Nitrates, nitrites, and N-nitroso compounds (EHC 5, 1978)

INTERNATIONAL PROGRAMME ON CHEMICAL SAFETY

ENVIRONMENTAL HEALTH CRITERIA 5

Nitrates, Nitrites and N-Nitroso Compounds

This report contains the collective views of an international group of
experts and does not necessarily represent the decisions or the stated
policy of either the World Health Organization or the United Nations
Environment Programme

Published under the joint sponsorship of the United Nations
Environment Programme and the World Health Organization

World Health Organization
Geneva, 1978

ISBN No. 92 4 154065 6

Publications of the World Health Organization enjoy copyright
protection in accordance with the provisions of Protocol 2 of the
Universal Copyright Convention. For rights of reproduction or
translation of WHO publications, in part or in toto, application
should be made to the Office of Publications, World Health
Organization, Geneva, Switzerland. The World Health Organization
welcomes such applications.

The designations employed and the presentation of the material in
this publication do not imply the expression of any opinion whatsoever
on the part of the Secretariat of the World Health Organization
concerning the legal status of any country, territory, city or area or
of its authorities, or concerning the delimitation of its frontiers or
boundaries.

The mention of specific companies or of certain manufacturers'
products does not imply that they are endorsed or recommended by the
World Health Organization in preference to others of a similar nature
that are not mentioned. Errors and omissions excepted, the names of
proprietary products are distinguished by initial capital letters.

CONTENTS

ENVIRONMENTAL HEALTH CRITERIA FOR NITRATES, NITRITES AND N-NITROSO
COMPOUNDS

1. SUMMARY AND RECOMMENDATIONS FOR FURTHER STUDIES

1.1. Summary
1.1.1. Analytical methods
1.1.1.1 Nitrates and nitrites
1.1.1.2 N-nitroso compounds
1.1.2. Sources and occurrence in the environment
1.1.2.1 Nitrates and nitrites
1.1.2.2 N-nitroso compounds
1.1.3. Metabolism
1.1.3.1 Nitrates and nitrites
1.1.3.2 N-nitroso compounds
1.1.4. Experimental studies in animals
1.1.4.1 Nitrates and nitrites
1.1.4.2 N-nitroso compounds
1.1.5. Epidemiological studies
1.1.5.1 Nitrates and nitrites
1.1.5.2 N-nitroso compounds
1.1.6. Evaluation of health risks
1.1.6.1 Nitrates and nitrites
1.1.6.2 N-nitroso compounds
1.2. Recommendations for further studies
1.2.1. Analytical problems
1.2.1.1 Nitrates and nitrites
1.2.1.2 N-nitroso compounds
1.2.2. Sources and levels in the environment
1.2.2.1 Nitrates and nitrites
1.2.2.2 N-nitroso compounds
1.2.3. Metabolism
1.2.3.1 Nitrates and nitrites
1.2.3.2 N-nitroso compounds
1.2.4. Experimental studies
1.2.5. Epidemiological and clinical studies
1.2.5.1 Nitrates and nitrites
1.2.5.2 N-nitroso compounds

2. CHEMISTRY AND ANALYTICAL METHODS

2.1. Chemical properties and reactions
2.1.1. Nitrates and nitrites
2.1.2. N-nitroso compounds
2.1.3. Formation of N-nitroso compounds in vitro
2.1.4. The effects of other substances on the formation of
N-nitroso compounds
2.2. Analytical methods
2.2.1. Nitrates and nitrites
2.2.2. N-nitroso compounds

3. SOURCES OF NITRATES, NITRITES AND N-NITROSO COMPOUNDS IN AIR,
WATER, SOIL, AND FOOD

3.1. Natural occurrence
3.1.1. Nitrates and nitrites
3.1.2. N-nitroso compounds
3.2. Sources related to man's activities
3.2.1. Nitrates and nitrites
3.2.1.1 Fertilizers
3.2.1.2 Animal wastes
3.2.1.3 Municipal, industrial, and transport
wastes
3.2.1.4 Deliberate addition of nitrates and
nitrites to food
3.2.2. N-nitroso compounds
3.2.2.1 Food
3.2.2.2 Tobacco
3.2.2.3 Industrial uses

4. TRANSPORT AND TRANSFORMATION IN ENVIRONMENTAL AND BIOLOGICAL
MEDIA

4.1. Nitrogen Cycle
4.2. Transformation in food
4.2.1. Reduction of nitrates to nitrites
4.2.2. Formation and degradation of N-nitroso compounds
4.3. Formation of N-nitroso compounds from drugs and pesticides
4.4. Formation of N-nitroso compounds in animal organisms
4.4.1. Formation of N-nitroso compounds in simulated
gastric juice
4.4.2. Formation of N-nitroso compounds in vivo
4.5. Formation of N-nitroso compounds by microorganisms
4.6. The effects of other chemicals on the formation of
N-nitroso compounds

5. ENVIRONMENTAL LEVELS AND EXPOSURES

5.1. Nitrates and nitrites
5.1.1. Ambient air
5.1.2. Water
5.1.3. Selected foods
5.1.4. Estimate of general population exposure
5.2. N-nitroso compounds
5.2.1. Ambient air
5.2.2. Water
5.2.3. Selected foods
5.2.4. Tobacco and tobacco smoke
5.2.5. Estimate of general population exposure
5.2.6. Occupational exposure to N-nitroso compounds

6. METABOLISM OF NITRATES, NITRITES, AND N-NITROSO COMPOUNDS

6.1. Gastrointestinal absorption
6.1.1. Nitrates and nitrites
6.1.2. N-nitroso compounds
6.2. Biotransformation and elimination
6.2.1. Nitrates and nitrites
6.2.2. N-nitroso compounds

7. EXPERIMENTAL ANIMAL STUDIES ON THE EFFECTS OF NITRATES, NITRITES,
AND N-NITROSO COMPOUNDS

7.1. Nitrates and nitrites
7.1.1. Acute and subacute toxicity studies
7.1.2. Chronic toxicity and carcinogenicity studies
7.1.3. Embryotoxicity
7.1.4. Mutagenicity
7.1.5. Interaction with nutritional factors
7.2. N-nitroso compounds
7.2.1. Acute and subacute toxicity studies
7.2.2. Carcinogenicity
7.2.2.1 Interspecies variation in response
7.2.2.2 Intraspecies variation in response
7.2.2.3 Dose-response relationships of N-nitroso
compounds
7.2.2.4 Tumour induction by combined
administration of nitrites, and amines or
amides
7.2.2.5 Dose-response relationship for
combinations of nitrites and amines
7.2.2.6 Transplacental carcinogenesis
7.2.2.7 Morphological studies
7.2.2.8 Biochemical mechanisms
7.2.2.9 Interaction with various chemical factors
7.2.2.10 Miscellaneous modifying factors
7.2.3. Embryotoxicity and teratogenicity
7.2.4. Mutagenicity

8. EFFECTS OF NITRATES, NITRITES, AND N-NITROSO COMPOUNDS ON MAN

8.1. Nitrates and nitrites
8.1.1. Epidemiological studies
8.1.1.1 Exposure through water
8.1.1.2 Exposure through vegetables
8.1.1.3 High accidental exposures
8.1.1.4 Ambient air exposures
8.1.2. Factors involved in susceptibility to nitrates
8.1.3. Dose-response relationships for nitrates and
nitrites
8.2. N-nitroso compounds

9. EVALUATION OF HEALTH RISKS TO MAN FROM EXPOSURE TO NITRATES,
NITRITES, AND N-NITROSO COMPOUNDS

9.1. Nitrates and nitrites
9.1.1. General considerations
9.1.2. Assessment of health risks
9.2. N-nitroso compounds
9.2.1. General considerations
9.2.2. Assessment of health risks
9.3. Reduction of exposure th

REFERENCES

NOTE TO READERS OF THE CRITERIA DOCUMENTS

While every effort has been made to present information in the
criteria documents as accurately as possible without unduly delaying
their publication, mistakes might have occurred and are likely to
occur in the future. In the interest of all users of the environmental
health criteria documents, readers are kindly requested to communicate
any errors found to the Division of Environmental Health, World Health
Organization, Geneva, Switzerland, in order that they may be included
in corrigenda which will appear in subsequent volumes.

In addition, experts in any particular field dealt with in the
criteria documents are kindly requested to make available to the WHO
Secretariat any important published information that may have
inadvertently been omitted and which may change the evaluation of
health risks from exposure to the environmental agent under
examination, so that the information may be considered in the event of
updating and re-evaluating the conclusions contained in the criteria
documents.

WHO TASK GROUP ON ENVIRONMENTAL HEALTH CRITERIA FOR NITRATES,
NITRITES, AND N-NITROSO COMPOUNDS

Lyons, France, 16-21 February 1976

Participants

Members

Dr. K. R. Bulusu, National Environmental Engineering Research
Institute, Nagpur, India

Mr T. J. Coomes, Food Chemistry, Composition and Safety, Ministry of
Agriculture, Fisheries and Food, London, England (Rapporteur)

Dr R. Kroes, Department of Oncology, National Institute of Public
Health, Bilthoven, Netherlands

Dr W. Lijinsky, Frederick Cancer Research Center, Frederick, MD, USA

Dr S.S. Mirvish, Eppley Institute for Research in Cancer, The
University of Nebraska College of Medicine, Omaha, NE, USA

Professor M. Nikonorow, Department of Food Research, State Institute
of Hygiene, Warsaw, Poland

Dr T. Petrova-Vergieva, Centre of Hygiene, Medical Academy, Sofia,
Bulgaria

Dr R. Preussmann, Institute for Toxicology and Chemotherapy, National
Cancer Research Centre, Heidelberg, Federal Republic of Germany

Dr P. Schmidt, Ministry of Health, Prague, Czechoslovakia

Professor K. Symon, Centre of General and Community Hygiene, Institute
of Hygiene and Epidemiology, Prague, Czechoslovakia (Vice
Chairman)

Professor R. Truhaut, Toxicological Research Centre, Faculty of
Pharmacy and Biology, Paris, France (Chairman)

Representatives of other organizations

Mme M.-Th. van der Venne, Commission of the European Communities,
Health Protection Directorate, Luxembourg

Mr G. Dorin, Natural Resources and Pollution Control Division,
Organization for Economic Cooperation and Development, Paris,
France

Secretariat

Dr L. Griciute, Chief, Unit of Environmental Carcinogenesis, IARC,
Lyons, France

Dr A. Kolbye, Bureau of Foods, US Food and Drug Administration,
Washington, DC, USA

Dr F. C. Lu, Chief, Food Additives, World Health Organization, Geneva,
Switzerland (Secretary)

Dr R. Montesano, Unit of Chemical Carcinogenesis, IARC, Lyons, France

Dr I. C. Munro, Toxicology Research Division, Health Protection
Branch, Department of National Health and Welfare, Ottawa, Canada

Mr E. A. Walker, Unit of Chemical Carcinogenesis, IARC, Lyons, France

^a Unable to attend:

Dr E. Arrhenius, Director for Experimental Biology, Wallenberg
Laboratory, Lilla Frescati, Stockholm, Sweden

Dr V. Okulov, Petrov Research Institute of Oncology, Leningrad,
USSR

List of Abbreviations

DMN N-methyl- N-nitrosomethanamine ( N-nitrosodimethylamine,
dimethylnitrosamine)

DEN N-ethyl- N-nitrosoethanamine ( N-nitrosodiethylamine,
diethylnitrosamine)

DMA N-methylmethanamine (dimethylamine)

DEA N-ethylethanamine (diethylamine)

TMA N,N-dimethylmethanamine (trimethylamine)

TMAO trimethyloxamine (trimethylamine oxide)

ENVIRONMENTAL HEALTH CRITERIA FOR NITRATES, NITRITES, AND N-NITROSO
COMPOUNDS

A WHO Task Group on Environmental Health Criteria for Nitrates,
Nitrites and N-nitroso compounds met in Lyons from 16 to 20 February
1976. Dr Higginson, Director of the International Agency for Research
on Cancer opened the meeting on behalf of the Director-General. The
Task Group reviewed and amended the second draft of the criteria
document and made an evaluation of the health risks from exposure to
these compounds.

