CARBONYL-AMINE REACTION
PRODUCTS AS POSSIBLE NITROSAMINE
PRECURSORS

* Dissertation for the Degree of Ph. D.

' MICHlGAN STATE UNIVERSITY

JAMES TILDEN MARSHALL JR.
1974

 

i
I
J)
U)

 

M :r. 3.15;". State

553 Unite i ity r l

 

This is to certify that the
thesis entitled
Carbonyl-Amine Reaction Products As
Possible Nitrosamine Precursors

presented by

James Tilden Marshall, Jr.

has been accepted towards fulfillment
of the requirements for

 

 

Ph.D. degreein Food Science and
Human Nutrition

74WQZMW

Major [ofessofl'

Date December 194 1973

 

0-7639

    
      

ambmc 3v ‘3‘
"MG & SUNS’
800K BINDEHY INC.

LIBRARY BINDE RS
"HIGH". Ila-IRA.

  

ABSTRACT

CARBONYL-AMINE REACTION PRODUCTS
AS POSSIBLE NITROSAMINE
PRECURSORS

BY

James Tilden Marshall, Jr.

Imines can be found in foods as a result of con-
densation of aldehydes with amino groups during non-
enzymatic browning. The purposes of this investigation
were to determine (1) if such imines would react directly
with nitrite to form N-nitrosamines; (2) if N-nitro-
samines would form from a reaction between sodium
nitrite and an intermediate in the synthesis of imines;
or (3) if nitrosamines could be formed in a model dry
food system, designed to enhance non-enzymatic browning
in the presence of aldehyde and/or sodium nitrite.

Aliphatic imines, synthesized from aldehydes and
primary amines, were exposed to sodium nitrite or
nitric oxide. Sodium nitrite was also added directly
to the aldehyde-amine reaction mixture. Reaction pro-
ducts were tested for nitrosamine content with Griess

and ninhydrin reagents; and, whenever possible, products

James Tilden Marshall, Jr.

were characterized by boiling point and by infrared spec-
trophotometry and gas chromatography-mass Spectrometry
(CC-MS). Nitrosamines were not found in any of the
reaction products.

Sodium nitrite was incorporated into emulsified
ground ham muscle. Half of the samples were freeze-
dried to remove water prior to a fifteen-hour heat
treatment; the other half were given only the heat
treatment. Colorimetric evidence of nitrosamines was
found in samples to which an excessive quantity of
sodium nitrite had been added. Non-freeze-dried samples
consistently tested positive with Griess and ninhydrin
reagents, while the freeze-dried meat contained only
traces of apparent nitrosamine. Samples which gave
positive colorimetric tests were analyzed by GC-MS
but the apparent nitrosamine compounds were not
identified when compared to mass spectra of known

nitrosamines and pyrazines.

CARBONYL-AMINE REACTION PRODUCTS
AS POSSIBLE NITROSAMINE

PRECURSORS

BY

James Tilden Marshall, Jr.

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Food Science and Human Nutrition

1974

ACKNOWLEDGMENTS

Sincere gratitude is extended to Dr. L. R.
Dugan, Jr. for his help, guidance, and suggestions
throughout the author's course of study at Michigan
State University.

The author would also like to thank Mr. Jack
Harten for operating the mass spectrometer; Dr. H. Roper
(Hamburg, Germany) and Dr. D. Osborne (Sharnbrook,
United Kingdom) for the nitrosamine standards which
they provided; and Dr. P. Markakis, Dr. R. A. Merkel,
Dr. L. L. Bieber, and Dr. W. W. Wells for their assis-
tance in the preparation of this manuscript.

To his wife, Cecelia, the author gives special

thanks for her encouragement and patience.

ii

TABLE OF CONTENTS

LIST OF TABLES . . . . . .
LIST OF FIGURES . . . . . .
INTRODUCTION . . . . . . .
REVIEW OF LITERATURE . . . .
EXPERIMENTAL PROCEDURES . . .

Safety Precautions . . . .

Synthesis of Nitrosamine Standards from

Secondary Amines . . . .
Synthesis of Imines. . . .

Attempted Syntheses of Nitrosamines

Imine and Sodium Nitrite .
Imine and Nitric Oxide. .

Aldehyde, Primary Amine, and Sodium

Ground Ham Model System . .
Preparation of Ham Samples
Extraction. . . . . .
Analysis of Ham Extract .

Gas Chromatography . . . .

Gas Chromatography--Mass Spectrometry.

Infrared Spectrophotometry .
pH 0 O O O O
Thin-Layer Chromatography. .

