- Email: [email protected]
microsome test
Mutation Research, 192 (1987) 91-94 Elsevier
91
MTRL 030
Mutagenicity of N-nitrosodiethanolamine in the Salmonella/microsome test Y. M o r i a, H . Y a m a z a k i a a n d Y. K o n i s h i b a Laboratory of Radiochemistry, Gifu Pharmaceutical University, 6-1, Mitahora-higashi 5-chome, Gifu 502 and b Department of Oncological Pathology, Cancer Center, Nara Medical College, 840 Shijo-cho, Kashihara, Nara 634 (Japan) (Accepted 19 May 1987)
Keywords." N-Nitrosodiethanolamine; Liver $9; Rat; Hamster; (Salmonella/microsome test).
Summary Mutagenicity of a commercially available N-nitrosodiethanolamine (NDELA) and purified NDELA was examined, using Salmonella typhimurium TA100 as a tester strain. Purified NDELA was positive in the presence of liver activation system from either rats or hamsters, but the mutagenicity was completely lost when dimethyl sulfoxide (DMSO) was used as a solvent. In contrast, the commercial NDELA which was chemically of 93.8°70 purity showed positive mutagenicity without metabolic activation, and the liver activation system and DMSO had no effect on the direct mutagenic activity. These results indicate that an apparent discrepancy among previous findings of several investigators with the mutagenic response of NDELA might be due to an impurity in NDELA samples and the solvent, DMSO.
NDELA is among the most widely distributed of the N-nitroso compounds in the United States and probably the one to which human exposure is the greatest (Scanlan, 1983; Lijinsky and Kovatch, 1985), since it occurs at high levels in synthetic cutting oils, cosmetics, lotions, shampoos (Fan et al.,
Correspondence: Dr. Yukio Mori, Laboratory of Radiochemistry, Gifu Pharmaceutical University, 6-1, Mitahorahigashi 5-chome, Gifu 502 (Japan).
Abbreviations: DMSO, dimethyl sulfoxide; HPLC, highpressure liquid chromatography; NDELA, N-nitrosodiethanolamine (CAS Registry Number 1116-54-7); $9, 9000 g supernatant.
1977), tobacco and pesticides (IARC Monographs, 1978). NDELA can be readily formed by interaction between sodium nitrite and diethanolamine or triethanolamine (Zingmark and Rappe, 1976) and induces liver, kidney, trachea and/or nasal cavity carcinomas in rats (Lijinsky and Kovatch, 1985) and hamsters (Hilfrich et al., 1977). However, different results with mutagenic effects of NDELA on S. typhimurium strains TA100, TA1535 and TA1530 were reported by various investigators. In order to clarify the mutagenicity of NDELA, we investigated the mutagenic activities of commercially available NDELA and purified NDELA in the absence or presence of liver $9 from rats and hamsters.
0165-7992/87/$ 03.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)
92 Materials and methods
Results
Chemicals Polychlorobiphenyls (mixture of hexa-isomers) were obtained from Wako Pure Chemical Industries, Ltd., Osaka, Japan. N D E L A (lot No. CSO2) was obtained from Tokyo Chemical Industry, Ltd. and stored at - 2 0 ° C before use. After purification by column chromatography on silica gel (Wakogel C-200, Wako Pure Chemical Industries) using ethyl acetate as an eluting solvent, the c o m m e r c i a l N D E L A appeared as a lightyellow, viscous oil. The purity and chemical structure of N D E L A were confirmed by H P L C and high-resolution mass spectrometry. All other commercial products were the purest grades available.
