Stability of mutagenic nitrosated products of indole compounds occurring in vegetables.

Levels of indolylglucosinolates in Brassica vegetables correlated significantly with the amounts of N-nitroso compounds formed in these vegetables after nitrite treatment. Nitrosation of indole-3-carbinol, indole-3-acetonitrile and indole, hydrolysis products of an indolylglucosinolate, resulted in formation of nitrosated products, which were directly mutagenic to Salmonella typhimurium TA100. The nitrosated products were unstable at pH 2 but stable at pH 8. Experiments to elucidate the mechanisms behind these differences in stability showed an equilibrium between the nitrosated indole compound and the free compound plus nitrite.

Several known indole compounds are not important precursors of direct mutagenic N-nitroso compounds in green cabbage.

In this study we investigated the role of indole-3-acetonitrile, indole-3-carbinol, indole and tryptophan in the formation of N-nitroso compounds in green cabbage extracts. Green cabbage extracts were separated by gel permeation chromatography. Fractions were treated with nitrite, tested for mutagenicity and analysed for total N-nitroso content. Fractions in which spiked indole-3-acetonitrile, indole-3-carbinol, indole and tryptophan eluted appeared to be low in mutagenic activity and contained relatively small amounts of N-nitroso compounds. To detect indole compounds other than the ones used in the gel permeation chromatography experiments, high-performance liquid chromatography and gas chromatography-mass spectrometry analyses were performed of green cabbage extracts. Indole-3-carboxaldehyde was found to be the most commonly occurring indole compound, but it did not show direct mutagenic activity upon nitrite treatment. Indole-3-acetonitrile was the second most common compound; although it was mutagenic after nitrite treatment, its contribution to the mutagenicity of nitrite-treated green cabbage was roughly estimated to be only 2%. No other indole compounds were detected. From this study we conclude that neither the tested indole compounds nor indole-3-carboxaldehyde play a significant role in the formation of direct mutagenic N-nitroso compounds in nitrite-treated green cabbage extracts.

The stability of the nitrosated products of indole, indole-3-acetonitrile, indole-3-carbinol and 4-chloroindole.

The nitrosation rates of indole-3-acetonitrile, indole-3-carbinol, indole and 4-chloroindole and the stability of their nitrosated products were investigated. Each of the nitrosated indole compounds was directly mutagenic to Salmonella typhimurium TA100 in the following order of potency: 4-chloroindole much greater than indole-3-carbinol greater than or equal to indole greater than indole-3-acetonitrile. Total N-nitroso determinations, carried out according to a modified method of Walters et al. (Analyst, Lond. 1978, 103, 1127), and Ames test results revealed that each of the indole compounds immediately formed mutagenic N-nitroso products upon nitrite treatment under acidic conditions. However, the nitrosation rates of indole and 4-chloroindole were higher than those of indole-3-acetonitrile and indole-3-carbinol. For indole-3-carbinol, indole-3-acetonitrile and indole, no change in the amount of nitrosated products was observed at increasing incubation times from about 15 up to 60 min. For 4-chloroindole the amount of nitrosated products decreased with increasing incubation times. In all cases the responses in the Ames test paralleled the amounts of nitrosated products. The stabilities of the nitrosated products of the indole compounds were investigated at pH 2 and 8. Both mutagenicity data and measurements by high-performance liquid chromatography using a photohydrolysis detector indicated that the nitrosation products of indole-3-acetonitrile, indole-3-carbinol and indole were more stable at pH 8 than at pH 2. Conversely, nitrosated 4-chloroindole was stable at pH 2 but not at pH 8. The pH 8 chromatograms showed a large nitrite peak. From this we hypothesized that the presence of free nitrite might be responsible for the stability of nitrosated indole-3-acetonitrile, indole-3-carbinol and indole at pH 8. Experiments confirmed the existence of an equilibrium between the nitrosated indole compound and the free indole compound plus nitrite.

Formation of mutagenic N-nitroso compounds in vegetable extracts upon nitrite treatment: a comparison with the glucosinolate content.