Griess Reagent . . . .
Ninhydrin Reagent . . .

iii

Page

vi

23
23

23
24
25

25
26
27

27

27
29
30

30
30
31
31
31

32
32

Page

RESULTS AND DISCUSSION . . . . . . . . . . 33
Synthesis and Identification of Standard NEBA . 33
Synthesis and Identification of Imines. . . . 37
Attempted Syntheses of Nitrosamines. . . . . 47

HMI and Sodium Nitrite . . . . . . . 47

BEI and Sodium Nitrite . . . . . . . . 49

Imine and Nitric Oxide . . . . 53
Aldehyde, Primary Amine, and Sodium Nitrite . 55
Ground Ham Model System. . . . . . . . . 56
SUMMARY AND CONCLUSIONS. . . . . . . . . . 61
APPENDIX. 0 O O O O O O O O O O O O O 63
REFERENCES CITED . . . . . . . . . . . . 74

iv

LIST OF TABLES

Table Page

1. Levels of N-nitrosamines reported in several
species of fish as confirmed by thin—
layer chromatography (TLC), gas chroma-
tography (GC), or mass spectrometry (MS) . . 9

2. Maximum levels of N-nitrosamines reported in
various meat products as confirmed by gas
chromatography (GC) or mass spectrometry
(MS) . . . . . . . . . . . . . . 12

3. Relative contribution of eight ions to the
total ionization produced by electron
impact fragmentation in synthesized NEBA
as compared with values previously
reported (60) . . . . . . . . . . . 36

4. Presence of nitrosamine-positive compounds in
the extract from "dry" and "wet" model
food systems as indicated by Griess and
ninhydrin reactions on thin-layer plates . . 58

LIST OF FIGURES

Figure Page

1. Mechanism for the reaction between phospha-
tidylethanolamine and nonanal . . . . . 22

2. Infrared spectrum of N-nitrosoethylbutylamine
(NEBA) standard prepared in this
laboratory . . . . . . . . . . . 34

3. Mass Spectrum of N-nitrosoethylbutylamine
(NEBA) standard prepared in this
laboratory . . . . . . . . . . . 35

4. Infrared spectrum of N-hexylidenemethylamine
(HMI) O O O O O O C O O O O O O 39

5. Infrared spectrum of N-butylidene-ethylamine
(BEI) O O O O O O O O O O O O O 39

6. Mass spectrum of N—hexylidenemethylamine
(HMI) . O O O O O O O O O O O O 4o

7. McLafferty rearrangement of HMI to form an
m/e 57 fragment ion . . . . . . . . 41

8. Mass spectrum of N-butylidene-ethylamine

(BEI) . . . . . . . . . . . . . 42

9. Mass Spectrum of HMI "dimer" . . . . . . 44
10. Mass spectrum of BEI "dimer" . . . . . . 45
ll. Infrared spectrum of HMI "dimer" . . . . . 46

12. Infrared spectra of HMI and the 30, 113, and
125 C/l7 mm fractions (A, B, C, and D,
respectively) distilled from a mixture
of HMI and sodium nitrite . . . . . . 48

vi

Figure

13A.

13B.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

Mass spectrum of the 113 C/20 mm fraction
distilled from a mixture of HMI and

sodium nitrite .

Mass spectrum of the 125 C/20 mm fraction
distilled from a mixture of HMI and

sodium nitrite .

Infrared spectra of BEI (A) and the 62—72
(B) and 108-112 (C) C/12 mm fractions
distilled from a mixture of BEI and

sodium nitrite .

Mass Spectrum of the 62-72 C/12 mm fraction
distilled from a mixture of BEI and

sodium nitrite .

Infrared spectra of N-nitrosodimethylamine
(A) purchased from Eastman Kodak Company

and (B) synthesized in this laboratory .

Mass Spectrum of N-nitrosodimethylamine
purchased from Eastman Kodak Company

Mass Spectrum of N-nitrosodimethylamine

synthesized in this laboratory.

Infrared Spectra of N-nitrosodimethylamine
-diethylamine (B), -dipropylamine
(C), -dibutylamine (D), -methylpropy1amine
(E), -methy1butylamine (F), —ethy1butyla—
(G), ~piperidine (H), and -pyrroli-

(A):

mine
dine (I) . . .

Mass spectrum of N-nitrosodiethylamine pur-

chased from Eastman Kodak Company.

Mass spectrum of N-nitrosodipropylamine pur-
chased from Eastman Kodak Company.

Mass spectrum of N-nitrosodibutylamine pur-
chased from Eastman Kodak Company.

Mass Spectrum of N—nitrosomethylpropylamine
Institut ffir

obtained from Dr.

Hamburg 13, Papendamm

vii

H. Roper,
Organische Chemie und Biochemie, 2

Page

50

51

52

54

63

64

65

66

67

68

69

70

Figure

24. Mass spectrum of N-nitrosomethylbutylamine
obtained from Dr. D. Osborne, Unilever
Research Laboratories, Colworth House,
Sharnbrook, Bedfordshire, UK . . . .

25. Mass spectrum of N-nitrosopyrrolidine pur-
chased from Eastman Kodak Company . .

26. Mass spectrum of N-nitros0piperidine pur-
chased from Eastman Kodak Company . .

viii

Page

. 71
. 72
. 73

INTRODUCTION

Inclusion of sodium nitrite in meat—curing
mixtures was approved by the United States Department
of Agriculture in 1925. Sodium nitrite stabilizes the
color of cured meat, contributes flavor, and aids in
preservation. The preservative function is especially
important in semi-perishable cured meats and in smoked
fish.