The preparation of the first draft of the criteria document was
based on national reviews of health effects research on nitrates,
nitrites, and N-nitroso compounds received from the national focal
points for collaboration in the WHO Environmental Health Criteria
Programme in Bulgaria, Canada, Czechoslovakia, the Federal Republic of
Germany, Netherlands, New Zealand, Poland, the United Kingdom, the
USA, and the USSR. Dr I. C. Munro, Toxicological Research Division,
Health Protection Branch of the Department of National Health and
Welfare, Ottawa, Ontario, Canada, prepared both the first draft and
the second draft which took into account the comments received from
the national focal points in Bulgaria, Canada, Czechoslovakia,
Finland, Japan, New Zealand, Poland, Sweden, USA, and the USSR; from
the United Nations Industrial Development Organization (UNIDO), the
Food and Agriculture Organization of the United Nations (FAO), the
United Nations Educational, Scientific and Cultural Organization
(UNESCO), the International Atomic Energy Agency (IAEA), the Health
Protection Directorate of the Commission of the European Communities
(CEC), and from the International Federation of Pharmaceutical
Manufacturers' Associations (IFPMA).

At the request of the Secretariat, comments were also received
from Dr S. Oden, Agricultural College, Department of Soil Science,
Division of Ecochemistry, Uppsala, Sweden.

The collaboration of these national institutions, international
organizations and individual experts is gratefully acknowledged.
Without their assistance this document could not have been completed.
The collaboration of the International Agency for Research on Cancer
in the preparation of the document and in acting as host to the Task
Group is also greatly appreciated.

The Secretariat wishes to thank Mr A. W. Kenny, Department of the
Environment, London, England and Mr D. A. H. Price, Chorley Wood,
Herts, England for their advice in the preparation of some sections of
the document and Dr Munro for his help in the final phases of editing.

This document is based primarily on national contributions and on
original publications listed in the reference section. In addition,
some recent publications reviewing the environmental health aspects of

nitrates, nitrites and N-nitroso compounds have been used. These
include reviews and symposia by the US National Academy of Sciences,
Washington, DC (Committee on Nitrate Accumulation, 1972), the US
Department of Health, Education and Welfare (1970), the US
Environmental Protection Agency,^a the International Agency for
Research on Cancer (Bogovski 8: Walker, 1974; Bogovski et al., 1972a;
Walker et al., 1970), Druckrey et al. (1967), Lee (1970a), Magee &
Barnes (1967), Magee et al. (1976), Montesano & Bartch (1976), and Sen
(1974).

Details concerning the WHO Environmental Health Criteria
Programme including the definition of some terms frequently used in
the documents may be found in the general introduction to the
Environmental Health Criteria Programme published together with the
environmental health criteria document on mercury (Environmental
Health Criteria 1 -- Mercury, World Health Organization, Geneva,
1976).

^a "US Environmental Protection Agency (1976) Scientific and
technical assessment report on nitrosamines. Preprint of document
submitted for publication in the STAR series, Washington DC,
Office of Reserch and Development, 210 pp.

1. SUMMARY AND RECOMMENDATIONS FOR FURTHER STUDIES

1.1 Summary

1.1.1 Analytical methods

1.1.1.1 Nitrates and nitrites

The Task Group recognized that interpretation of the results of
analyses of nitrates and nitrites in environmental and biological
media would vary according to the analytical methods employed (e.g.
spectrophotometry, spectrofluorimetry, nitrate specific electrode) and
that this would make meaningful comparisons of much data in the
literature difficult.

It was considered, however, that reported figures for water and
for meat products could be compared as far as the assessment of health
hazards was concerned. The Group also noted that, although the
principle underlying a particular method of analysis may well be the
same for a variety of substrates, difficulties usually arise during
the sampling, extraction, and clean-up procedures which vary in
complexity according to the nature of the substrate being analysed.

1.1.1.2 N-nitroso compounds

The detection and estimation of volatile N-nitroso compounds is
complicated by the following basic issues: they are likely to be
present in environmental media in concentrations of only 1 part in
10⁹ parts; they occur in a complex matrix in food and biological
samples many components of which will contain nitrogen and will react
in a similar manner chemically; they must be isolated from this matrix
in a form that permits their estimation and unequivocal
identification. Whilst removal from the matrix is easy in the case of
N-nitroso compounds of low molecular weight because they are steam-
volatile, this approach cannot be used for non-volatile N-nitroso
compounds. Analysis of these compounds has received scant attention so
far, although some work on clean-up procedures exists, and the use of
the liquid chromatography technique is now under investigation
following its successful application to the separation of model
mixtures of N-nitroso- N-alkyl ureas and the analogous urethanes.

Irrespective of the isolation techniques used, the quantitative
determination of N-nitroso compounds requires a concomitant positive
identification of the molecular species determined. For this reason,
the preferred method of analysis, gas-liquid chromatography allied to
a nitrogen-sensitive detector, must be linked to high-resolution mass
spectrometry to confirm the presence of N-nitroso compounds. Results
should be considered positive only when or if mass-spectroscopic
techniques have confirmed their unequivocal presence.

1.1.2 Source and occurrence in the environment

1.1.2.1 Nitrates and nitrites

Nitrates are present naturally in soils, waters, all plant
materials, and in meats. They are also found in small concentrations
(1-40 µg/m³) in air as a result of air pollution. Levels in
cultivated soils and thus, levels in water, (which normally do not
exceed 10 mg/litre) may be increased by the use of commercial
nitrogenous fertilizers and by the return of wastes, derived from
animal husbandry or other sources, to the soil. Nitrate contents of
crops are influenced by the plant species, by genetic and
environmental factors, and by agricultural management practices. In
certain crops the levels may be very high (1000 mg/kg or more).

Nitrites are formed in nature by the action of nitrifying
bacteria as an intermediate stage in the formation of nitrates, but
concentrations in plants and water are usually very low. However,
microbiological conversion of nitrate to nitrite may occur during the
storage of fresh vegetables, particularly at room temperature, when
nitrite concentrations may rise to exceptionally high levels (about
3600 mg/kg dry weight). Both nitrates and nitrites are widely used in
the production and preservation of cured meat products and of some
fish. Such uses, which are controlled by law in many countries, are
considered vital for the prevention of botulism caused by the growth
of the toxin-producing strains of Clostridium botulinum that are
sometimes present in raw meat and that may persist in cooked meats.
The weekly intake of nitrates by a member of the general population in
England or in the USA has been roughly estimated to average about
400-500 mg but these figures cannot be applied generally because of
variations in feeding habits and in the nitrate concentrations in food
and water.

1.1.2.2 N-nitroso compounds

Low concentrations of N-nitroso compounds have been detected in
air, water, and food, notably in nitrite-treated meat products and
certain fish products. In most cases, the concentrations found in food
have been in the µg/kg range. No effective estimate of general
population exposure to N-nitroso compounds can be made on the basis
of these limited data. N-methyl- N-nitrosomethanamine
( N-nitrosodimethylamine, DMN) has been detected in urban air
samples and the presence of N-nitroso compounds, tentatively
identified as N-nitroso derivatives of some pesticides, has been
reported in samples from water treatment plants and river water in
the USA.

The in vivo formation of N-nitroso compounds from nitrates or
nitrites and amines or amides has been demonstrated in experimental
animals and in one case in man.

1.1.3 Metabolism

1.1.3.1 Nitrates and nitrites

In normal healthy individuals, nitrates and nitrites are rapidly
absorbed from the gastro-intestinal tract. Absorbed nitrite reacts
with haemoglobin to form methaemoglobin which, in adults, is rapidly
converted to oxyhaemoglobin by reducing systems such as NADH-
methaemoglobin reductase. In infants up to three months old and in
very young animals this enzyme system is not completely developed.
Under these conditions, the methaemoglobin formed may increase in the
body resulting in a characteristic clinical condition
(methaemoglobinaemia). Microorganisms present in the food and
gastrointestinal tract of very young infants may convert nitrates to
nitrites and thus exacerbate the problem in this age group. In healthy
individuals, absorbed nitrates are rapidly excreted by the kidneys.

1.1.3.2 N-nitroso compounds

Published information on the absorption, metabolism, and
elimination of N-nitroso compounds is limited. In cases where
experimental animal data are available, they demonstrate that
N-nitroso compounds are rapidly absorbed from the gastrointestinal
tract and that their biological half-time appears to be less than
24 h. A part of some compounds may be excreted unchanged via the
kidney, or even exhaled, but the greater part is metabollically
transformed (hydroxylation, chain shortening, ring-opening etc.) and
several metabolites of N-nitroso compounds have been identified
in urine. Significant amounts of some compounds such as DMN may be
degraded completely and the resulting carbon dioxide exhaled. The
extent of such degradation varies depending on the structure of the
compound and the animal species involved.

1.1.4 Experimental studies in animals

1.1.4.1 Nitrates and nitrites

The major effect of nitrates and nitrites is the induction of
methaemoglobinaemia, mostly readily observed in very young animals.
Most experimental work has been connected with this problem although
embryotoxic effects resulting in prenatal mortality, resorptions, and
decreased birthweights have been noted in rat pups whose mothers
received drinking water containing sodium nitrite. In adult rats given
drinking water containing nitrite for 24 months, methaemoglobin levels
were elevated but not to the point of producing overt toxic effects.
Animal species studied appeared to be fairly resistant to the
induction of methaemoglobinaemia by nitrites, since high doses were
required to induce even minimal changes. However, very young animals
have not been studied extensively or sufficiently. Nitrates and
nitrites do not appear to be carcinogenic but nitrite mutagenicity has
been demonstrated in several non-mammalian test systems.

1.1.4.2 N-nitroso compounds

In experimental animals, the most important biological actions of
N-nitroso compounds are their carcinogenicity and teratogenicity.

The carcinogenic action of N-nitroso compounds in animals is
known to occur in many different organs. In general, the routes of
administration do not influence the localization of the tumours.
However, both dose level and dose rate may affect the organ involved
and the type of tumour produced. Specific structural changes in both
dialkyl nitrosamines and cyclic nitrosamines affect their
carcinogenicity. N-nitroso compounds have been shown to be
transplacentally carcinogenic, when given to animals in the second
part of gestation, irrespective of the route of administration.
Carcinogenicity following the combined administration of amines or
amides and nitrites to animals has also been reported indicating the
in vivo formation of N-nitroso compounds.

The mutagenic action of nitrosamides, noted in test systems,
differs from that of nitrosamines in that the first group of compounds
has been found to be mutagenic in almost all test systems, whereas
nitrosamines seem only to be active in systems where metabolic
activation occurs.

Nitrosamines are known to have toxic and sometimes lethal effects
on animal embryos, whereas nitrosamides cause malformations in several
organs and systems.

1.1.5 Epidemiological studies

1.1.5.1 Nitrates and nitrites

Adults do not appear to be harmed directly by exposure to the
prevailing concentrations of nitrates and nitrites in the environment,
although some recent studies have indicated that nitrate aerosols in
the ambient air may act as respiratory irritants. However infants and
very young children are particularly susceptible to the induction of
methaemoglobinaemia by nitrates and nitrites, ingested in water and
food, and several cases of illness and death have been reported. In
most cases of methaemoglobinaemia, well-water containing high
concentrations of nitrates had been used in the reconstitution of
infant dried milk preparations. Most instances have been associated
with water containing more than 90 mg per litre but a few cases of
methaemoglobinaemia in infants have been associated with the
consumption of water containing less than 50 mg per litre. Cases of
methaemoglobinaemia in babies fed with spinach purée or carrot juice
(both of which may contain very high levels of nitrates) have been
reported, but there are too few data to establish dose-response
relationships.

1.1.5.2 N-nitroso compounds

So far, correlations have not been established that link cancer
in man with exposure to N-nitroso compounds or their precursors, but
the possible role of N-nitroso compounds and in particular their
in vivo formation in the development of nasopharyngeal, oesophageal,
and stomach cancer has been suggested.