As shown in Fig. 1A, H P L C analysis of commercially available N D E L A revealed the presence of 3 impurities: the compound was chemically of 93.8°7o purity, as calculated from the chromatogram. Similar results were also observed after the normal phase chromatography using acetonitrile as the mobile phase (data not shown). Commercially available N D E L A was purified by column chromatography (Fig. 1B) and found by high-resolution mass spectrometry to show the following fragments characteristic of N D E L A (Fan et al., 1977a): m / z 135.07682 (calculated for C4HllN203: 135.07695), m / z 104.07050 (calculated for C4HIoNO2: 104.07114), m / z 103.04958 (calculated for C3HTN202: 103.05074) and m / z 91.05433 (calculated for C2HTN202: 91.05074). Fig. 2 indicates the results of the mutagenicity of commercially available N D E L A and purified N D E L A with or without activation by liver $9 from rats and hamsters. In the absence of liver $9, the number of revertant colonies with the purified N D E L A was the same as the spontaneous rates (Fig. 2B). The purified N D E L A proved to be mutagenic in the presence of liver $9 from rats (Fig. 2B) and hamsters (Fig. 2C) and the activities
Purity of the commercial NDELA and purified NDELA H P L C was performed using a Jasco model TRIROTOR-II (Japan Spectroscopic Co. Ltd., Tokyo) equipped with a solvent programmer (Jasco model GP-A40). N D E L A was separated by a Jasco Finepak SIL Cl8 (4.6 mm inside diameter x 25 cm) and measured at 239 nm. N D E L A (1.34 mg) was dissolved in 1 ml of 30o70 CH3CN, and 3 /zl of the solution was injected into H P L C directly and eluted at 0.5 ml/min for 0-14 min with 20-30°70 acetonitrile; 14-24 min with 30°7o acetonitrile. A J E O L model JMS-D300 mass spectrometer equipped with a JMA-2000S mass data analysis system (Nippon Denshi Ltd., Akishima) was operated at an ionization energy of 70 eV, ionization current of 300 A and an accelerating voltage of 5 kV in the electron impact mode. Mutagenicity test Each test was carried out, in triplicate, by Ames preincubation assay (Maron and Ames, 1983), with TA100 as a tester strain. The mutagenic response of N D E L A dissolved in either water or DMSO was checked as described previously (Mori et al., 1985b, 1986), using liver $9 from male Wistar rats (180 g) or male Syrian golden hamsters (80 g) pretreated with polychlorobiphenyls.
~D
o
,
4
I
,
12
i
16 20 24
i
8
12
16 20 24
Fig. 1. HPLC chromatograms of commercially available NDELA (A) and purified NDELA (B). AUFS, absorbance unit full scale.
93
400
400
400
~"
300
300
300
>
200
200
2OO
o
100
100
lO0
0
'
'
'
l
5
lO
' , 15 20
0
,
,
,
l
5
lO
, 15
, 20
Amount o f NDELA ( m g / p l a t e
I
I
1
5
I
I
lO 15
I
20
)
Fig. 2. Mutagenicity of commercially available NDELA (A) and purified NDELA (B, C) in S. typhimurium TA100 in the absence (circles) or presence of liver $9 from rats (triangles) or hamsters (squares). NDELA was dissolved in water (solid marks) or DMSO (open marks). Dotted line indicates the number of spontaneous revertants. increased linearly within the concentration ranges of 1-15 mg and 1 - 1 0 m g / p l a t e , respectively. However, when D M S O was used as a solvent, the mutagenic activities of purified N D E L A were completely decreased to the background level in both cases. In contrast, commercially available N D E L A was weakly mutagenic without metabolic activation, and DMSO and liver $9 f r o m either rats or hamsters had no effect on the activities (Fig. 2A). Discussion There are conflicting reports in the literature concerning the mutagenic response o f N D E L A . Hesbert et al. (1979) and Dahl 0985) have reported that synthesized or commercially available N D E L A up to 13 m g / p l a t e shows direct-acting mutagenicity in strains T A I 0 0 and TA1535, and that liver $9 f r o m mice (Hesbert et al., 1979), rats and rabbits (Dahl, 1985) does not affect the activities. On the other hand, synthesized N D E L A is reported to be non-mutagenic at doses up to 10 m g / p l a t e towards the same strains in either absence or presence o f liver $9 from rats or hamsters (Lijinsky and Andrews, 1983; Gilbert et al., 1981; Rao et al., 1979). Prival et al. (1982) have shown that synthesized N D E L A (5-15 mg) is mutagenic on strain TA100 and TA1535 in the
presence of hamster-liver $9, but not rat-liver $9. The findings of our present results with commercially available N D E L A (Fig. 2A) are coincident with those of Hesbert et al. (1979) and Dahl (1985). However, this commercial sample was shown to contain 3 impurities which were determined by H P L C (Fig. 1). Purified N D E L A (1-20 mg) showed positive mutagenicity only after metabolic activation by liver $9 f r o m either rats or hamsters (Fig. 2B and C); these results were inconsistent with all the findings of the other investigators, except for the positive results in the presence of hamster-liver $9 (Prival et al., 1982). Recently, we have reported that the mutagenic activities of eleven N-nitrosodialkylamines in Salmonella were considerably or completely decreased by addition o f DMSO to the incubation mixture (Mori et al., 1985a). Actually, DMSO completely inhibited the mutagenic activities of purified N D E L A (Fig. 2B and C). Accordingly, observed direct-acting mutagenicity of N D E L A might be due to an impurity(s) in the commercial sample, and the apparent discrepancy between our results with purified N D E L A and previous findings of several investigators (Lijinsky and Andrews, 1983; Gilbert et al., 1981; Rao et al., 1979) who all used DMSO might be due to methodological differences. Consequently, we conclude that N D E L A which is car-
94
cinogenic to rats and hamsters is mutagenic towards strain TA100 in the presence of the metabolic activation system with either rat- or hamster-liver $9.