The addition of sodium nitrite to foods has
been challenged, because of the participation of nitrite
in formation of N-nitrosamines. The N-nitrosamines are
toxic and carcinogenic for a wide range of animal
species, and may be readily synthesized from secondary
or tertiary amines and sodium nitrite under acidic con-
ditions.

In order to establish whether nitrosamines are
formed in foods which contain nitrite and/or nitrate,
many investigators have analyzed food products directly
for the volatile nitrosamines. Since non-volatile
nitrosamines are difficult to detect, due to inter-
ference by other food constituents, their occurrence

in foods has not been established. The concentrations

of volatile nitrosamines found have seldom exceeded a
few parts per billion. Detection of a given nitrosamine
does not indicate the source of the precursor; therefore,
the conditions or events resulting in an increased
occurrence cannot be controlled.

The likely occurrence of both volatile and non-
volatile nitrosamines could be inferred if the necessary
precursors are present in a food product. For example,
imines can be found in foods as a result of condensation
of aldehydes with amino groups during non-enzymatic
browning. The purposes of this investigation were to
determine (1) if such imines would react directly with
nitrite to form N-nitrosamines; (2) if N—nitrosamines
would form from a reaction of sodium nitrite with an
intermediate in the synthesis of imines; or (3) if
nitrosamines could be formed in a dry model food system,
designed to enhance non-enzymatic browning in the

presence of aldehyde and/or sodium nitrite.

REVIEW OF LITERATURE

The organic compounds R-N=O may be nitrites,
-O-N=O; C-nitroso compounds, -C-N=O; or N-nitroso com-
pounds, -N-N=O. The latter group of compounds has been
investigated extensively in recent years, and several
members of this class have been demonstrated to be
toxic, carcinogenic, mutagenic, or teratogenic (12,

49, 80).

The hazard posed by these N—nitroso compounds
was not well known prior to a report in 1954 of two men
suffering with liver cirrhosis after having worked with
dimethylnitrosamine (NDMA) in a research laboratory (3).
Also reported were the results of NDMA toxicity studies
using rats, rabbits, mice, guinea pigs, and dogs. With
these animals, doses of 20-40 mg/kg of body weight
resulted in severe liver necrosis leading to death.

This early report of the hazardous properties of~
nitrosamines did not signal an instantaneous curtail-
ment of their use by the chemical industry. As late

as 1963, patents were issued for various nitrosamines

for use as gasoline and lubricant additives, antioxidants,

stabilizers, rubber additives, bacteriocides, etc. (81).

Another indication of N-nitrosamine toxicity
resulted from a report in 1964 describing the circum-
stances surrounding the illness and death of several
cattle, sheep, and goats in Norway (37). An investi-
gation into the cause of an unusual liver disease
which began to appear in these animals in 1961, estab-
lished that all of the animals had been fed herring
meal, and that most of the meal was produced by only
one of approximately 100 herring meal factories in
Norway. Koppang gt_gl. described production of herring
meal from herring preserved with nitrite and formal-
dehyde, and concluded that excessive levels of nitrite
present in herring during processing can result in the
formation of an unidentified toxic agent (38). The
toxic compound was not nitrite itself, as addition of
large amounts of nitrite to non-nitrite-treated meal
just prior to feeding did not result in toxic hepatosis.
The formaldehyde did not produce toxic hepatosis either,
but excessively high concentrations did produce a meal
which had some harmful effects of a different nature.
The toxic substance was subsequently isolated and
identified as dimethylnitrosamine; and its formation
from methyl amines present in nitrite-treated herring
meal was demonstrated (16, 17, 18, 58).

The acute oral LD50 in rats has been determined

for several of the N-nitroso compounds (49); few have

been found to be more toxic than N-nitrosodimethylamine,
the first to be tested. The manifestation of toxicity
seems to be linked with the chemical structure of the
various N-nitroso compounds. The two basic classes for
N-nitroso compounds, by structure, are nitrosamines and
nitrosamides.

The nitrosamines are fairly stable and appear to
remain intact under physiological conditions (46).
Nitrosamines may not be toxic as such; but if broken
down by enzymes, an active metabolite, such as a
diazoalkane or a carbonium ion, may be released (13,
49). Since the liver metabolizes foreign compounds
at a high rate, more of the toxic component is released,
and thus more damage results in the liver than in less
enzymatically active organs (46).

The nitrosamides are much less stable and will
decompose Spontaneously in neutral to alkaline pH
ranges (12, 47, 80). Thus, the nitrosamides do not
exhibit a marked selective toxicity towards the liver
as do nitrosamines.