1.1.6 Evaluation of health risks

1.1.6.1 Nitrates and nitrites

Epidemiological and clinical studies on man have shown that the
main toxic manifestation resulting from the ingestion of nitrates and
nitrites is methaemoglobinaemia. This has been confirmed by
experimental animal studies. On the basis of available data, the Task
Group concluded that the prevailing concentrations of nitrates and
nitrites in food and water did not constitute a health risk for adult
members of the general population and older children, but that the
risk may be higher for infants under 6 months of age and particularly
under 3 months. For this reason, the Group recommended that infant
dried milk preparations should be reconstituted with low-nitrate water
(at least below 45 mg/litre) and that low-nitrate vegetables should be
used in baby foods.

Also, the use of nitrates and nitrites as food additives should
be reduced to the minimum, and avoided in fresh meat and fish. Nitrate
levels in public water supplies should comply with the tentative limit
of 45 mg/litre recommended by the 1971 International Standards for
Drinking Water.

1.1.6.2 N-nitroso compounds

Although the precursors of N-nitroso compounds (nitrites,
amines, and amides) are known to be widely distributed in various
environmental media, information concerning N-nitroso compounds is
limited. However, they are known to be present in certain foods and
experimental animal studies have shown that they are formed in the
body from a variety of precursors. This may also occur in man.

N-nitroso compounds are carcinogenic in a wide range of animal
species, most are mutagenic in test systems, and some have been shown
to be teratogenic in animals.

Although there is no epidemiological or clinical evidence at
present, it is highly probable that these compounds may also be
carcinogenic in man. A quantitative estimation of the carcinogenic
risk to man associated with different levels of exposure is not
possible, at this time, because of inadequate data. For these reasons,
exposure to N-nitroso compounds and their precursors, (nitrites,
amines, and amides) should be kept as low as practically achievable.

1.2 Recommendations for Further Studies

1.2.1 Analytical problems

1.2.1.1 Nitrates and nitrites

The major need is for standardization of analytical methods. At
present, it is difficult to compare the studies reported by one
laboratory with those reported by another. While in many instances the
principle underlying the determination is the same for many of the
studies reported, the large variety of substrates containing nitrates
and nitrites gives rise to difficulties with respect to sampling,
extraction, and cleanup procedures. Further efforts are needed to
standardize these analytical procedures on an international basis. To
this end, the efforts of international and regional groups should be
supported.

1.2.1.2 N-nitroso compounds

The principal problem associated with the determination of
N-nitroso compounds in food and other environmental media results
from interference by other components of the substrates. At present,
positive identification of N-nitroso compounds can be made only by
mass spectroscopie techniques. Since such techniques are expensive and
not generally available, alternative methods are required. In
addition, methods for the detection and determination of nonvolatile
N-nitroso compounds should be developed further.

1.2.2 Sources and levels in the environment

1.2.2.1 Nitrates and nitrites

Research should be undertaken to find acceptable substitutes for
nitrates and nitrites in the preservation of certain foods such as
canned meats.

National surveys of nitrate levels in soils, water, plant
materials, foods, especially meat and milk products, and air are
required together with quantitative data concerning other factors
considered to have an effect on these levels. Similar information on
nitrite levels is required with particular reference to foods and to
areas where significant microbiological reduction of nitrates is
likely.

It is important that levels determined in survey work of this
nature should be reported on the basis of standardized analytical
methods to facilitate the eventual comparison of data from all
sources. National authorities should be encouraged to publish survey
data or to communicate them to the World Health Organization.

1.2.2.2 N-nitroso compounds

National surveys of food, air, and water for the presence of
volatile and, where possible, nonvolatile N-nitroso compounds are
required and any results reported should be confirmed by mass
spectroscopy. More studies are needed on the chemical conditions under
which N-nitroso compounds are formed (e.g. in mixtures of nitrites
and amines or amides). The use of ascorbic acid for the prevention of
nitrosamine formation and the inhibitory or catalytic effect of food
constituents on the formation of N-nitroso compounds also require
studies. The role of oxides of nitrogen as possible nitrosating agents
should be investigated in relation to the occurrence of N-nitroso
compounds in the environment (e.g. in the ambient and workroom air).

1.2.3 Metabolism

1.2.3.1 Nitrates and nitrites

Further work on the influence of ascorbic acid and other
ingredients of the stomach contents on the metabolism of nitrates and
nitrites is required. The treatment of infant dried milk formulae with
ascorbic acid or by the introduction of Lactobacilli to prevent
nitrate reduction should also be studied.

Other areas requiring investigation include: the influence of
gastro-enteric disease on the development of methaemoglobinaemia; the
influence of the total gut flora on nitrate metabolism in vivo; the
relationship between ingested nitrate and salivary nitrate and nitrite
levels.

1.2.3.2 N-nitroso compounds

More knowledge should be gained on the in vivo formation of
N-nitroso compounds in man and the factors involved. Studies
comparing the metabolism of N-nitroso compounds in experimental
animals and in man are considered to be of the greatest importance.

1.2.4 Experimental studies

Further research on the biological action of N-nitroso
compounds should concentrate on dose-response relationships especially
at low levels, and on their combined effects with other carcinogens,
and environmental pollutants. The influence of nutritional factors on
the carcinogenicity of N-nitroso compounds should be studied in more
detail.

More inhalation studies are necessary to assess the importance of
the recently reported occurrence of N-nitroso compounds in air and
further research is needed on the quantitative aspects of the
mutagenic activity of N-nitroso compounds and its possible
significance for man.

1.2.5 Epidemiological and clinical studies

1.2.5.1 Nitrates and nitrites

With respect to the adverse effects of nitrates and nitrites on
infants, there is a need to investigate the relationship between
methaemoglobinaemia and sudden infant death and to make further
studies on the role of gastroenteritis in increasing infant
susceptibility to nitrate poisoning. The role of acidified milk
preparations and Lactobacilli in protecting infants against
methaemoglobinaemia, and the possible protective role of ascorbic acid
fortification of infant milk preparations should also be elaborated.

1.2.5.2 N-nitroso compounds

Prospective and retrospective epidemiological studies in man,
exposed to N-nitroso compounds, are needed. Efforts should be made
to determine whether cancers, that are peculiar to special areas of
the world, might be due to exposure to N-nitroso compounds. Chemical
analyses of the environment for N-nitroso compounds and their
precursors should be carried out in conjunction with these
epidemiological studies.

2. CHEMISTRY AND ANALYTICAL METHODS

2.1 Chemical Properties and Reactions

2.1.1 Nitrates and nitrites

The nitrate ion (NO₃^-) is the conjugate base of nitric acid
(HNO₃). Nitric acid is a strong acid (pK_a = -1.37) which
dissociates in water yielding nitrate ions and hydroxonium ions
(H₃O⁺). Salts of nitric acid (nitrates) are readily soluble in
water with the exception of the basic nitrates of mercury and bismuth.

The nitrite ion is the conjugate base of nitrous acid (HNO₂)
which is a weak acid (pk_a = 3.37) and exists only in cold dilute
aqueous solution because it decomposes readily to give water and
dinitrogen trioxide (N₂O₃) or nitric acid, nitric oxide (NO), and
water. Salts of nitrous acid (nitrites) are much more stable than the
acid itself and are readily soluble in water with the exception of
silver nitrite.

In the environment (e.g. surface waters, soil) both nitrite and
nitrate ions can be formed from the ammonium ion (NH₄⁺) in a two
step biological oxidation (nitrification) process:

2 NH₄⁺ + 2OH^- + 3O₂ <=> 2 NO₂^- + 2H⁺ + 4H₂O (1)

2 NO₂^- + O₂ <=> 2 NO₃^- (2)

These two reactions are mediated by different microorganisms:
reaction (1) by an aerobic chemolithotroph Nitrosomonas; reaction
(2) by Nitrobacter which obtains almost all its energy from the
oxidation of nitrites.

Higher plants assimilate nitrite from the soil by (a) reduction
of nitrate to nitrite which is catalysed by nitrate reductase (NADPH)
(1.6.6.3), and (b) reduction of nitrite to ammonia catalysed by
nitrite reductase (1.7.99.3). Bacteria of many kinds can also reduce
nitrate to nitrite. However, because nitrite is easily oxidised to
nitrate the concentration of nitrites in environmental media such as
surface waters is usually very low (about 1 mg/litre) even when the
nitrate concentration is high (50-100 mg/litre).

These biochemical reactions are a part of the nitrogen cycle
which is further discussed in section 3.1.

2.1.2 N-nitroso compounds

N-nitroso compounds have a general structure

R₁
\ N-N=O
R₂/

They can be divided into two classes with different chemical
properties (Druckrey et al., 1967; Fridman et al., 1971):

(1) nitrosamines where R₁ and R₂ are alkyl or aryl groups;
and

(2) nitrosamides where R₁ is an alkyl or aryl group, and R₂
is an acyl group.

Nitrosamines are generally stable compounds that only slowly
decompose in the light or in aqueous acid solutions.

In contrast, nitrosamides are much less stable in aqueous acids
and unstable in basic solutions. Examples of nitrosamides are N-
alkyl- N-nitrosoureas (3) and N-alkyl- N-nitrosourethanes (4).

R - N - C - NH₂ (3)
' "
NO O

R - N - C - OC₂ H₅ (4)
' "
NO O

The physical properties of N-nitroso compounds vary widely depending
on the substituent groups. Some like N-methyl- N-nitrosomethanamine
(dimethylnitrosamine, DMN) are oily liquids miscible with polar
solvents. Some are solids e.g. N-nitroso- N-phenylbenzenamine
(diphenylnitrosamine) and are only slightly soluble in ethanol and
practically insoluble in water. The lipid/water partition coefficients
vary widely. Nitrosamines show ultraviolet absorption peaks in water
at 230-240 nm and 330-350 nm. For nitrosamides, the long-wavelength
absorption peak in water is at 390-420 nm. Some N-nitroso compounds
are volatile (Mirvish, 1975, 1976; Sen, 1974). Physical properties of
N-nitroso compounds have been listed by Druckrey et al. (1967),
Fieser & Fieser (1967), and Weast (1976).

Nitrosamines may react by "transnitrosation" i.e. as nitrosating
agents to nucleophilic^a species (Buglass et al., 1974). This
reaction may have important biological implications.

^a i.e. electron-rich.

2.1.3 Formation of N-nitroso compounds in vitro

The formation of N-nitroso compounds from amines and nitrites
has been reviewed by Mirvish (1975), Sander (1971a, 1971b), and Sander
& Schweinesberg (1972).

For example, for N-methylmethanamine (dimethylamine) (DMA) and
sodium nitrite in dilute hydrochloric acid solutions, nitrositation is
considered to proceed as follows (Mirvish, 1970):

NaNO₂ + HCl <=> HNO₂ + NaCl (5)

2HNO2 <=> N₂O₃ + H₂O (6)

(CH₃)₂NH + N₂O₃ <=> (CH₃)₂HN - NO + NO₂^- (7)

The reaction rate depends on the concentration of nonionized
amine and nitrous acid. At pH > 1, the main nitrosating agent is
dinitrogen trioxide which is formed reversibly from 2 molecules of
nitrous acid. The rate of reaction (7) is proportional to the
concentration of dinitrogen trioxide, [N₂O₃], and hence to the
square of nitrous acid concentration, [HNO₂]², i.e.

rate (7) = k₁[N₂O₃] [HNO₂]² (8)

The concentrations of nonionized amine and of free nitrous acid vary
with pH but k₁, is independent of pH. For practical purposes it is
more convenient to rewrite equation (8) in terms of the total
concentrations of nitrite and DMA i.e.

rate (3) = k₂ [total amine] [total nitrite]² (9)

where k₂ depends on pH; k₂ and the reaction rate show maximum
values at pH = 3.4 corresponding to the strength of nitrous acid
(pk_a = 3.37). The reaction rate decreases tenfold for each 1-unit
increase in pH above pH = 3.4. Below this pH level, the nitrite is
almost completely converted to nitrous acid. The main effect of a
further reduction in pH is a continuous drop in nonionized amine
concentration, causing a decrease in the reaction rate. There is no
sharp pH limit for nitrosation. It can occur slowly at a pH of 5 or
even 6, as observed for DMA (Mirvish, 1970).