References Dahl, A.R. (1985) Mutagenicity of some dialkylnitrosamines, cyclic nitrosamines and N,N-diethanolnitrosamine in Salmonella typhimurium with rat and rabbit nasal, lung and liver $9 homogenates, Mutation Res., 158, 141-147. Fan, T.Y., U. Goff, L. Song, D.H. Fine, G.P. Arsenault and K. Biemann (1977a) N-Nitrosodiethanolamine in cosmetics, lotions and shampoos, Food Cosmet. Toxicol., 15,423-430. Fan, T.Y., J. Morrison, D.P. Rounbehler, R. Ross, D.H. Fine, W. Miles and N.P. Sen (1977b) N-Nitrosodiethanolamine in synthetic cutting fluids: a part-per-hundred impurity, Science, 196, 70-71. Gilbert, P., L. Fabry, B. Rollmann, P. Lombart, J. Rondelet, F. Poncelet, A. Leonard and M. Mercier (1981) Mutagenicity of N-nitrosodiethanolamine and its acetyl derivatives, Mutation Res., 89, 217-228. Hesbert, A., M. Lemonnier and C. Cavelier (1979) Mutagenicity of nitrosodiethanolamine on Salmonella typhimurium, Mutation Res., 68, 207-210. Hilfrich, J., I. Schmeltz and D. Hoffmann (1977) Effects of Nnitrosodiethanolamine and 1,1-diethanolhydrazine in Syrian golden hamsters, Cancer Lett., 4, 55-60. IARC (1978) Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, Some N-Nitroso Compounds, Vol. 17, pp. 77-82. Lijinsky, W., and A.W. Andrews (1983) The superiority of hamster liver microsomal fraction for activating nitros-
amines to mutagens in Salmonella typhimurium, Mutation Res., 111, 135-144. o, -.~ Lijinsky, W., and R.M. Kovatch (1985) Induction of liver tumors in rats by nitrosodiethanolamine at low doses, Carcinogenesis, 6, 1679-1681. Maron, D.M., and B.N. Ames (1983) Revised methods for the Salmonella mutagenicity test, Mutation Res., 113, 172-215. Mori, Y., H. Yamazaki, K. Toyoshi, Y. Emi, K. Uchida, M. Tsutsumi and Y. Konishi (1985a) Inhibitory effect of organic solvents on the mutagenicity of N-nitrosodialkylamines in Salmonella, Mutation Res., 142, 153-158. Mori, Y., H. Yamazaki, K. Toyoshi, T. Makino, T. Obara, Y. Yokose and Y. Konishi (1985b) Mutagenic activation of carcinogenic N-nitrosopropylamines by rat liver: evidence for a cytochrome P-450 dependent reaction, Carcinogenesis, 6, 415-420. Mori, Y., H. Yamazaki, K. Toyoshi, A. Denda and Y. Konishi (1986) Mutagenic activation of carcinogenic N-nitrosopropylamines by liver $9 fractions from mice, rats and hamsters: evidence for a cytochrome P-450-dependent reaction, Carcinogenesis, 7, 375-379. Prival, M.J., A.T. Sheldon Jr. and D. Popkin (1982) Evaluation, using Salmonella typhimurium, of the mutagenicity of seven chemicals found in cosmetics, Food Cosmet. Toxicol., 20, 427-432. Rao, T.K., J.A. Young, W. Lijinsky and J.L. Epler (1979) Mutagenicity of aliphatic nitrosamines in Salmonella typhimurium, Mutation Res., 66, 1-7. Scanlan, R.A. (1983) Formation and occurrence of nitrosamines in food, Cancer Res. (Suppl.), 43, 2435s-2440s. Zingmark, P.-A., and C. Rappe (1976) On the formation of Nnitrosodiethanolamine from a grinding fluid under simulated gastric conditions, Ambio, 5, 80-81. Communicated by T. Sugimura
91
MTRL 030
Mutagenicity of N-nitrosodiethanolamine in the Salmonella/microsome test Y. M o r i a, H . Y a m a z a k i a a n d Y. K o n i s h i b a Laboratory of Radiochemistry, Gifu Pharmaceutical University, 6-1, Mitahora-higashi 5-chome, Gifu 502 and b Department of Oncological Pathology, Cancer Center, Nara Medical College, 840 Shijo-cho, Kashihara, Nara 634 (Japan) (Accepted 19 May 1987)
Keywords." N-Nitrosodiethanolamine; Liver $9; Rat; Hamster; (Salmonella/microsome test).