Magee and Barnes, in 1956, reported the
induction of malignant liver tumors in the rat by NDMA
(48). Discovery of the strongly carcinogenic nature
of the N-nitroso compounds led to studies by a number
of scientists working in the field of cancer research,

particularly those interested in eXperimental tumor

production (45, 49, 80). Approximately 100 N-nitroso
compounds have been tested for carcinogenicity, and
about 75% were active carcinogens for several animal
species. Druckrey g£_§l. reviewed data concerning the
carcinogenic action of 65 different nitroso compounds
in the rat (12). The nitrosamines appear to affect
primarily those organs capable of enzymatic alteration
or breakdown of the parent nitrosamine into a more
active metabolite which Magee refers to as a "proximate
carcinogen" (45). However, nitrosamines have induced
cancer in the esophagus, kidney, bladder, lung, and
nasal cavity (49). In addition, a wide range of Species
including the rat, mouse, guinea pig, hamster, pig,
rabbit, trout, dog, and monkey have proven susceptible
to nitrosamines (46, 49). Thus, while cancer in man
has not been directly connected with nitrosamines, it
would seem probable that man is susceptible.
Nitrosamides produce local sarcomas when injected
subcutaneously, and if administered systemically, may be
carcinogenic in virtually every organ of the rat,
including the nervous system (49, 80). Tumors of the
nervous system are especially pronounced in the case
of transplacental induction of malignant tumors. For
example, ethylnitrosourea has been shown to induce
tumors of the nervous system in over 90% of the off-
spring when given as a single low dose to the pregnant

adult (13, 80).

Initially, the toxic and carcinogenic properties
of the N—nitroso compounds were considered a hazard
primarily for workers in industrial and laboratory
situations. However, the discovery of NDMA in nitrite-
treated herring meal raised the specter of carcinogen
formation occurring in foodstuffs containing nitrate
and/or nitrite. Nitrates and/or nitrites are added to
cured meats to enhance flavor, develop color, and aid
in preservation. In countries other than the United
States, these salts may also be added to certain types
of cheese, as a preservative (2). Nitrates are also
commonly found in plants, especially forage crops and
vegetables, water supplies, and soil (2, 84). Thus,

a source of nitrous acid, one of the major reactants
necessary for nitroso synthesis, is common to many
foods. If, in addition, a food contained amines or
other compounds suitable for N-nitrosation, and if
processing or storage conditions were favorable, the
potential for N-nitrosamine formation would seem very
real. It is not surprising, therefore, that a number of
food products have been analyzed for N-nitrosamine con-
tent. A partial list of the commodities examined
includes fish (7, 19, 21, 34, 39, 63, 65, 66), cured
meats (6, 7, 22, 25, 26, 43, 52, 64, 68, 69, 74),
cheese (7, 29, 30, 33, 65), milk (33), wheat flour

(33, 65), mushrooms (l9), ryebread (30), tobacco (35, 70),

spinach (7, 65), alcoholic beverages (67, 82), cereal
grains (S7), and fruit (15). It should be emphasized
that (1) not all of these products were found to contain
N-nitrosamines; (2) when N-nitrosamines were found,

they were seldom present in concentrations more than

100 ppb (100 ug/kg) and usually less than 10 ppb; and
(3) some of these reports are of questionable value,
especially those published prior to about 1969, because
the methods used to confirm the presence of nitroso
compounds lacked the required specificity.

Fish was one of the first food products to be
analyzed for nitrosamine content. If nitrosamines
could form.in fish meal, it seemed logical to suspect
a Similar occurrence in smoked fish intended for human
consumption, as the processing entails immersion in a
brine containing sodium nitrite, followed by a heat
treatment. For example, smoke processing of chub, a
freshwater fish, includes a brining treatment resulting
in no less than 3.5% salt (NaCl) and 100-200 ppm sodium
nitrite in the edible portion of the loin muscle.
Thermal processing must be sufficient to assure a
temperature of 160 F throughout the fish for a minimum
of 30 minutes (23).

Nitrosamine quantities reported in several fish
species and the methodology used to confirm their

presence are presented in Table l. Relying solely on

 

 

mzuoo H v H v H v mus mua and “so unmouo
mzuoo «are sane omnv Lame oanmm
oo m.~ v ovuo mane m m.~a w Ammo new
00 OH v Lame eumsom
we owe v ammo new
one com v Ammo axmsoumos
one m.o m.m w ma “mac amuse
m. m m. m m w m m m m
a. I a. a 1 X. P n P
3. P m T. x. 1 a D. a.
a 0 a a T. o
u .1 a a.
convex I mocmuomom

 

 

 

 

 

 

 

 

 

 

 

non .mocwfimmouuwclz

 

 

away muuoeouuuoom mmmfi no

.AUUV Sommumouosonno mom .Aoqav hammnooumsouno Homwatcwnu an oofiuwm
Icoo mm amen mo moaoomm amuo>om cw oouuooou mocfleomonuwclz mo mao>oq||.a momma

10

thin-layer chromatography (TLC), Ender and Ceh found in
smoked haddock as much as 15 ppb of a nitroso compound
which appeared to be NDMA (19). Koprowski investigated
the effect of two smoking temperatures and three nitrite
concentrations on the formation of NDMA in smoked chub
(39). He did not find NDMA; however, the minimum
detectable nitroso concentration with his TLC method
was 500 ppb. Using TLC, Sen gt_gl. occasionally found
Griess-positive compounds in the extract from herring,
but the quantities were too small to confirm (65).