The nitrosation of amides, such as N-alkylureas and
N-alkylurethanes proceeds rapidly (Challis & Challis, 1970;
Mirvish, 1971; Sander & Burkle, 1969). In this case the nitrosating
agent is probably the nitracidium ion (H₂NO₂)⁺:

2HNO₂ <=> (H₂NO₂)⁺ + NO₂^- (10)

HNO₂ + H⁺ <=> (H₂NO₂)⁺ (10a)

and nitrosation is accomplished by the following reaction:

RNH.COR' + (H₂NO₂)⁺ -> RN(NO).COR' + H₂O + H⁺ (12)

The reaction rate is again proportional to the concentrations of
nonionized alkylurea and nitracidium ions the formation of which can
be considered to proceed by equation (10a). Hence

rate (12) = k₃ [RNH.COR'] [HNO₂] [H⁺] (13)

rate (12) = k₄ [total amide] [total nitrite] [H⁺] (14)

The reaction rate, which increases about tenfold for each 1-unit drop
in pH from 3 to 1, does not show a pH maximum; k₄ depends on the
ionization of nitrite and, hence, on pH but it does not depend on the
ionization of amides, which are only slightly ionized above pH = 2.

Tables giving the rate constants for 15 amines and 21 amides
according to the above equations (Mirvish, 1975), indicate that the
most rapidly nitrosated classes of compounds are the N-alkylureas,
N-arylureas, N-alkylcarbamates, secondary aromatic amines,
secondary amine piperazine, morpholine derivatives, and tertiary
enamines.

It has been suggested that under mildly acidic conditions
tertiary amines also react with nitrous acid to produce nitrosamines
(Hein, 1963; Lijinsky, 1974; Lijinsky & Greenblatt, 1972; Lijinsky &
Singer, 1974; Lijinsky et al., 1972b; Roberts & Caserio, 1964; Smith &
Loeppky, 1967). Ender et al. (1967) studied the reaction between
nitrites and various methylamines including: methanamine
(monomethylamine); DMA; N,N-dimethylmethanamine (trimethylamine,
TMA); and trimethyloxamine (trimethylamine oxide, TMAO); they found
that DMN was produced in all cases. However, the rate of production
was proportional to the amount of nitrite present and increased with
decreasing pH values and increasing temperature. DMA was the most
reactive followed by TMA. Small amounts of DMN were formed from DMA
and sodium nitrite under very mild conditions (e.g. at 4°C). At

pH = 6.0, 2 to 2.5 times more DMN was formed than at pH = 6.5.
However, Malins et al. (1970), who failed to detect DMN formation at
pH levels of 5.8-6.4 after heating an aqueous mixture of sodium
nitrite and TMAO or DMA, found that trace amounts of DMN were
detectable in reaction mixtures consisting of TMA at concentrations of
400-2000 mg/litre and sodium nitrite at 400 mg/litre.

Fiddler et al. (1972) demonstrated the formation of DMN from
quaternary ammonium compounds and nitrite. The compounds studied
included N,N,N-trimethylethaminium chloride (neurine chloride),
2-(acetyloxy)- N,N,N-trimethylethanaminium chloride (acetylcholine
chloride), choline chloride, 1-carboxy- N,N,N-trimethylmethanaminium
hydroxide (betaine), and 3-carboxy-2-hydroxy- N,N,N-trimethyl-1-
propan-aminium chloride (carnitine chloride).

Nitrites are present in various foods and in saliva (Tannenbaum
et al., 1974) and can be formed in the infected bladder by bacterial
reduction. They may also be present in the stomach of infants and of
achlorhydric subjects where they are formed from nitrates, lower
acidity allowing the growth of nitrate-reducing bacteria.

Secondary amines are widely distributed in foods and have been
found in fish, eggs, rolls, biscuits, chocolate, soup cubes, meats,
and potatoes (Heyns, 1973, Lijinsky & Epstein, 1970). Tobacco and
tobacco smoke contain several secondary amines including pyrrolidine,
DMA, and piperidine (Neurath, 1972). Some aliphatic and heterocyclic
amines were identified in human blood and urine (Asatoor & Simenhoff,
1965; Perry et al., 1962). Other sources of secondary amines have been
given by Sander et al. (1971).

Methylguanidine, a natural constituent of beef (Kapeller-Adler &
Krael, 1930a) and shark, rayfish, and cod (Kapeller-Adler & Krael,
1930b), reacted with nitrite to produce N-methyl- N-nitrosourea and
N-methyl- N-nitrosocyanamide (Mirvish, 1971). The amino acids
1-proline, 1-hydroxyproline, and N-methylglycine (sarcosine) were
nitrosated 140-230 times more quickly than DMA at pH = 2.2-2.5
(Mirvish et al., 1973a).

N-nitroso compounds formed from 22 natural compounds were
listed by Mirvish (1975). In addition, nitrosation of N,N-bis (3
aminopropyl)-1, 4-butanediamine (spermine) and spermidine, two
polyamines, was reported by Ferguson et al. (1974) and Hildrum et al.
(1975).

2.1.4 The effects of other substances on the formation of N-nitroso
compounds

Several substances have been shown to catalyse the formation of
nitroso compounds from secondary amines and nitrite. Boyland & Walker
(1974) and Fan & Tannenbaum (1973) noted that chloride, bromide,
iodide, and thiocyanate catalysed the reaction while sulfate and
perchlorate ions did not have any effect. The effects of thiocyanate

have been studied more extensively; in its absence, the nitrosation of
N-methylbenzenamine ( N-methylaniline) and other secondary amines
is at a maximum at pH = 3, but in its presence, the reactions proceed
much more rapidly between pH levels of 1 and 2. Thiocyanate is present
in amounts of 110-330 mg/litre in human saliva. It has been estimated
that the thiocyanate concentration in the stomach is 3 times higher in
smokers than in nonsmokers.

Roller & Keefer (1974) reported a pronounced increase in the rate
of formation of DMN from DMA and nitrite in the presence of certain
carbonyl compounds and at a pH level higher than 3. Formaldehyde was
the most effective catalyst and the effect was appreciable even at
pH = 9. Challis & Bartlett (1975) reported that 3-[[3-(3,4-
dihydroxyphenyl)-1-oxo-2-propenyl]oxy]-1,4,5-trihydroxycyclohexane-
carboxilic acid (chlorogenic acid), a constituent of coffee was a
potent catalyst and in studies by Walker et al. (1975) 3,4,5-
trihydroxybenzoic acid (gallic acid) catalysed the nitrosation of
amines but only within a restricted pH range (around pH = 4).

On the other hand, Bogovski et al. (1972b) noted that tannins,
which are present in many foods, competed with secondary amines for
nitrite and thus led to a reduction in the amount of nitrosamine
formed. Similarly Challis (1973) demonstrated the preferential
nitrosation of phenols in the presence of amine to form
p-nitrosophenols suggesting a scavenging effect of phenols at
low pH.

Ascorbic acid inhibited the formation of DMN from oxytetracycline
and nitrite and also from aminophenazone (aminopyrine) and nitrite
(Mirvish et al., 1972, 1974). The same authors reported that gallic
acid, the active ingredient in tannins, completely inhibited
nitrosomorpholine formation from the parent amine and nitrite and that
sodium sulfite had a similar blocking activity.

The inhibitory effects of ascorbic acid and other inhibitory
agents on chemical nitrosation have recently been compared by Mirvish
et al., 1975 and it would seem, at present, that ascorbic acid is the
most effective and useful inhibitor of amine nitrosation.

2.2. Analytical Methods

2.2.1 Nitrates and nitrites

Methods for the determination of nitrates and nitrites in surface
and waste waters have been reviewed by Marculescu (1971). The most
suitable methods are colorimetric procedures using sodium salicylate
for nitrates and 4-aminobenzenesulfonic acid (sulfanilic acid) and
1-naphthalenamine (l-naphthylamine) for nitrites.

A standard procedure for determining nitrates in plants (HMSO,
1973) is based upon the reduction of nitrates to ammonia which is

removed by steam distillation and determined titrimetrically. The
nitrate electrode has been used in the determination of nitrates in
extracts of soils and herbage, and drainage water in the United
Kingdom (HMSO, 1974) and in the Federal Republic of Germany (Weil &
Quentin, 1973). Results indicated that several extraction procedures
applied to herbage gave higher values with the nitrate electrode than
with the standard distillation procedure. For drainage waters, better
agreement was obtained between the electrode and a spectrophotometric
procedure involving, 3,4-xylenol.

A variety of methods is available for the determination of
nitrates and nitrites in foods. A nitrate specific electrode for the
electrochemical determination of nitrate in spinach suspensions was
tested by Voogt (1969). Other anions Present in the spinach did not
have any direct influence on the precision of the results. Variations
in nitrate activity due to variations in the ionic strength of the
spinach extracts could be minimized by measuring the potential of the
extract in a 1% sodium sulfate solution. The precision of the method
was ± 2%. Kamm et al. (1965) developed a new method for the
determination of nitrates and nitrites in foods that would accurately
determine concentrations as low as 1 mg/kg. 1-Naphthylamine was
diazotized by nitrite and coupled with excess amine to give 4-(1-
naphthylazo)-1-naphthylamine which was measured spectrophoto-
metrically. Nitrate was quantitatively reduced by passage through a
cadmium column and determined as nitrite. Nitrite passed through the
column unaltered; thus nitrate was determined by difference.
Spectrophotometric and spectrofluorimetric methods for the
determination of low levels of nitrite in cheese were developed by
Rammel & Joerin (1972). The limits of detection for the two methods
were 50 µg and 3.0 µg of nitrite-N respectively, per kg of cheese or
milk products. A method to determine free and bound nitrite in meats
was published by Mirna (1974). Free nitrite was determined with the
Griess reagent whereas bound nitrite was liberated with Hg²⁺ in
aqueous acetone solution prior to diazotization. Methods of analysis
for nitrates and nitrites in several food products including meats,
cured meats, dry cure mix of curing pickle, flours, and baby foods
have been described (Horwitz, 1975) and adequate methods for the
determination of nitrates and nitrites in urine and blood are also
available (Shechter et al., 1972; Schneider & Yeary, 1973; Wegner,
1972).

2.2.2 N-nitroso compounds

The problems of estimating N-nitroso compounds in food and
other environmental media were recently reviewed by Bogovski & Walker
(1974), Bogovski et al. (1971a), Eisenbrand (1973), Fiddler (1975),
Scanlan (1975), Sen (1974), and Walker et al., (1976). The analytical
process can be divided into three major steps: extraction and
distillation from the specimen; purification; and qualitative and
quantitative determination.

The main difficulties in such analyses arise from the fact that
nitrosamines occur at very low concentrations and that they lack
suitable characteristics for trace analysis. They also suffer from
interference from other chemicals in the substrate which gives rise to
a considerable number of false-positive reports of the presence of
nitrosamines.

Most of the nitrosamines so far detected in foods are steam
volatile. Many analytical methods take advantage of this fact and, in
most of them, nitrosamines are isolated by distillation from an
aqueous, acidic, or basic solution. Distillation from an acidic
solution has the additional advantage of removing interfering amines.
Howard et al. (1970) digested fish samples with methanolic potassium
hydroxide before subjecting them to distillation. Telling et al.
(1971) reported improved recoveries of nitrosamines by vacuum
distillation. Other workers (Kröller, 1967; Sen et al., 1969a, 1972)
preferred initial extraction of the nitrosamines with ether or
methylene chloride prior to aqueous distillation. However, these
techniques cannot be used for the analysis of nonvolatile nitrosamines
such as nitrosoproline and nitrososarcosine.