Summary Mutagenicity of a commercially available N-nitrosodiethanolamine (NDELA) and purified NDELA was examined, using Salmonella typhimurium TA100 as a tester strain. Purified NDELA was positive in the presence of liver activation system from either rats or hamsters, but the mutagenicity was completely lost when dimethyl sulfoxide (DMSO) was used as a solvent. In contrast, the commercial NDELA which was chemically of 93.8°70 purity showed positive mutagenicity without metabolic activation, and the liver activation system and DMSO had no effect on the direct mutagenic activity. These results indicate that an apparent discrepancy among previous findings of several investigators with the mutagenic response of NDELA might be due to an impurity in NDELA samples and the solvent, DMSO.
NDELA is among the most widely distributed of the N-nitroso compounds in the United States and probably the one to which human exposure is the greatest (Scanlan, 1983; Lijinsky and Kovatch, 1985), since it occurs at high levels in synthetic cutting oils, cosmetics, lotions, shampoos (Fan et al.,
Correspondence: Dr. Yukio Mori, Laboratory of Radiochemistry, Gifu Pharmaceutical University, 6-1, Mitahorahigashi 5-chome, Gifu 502 (Japan).
Abbreviations: DMSO, dimethyl sulfoxide; HPLC, highpressure liquid chromatography; NDELA, N-nitrosodiethanolamine (CAS Registry Number 1116-54-7); $9, 9000 g supernatant.
1977), tobacco and pesticides (IARC Monographs, 1978). NDELA can be readily formed by interaction between sodium nitrite and diethanolamine or triethanolamine (Zingmark and Rappe, 1976) and induces liver, kidney, trachea and/or nasal cavity carcinomas in rats (Lijinsky and Kovatch, 1985) and hamsters (Hilfrich et al., 1977). However, different results with mutagenic effects of NDELA on S. typhimurium strains TA100, TA1535 and TA1530 were reported by various investigators. In order to clarify the mutagenicity of NDELA, we investigated the mutagenic activities of commercially available NDELA and purified NDELA in the absence or presence of liver $9 from rats and hamsters.
0165-7992/87/$ 03.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)
92 Materials and methods
Results
Chemicals Polychlorobiphenyls (mixture of hexa-isomers) were obtained from Wako Pure Chemical Industries, Ltd., Osaka, Japan. N D E L A (lot No. CSO2) was obtained from Tokyo Chemical Industry, Ltd. and stored at - 2 0 ° C before use. After purification by column chromatography on silica gel (Wakogel C-200, Wako Pure Chemical Industries) using ethyl acetate as an eluting solvent, the c o m m e r c i a l N D E L A appeared as a lightyellow, viscous oil. The purity and chemical structure of N D E L A were confirmed by H P L C and high-resolution mass spectrometry. All other commercial products were the purest grades available.