The sensitivity of their TLC and gas chromatography
(GC) methods was 150 ppb for NDMA and 50 ppb for
diethyl- and di-n—prOpylnitrosamines (NDEA and NDPA).

The first investigators of nitrosamine formation
in foods relied primarily upon TLC and/or GC data for
the final identification of nitrosamines. However,
beginning in about 1971, mass spectrometry (MS) began
to find application in this area, and has increasingly
been the preferred method for final nitrosamine identifi-
cation. Fazio EE_El' confirmed by mass spectrometry (MS)
levels of NDMA ranging from 4 to 26 ppb in samples of
raw, smoked, and smoked nitrite- and/or nitrate-treated
sable, salmon, and shad (21). The highest value, 26 ppb,
was found in a smoked, nitrite- and nitrate-treated
sable. Raw sable contained 4 ppb and smoked, uncured

sable contained 9 ppb. Crosby et al. analyzed several

11

fish species and found from 1 to 9 ppb, which they con-
firmed with MS (7). It is interesting to note that

the only fish containing more than 4 ppb were those
which did not contain nitrite. Moreover, these fish
had been cooked, either fried or baked, and the authors
mention problems with quantitation in some of their
cooked samples due to interference from nitrogenous
compounds.

Maximum nitrosamine concentrations reported by
several research groups which have analyzed cured meat
products are presented in Table 2. Some authors reported
finding no nitrosamines at or above the minimum concen-
tration detectable with their methods; in these cases,
the minimum detectable level is listed in Table 2 pre-
ceded by the symbol "<,"

Fazio gt_al. analyzed 51 samples of cured.meats,
including (1) cold cuts, (2) sausages, (3) baby foods,
(4) canned meats, (5) bacon, hams, other pork products,
and (6) miscellaneous beef products (22). The maximum
nitroso concentration they found was 5 ppb in a sample
of smoked ham. This substance was identified by MS as
NDMA. The other samples were considered to contain
NDMA also, on the basis of GC retention times; but
since these concentrations were too low to confirm by
MS, the authors referred to these values as "apparent

NDMA.” This serves to illustrate the problem encountered

 

12

 

 

 

 

 

 

 

 

 

 

 

 

oo ma m m m m as Ammo mamamcmm
ms oav Loo newsmaumanao
m2 ooo.m~ om om om “mm .mm .emo new
we om Lmvo :eumoq
mg as q a any Shmouo
ms mmv mmv may mmv mmv lead mandama
ow oav em “SN .mmo umaeoae
m: N N m m m m m e INNS oases

s .s 1: Adm Ado o. o q s o

a I 1 IT. 11. m. p. E E 9 O

E a a OS on. a u a. n T.

a a. a a. an m w .. a. p

w q a... m p I. .1... m

u 8 .m 83 8w m m S 1.

ooauoz mm. m. m. m. m. % W s mucoummom
a
one .mocasmmouuwclz

 

 

 

.mzv muuosonuooom name no ”Dov mommuooumsouno mom mo ooeuwm
noon on muoooonm vows msowum> SH oouuomou mocwesmouuwclz mo mam>oa ESEmezll.m names

13

by many persons attempting to isolate and identify such
low nitrosamine concentrations. Very often, an indi—
cation of possible nitrosamine presence is seen, such

as a positive color appearing on a thin-layer plate, or
a GC retention time that is identical to the nitrosamine
standard; but unless the suspected compound is present
in a concentration of 10 ppb or greater, it is usually
impossible to isolate enough of the unknown to obtain

a mass spectrum,

Crosby gt_al. examined extracts of bacon, fish,
cheese, spinach, frying oil, and an assortment of cured
meats (7). If a sample showed retention times cor-
re3ponding to one of seven nitrosamine standards they
used, it was further analyzed by high-resolution mass
spectrometry. While most of the bacon samples analyzed
had no detectable nitrosamines, a few had 1 to 4 ppb,
and one sample had 40 ppb.

Fiddler gt_al. reported finding 11 to 84 ppb
NDMA in three of forty commercial frankfurter samples
(26). As 16 billion frankfurters were consumed in the
United States in 1971, the authors emphasized the sig-
nificance of finding even a few ug/kg. They also
investigated the relationship between NDMA formation
and the addition of excessive levels of sodium nitrite
to frankfurter formulations. Even when 750 mg nitrite

were added per kg of meat (5 times the legal limit of

14

156 mg/kg), no significant NDMA formation occurred.
With 1500 mg nitrite/kg or more, NDMA concentrations
of 10 ppb or more were found in the franks. A reduction
in NDMA formation occurred when sodium ascorbate was
added to the frankfurter formulation, which the authors
attributed to the reduction of nitrite to nitric oxide
during curing. The blocking of nitroso formation with
ascorbate was also demonstrated by Mirvish g£_gl, (50).
In a study which illustrates the importance of
adding sodium nitrite to cured meat, Christiansen gE_§l.
varied the nitrite levels in samples of canned com-
minuted pork between 0 and 500 ug/kg (6). The cans of

pork were inoculated with types A and B Clostridium

 

botulinum, given a normal thermal process, and then

 

placed in storage for as long as six months, at 7 C
or 27 C. Since this is a product which is labeled

