N-Nitroso compounds in the diet

https://doi.org/10.1016/S1383-5742(99)00015-0Get rights and content

Abstract

N-Nitroso compounds were known almost 40 years ago to be present in food treated with sodium nitrite, which made fish meal hepatotoxic to animals through formation of nitrosodimethylamine (NDMA). Since that time, N-nitroso compounds have been shown in animal experiments to be the most broadly acting and the most potent group of carcinogens. The key role of nitrite and nitrogen oxides in forming N-nitroso compounds by interaction with secondary and tertiary amino compounds has led to the examination worldwide of foods for the presence of N-nitroso compounds, which have been found almost exclusively in those foods containing nitrite or which have become exposed to nitrogen oxides. Among these are cured meats, especially bacon—and especially when cooked; concentrations of 100 μg kg−1 have been found or, more usually, near 10 μg kg−1. This would correspond to consumption of 1 μg of NDMA in a 100-g portion. Much higher concentrations of NDMA (but lower ones of other nitrosamines) have been found in Japanese smoked and cured fish (more than 100 μg kg−1). Beer is one source of NDMA, in which as much as 70 μg l−1 has been reported in some types of German beer, although usual levels are much lower (10 or 5 μg l−1); this could mean a considerable intake for a heavy beer drinker of several liters per day. Levels of nitrosamines have been declining during the past three decades, concurrent with a lowering of the nitrite used in food and greater control of exposure of malt to nitrogen oxides in beer making. There have been declines of N-nitroso compound concentrations in many foods during the past two decades. The small amounts of nitrosamines in food are nonetheless significant because of the possibility—even likelihood—that humans are more sensitive to these carcinogens than are laboratory rodents. Although it is probable that alkylnitrosamides (which induce brain tumors in rodents) are present in cured meats and other potentially nitrosated products in spite of much searching, there has been only limited indirect evidence of their presence.

Introduction

In their examination of the causes of human cancer, Doll and Peto [1]ascribe a major role (60%) to `diet', based on epidemiological observations; they did not particularize which agents in the food might be responsible, or those that might be more important than others. Among the numerous chemical carcinogens that have been detected in human food and drink, N-nitroso compounds are among the most recent and are distinguished by being very potent. They are seldom present by deliberate (if coincidental) addition. Since N-nitroso compounds are easily formed by interaction of a secondary amino compound with a nitrosating agent (commonly a nitrite salt, but also `nitrous gases'), it might have been expected that foods treated with nitrites for colouring, flavouring and preservation would likely contain N-nitroso compounds, but this apparently was not thought of.

The first inkling of such a connection was early in the 1960s when some sheep in Norway died of liver toxicity after feeding on fish meal that had been treated with sodium nitrite 2, 3, 4. It would not have been expected that a nitrosamine, nitrosodimethylamine (NMDA), would be present in the nitrite-treated fish meal, since nitrosation of amines (in this case, dimethylamine and trimethylamine) was considered to take place only in acid solution, and the fish meal milieu was neutral or alkaline. Much later, Keefer and Roller [5]demonstrated that interaction of a secondary amine with nitrite readily took place in alkaline medium in the presence of a carbonyl compound (especially an aldehyde–formaldehyde in the case of fish meal). The presence of considerable quantities of nitrosamines in cured tobacco is also a consequence of nitrosation in non-acid medium [6]. It is noteworthy that the amount of NMDA in the fish meal was large, even considering the potent hepatotoxicity of NDMA, because some of the sheep died.

These observations followed closely the first report of toxicity and carcinogenicity of NDMA by Magee and Barnes [7], after which there was an explosion of interest in the toxicology of the N-nitroso compounds, which are among the simplest of chemical carcinogens. Most of the experiments done subsequently involved the testing of N-nitroso compounds of various structures (several hundred compounds), the results of which provided considerable insight into the mechanisms by which N-nitroso compounds were activated and induced cancer 8, 9. Among the findings were that 40 or more species that have been treated with, e.g., nitrosodiethylamine (NDEA), were all susceptible to its carcinogenic action [10], suggesting strongly that, unlike other types of carcinogen, there was probably no non-susceptible species (e.g., humans). Another finding was that many species differed as to which organ(s) responded to a particular N-nitroso compound, which also depended on the chemical structure of the carcinogen. There were some N-nitroso compounds that appeared to be inactive (i.e., non-carcinogenic), which also gave insight into mechanisms of carcinogenesis by these compounds. It would, however, be misleading to claim that the mechanism by which any N-nitroso compound induces tumors is understood (or, indeed, any other carcinogen), particularly so, since the size of the dose, the frequency of the dose and the route of administration to a certain species (e.g., rats) can change the affected organ from lung to kidney to liver, even changing the target cell from which tumors arise from hepatocytes to endothelial cells [11]. These modulations make it difficult—if not impossible—to predict, based on experiments in rats or mice, which would be the target organ of a particular N-nitroso compound in humans (or, in fact, in any other species). On the other hand, NDEA has induced liver tumors in most of the species in which it has been tested, including rodents, snakes, birds, molluscs and monkeys 10, 12.