Various column chromatographic procedures have been reported for
the clean-up of nitrosamines isolated from foods and biological
materials. These include ion-exchange (Alam et al., 1971a, 1971b; Sen
et al., 1969a, 1969b) or basic alumina columns (Sen, 1970; Sen et al.,
1970; Telling et al., 1972). These preliminary clean-ups have proved
to be extremely useful in cases where nitrosamines were estimated by
the conventional thin layer chromatography (TLC) and gas-liquid
chromatography (GLC) methods but such clean-ups were thought to be
unnecessary when a highly specific method such as high resolution gas-
liquid chromatography-mass spectroscopy (GLC-MS), was used for the
analysis.

Detection techniques can be divided into screening methods
suitable for routine surveys, and confirmation techniques to be used
if the results of the preliminary screening technique are positive.
Combined high resolution GLC-MS is believed to be the only reliable
confirmation technique available at the moment.

Eisenbrand & Preussman (1970) have described a colorimetric
technique in which nitrosamines are cleaved to nitrosyl bromide and
secondary amines, and the liberated NO⁺ ion is measured
colorimetrically after reacting with N-1-naphthalenyl-1,2-
ethanediamine (N(naphthyl-(1)-) ethylenediamine). The technique
appears to be reliable and applicable to a wide variety of
nitrosamines. The aminies formed after splitting may also be used to
estimate nitrosamines through the formation of fluorescent
5-(dimethylamino)-1-naphthalenesulphonyl (dansyl) derivatives.

Various TCL methods have been used for the detection and semi-
quantitative estimation of nitrosamines (Eisenbrand & Preussman, 1970;
Kröller, 1967; Möhler & Mayrhofer, 1968; Sen & Dalpé, 1972; Sen et
al., 1969a, 1973a; Yang & Brown, 1972). Most of these methods are
based on the principle of splitting the nitrosamines by UV radiation
into the parent secondary amines and nitrous acid, and subsequently
detecting these breakdown products with 2,2-dihydroxy-IH-indene-
1,3(2H)-dione (ninhydrin) N-phenylbenzeneamine (diphenylamine), and
Griess reagent, respectively. In some methods, nitrosamines are
reduced to hydrazines which are detected on TLC plates after the
formation of suitable derivatives.

GLC offers a rapid and sensitive technique for the analysis of
nitrosamines. In earlier work, the flame ionization detector was used
but it was later abandoned because of the lack of sensitivity, and
specifity. In more recent studies, various nitrogen-specific detectors
have been used such as the alkaline-flame ionization detector (Fiddler
et al., 1971; Howard et al., 1970; Kawabata, 1974), the Coulson
electrolytic conductivity detector (Crosby et al., 1972; Issenberg &
Tannenbaum, 1972; Panalaks et al., 1972; Rhoades & Johnson, 1970; Sen
et al., 1972, 1973a), and the microcoulometric detector (Newall &
Sisken, 1972). Each has some advantages and disadvantages, and the
reader is advised to consult the original papers for further details.
In one technique, nitrosamines were oxidized to the corresponding
nitramines which were then detected by an electron capture detector
(Althorpe et al., 1970; Castegnaro et al., 1972; Sen, 1970; Telling,
1972). Alliston et al. (1972) and Eisenbrand (1972) converted the
nitrosamines to the parent amines from which the heptafluoro-butyryl
derivatives were prepared and determined by electron capture detector.

Recently, Fine & Rufeh (1974) and Fine et al. (1974) have
reported a new instrument which is specific to the N-nitroso
functional group and is capable of detecting N-nitroso compounds in
foodstuffs at the µg/kg level with little or no concentration or
purification. In this technique, the N-nitroso compounds are cleaved
at the N-NO bond in the presence of a specific catalyst, and the
liberated NO is converted to excited nitrogen dioxide (NO₂*) by
reaction with ozone. As the excited nitrogen dioxide rapidly decays,
it emits light in the near infrared region of the spectrum which can
be detected and measured. The instrument can be coupled either to a
GLC or a high pressure liquid chromatograph thus making it suitable
for the analysis of both volatile and nonvolatile nitrosamines.

The nonvolatile nitrosamines constitute a more varied group of
compounds than the volatile nitrosamines and, as such, cause
additional analytical problems. Nitrosoamino acids, such as
nitrososarcosine, nitrosoproline, and nitroso-2-hydroxyproline may be
analysed by conversion to volatile derivatives such as silyl ethers
(Eisenbrand et al., 1975).

Methods have been proposed for the determination of total
N-nitroso compounds, the general approach being to cleave the nitroso
group and measure the nitric oxide formed. Fan & Tannenbaum (1971)
eliminated the problem of nitrate interference by using long-wave
ultraviolet irradiation (360 nm) to split the nitroso group. The
released nitrite was diazotized and coupled to form a dyestuff before
colorimetric estimation. The method was designed for automation.
Eisenbrand & Preussmann (1970) used hydrobromic acid in a nonaqueous
medium to split the nitroso group. A method for splitting the nitroso
group which does not require any anhydrous medium has been proposed by
Fine et al. (1976a).

3. SOURCES OF NITRATES, NITRITES AND N-NITROSO COMPOUNDS IN AIR,
WATER, SOIL AND FOOD

3.1 Natural Occurrence

3.1.1 Nitrates and nitrites

Nitrates in soil and in surface and groundwater result from the
natural decomposition by microorganisms of organic nitrogenous
material such as the protein in plants, animals, and animal excreta.
The ammonium ion formed is oxidized to nitrites and nitrates (section
2.1.1). Natural occurrence of nitrates and nitrites in the environment
is a consequence of the nitrogen cycle (section 4.1) but normally
nitrites are only found in very low concentrations.

3.1.2 N-nitroso compounds

Systematic studies on the natural occurrence of N-nitroso
compounds have not been reported but a few studies show that these
compounds may occur in certain microorganisms (Murthy et al., 1966;
Vavra et al., 1960) and in one variety of mushroom (Hermann, 1960). At
least one of these compounds, strephozotrin, is a potent carcinogen
(Arison & Feudale, 1967; Sibay & Hayes, 1969). Other reports on the
natural occurrence of diethylnitrosamine (DEN) in certain plants have
still to be confirmed by modern analytical methods.

3.2 Sources Related to Man's Activities

3.2.1 Nitrates and nitrites

3.2.1.1 Fertilizers

Artificial fertilizers, a major source of environmental nitrates,
may be composed of a variety of chemicals including ammonium, calcium,
potassium, and sodium nitrates, and urea. The production of
nitrogenous fertilizers in the world has increased in terms of N from
15.8 million tonnes in 1961/62-1965/66 to 42.3 million tonnes in
1974/75 (United Nations, 1976).

The fact that plants cannot use soil nitrogen completely is of
great significance; nitrogen utilization may vary from 25 to 85%
depending on the crop and on agricultural techniques. Thus, to obtain
maximum production, a great excess of nitrogen fertilizer must be
applied to the soil and the resulting nitrogen runoff will be
substantially increased. For example, Kohl et al. (1971) showed that
as much as 55-60% of the nitrogen input in the Sangamon River feeding
Lake Decatur, IL, USA, was of fertilizer origin. Lee (1970b), Sawyer
(1947), and Sylvester (1961), have all published data showing that
nitrogen runoff is 3-10 times higher from fertilized areas than from
unfertilized areas in the same region. However, analysis of stream

waters did not show a clear relationship between the nitrate
concentrations in British rivers and the amounts of fertilizers used
on adjacent land (Tomlinson, 1970).

Brown & Smith (1967) observed that nitrogen fertilization tended
to increase the nitrate content of vegetables and attempts have been
made to correlate nitrogen application rates with the nitrate contents
of lettuce, radish, and spinach. In studies in Bulgaria, Biocev &
Pocinkova (1972) noted that the nitrate levels in spinach increased
when as little as 20 kg of nitrogen per ha was added to the soil.
Schuphan (1969) observed that application of four times the normal
amount of fertilizer resulted in considerably higher nitrate levels in
spinach but that the nitrite levels remained low.

3.2.1.2 Animal wastes

Another major source of nitrates is farm animal wastes which
contain large amounts of nitrogenous materials that may be converted
into nitrates. The problem is more acute where farming is carried out
intensively, a common practice in North America for both livestock and
poultry. Since a 450 kg steer excretes about 43 kg of nitrogen per
year, a 3200 head feedlot would produce 1400 tonnes annually on a
relatively small area--an amount equivalent to about 260 000 people.
Thus, such feedlots become "small area" sources of nitrogen runoff.
Only 10% of these wastes is returned to cultivated land (Standford et
al., 1969) and runoff studies demonstrate a considerable problem of
environmental pollution. Nye (1973) reported Gilbertson et al. (1970),
who found that the total nitrogen concentration in runoffs ranged from
about 50 to over 5500 mg/litre. Animal husbandry, even when carried
out on pastures or with the return of the animal wastes to cultivated
land, may still impose problems. Adriano et al. (1971) concluded that
wastes from a maximum of 7-8 cows could efficiently be used per
hectare of farmland or pasture and that higher application rates might
raise nitrate levels above 10 mg/litre in the subsoil waters.

3.2.1.3 Municipal, industrial, and transport wastes

Discharges of municipal and industrial wastes are concentrated
sources of nitrogen compounds that are, to a large extent, released
directly into surface waters. The amount of nitrogen in human wastes
is estimated to be about 5 kg per person per year (Committee on
Nitrate Accumulation, 1972). Even if treated, this waste will
represent a heavy water pollution load since secondary treatment
removes less than half of the nitrogen. Ammonium ions in the effluent
of septic tanks may be rapidly converted to nitrate which may
penetrate some distance from the tank. Sludge from treatment plants
and septic tanks has also to be disposed of and represents another
significant source of nitrogen pollution. Solid waste disposal
practices, particularly sanitary landfills and dumps, may represent a
source of water pollution by nitrogen compounds.

The nitrogen content of industrial wastes is highly variable;
fuel and food processing industries and petroleum refineries may
constitute important sources of nitrogen pollution. The
nitrogen/BOD^a ratio of food processing plant wastes is about 0.05
while for animal processing wastes this ratio amounts to 0.5.
(Committee on Nitrate Accumulation 1972). Oxides of nitrogen released
into the atmosphere from man-made sources such as motor vehicles,
fossil fuel combustion, and industrial processes amount to about 50
million tonnes per year on a global scale (Robinson & Robbins, 1972).
A considerable proportion of this fixed nitrogen is eventually
returned to the earth's surface as nitrate.

3.2.1.4 Deliberate addition of nitrates and nitrites to food

Nitrates and nitrites are widely used in the production of
certain meat products and in the preservation of fish in some
countries. Reasons for using these salts in food production have been
reviewed by Ingram (1974). Nitrite is used in meat curing to obtain
the characteristic pink colour and flavour of cured meat. While a
nitrite content of less than 5 mg/kg of meat is sufficient to give a
satisfactory colour for a limited period of time, up to 20 mg/kg may
be necessary to give commercially adequate colour stability and about
50 mg/kg to produce the characteristic flavour. However, detailed
experimental confirmation of these figures is lacking.

Curing meat gives an important degree of protection against
botulism and may provide similar protection against other bacteria
such as Clostridium welchii and staphylococci, although the
importance of this has not yet been assessed. The question of how much
nitrite is necessary to protect against botulism is very complex
because of several associated factors.

The addition of nitrates and nitrites to meats, meat products,
and cheese is governed by legislation in most countries, some of which
also allow the addition of these salts to fish products.

3.2.2 N-nitroso compounds

3.2.2.1 Food

The formation of N-nitroso compounds from nitrites and amines
during the storage and processing of food is discussed in section
4.2.2.

^a BOD = biological oxygen demand.

3.2.2.2 Tobacco

Nitrosonornicotine has been found in unburned smoking tobacco,
chewing tobacco, and in snuff. The same compound has been identified
in the mainstream smoke of a popular nonfilter cigarette in the USA
(Hoffman et al., 1974).