”keep under refrigeration," the latter temperature

would test the keeping quality of the product and

favor the growth of g. botulinum. This was intended

 

to simulate the abuse which this product is believed

to receive at the retail and consumer level. No I
nitrosamines were found in any of the cans. Toxin pro-
duction was reduced or completely inhibited, depending
upon the number of spores inoculated at a given level

of nitrite. In addition to demonstrating the protective

effect of nitrite against toxin production, this

15

experiment also illustrated the value of sodium nitrite
in preventing non-toxic spoilage. Those cans to which
no nitrite was added and which did not receive spores
underwent non-toxic spoilage within one month, even when
stored at 7 C.

Sen gt_gl. reported a possible explanation for
the occasional presence of nitrosamines in cured
sausages (69). They confirmed the presence of 2500 ppb
nitrOSOpyrrolidine (NPYR), 850 ppb NDMA, and 25,000 ppb
nitrosopiperidine (NPIP) in a dry curing powder which
consisted of black pepper, paprika, two secret spices,
salt, sodium nitrate, and sodium nitrite (0.96%). In
further tests they demonstrated that both black pepper
and paprika would react with nitrite to produce the
NPYR and NPIP, while the paprika was apparently
responsible for the NDMA formation. The two secret
ingredients did not react to form either NPYR or NPIP.
The authors suggest that pre-mixing of the dry spices
with nitrite, followed by prolonged storage before use,
could be a hazardous practice.

Methods for the recovery, quantitation, and con-
firmation of N-nitrosamines in foods were reviewed in
1971 by Walters (76) and in 1972 by Wasserman (79).
Many researchers have analyzed foods for N-nitrosamine
content, but the methods they have used were usually

developed independently and for a specific application.

16

Unfortunately, there is no standard method which can be
used for all N-nitrosamines and all foods.

Thin-layer chromatography has been used exten-
sively for detection of nitrosamines (28). Some of the
earlier reports of nitrosamine occurrence in food
extracts relied solely on TLC for detection and con-
formation. A primary application of TLC in more recent
investigations has been the screening of food extracts.
Those samples which TLC indicates as containing nitro-
samines are then subjected to more extensive clean—up
and analytical procedures. Quite often, TLC is used to
isolate nitrosamines from contaminants so that a "clean"
mass spectrum may be obtained.

The TLC visualization method most often used is
the photolytic cleavage by ultraviolet light of the
-N=O group, which then combines with Griess reagent
(sulfanilic acid and l-naphthylamine, or one of several
similar diazotizing reagents) to produce a colored spot
on the plate (9, 54). Compounds other than nitrosamines,
such as esters of nitrous acid, C-nitro- compounds,
inorganic nitrate, inorganic nitrite, and pyrazines can
cause development of a positive Griess color (77). It
is, therefore, a good practice to confirm a positive
Griess test by using some other visualizing reagent.

Since the photolytic cleavage of nitrosamines produces

17

the parent amine (24), the ninhydrin test is often used
in conjunction with the Griess test (65, 67).

Infrared spectrophotometry (IR) has found little
application in the analysis of foods for nitrosamine
content, as the quantities extracted are insufficient
to obtain IR spectra. However, IR is used in confirming
the identity of nitrosamine standards. Aliphatic nitro-
samines Show N=O stretching absorption in the 1425-

1460 cm.1 region; N-N stretching absorbs around 1030-

1150 cm"1

(83).

Polarography has found a very limited use in
nitrosamine analysis. Devik used TLC, GC, and polar-
ography to analyze products of a Maillard reaction
between glucose and amino acids (10). He reported
the formation of nitrosamines in this model system.
However, this statement was retracted in a subsequent
paper by Kadar and Devik (36). Re-examination of the
products of the amino acid-glucose reaction by MS con-
firmed the apparent nitrosamines were actually pyrazines.
This agrees with the results found by Havre and Ender
(32). Walters gt_gl. state that unsaturated aldehydes
and ketones also give polarographic results similar to
nitrosamines (77).

The nitrosamine concentrations in foods reported

to date have been quite low, often less than 10 ppb

(pg/kg). To isolate, identify, and quantitate such low

18

levels of nitrosamine is.a difficult task, made.more so
by the complexity and variety of potentially interfering
substances present in various food products. The minimum
nitrosamine quantity which can be visualized on a thin-
layer plate when using Griess or ninhydrin reagents is,
at best, 0.2 to 0.5 pg (65, 76). A 1 ug sample is a
reasonable quantity to spot on a thin-layer plate if
consistent sample visualization is to be obtained.