As shown in Fig. 1A, H P L C analysis of commercially available N D E L A revealed the presence of 3 impurities: the compound was chemically of 93.8°7o purity, as calculated from the chromatogram. Similar results were also observed after the normal phase chromatography using acetonitrile as the mobile phase (data not shown). Commercially available N D E L A was purified by column chromatography (Fig. 1B) and found by high-resolution mass spectrometry to show the following fragments characteristic of N D E L A (Fan et al., 1977a): m / z 135.07682 (calculated for C4HllN203: 135.07695), m / z 104.07050 (calculated for C4HIoNO2: 104.07114), m / z 103.04958 (calculated for C3HTN202: 103.05074) and m / z 91.05433 (calculated for C2HTN202: 91.05074). Fig. 2 indicates the results of the mutagenicity of commercially available N D E L A and purified N D E L A with or without activation by liver $9 from rats and hamsters. In the absence of liver $9, the number of revertant colonies with the purified N D E L A was the same as the spontaneous rates (Fig. 2B). The purified N D E L A proved to be mutagenic in the presence of liver $9 from rats (Fig. 2B) and hamsters (Fig. 2C) and the activities
Purity of the commercial NDELA and purified NDELA H P L C was performed using a Jasco model TRIROTOR-II (Japan Spectroscopic Co. Ltd., Tokyo) equipped with a solvent programmer (Jasco model GP-A40). N D E L A was separated by a Jasco Finepak SIL Cl8 (4.6 mm inside diameter x 25 cm) and measured at 239 nm. N D E L A (1.34 mg) was dissolved in 1 ml of 30o70 CH3CN, and 3 /zl of the solution was injected into H P L C directly and eluted at 0.5 ml/min for 0-14 min with 20-30°70 acetonitrile; 14-24 min with 30°7o acetonitrile. A J E O L model JMS-D300 mass spectrometer equipped with a JMA-2000S mass data analysis system (Nippon Denshi Ltd., Akishima) was operated at an ionization energy of 70 eV, ionization current of 300 A and an accelerating voltage of 5 kV in the electron impact mode. Mutagenicity test Each test was carried out, in triplicate, by Ames preincubation assay (Maron and Ames, 1983), with TA100 as a tester strain. The mutagenic response of N D E L A dissolved in either water or DMSO was checked as described previously (Mori et al., 1985b, 1986), using liver $9 from male Wistar rats (180 g) or male Syrian golden hamsters (80 g) pretreated with polychlorobiphenyls.
~D
o
,
4
I
,
12
i
16 20 24
i
8
12
16 20 24
Fig. 1. HPLC chromatograms of commercially available NDELA (A) and purified NDELA (B). AUFS, absorbance unit full scale.
93
400
400
400
~"
300
300
300
>
200
200
2OO
o
100
100
lO0
0
'
'
'
l
5
lO
' , 15 20
0
,
,
,
l
5
lO
, 15
, 20
Amount o f NDELA ( m g / p l a t e
I
I
1
5
I
I
lO 15
I
20
)
Fig. 2. Mutagenicity of commercially available NDELA (A) and purified NDELA (B, C) in S. typhimurium TA100 in the absence (circles) or presence of liver $9 from rats (triangles) or hamsters (squares). NDELA was dissolved in water (solid marks) or DMSO (open marks). Dotted line indicates the number of spontaneous revertants. increased linearly within the concentration ranges of 1-15 mg and 1 - 1 0 m g / p l a t e , respectively. However, when D M S O was used as a solvent, the mutagenic activities of purified N D E L A were completely decreased to the background level in both cases. In contrast, commercially available N D E L A was weakly mutagenic without metabolic activation, and DMSO and liver $9 f r o m either rats or hamsters had no effect on the activities (Fig. 2A). Discussion There are conflicting reports in the literature concerning the mutagenic response o f N D E L A . Hesbert et al. (1979) and Dahl 0985) have reported that synthesized or commercially available N D E L A up to 13 m g / p l a t e shows direct-acting mutagenicity in strains T A I 0 0 and TA1535, and that liver $9 f r o m mice (Hesbert et al., 1979), rats and rabbits (Dahl, 1985) does not affect the activities. On the other hand, synthesized N D E L A is reported to be non-mutagenic at doses up to 10 m g / p l a t e towards the same strains in either absence or presence o f liver $9 from rats or hamsters (Lijinsky and Andrews, 1983; Gilbert et al., 1981; Rao et al., 1979). Prival et al. (1982) have shown that synthesized N D E L A (5-15 mg) is mutagenic on strain TA100 and TA1535 in the
presence of hamster-liver $9, but not rat-liver $9. The findings of our present results with commercially available N D E L A (Fig. 2A) are coincident with those of Hesbert et al. (1979) and Dahl (1985). However, this commercial sample was shown to contain 3 impurities which were determined by H P L C (Fig. 1). Purified N D E L A (1-20 mg) showed positive mutagenicity only after metabolic activation by liver $9 f r o m either rats or hamsters (Fig. 2B and C); these results were inconsistent with all the findings of the other investigators, except for the positive results in the presence of hamster-liver $9 (Prival et al., 1982). Recently, we have reported that the mutagenic activities of eleven N-nitrosodialkylamines in Salmonella were considerably or completely decreased by addition o f DMSO to the incubation mixture (Mori et al., 1985a). Actually, DMSO completely inhibited the mutagenic activities of purified N D E L A (Fig. 2B and C). Accordingly, observed direct-acting mutagenicity of N D E L A might be due to an impurity(s) in the commercial sample, and the apparent discrepancy between our results with purified N D E L A and previous findings of several investigators (Lijinsky and Andrews, 1983; Gilbert et al., 1981; Rao et al., 1979) who all used DMSO might be due to methodological differences. Consequently, we conclude that N D E L A which is car-
94
cinogenic to rats and hamsters is mutagenic towards strain TA100 in the presence of the metabolic activation system with either rat- or hamster-liver $9.