3.2.2.3 Industrial uses

Although N-nitroso compounds do not appear to be extensively
used at present, Magee (1972) reports that patent applications have
been made in the UK for their use in the manufacture of dyestuffs,
lubricating oils, explosives, insecticides, and fungicides. Some
nitrosamines (nitrosodiphenylamine, N-N-dinitrosopentamethylenete-
tramine, polymerized N-nitroso 2,2,4-trimethyl-1,2-dihydroquinoline
and N-methyl- N, 4-dinitrosoaniline) are used as organic
accelerators and antioxidants in the production of rubber (Boyland et
al., 1968). DMN has been used as an industrial solvent, as a
nematocide, and in the synthesis of the rocket fuel 1,1-
dimenthylhydrazine. There is some evidence that DMN might also be
formed during the combustion of this rocket fuel (Simoneit &
Burlingame, 1971). There are patents for the use of DMN as a solvent
in the plastics and fibre industry, as an additive for lubricants, and
to increase the dielectric constant in condensers (Daiber, 1966).

Industrial uses may result in the occurrence of N-nitroso
compounds in the work environment and in industrial effluents. Fine et
al. (1976b) reported point sources of air pollution by DMN in
Baltimore, MD, and Belle, WV, in the USA. A factory using DMN as an
intermediate was shown to be the source in Baltimore and was shut
down; in Belle the source of DMN was an amine-manufacturing facility.

4. TRANSPORT AND TRANSFORMATION IN ENVIRONMENTAL
AND BIOLOGICAL MEDIA

4.1. Nitrogen Cycle

The continuous interchange between atmospheric and terrestrial
nitrogen takes place along a number of different pathways including
air, water, soil, microorganisms, plants, animals, and man. This
transfer and transformation of nitrogen is referred to as the nitrogen
cycle (Fig. 1).

The main factors affecting it are the climatic conditions, the
type and density of animal and plant populations, agricultural
practices, and animal husbandry. The nitrogen cycle has undergone
profound modifications through the agricultural and industrial
activities of man (Bolin & Arrhenius, 1977; Committee on Nitrate
Accumulation, 1972; Commoner, 1970; FAO/IAEA Panel of Experts, 1974).

Atmospheric nitrogen is in the form of dinitrogen (N₂); the
great strength of the N = N bond is mainly responsible for its
chemical inertness. A part of atmospheric nitrogen is transformed by
microbial action and incorporated into living organisms. This process
is called nitrogen "fixation" and is estimated to amount globally to
150 million tonnes of fixed nitrogen per year. In industrial nitrogen
fixation, atmospheric nitrogen is combined with hydrogen at high
temperatures and pressures in the presence of suitable metal catalysts
(Haber-Bosch process) to produce ammonia. Industrial nitrogen fixation
accounts for about one quarter of the total world production of fixed
nitrogen (Bolin & Arrhenius, 1977). Various atmospheric processes
which have been discussed elsewhere (WHO, 1977) are minor sources of
fixed nitrogen.

Biological nitrogen fixation, i.e. its reduction to ammonia, can
be accomplished by only a limited number of organisms. Symbiotic
nitrogen fixation takes place in the root nodules of legumes such as
soya bean, clover, and alfalfa, which contain bacteria of the
Rhizobium species. There are also symbiotic processes with plants
other than legumes involving, for example, some cyanobacteria. A
number of free living bacteria and algae can also fix nitrogen (Burns
& Hardy, 1975; Quispel, 1974). The fixation is catalysed by a complex
enzyme nitrogenase (1.7.99.2). Ammonia produced by biological fixation
is then converted to nitrite and nitrate by the process of
nitrification (see section 2.1.1). Plants can assimilate only a part
of the nitrates present in soils; some leaches into ground water and
rivers and may reach estuaries and oceans, the rest is subjected to
denitrification, another natural biochemical process that degrades
nitrates to nitrogen or nitrous oxide (dinitrogen oxide) which are
released into the atmosphere. Denitrification takes place in the soil
and also at the interface between water and sediment in oceans,
rivers, lagoons, and lakes. Nitrates from natural fixation and

artificial fertilizers are ultimately used for the synthesis of
biological molecules, particularly proteins. Plants and animal waste
and dead tissues return fixed nitrogen to the soils, where part of it
is recycled and part returned to the atmosphere, thus completing the
nitrogen cycle. According to Delwiche (1970), nitrogen fixation on a
world basis may exceed denitrification by about 10%. The increased use
of industrial fertilizers has resulted in some areas in increased
concentrations of nitrates in bodies of water, resulting, in some
cases, in eutrophication.

4.2 Transformation in Foods

4.2.1 Reduction of nitrates to nitrites

Because of the ability of spinach to accumulate large quantities
of nitrates and reported cases of intoxication associated with the
consumption of this vegetable, several studies have been undertaken on
the conversion of nitrates to nitrites in spinach.

Data presented by Phillips (1968a) indicated that initial
nitrite-N contents of fresh, frozen, canned, and baby-food spinach
were generally less than 1 mg/kg fresh weight. However, several
authors have reported a rapid fall in nitrate levels and increase in
nitrite levels in fresh spinach during the first 4 days of storage at
room temperature (Achtzehn & Hawat, 1970; Phillips, 1968a; Schuphan,
1965). Higher nitrite levels occurred in spinach from fertilized
ground (Brown & Smith, 1967) and these could reach exceptionally high
values (3600 mg/kg dry weight) with excessive fertilization (Schuphan,
1965).

Under refrigeration, the nitrite-N contents of fresh spinach
increased very gradually throughout a storage period of 28 days
(Phillips, 1968a). Significant increases in nitrite-N levels did not
occur during the storage of frozen, canned, or baby-food spinach, but
increased concentrations were found in frozen spinach, that had been
left to thaw at room temperature for an excessively long period (39 h)
(Phillips, 1968a).

There was also a slight rise in nitrite levels when partially
consumed jars of commercial baby foods containing nitrates were stored
for 7 days at room temperature instead of under refrigeration
(Phillips, 1969). Selenka (1970) noted that nitrite formation in baby
foods was rapid in the presence of Escherichia coli and Pseudomonas
fluorescens, less rapid with Bacillus subtilis, and very slow with
Staphylococcus albus.

When foods consisted of a solid immersed in a liquid (e.g. canned
foods or frozen foods after thawing) nitrates were partially
transferred to the liquid portion or into the water in which the food
was cooked (Bodiphala & Ormrod, 1971). When large volumes of water and

long cooking times were employed (Kilgore et al., 1963), and when
canned vegetables were blanched in hot water instead of steam
(Johnson, 1966), significant amounts of nitrates were leached out of
the foods. Nitrate reductase (NADPH) (1.6.6.3) activity was rapidly
destroyed during cooking, thereby greatly diminishing further
conversion of nitrates to nitrites (Bodiphala & Ormrod, 1971). It is
also well known that the sterilization treatments necessary for
canning destroy microorganisms that could convert nitrates to
nitrites.

Conversion of nitrates to nitrites occurred more slowly in
vacuum-packed bacon than in unpackaged bacon, presumably due to the
low reducing ability of anaerobes (Cavett, 1962). Spencer (1967) found
that the nitrite content of vacuum-packed bacon decreased slowly on
storage. It has also been reported by Sebranek et al. (1974) that
nitrite levels in meat, determined 2 days after processing, were less
than half those originally added to frozen samples and samples
processed at 71°C, and that they decreased further during storage.
Frying, grilling, or boiling bacon or ham reduced the nitrites content
by 20-90% (Food Standards Committee, 1959).

When direct gas firing of spray dryers was employed, the nitrate-
N contents of dried milk products increased by 1-3 mg/kg compared with
those obtained using indirectly heated sprayers but nitrite-N levels
were unaffected (Manning et al., 1968). Air drying of potato and corn
starch led to the formation of only trace amounts of nitrite
(Gerritsen & De Willingen, 1969).

Nitrates may be reduced to nitrites when cooking is carried out
in aluminium utensils (Osteryoung, unpublished data)^a. This
observation appears to be significant since some countries use
aluminium utensils for boiling milk and water, a practice which could
lead to the formation of sizeable quantities of nitrites. This effect
of aluminium should be investigated further.

4.2.2 Formation and degradation of N-nitroso compounds

The conditions under which various amines, amino acids, and
proteins in food could react with nitrite to form nitrosamines were
studied by Ender & Ceh (1971) and by Sen et al. (1970) who showed that
when cod, herring, hake, halibut, mackerel, or salmon were treated
with sodium nitrite at 200 mg/kg and cooked at 110°C for 60-70 min
there were only trace amounts (2.5-25 µg/kg) of DMN in the cooked
product. The highest levels were found in mackerel and hake, both of
which contain large amounts of DMA and TMA. Samples without added
nitrite did not contain any detectable nitrosamines.

^a Nitrates as human and animal health hazards, Paper presented at
the Second Conference on Environmental Chemicals, Colorado State
University, 1973.

The formation of DMN was studied in aqueous model systems
containing methyl amines and sodium nitrite under conditions which
were more severe than those employed in the commercial processing of
nitrite-treated smoked chub (a fresh water fish containing small
amounts of TMA, TMAO, and DMA). The results of the studies showed that
not more than 10 µg of DMN per kg of the final product would be formed
during the smoking process (Malins et al., 1970).

Recently, Sen et al. (1973b) suggested that a major source of
nitrosamines in cured meat might arise from an interaction between the
nitrites and spices, such as black pepper and paprika, that are
present in curing mixtures. Nitrosopyrrolidine, nitrosopiperidine, and
DMN were found in a curing mixture used by a meat manufacturer in
Canada. The same authors (Sen et al. 1974a) have also studied the
effect of sodium nitrite concentration on the formation of
nitrosopyrrolidine and DMN in fried bacon. Bacon samples prepared with
sodium nitrite at 0, 50, 100, 150, or 200 mg/kg were analyzed for
nitrosopyrrolidine and DMN. No nitrosamine was detected in samples
prepared without nitrite but all treated samples contained 2-20 µg/kg
of nitrosamines. The level of nitrosopyrrolidine was related to the
initial concentration of nitrite in the bacon. It has also been shown
that nitrosamine formation in bacon increases with increasing
temperature and time of frying and that whereas baking, broiling, or
frying produce variable amounts of nitrosamines none is produced with
cooking in a microwave oven (Pensabene et al., 1974).

After deliberate nitrosation of eggs and meat with unusually
large amounts (1%) of sodium nitrite, some N-nitroso compounds
appeared to have been formed but the chemical nature of the compounds
detected was not clear (Walters, 1971).

No systematic studies on the formation of N-nitroso compounds
in cheese have been performed, although some types of cheese are known
to be processed with nitrate and nitrite.

There are few data on the fate of N-nitroso compounds during
the cooking, processing, or storage of food but some studies have
demonstrated that the volatile nitrosamine DMN and nitrosopyrrolidine
may be lost during the frying of bacon (Sen et al., 1973a).

4.3 Formation of N-nitroso Compounds from Drugs and Pesticides

Reaction with nitrite to form nitrosamines is not restricted to
food components. Lijinsky & Greenblatt (1972) and Lijinsky (1974)
reported that some antibiotics and other drugs that are widely used
can react with nitrites to form nitrosamines in alarmingly high
quantities. The drugs examined included oxytetracycline,
aminophenazone (aminopyrine), disulfiram, N,N-diethyl-3-
pyridinecarboxamide (nikethamide), tolazamide, and (E,E)-1-[5-(1,3-
benzodioxol-5-yl)-1-oxo-2,4-pentadienyl] piperidine (piperine).

Optimum conditions of temperature, pH, and concentration for these
reactions have been reported by Lijinsky et al. (1972a, 1972b, 1972c)
who more recently (Lijinsky, 1974) studied the reactions of
aminopyrine and other commonly used drugs with nitrous acid at rather
low concentrations to assess the magnitude of the hazard to man from
such interactions. The topic has been reviewed by Mirvish (1975) who
has listed 41 drugs and pesticides that have been nitrosated.
Pesticides listed include atrazine, simazine, ziram, and thiram.

4.4 Formation of N-nitroso Compounds in Animal Organisms

4.4.1 Formation of N-nitroso compounds in simulated gastric juice

Formation of DEN was demonstrated when DEA and nitrite were
incubated in the gastric juice of the rat, rabbit, cat, dog, and man.
More DEN was formed in human and rabbit gastric juices (pH 1-2 in both
cases) than in rat gastric juice (pH 4-5). (Sen et al., 1969a, 1969b).