Thus, to test an extract with Griess and ninhydrin,

in duplicate, would require 4 ug of nitrosamine. If

the original food contained 4 ppb (ug/kg) nitrosamine,
the sample size prior to extraction would have to be

at least 1000 g and the nitrosamine recovery 100%. It
is apparent that examination of a food containing 10 ppb
nitrosamine entails very large samples if the method of
determination is thin-layer chromatography.

One of the major advantages of gas chromatography
as compared to TLC is its greater sensitivity. Howard
g£_21. described a GC method capable of detecting 10 ppb
NDMA in smoked fish (34). The quantity of NDMA required
to produce a clearly pronounced peak was 4.5 ng. This
is about 1000 times less than the quantity needed for
TLC visualization. Sen obtained another 1000-fold
increase in sensitivity by oxidizing NDMA to dimethyl-
nitramine (C2H6NN02), which is extremely sensitive to
electron capture detection (63). However, the conversion

to the nitramine is not quantitative and may be variable.

19

Gas chromatography is clearly more sensitive
than TLC, but it does have certain limitations. When
using TLC, nitrosamines are distinguished from other
compounds which may be present in a food extract by
means of a fairly specific colorimetric reaction. With
gas chromatography, one must compare the retention time
of a peak in question to the retention time of a
standard nitrosamine. This immediately limits GC
analysis to those nitrosamines for which standards
are available. In addition, there is always the chance
that some other compound present in the extract may have
the same retention time. For example, pyrazines commonly
found in cooked foods often have retention times similar
to nitrosamines (59, 61).

Williams et_al. used a gas chromatograph
coupled with a mass spectrometer (GC-MS) to demonstrate
that the aldehyde furfural was the compound previously
mistaken for a nitrosamine in an extract from an
alcoholic beverage (82). GC-MS combines the sensi-
tivity and resolving power of gas chromatography with
the ability of a mass Spectrometer to unambiguously
characterize the sample. Mass spectra of several
N-nitrosamines have been deposited with the ACS Micro-
film Depository Service (53). Saxby investigated the
fragmentation by electron impact of 24 dialkyl-N-

nitrosamines, and proposed decomposition mechanisms (60).

20

N-nitrosamines are generally yellow, neutral to
weakly basic compounds, and are insoluble in dilute
aqueous mineral acids (51, 71). The stability of four
representative N-nitrosamine structures has been reported
by Fan and Tannenbaum (20). Secondary amines, whether
aliphatic or aromatic, readily combine with nitrous acid
to yield N-nitrosamines. Aliphatic tertiary amines
will also form N-nitrosamines, but aromatic tertiary
amines undergo ring substitution to yield C-nitroso-
compounds (51). When primary amines react with nitrous
acid, the product is a diazonium salt. The diazonium
salts of primary aliphatic amines are very unstable
and break down to yield a mixture of alcohols and
alkenes (51). However, Ender g£_al, reported a 9%
yield of NDMA from monomethylamine and nitrous acid (18).

The presence in foods of substantial quantities
of nitrogen-containing compounds which could serve as
precursors to N-nitrosamine formation has been questioned.
Abundant quantities of volatile secondary amines are
not common in most biological systems, although marine
fish contain more of these amines than do fresh water
fish (39). Fiddler §£_gl, have demonstrated formation
of NDMA from naturally occurring quarternary ammonium
compounds (27). Archer gt_§l. were able to Show for-
mation of N-nitrososarcosine from creatine, a normal

constituent of meat (1). Various amino acids have

21

served as precursors of N-nitrosamines (41, 78). Lijinsky
§£_gl, obtained measurable yields of nitrosamine from
each of six drugs which had tertiary nitrogen structures
(42).

Still other sources of nitrogen compounds which
have not received attention as nitrosamine precursors
are the reaction intermediate and end-product resulting
from the interaction of primary amines and aldehydes.
Dugan and Rao have investigated the Maillard browning
reaction in dry model systems containing phospholipids
and aldehydes (14). They were primarily interested in
the effect of such Maillard reactions on the flavor of
dry foods in prolonged storage situations. However,
they demonstrated the ready formation of tertiary
nitrogen Schiff bases which.may provide another source
of N-nitrosamine precursors. The proposed mechanism
for the reaction between phosphatidylethanolamine (PE)
and nonanal is shown in Figure l. The intermediate
compound contains a secondary amino group which could
possibly react with nitrous acid, provided the inter-
mediate exists as a secondary amine for a finite time
span. The end-product is a tertiary amine, and it
could also prove to be a site for attack by nitrous

acid.

22

l . -
O-f-O-CHz-CHZ—NHZ + men-can"

{ 'l
H

o-h-o-cnz-cnz-h-gn-cau

. I7

LSECONDARY NITROGEN INTERMEDIATE-J

l-HOH

{ ' -
O-f-O-CHZ-CHZ-N-CH-Cal-ln
TERTIARY NITROGEN SCHIFF BASE

 

 

Figure l.--Mechanism for the reaction between phospha-
tidylethanolamine and nonanal

EXPERIMENTAL PROCEDURES

Safety Precautions

 

N-nitroso compounds are both toxic and carcino-
genic; therefore, rubber gloves were always used when
working with them. All experiments involving these
compounds or nitric oxide gas were conducted under an
exhaust hood.