References Dahl, A.R. (1985) Mutagenicity of some dialkylnitrosamines, cyclic nitrosamines and N,N-diethanolnitrosamine in Salmonella typhimurium with rat and rabbit nasal, lung and liver $9 homogenates, Mutation Res., 158, 141-147. Fan, T.Y., U. Goff, L. Song, D.H. Fine, G.P. Arsenault and K. Biemann (1977a) N-Nitrosodiethanolamine in cosmetics, lotions and shampoos, Food Cosmet. Toxicol., 15,423-430. Fan, T.Y., J. Morrison, D.P. Rounbehler, R. Ross, D.H. Fine, W. Miles and N.P. Sen (1977b) N-Nitrosodiethanolamine in synthetic cutting fluids: a part-per-hundred impurity, Science, 196, 70-71. Gilbert, P., L. Fabry, B. Rollmann, P. Lombart, J. Rondelet, F. Poncelet, A. Leonard and M. Mercier (1981) Mutagenicity of N-nitrosodiethanolamine and its acetyl derivatives, Mutation Res., 89, 217-228. Hesbert, A., M. Lemonnier and C. Cavelier (1979) Mutagenicity of nitrosodiethanolamine on Salmonella typhimurium, Mutation Res., 68, 207-210. Hilfrich, J., I. Schmeltz and D. Hoffmann (1977) Effects of Nnitrosodiethanolamine and 1,1-diethanolhydrazine in Syrian golden hamsters, Cancer Lett., 4, 55-60. IARC (1978) Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, Some N-Nitroso Compounds, Vol. 17, pp. 77-82. Lijinsky, W., and A.W. Andrews (1983) The superiority of hamster liver microsomal fraction for activating nitros-
amines to mutagens in Salmonella typhimurium, Mutation Res., 111, 135-144. o, -.~ Lijinsky, W., and R.M. Kovatch (1985) Induction of liver tumors in rats by nitrosodiethanolamine at low doses, Carcinogenesis, 6, 1679-1681. Maron, D.M., and B.N. Ames (1983) Revised methods for the Salmonella mutagenicity test, Mutation Res., 113, 172-215. Mori, Y., H. Yamazaki, K. Toyoshi, Y. Emi, K. Uchida, M. Tsutsumi and Y. Konishi (1985a) Inhibitory effect of organic solvents on the mutagenicity of N-nitrosodialkylamines in Salmonella, Mutation Res., 142, 153-158. Mori, Y., H. Yamazaki, K. Toyoshi, T. Makino, T. Obara, Y. Yokose and Y. Konishi (1985b) Mutagenic activation of carcinogenic N-nitrosopropylamines by rat liver: evidence for a cytochrome P-450 dependent reaction, Carcinogenesis, 6, 415-420. Mori, Y., H. Yamazaki, K. Toyoshi, A. Denda and Y. Konishi (1986) Mutagenic activation of carcinogenic N-nitrosopropylamines by liver $9 fractions from mice, rats and hamsters: evidence for a cytochrome P-450-dependent reaction, Carcinogenesis, 7, 375-379. Prival, M.J., A.T. Sheldon Jr. and D. Popkin (1982) Evaluation, using Salmonella typhimurium, of the mutagenicity of seven chemicals found in cosmetics, Food Cosmet. Toxicol., 20, 427-432. Rao, T.K., J.A. Young, W. Lijinsky and J.L. Epler (1979) Mutagenicity of aliphatic nitrosamines in Salmonella typhimurium, Mutation Res., 66, 1-7. Scanlan, R.A. (1983) Formation and occurrence of nitrosamines in food, Cancer Res. (Suppl.), 43, 2435s-2440s. Zingmark, P.-A., and C. Rappe (1976) On the formation of Nnitrosodiethanolamine from a grinding fluid under simulated gastric conditions, Ambio, 5, 80-81. Communicated by T. Sugimura