The formation of nitrosamines by the interaction of some drugs
with nitrite in the presence of human gastric juice have been studied
by Scheunig & Ziebarth (1976). At 37°C and a pH = 2, for 1 h,
aminopyrine, sodium [(2,3 dihydro-1,5-dimethyl-3-oxo-2-phenyl-IH-
pyrazol-4-yl)methylamino] methanesulfonate (analgin), and piperazine
gave nitrosamine yields (calculated on the basis of nitrite used) of
69%, 11%, and 74.8% respectively.

In vitro studies have been carried out (Wells et al., 1974) in
which several foods (pork, egg, bread, milk, and cheese) were
incubated under simulated gastric conditions with concentrations of
nitrite similar to those used as food preservatives. The effect of the
thiocyanate ion as a catalyst for nitrosation was also studied since
it is secreted in saliva. Of the foods studied, only cheese produced
detectable amounts of volatile nitrosamines. The identity of the
nitrosamines was not indicated.

4.4.2 Formation of N-nitroso compounds in vivo

When DEA and nitrite were fed to cats and rabbits, considerable
amounts of DEN were detected in the stomach of the experimental
animals (Sen et al, 1969a, 1969b). Similar results were reported by
Sander & Sief (1969). Epstein (1972) reported the formation of
nitrosopiperidine in the gastrointestinal tract of rats treated with
nitrite and piperidine hydrochloride. When the nitrite concentration
was constant, nitrosopiperidine formation in the small intestine
increased with increasing concentrations of piperidine.
Nitrosopiperidine was also found in the stomach. Recently, Sander et
al. (1974a) demonstrated the formation of N-N'-dinitrosopiperazine,
DMN, and N-nitroso- N-methylbenzylamine in the stomach contents of
rats given the parent amines combined with nitrite. Considerable
individual variation in the degree of synthesis of N-N'-
dinitrosopiperazine was noted in the animals. In another recent

report, N-nitrosopyrrolidine was formed very rapidly in the stomach
of dogs from sodium nitrite and pyrrolidine (within 2-6 min) but after
30 min nearly all of it had disappeared, presumably due to its rapid
absorption (Mysliwy et al., 1974).

Indirect evidence of in vivo formation of N-nitroso compounds
has also been provided by some toxicity studies. Thus, hepatic
lesions, formed following administration of nitrite and some amines,
were similar to those produced by DMN or N-nitroso- N-
methylbenzylamine (Asahina et al., 1971). Similar effects were noted
where nitrite was administered up to 3 h after DMA, but the effect was
markedly reduced if the nitrite was given prior to the amine.

4.5 Formation of N-nitroso Compounds by Microorganisms

Studies conducted by Hawksworth & Hill (1971a), Klubes & Jondorf
(1971), and Sander & Sief (1969) suggested that nitrosamines could be
synthesized from secondary amines and nitrates or nitrites by
Escherichia coli and some species of streptococci. Fong & Chan
(1973b) demonstrated that homogenized Chinese salt fish inoculated
with Staphylococcus aureus (a nitrate-reducing bacterium) produced
considerable amounts of DMN.

Formation of nitrosamines in the presence of bacteria is unlikely
to occur in the large intestine, but the infected bladder and
achlorhydric stomach are likely sites (Hawksworth et al., 1974).

The ultimate mechanism of bacterial production of nitrosamines
remains to be ascertained. According to Hawksworth et al. (1974),
certain bacteria do reduce nitrate to nitrite but the formation of the
nitrosamine may be nonenzymatic and involve some heat-resistant
metabolite.

4.6 The Effects of Other Chemicals on the Formation of N-nitroso
Compounds

Fiddler et al. (1973) and Greenberg (1974) showed that high
levels of ascorbic acid reduced nitrosamine formation in frankfurter
sausages and in fried bacon. On the other hand, Nagata & Mirna (1974)
reported an increase in nitrosamine formation in meat products in the
presence of ascorbic acid. Other studies conducted on the inhibition
of nitrosamine formation by various compounds include a report by Sen
& Donaldson (1974) in which nitrosamine formation in human saliva was
inhibited by ascorbic acid. Ziebarth & Scheunig (1976) tested a number
of substances and beverages for the inhibition of the nitrosation of
several drugs under simulated gastric conditions. Of all the
substances investigated, ascorbic acid was regarded as the best
inhibitor because of its pronounced activity at pH values occurring in
the stomach and because it was not toxic in the amounts used.

5. ENVIRONMENTAL LEVELS AND EXPOSURES

5.1 Nitrates and Nitrites

5.1.1 Ambient air

Nitrate aerosols are the final stage in the atmospheric oxidation
of gaseous oxides of nitrogen, and substantial amounts of particulate
nitrates may be formed in urban areas affected by photochemical
pollution (Pitts & Lloyd, 1973). The concentration of nitrates in air
may range from about 1 to 40 µg/m³, depending on the sampling and
averaging periods. For example, the estimated annual mean values
(1968-1972) in Chattanooga, TN, USA, were between 1 and 6 µg/m³
(French et al., unpublished)^a. The daily mean concentrations of
airborne nitrates in the central part of Tokyo ranged, in 1973, from
0.9 to 41.8 µg/m³ with an annual mean of 8.2 µg/m³. On the other
hand, in a small city with few industries (Matsue City) the daily
means were in the range of 1.1 -- 9.2 µg/m³, with an annual mean of
2.6 µg/m³ (Japan Environmental Sanitation Center, 1974).

5.1.2 Water

The concentrations of nitrates and nitrites in surface and ground
waters vary within wide limits, depending on geochemical conditions,
human and animal waste management practices, the extent to which
nitrogen-containing agriculture fertilizers are used locally, and on
industrial discharges of nitrogen compounds (section 3.2.1.).

In general, surface waters do not usually contain nitrate in
concentrations higher than 10 mg/litre, and nitrite concentrations
rarely exceed 1 mg/litre. However, a steady upward trend of nitrate
levels has been reported in recent years in some countries, both in
surface and ground waters. Thus, for example, in the River Thames,
England, nitrate concentrations increased from an average of
4 mg/litre in 1968 to an average of 9 mg/litre for the last quarter of
1973 (Water Research Centre, 1974). Similar increases have been
observed in several other English rivers (Casey, 1975; Owens, 1970;
Tomlinson, 1970). The nitrate concentrations are increasing in some
rivers that drain the great agricultural section of the centre of the
USA, and in selected small rivers the 45 mg/litre limit is sometimes
exceeded (Viets & Aldrich, 1973). A small increase in the nitrate
concentration of the Tamagawa River, Tokyo, Japan has also been
reported. From 1951-1965, the nitrate ion concentration rose from
7.9 mg/litre to 9.1 mg/litre. During the same period, the nitrite
concentration increased from 0.049 mg/litre to 0.53 mg/litre, i.e by a
factor of about 10 (Goto, 1973).

^a French, J. G., Hasselblad, V., & Johnson, R. Aggravation of
asthma by air pollutants. 1971 -72 Southeastern CHESS studies.

Studies of 991 settlements in Bulgaria indicated that only 64
towns and villages had drinking water levels of nitrates between
30 mg/litre (Bulgarian standard) and 50 mg/litre. In 20 settlements,
situated in areas with intensive agriculture and stock breeding, the
nitrate concentrations exceeded 50 mg/litre. The report^b points out
that such problems did not exist some 10 years ago when smaller
quantities of nitrogen fertilizers were used in agriculture.

Much higher concentrations of nitrates are sometimes found in
ground water, particularly in water derived from dug wells. A survey
of over 2000 rural wells in Saskatchewan, Canada, revealed that 18.8%
contained nitrate concentrations of more than 50 mg/litre and 5.3% had
nitrate levels exceeding 300 mg/litre (Robertson & Draycott, 1948).
Hedlin (1971) also reported levels above 45 mg/litre in some wells in
a rural area of Manitoba, Canada. In many farm wells in central USA,
nitrate concentrations may range from 45-450 mg/litre. This problem is
neither new nor local, since such conditions have been recorded from
1895 to 1970 in Illinois, in 1939 in Iowa, and in 1970 in Minnesota
(Viets & Aldrich, 1973). The mean nitrate concentration in ground
water consumed by children affected by methaemoglobinaemia in
Czechoslovakia ranged from 18-257 mg/litre (Schmidt & Knotek, 1970).
According to Gruenar & Shuval (1970), about 180 wells for community
water supplies in the densely populated central and southern coastal
plain in Israel had nitrate concentrations exceeding 45 mg/litre. In
England, nitrate concentrations in some ground waters have been
reported to range from 12 mg/litre (Foster & Crease, 1974) to over
22 mg/litre (Reeves et al., 1974). Nitrate concentrations exceeding
45 mg/litre have not been reported in centralized water supplies in
the USSR^c. However, high concentrations have been found from time
to time in dug wells, for example, 310-400 mg/litre in Leningrad
Oblast (Motylev, 1969), 110-200 mg/litre in the Tatar SSR (Petukhov
et al., 1972) and up to 430 mg/litre in the Moldvian SSR (Diskalenko,
1969).

5.1.3 Selected foods

According to the data compiled by the National Institute of
Environmental Health Sciences (NIEHS, 1970), the levels of nitrates in
vegetables vary considerably. The highest levels were found in beets,
egg plant, kale, and spinach and the lowest in tomatoes and peas;
similar findings were obtained in the German Democratic Republic by
Achtzehn & Hawat (1969). It is of interest to note that the nitrate
levels in vegetables reported by Jackson in 1967 were similar to those
reported by Richardson in 1907, when manure was used instead of
chemical fertilizers.

^b Contribution to the WHO environmental health criteria document on
nitrates, nitrites and nitrosamines, Sofia, 1974.

^c Contribution to the WHO environmental health criteria document on
nitrates, nitrites and nitrosamines, Moscow, 1974.
Nitrate contents vary not only between vegetable species but also
widely within a given species. This variation within a species may be
accounted for by such factors as temperature, sunlight, soil moisture,
and the level of available nitrogen in the soil (US Department of
Agriculture, 1965). A relationship between nitrate accumulation in
spinach and levels of fertilizer applied to the soil has been
demonstrated by a number of authors (Brown & Smith, 1967; Phillips,
1971). Furthermore, Schuphan (1965) reported exceptionally high levels
of nitrites (3600 mg/kg dry weight) in excessively fertilized fresh
spinach stored at room temperature.

A survey of the nitrate contents of fruits in the German
Democratic Republic, revealed that they were high in bananas and
strawberries but could not be detected in the other fruits examined
(Achtzehn & Hawat, 1969).

Cow's milk contained nitrate levels of 0-0.5 mg/litre (Simon et
al., 1964).

The levels of nitrates and nitrites in baby foods are of special
concern since infants are considerably more sensitive to the toxic
effects of nitrates than adults. Kamm et al. (1965) studied 194
prepared infant foods and found that, on average, fruits, dairy
products, puddings, egg products, meats, dry and concentrated food
supplements, and precooked cereal products contained nitrate levels of
less than 90 mg/kg. Vegetables, however, had a wide range of nitrate
contents varying from 0.9 to 2165 mg/kg, but nitrite levels never
exceeded 7 mg/kg. In studies on a number of canned foods, baby foods,
frozen foods, and vegetables, several varieties of fruit, spinach, and
beets generally had the highest nitrate contents (Bodiphala & Ormrod,
1971). Additional information may be found in articles by NIEHS (1970)
and Ashton (1970).

In a survey of various cured meats (Table 1, Ashton, 1970), the
highest nitrate content of 370-511 mg/kg was found in ham. In
analysing 197 samples of cured meat products, Panalaks et al. (1972)
found that the levels of nitrates and nitrites ranged from
0-3467 mg/kg and 0-252 mg/kg, respectively.

