity of not only itself but also its oxidation products.
A
C
D
c
DAPz o ,:j
!
DAPz
I 0
I 4
I 8
I 12
I 16
I 0
I 4
I 8
I 12
Retention
I 0 time
I 4
I 8
I 12
I 16
I
I
I
I
0
4
8
12
(rain)
Fig. 1. HPLC chromatograms of blue-rayon extracts obtained from oxidized hair dye A (A), B (B), C (C) and D (D) by detecting fluorescence. DAPz, 2,7-diaminophenazine.
308
Acknowledgements The authors are grateful to Prof. B.N. Ames, University of California, Berkeley, CA (U.S.A.) who kindly supplied Salmonella typhimurium TA98. We also wish to thank Mr. T. Nishigaki and Mr. Y. Tada for their excellent technical assistance in this investigation.
References Ames, B.N., H.O. Kammen and E. Yamasaki (1975a) Hair dyes are mutagenic: identification of a variety of mutagenic ingredients, Proc. Natl. Acad. Sci. (U.S.A.), 72, 2433-2437. Ames, B.N., J. McCann and E. Yamasaki (1975b) Methods for detecting carcinogens and mutagens with the Salmonella/ mammalian-microsome mutagenicity test, Mutation Res., 31, 347-364. Blijleven, W.G.H. (1977) Mutagenicity of four hair dyes in Drosophila melanogaster, Mutation Res., 48, 181-185. Burnett, C., C. Fuchs, J. Corbett and J. Menkart (1982) The effect of dimethyl sulfoxide on the mutagenicity of the hair dye p-phenylenediamine, Mutation Res., 103, 1-4. Corbett, J.F. (1969) Benzoquinone imines. Part VI. Mechanism and kinetics of the reaction ofp-benzoquinone di-imines with m-phenylenediamines, J. Chem. Soc. (B), 827-835. Corben, J.F. (1973) Role of m-difunctional benzene derivatives in oxidative hair dyeing, I. Reaction with p-diamines, J. Soc. Cosmet. Chem., 24, 103-134. Crebelli, R., L. Conti, A. Carere and R. Zito (1981) Mutagenicity of commercial p-phenylenediamine and of an oxidation mixture of p-phenylenediamine and resorcinol in Salmonella typhimurium TA98, Food Cosmet. Toxicol., 19, 79-84. Hayatsu, H., T. Oka, A. Wakata, Y. Ohara, T. Hayatsu, H. Kobayashi and S. Arimoto (1983) Adsorption of mutagens to
cotton bearing covalently bound trisulfo-copper-phthalocyanine, Mutation Res., 119, 233-238. Kirkland, D.J., and S. Venitt (1976) Cytotoxicity of hair colourant constituents: chromosome damage induced by two nitrophenylenediamines in cultured Chinese hamster cells, Mutation Res., 40, 47. Maron, D.M., and B.N. Ames (1983) Revised methods for the Salmonella mutagenicity test, Mutation Res., 113, 173-215. Mayer, V.W., and C.J. Goin (1980) Induction of mitotic recombination by certain hair-dye chemicals in Saccharomyces cerevisiae, Mutation Res., 78,243-252. Nishi, K., and H. Nishioka (1982) Light induces mutagenicity of hair-dye p-phenylenediamine, Mutation Res., 104, 347-350. Reznik, G., and J.M. Ward (1979) Carcinogenicity of the hairdye component 2-nitro-p-phenylenediamine: induction of eosinophilic hepatocellular neoplasms in female B6C3F1 mice, Food Cosmet. Toxicol., 17, 493-500. Rojanapo, W., P. Kupradinun, A. Tepsuwan, S. Chutimatae and M. Tanyakaset (1986) Carcinogenicity of an oxidation product of p-phenylendiamine, Carcinogenesis, 7, 1997 -2002. Watanabe, T., M. Ono, T. Hirayama and S. Fukui (1987) 2,7-Diamino-3,8-dimethylphenazine as the major mutagenic product from the reaction of 2,4-diaminotoluene with hydrogen peroxide, Mutation Res., 190, 113-117. Watanabe, T., T. Hirayama and S. Fukui (1989a) Phenazine derivatives as the mutagenic reaction product from o- or mphenylenediamine derivatives with hydrogen peroxide, Mutation Res., 227, 135-145. Watanabe, T., Y. Hanasaki, T. Hirayama and S. Fukui (1989b) Mutagenicity of nitro- and amino-substituted phenazines in Salmonella typhimurium, Mutation Res., 225, 75-82. Watanabe, T., T. Hirayama and S. Fukui (1990) The mutagenic modulating effect of p-phenylenediamine on the oxidation of o- or m-phenylenediamine with hydrogen peroxide in Salmonella test, Mutation Res., in press. Yahagi, T. (1975) Screening methods using microbes for the environmental carcinogen, Tanpakushitsu Kakusan Koso (Protein, Nucleic Acid and Enzyme), 20, 1178-1189. Communicated by H.S. Rosenkranz