Of the several hundred N-nitroso compounds that have been examined, only a few are likely to be encountered by humans outside the laboratory and almost all are nitrosamines, which are stable and not directly acting. As a group, nitrosamines induce tumors in a variety of organs [13], including liver, lung, kidney, bladder, pancreas, esophagus and tongue—depending on the species—but not in skin, brain, colon or bone. On the other hand, the unstable and directly acting nitrosamides (alkylnitrosoureas or alkylnitrosocarbamates) induce tumors of the nervous system (in rats but not in hamsters), stomach, gastrointestinal tract and bone. Many of the tumors induced by N-nitroso compounds are similar to the analogous human tumors and it is tempting to conclude that the latter might be caused by human exposure to N-nitroso compounds. Such human exposures occur in certain work environments, as a consequence of some habits (smoking and chewing tobacco), in the use of particular household products (cosmetics and shampoos), but probably the most widespread exposure is in food and drink. A related exposure (which will not be gone into here) is through the formation of N-nitroso compounds by interaction in the stomach of secondary and tertiary amino compounds (in food or medicines) with nitrite (in food and in saliva) or other nitrosating agents. This endogenous nitrosation, probably an important source of N-nitroso compounds, has been extensively discussed for many years [14]and has been demonstrated experimentally 15, 16, 17. It has also been studied epidemiologically in connection with human cancer [18]. Most human cancers, with the exception of those caused by tobacco use, have no known cause, and widespread exposure to N-nitroso compounds in food provides one link that should not be ignored.

Section snippets

Methods of analysis for N-nitroso compounds

At a meeting in Jamaica in 1968 sponsored by the World Health Organization dealing with distribution and occurrence of carcinogens, it was concluded that, while trace analysis of polynuclear hydrocarbons was possible to hundredths of a part per million or less, detection of nitrosamines was unlikely to be feasible even at 1 ppm, except in unusual circumstances. At about the same time, there was underdevelopment by the Thermo-Electron Corporation (Waltham, Massachusetts, USA), a device (thermal

Occurrence of N-nitroso compounds in food and beverages

The use of sodium nitrite to colour, flavour and preserve (prevention of botulism) meat and fish, replacing sodium nitrate which had been used since time immemorial, led to the search for N-nitroso compounds in such foods, once nitrosamines were classified as carcinogens in 1956 [7]. Many investigators in the field of food toxicology developed a great interest in this topic, as did the meat packing industry. There was a special interest in bacon, because it is so widely consumed. There

Quantitative considerations

The concentrations of N-nitroso compounds in foods are often tiny, and human exposure to them from this source is quite small, but the consequences to people are not to be ignored for two reasons. Firstly, because of the great carcinogenic potency of this group of carcinogens (in animal studies), particularly some of the simpler ones and, secondly, because nitrosamines might well be more effective (dose-for-dose) in humans than they are in experimental rodents, as they are more effective in

The future control of nitrites

The focal point of attempts to reduce the concentration of N-nitroso compounds in foods is the control of nitrites (or other nitrosating agents) which can come into contact with nitrosatable amino compounds. There has been notable success in reducing the concentration of NDMA in beer by controlling the exposure of malt to nitrogen oxides, effectively stopping nitrosation of the alkaloids in the malt. There has also been some success in reducing the amount of nitrite used in meat curing, which

References (50)

  • P.N. Magee et al.

    The production of malignant primary hepatic tumours in the rat by feeding dimethylnitrosamine

    Br. J. Cancer

    (1956)
  • W. Lijinsky, Chemistry and Biology of N-nitroso Compounds, Cambridge Univ. Press, Cambridge, UK,...
  • R. Preussmann, B.W. Stewart, N-Nitroso carcinogens, in: C.E. Searle (Ed.), Chemical Carcinogens, American Chemical...
  • P. Bogovski et al.

    Animal species in which N-nitroso compounds induce cancer

    Int. J. Cancer

    (1981)
  • W. Lijinsky et al.

    Carcinogenesis by nitrosamines and azoxyalkanes by different routes of administration to rats

    Biomed. Env. Sci.

    (1989)
  • W. Lijinsky

    Lifespan and cancer: the induction time of tumors in diverse animal species treated with nitrosodiethylamine

    Carcinogenesis

    (1993)
  • W. Lijinsky, In vivo testing for carcinogenicity, in: P.L. Grover, C.S. Cooper (Eds.), Chemical Carcinogenesis and...
  • W. Lijinsky et al.

    Nitrosamines as environmental carcinogens

    Nature (London)

    (1970)
  • J. Sander et al.

    Untersuchungen über die Enstehung von Nitrosaminen und Nitrosamiden im Menschen, im Tier und in Nahrungsmitteln

    Zbl. Bakt. Hyg. I Abt. Orig. B

    (1972)
  • J. Sander et al.

    Induktion maligner Tumoren bei Ratten durch gleichzeitige Verfütterung von Nitrit und sekundaren Aminen

    Z. Krebsforsch.

    (1969)
  • W. Lijinsky, H.W. Taylor, Nitrosamines and their precursors in food, Cold Spring Harbor Symposium on the Origins of...
  • S. Preston-Martin et al.

    Maternal consumption of cured meats and vitamins in relation to pediatric brain tumors

    Cancer Epidemiol., Biomarkers Prev.

    (1996)
  • D.H. Fine, F. Rufeh, Description of the thermal energy analyzer for N-nitroso compounds, IARC Scientific Publications...
  • C.L. Walters, Group analysis of N-nitroso compounds, Das Nitrosamin-Problem, Verlag Chemie, Weinheim, 1983, pp....
  • S. Lu et al.

    A new N-nitroso compound, N-3-methylbutyl-N-1-methylacetonylnitrosamine, in corn bread inoculated with fungi

    Scientia Sinica

    (1979)
  • Cited by (0)

    View full text