More than 30 vegetables were screened for their potential to form biologically active N-nitroso compounds upon treatment with nitrite under acidic conditions. The total N-nitroso content was determined in the nitrite-treated and untreated extracts of the vegetables according to a modified method of Walters et al. (Analyst, Lond. 1978, 103, 1127). All treated extracts contained N-nitroso compounds at levels ranging from 23 to 789 nmol/25 mg dry matter. In the same samples the mutagenic activity was determined using the Salmonella typhimurium assay. About half of the vegetables were found to be mutagenic upon nitrite treatment. (Nitrite-treated extracts were considered to be mutagenic if the number of induced revertants was at least twice as high as that induced by the corresponding untreated extract). The content of different glucosinolates in the dry matter of the vegetables was also determined. Glucosinolates could be detected only in cruciferous vegetables, at levels ranging from 1.8 to 26.0 mumol/g dry matter. Although the nitrite-treated extracts of brassica species contained more N-nitroso compounds and induced more revertants than did other vegetables, there was no significant correlation between these parameters. However, the amounts of N-nitroso compounds formed upon nitrite treatment (expressed per fresh weight) did correlate significantly (P less than 0.01) with the amounts of glucosinolates (r = 0.95). When the glucosinolates were divided into aryl/alkyl- and indolyl-glucosinolates, the significant correlation was maintained for both subgroups (r = 0.93 and 0.95, respectively). From this it can be concluded that glucosinolates are probably involved in the formation of N-nitroso compounds in certain nitrite-treated vegetables.

Inhibitory effect of cheese and some food constituents on mutagenicity generated in Vicia faba after treatment with nitrite.

The inhibitory potential of various food constituents on the mutagenicity generated in fava beans after treatment with nitrite has been investigated. Cheese was able to inhibit this direct-acting mutagenicity. The antimutagenic factor was not extractable from cheese; solvents of different polarity were used for the extraction. Casein, pectin, gelatin, Vicia faba protein and, to a lesser extent, whey protein and starch could also inhibit mutagenicity. A decrease in mutagenicity was always accompanied by a decrease in total N-nitroso content, as measured analytically. The mutagenic principles appeared to bind more strongly onto cheese than onto V. faba. The implications of this finding are discussed.

The role of various food products in the formation of N-nitroso compounds under acidic conditions.

The effect of lettuce cultivars on the nitrosation rate of proline was investigated. The lettuce was analysed for the presence of phenolic compounds. Lettuce and/or fish was incubated with nitrite under acidic conditions, and the incubation mixtures were investigated for the presence of N-nitroso compounds and mutagenic activity. Both volatile N-nitrosamines and mutagenic nonvolatile N-nitrosamines were detected. The formation of mutagenic N-nitroso compounds was also studied in selected cheese products after treatment with nitrite under acidic conditions. No direct relationship was observed between the total N-nitroso content of the samples and the corresponding mutagenicity. The ability of cheese to inhibit the direct mutagenicity occurring in fava beans after treatment with nitrite under acidic conditions was investigated. The antimutagenic factor, possibly casein, in cheese was not extractable with different solvents.

N-nitrosoproline excretion in urine, faeces and milk from cows in relation to feed composition.

The technique of Ohshima and Bartsch for measuring the excretion of N-nitrosoproline (NPRO) was applied in a balance trial with cows to estimate the in-vivo formation of N-nitrosamines. In the first trial, a cow with a high milk production received rations of concentrates and hay with low or high contents of nitrate and free proline. Samples were taken from the urine, faeces and milk. In the second trial, two non-lactating cows were put on hay rations with different nitrate and free proline contents, and samples were taken from the rumenal fluid every 15 min to measure the formation of nitrite from the ingested nitrate. Samples of urine and faeces were also taken. It was found that dried roughage may contain considerable amounts of free proline, and the NPRO formed is related to the nitrate content. No NPRO was found in the urine and faeces when the NPRO content of the hay was low, not even when the ingested nitrate was reduced drastically to nitrite or when the concentration of free proline was high. When NPRO was present in the feed, it was recovered from the urine and faeces but was not transmitted to the milk.

N-nitrosamines in grass silages.

During the fermentation of silages from nitrate-rich grass, the conditions are suitable for the formation of N-nitrosamines. Earlier investigations had shown that only low concentrations of volatile N-nitrosamines were formed. The first ten days of ensilage were investigated. The formation of nitrite was accompanied by the formation of volatile N-nitrosamines. NDMA and NDEA were detected in concentrations of up to about 2 micrograms/kg. After stabilisation of the silage, these concentrations dropped to about 0.6 microgram/kg. Preliminary results are presented concerning the presence of non-volatile N-nitroso compounds. The method of Walters et al. (1980) indicates that non-volatile N-nitroso compounds were present in amounts equivalent to 85 mg NPIP/kg sample.