Dubrow & Kakisch (1960) analysed 338 samples of cheese and
reported that all were free of nitrites (less than 1 mg/kg). However,
40% of the samples contained nitrate levels of more than 1 mg/kg.
Rammell & Joerin (1972) also found low nitrite levels in cheese.

Table 1. Nitrate and nitrite contents of cured meats

Meat type Nitrate (mg/kg) Nitrite (mg/kg)

silverside 133-303 9-26

ham 370-511 7-150

luncheon meat 59-214 3.1-47

chopped ham & pork 53-101 22-62

corned beef 118-135 18-208

frankfurter sausage 119-141 8.5-10.3

From: Ashton (1970)

5.1.4 Estimate of general population exposure

One of the important sources of exposure to nitrates for man is
water. The level of nitrates in water may vary from practically nil to
over 200 mg/litre. In water from municipal supplies, however, it is
likely to be under 10 mg/litre. Thus, assuming an intake of 2 litres
of water per day, the daily intake of nitrates from this source would
normally be less than 20 mg, but with extremes of 0 and over 400 mg.

The other main sources of nitrates and nitrites are certain
vegetables and meat products. The intake from these sources is even
more variable because of marked differences not only in levels in
these foods but also in dietary patterns. However, Ashton (1970)
estimated the weekly intake of nitrates for a member of the general
population in the USA to be about 400 mg including 210 mg from
vegetables, 110 mg from meat products, and 85 mg from water (7 litres
per week). The estimate of Hill et al. (1973) for a member of the
general population in England included 225 mg from vegetables, 110 mg
from meat, and 105 mg from water in "control towns" and 645 mg from
water in Worksop, England. However these figures cannot be applied
generally because of variations in feeding habits and nitrate levels
in environmental media. Since the intake of nitrites is even more
variable, no estimates have been reported.

5.2 N-nitroso Compounds

5.2.1 Ambient air

The occurrence of N-nitroso compounds in urban air was reported
first by Bretschneider & Matz (1973) and confirmed by Fine et al.
(1976b) who found DMN at concentrations of about 1.2-3.5 µg/m³
(0.33-0.96 ppb) in an industrial district in Baltimore, MD, USA, and
about 0.06-0.17 µg/m³ (0.014 ppb - 0.051 ppb) in Belle, WV, USA.
N-nitroso compounds may be present in air, either due to their
formation in the air from secondary amines and oxides of nitrogen
(Neurath, 1972) or due to industrial omissions as in the instances
referred to.

5.2.2 Water

There are few reports on the occurrence of N-nitroso compounds
in water. Fine et al. (1976b) analysed samples from the Mississippi
river (New Orleans, LA) and from 3 water treatment plants in
Louisiana. Using the new N-nitroso compound-specific thermal energy
analyser (TEA) interfaced to both a gas chromatograph (GC) and a high
performance liquid chromatograph (LC), several peaks were tentatively
identified as belonging to N-nitroso derivatives of some pesticides.
The estimated concentrations were of the order of 0.1 µg/kg.

5.2.3 Selected foods

A summary of the reported occurrences of nitrosamines in meat and
fish products, adapted from Sen (1974) and updated, is presented in
Table 2. Only the results that were confirmed by mass spectroscopy are
quoted in this table. It was noted that the majority of some 50
publications dealt with the determination of nitrosamines,
particularly DMN, in processed pork meat. The methods employed for
analysis mainly involved gas chromatography. A few results were
confirmed by mass spectroscopic techniques. However, mass spectroscopy
confirmation is currently being employed more frequently than in the
past.

Table 2. Levels of nitrosamines in various meat and fish products^a

N-nitroso
Meat Country compounds Levels (^c) Reference
or area found^b

dry sausage Canada DMN 10-20 µg/kg Sen (1972)
uncooked salami sausage Canada DMN 20-80 µg/kg Sen (1972)
salami sausage Netherlands DMN 0.3 µg/kg(^e) Stephany et al. (1976)
DMN 0.1(^e) Stephany et al. (1976)
NDBA 1.1(^e) Stephany et al. (1976)
Netherlands NPY 0.4(^e) Stephany et al. (1976)
NPIP 0.3(^e) Stephany et al. (1976)

bacon Canada NPY 4-25 µg/kg Sen et al. (1973a)
Canada NPY 25-40 µg/kg Sen et al. (1974a)
Netherlands DMN 0.8 µg/kg(^e) Stephany et al. (1976)

bacon DEN 0.2(^g)
Netherlands NDBA 0.6(^g) Stephany et al. (1976)
NPY 0.4(^g)
Netherlands NPIP 0.6(^g) Stephany et al. (1976)
USA NPY 7-35 µg/kg Pensabene et al. (1974)

bacon USA NPY 2, 28, 13 Fiddler et al. 1974

uncooked bacon Canada DMN 30 µg/kg Sen et al. (1973a)

fried bacon Netherlands DMN 2.4 µg/kg Groenen et al. (1976)
DEN 4.43 µg/kg
Netherlands DMN 1.1 µg/kg(^e) Stephany et al. (1976)
DEN 0.2(^g)
Netherlands NDBA 0.7(^g) Stephany et al. (1976)
NPY 16.4(^g)

Table 2 (Cont'd)

N-nitroso
Meat Country compounds Levels (^c) Reference
or area found^b

Netherlands NPIP 3.9(^g) Stephany et al. (1976)
UK NPY 1-40 µg/kg Crosby et al. (1972)
USA NPY 20-207 µg/kg Fazio et al. (1973)

smoked meat Netherlands DEN 7.91 µg/kg Groenen et al. (1976)
Netherlands DMN 3 Groenen et al. (1976)

smoked horse and Netherlands DMN 7.3 µg/kg(^d) Stephany et al. (1976)
beef meat DEN 0.6(^d) Stephany et al. (1976)
Netherlands NDBA 0.4(^d) Stephany et al. (1976)
NPY 0.1(^d) Stephany et al. (1976)
Netherlands NPIP 0.1(^d) Stephany et al. (1976)

ham with layer Germany DMN 3 µg/kg Eisenbrand et al. (1975)
of pepper grains (Federal Republic of)
on the outside.
Only fat portion.

ham with layer Germany NPIP 6 µg/kg Eisenbrand et al. (1975)
of pepper grains (Federal Republic of)
on the outside.

Whole product Germany NPY 6
homogenized. (Federal Republic of)

ham, fried Germany DMN 1 µg/kg Eisenbrand et al. (1975)
(Federal Republic of) NPIP 8
NPY 19

Table 2 (Cont'd)

N-nitroso
Meat Country compounds Levels (^c) Reference
or area found^b

ham with layer of Germany NPIP 4 µg/kg Eisenbrand et al. (1975)
pepper grains (Federal Republic of)
on the outside. Germany NPY 9
Whole product (Federal Republic of)
homogenized.

German type of Germany DMN 1 µg/kg Eisenbrand et al. (1975)
bacon, raw (Federal Republic of)

German type of Germany DMN 1 µg/kg Eisenbrand et al. (1975)
bacon, fried (Federal Republic of) NPIP 5 Eisenbrand et al. (1975)
Germany NPY 19 Eisenbrand et al. (1975)
(Federal Republic of)

smoked raw meat Germany DMN 2 µg/kg Eisenbrand et al. (1975)
(Federal Republic of)

smoked ham Germany DMN 8 µg/kg Eisenbrand et al. (1975)
(Federal Republic of)

smoked ham USA DMN 5 µg/kg Fazio et al. (1971b)

smoked ham USA DMN 5 µg/kg Fiddler et al. (1974)

cooked and smoked Netherlands DMN 0.4 µg/kg(^f) Stephany et al. (1976)
ham DEN 0.6(^f) Stephany et al. (1976)

Table 2 (Cont'd)

N-nitroso
Meat Country compounds Levels (^c) Reference
or area found^b

Netherlands NDBA 0.4(^f) Stephany et al. (1976)
NPY 0.3(^f) Stephany et al. (1976)
Netherlands NPIP 0.4(^f) Stephany et al. (1976)

cooked ham Netherlands DMN 6 µg/kg Groenen et al. (1976)

Bologna sausage Canada DEN 25 µg/kg Panalaks et al. (1974)
Canada NPY 20,100,105 Panalaks et al. (1974)

frankfurter sausage USA NPIP 50, 50, 60 µg/kg Wasserman et al. (1972)

spiced meat Canada DMN 5-48 µg/kg Sen et al. (1976)
products DEN 6-16
Canada NPIP 14-50 Sen et al. (1976)
Canada NPY 7-33 Sen et al. (1976)

fish meal Canada DMN 0.35-0.5 mg/kg Sen et al. (1972)

smoked, nitrate/ USA DMN 4-26 µg/kg Fazio at el. (1971a)
or nitrite treated
sable, salmon shad

fresh, smoked or UK DMN 1-9 µg/kg Crosby et al. (1972)
salted fish

salted white Hong Kong DMN 40-100 µg/kg Fong & Chan (1973a, 1973b)
herring

Table 2 (Cont'd)

N-nitroso
Meat Country compounds Levels (^c) Reference
or area found^b

salted yellow Hong Kong DMN 10-60 µg/kg Fong & Chan (1973a, 1973b)
croakers

crude salt salted Hong Kong DMN 400 µg/kg Fong & Chan (1973a, 1973b)
white herring

crude salt salted Hong Kong DMN 200 µg/kg Fong & Chan (1973a, 1973b)
yellow croakers

prime salt salted Hong Kong DMN 10 µg/kg Fong & Chan (1973a, 1973b)
white herring

prime salt salted Hong Kong DMN 5 µg/kg Fong & Chan (1973a, 1973b)
yellow croakers

salted anchovies Hong Kong DMN 20 µg/kg Fong & Chan (1973a, 1973b)

whole herring meal Hong Kong DMN 300 µg/kg Fong & Chan (1973a, 1973b)

^a Adapted from Sen (1974)
^b DMN -- N-methyl- N-nitrosomethanamine ^c All values confirmed by mass spectroscopy
DEN -- N-ethyl- N nitrosoethanamine ^d mean of 4 samples
NDBA -- Nitroso- N-butylamine ^e mean of 5 samples
NPY -- Nitrosopyrrolidine ^f mean of 6 samples
NPIP -- Nitrosopiperidine ^g mean of 10 samples

Several authors who detected nitrosamines in foods by screening
methods but did not confirm these results by mass spectrometry include
Ender et al. (1964, 1967), Ender & Ceh (1967), Fong & Walsh (1971),
Freimuth & Glaser (1970), Hedler & Marquardt (1968), Kröller (1967),
Lembke & Moebus (1970), Möhler & Mayrhofer (1968, 1969), and Sakshaug
et al. (1965). Nevertheless, results by screening methods should not
be entirely ignored.

Considerable variations have been found in the levels of volatile
N-nitroso compounds in fried and grilled bacon. Attempts to
correlate these levels with levels of nitrates and nitrites did not
reveal any definite pattern. Telling et al. (1974) studied the effect
of various cooking temperatures on the levels of N-nitroso compounds
in grilled bacon. The results indicated that the levels of DMN
remained fairly constant as the cooking temperature was raised but
those of nitrosopyrrolidine increased.

Lipid soluble nitrosamines have an affinity for the fatty
portions of food (Sen et al. 1973a).

5.2.4 Tobacco and tobacco smoke

Since precursors for the formation of nitrosamines occur in
tobacco, Druckrey & Preussman (1962) thought it likely that tobacco or
tobacco smoke might contain trace amounts of nitrosamines. Initially,
studies on the nitrosamine content of tobacco products were hampered
due to interference from other compounds. Later, evidence suggesting
the presence of DMN, nitrosopyrrolidine, methylbutylnitrosamine, and
nitrosopiperidine in tobacco smoke was obtained (Kröller, 1967;
Neurath et al., 1964; Neurath, 1972, Serfontein & Hurter, 1966).
Although anabasine and nornicotine are constituents of tobacco smoke,
the corresponding nitroso derivatives were not detected by Neurath
(1972). Recently, however, Hoffman et al. (1974) reported the presence
of N-nitrosonornicotine at levels of up to 88 mg/kg in unburned
tobacco.