Synthesis of Nitrosamine Standards
from Secondary Amines

N-nitrosamine standards not purchased from com-
mercial sources were prepared from secondary amines by
a method similar to that described by Hatt (31). The
synthesis of N-nitrosoethylbutylamine (NEBA) is a
typical example.

In a 500-ml round-bottom flask provided with
an oval-shaped magnetic stirring bar were placed 101 g
(1 mole) of N—ethyl-n-butylamine and sufficient 2N HCl
to obtain a pH of 6.0 to 6.5. This solution was stirred
vigorously and maintained at 70 to 75 C by heating on a
water bath, while 100 g (1.4 moles) of 97%-pure NaNO2

suspended in 100 ml water were added from a dropping

funnel over a period of an hour. The reaction mixture

23

24

was tested frequently and maintained at pH 6.0 to 6.5
by further addition of 2N HCl when necessary. Stirring
and heating were continued for two hours after all the
sodium nitrite had been added. The flask was arranged
for distillation, and the reaction mixture distilled
under vacuum until the residue was practically dry. To
the residue 140 ml water were added and the process of
distillation to dryness repeated. The distillates were
combined and saturated with potassium carbonate (approxi-
mately 1 lb was required). The organic layer was
removed, dried over anhydrous potassium carbonate, then
purified by another vacuum distillation.

The identity of the nitrosamine standards was
confirmed by Griess and ninhydrin tests, and by com-
parison of boiling points and infrared and mass spectra

with data reported in the literature (53, 55, 60, 83).

Synthesis of Imines

 

Imines were prepared from aliphatic aldehydes
and primary amines using a modification of the procedure
used by Campbell g£_al. (5). A typical example is the
synthesis of N-n-butylidene-ethylamine (C3H7CH:NC2H5).

In a SOD-ml round-bottom flask provided with an
oval-shaped magnetic stirring bar, 81.5 g (1 mole)

ethylamine hydrochloride and 100 ml distilled water

were stirred into solution and chilled in an ice bath.

25

A solution of 56 g (1 mole) KOH in 100 ml water, pre-
viously prepared and cooled to 5 C, was added and the
mixture stirred for thirty minutes to ensure chilling.
By means of a dropping funnel, 72 g (1 mole) butanol,
previously chilled to -26 C, were added gradually over
a period of one hour. Stirring was continued for fif-
teen minutes after addition was completed; then the
organic layer was removed and stored over KOH pellets
overnight. The following morning, the reaction products
were filtered into a 250-ml round-bottom flask, fresh
KOH pellets were added, and the mixture was distilled
at atmospheric pressure.

The imines prepared in this manner were char-
acterized by boiling points and by infrared and mass

Spectra.
Attempted Syntheses of Nitrosamines

Imine and Sodium Nitrite

Either N-n-hexylidenemethylamine (HMI) or N-n-
butylidene-ethylamine (BEI) were mixed with an excess
of acidified sodium nitrite. In a typical case, 69 g
(1 mole) NaNO2 and 200 ml water were placed in a 500-ml
two-necked round-bottom flask. The solution was
acidified with 6N HCl to pH 6.0 to 6.5 and maintained
in this range during the dropwise addition of 25 ml

N-n-butylidene-ethylamine. Stirring was continued for

26

one hour after all the imine had been added. The aqueous
layer was removed and washed with three 50—m1 aliquots

of ether. The ether extracts and the organic layer were
combined and fractionally distilled in vacuo. The
fractions, separated according to boiling point, were
further characterized by infrared Spectra and by the
Griess test. Based on these results, selected fractions

were analyzed by gas chromatography-mass spectrometry.

Imine and Nitric Oxide

Ray and 099 (56) have shown that a powdered
mixture of sodium nitrite, sodium nitrate, chromic oxide,
and ferric oxide (3:1:2:3 molar ratio) will evolve nitric
oxide gas which is 99.78% pure. Nitric oxide generated
by heating this powder was bubbled through either BEI or
N-benzylidenemethylamine. The apparatus used for this
treatment consisted of two small Pyrex test tubes
(20 mm x 150 mm) and a cylinder of nitrogen. A 109ml
aliquot of the imine was placed in the first tube; 10 g
of the dry powder in the second. The tubes and the
nitrogen tank were connected in series, with the nitric
oxide tube upstream from the tube containing imine.
The tubes were flushed with nitrogen, and a small
stream of nitric oxide was evolved from the powder by
mild heating with a Bunsen burner. Heating was termi—
nated when gas production was no longer evident, usually

after two hours. The sample was purged with nitrogen

27

for fifteen minutes, then transferred to a storage vial.
The infrared spectra and Griess reaction of each product
was recorded.

Aldehyde, Primary Amine,
and Sodium Nitrite

 

 

To a SOD-ml two-necked round-bottom flask were
added 32 m1 of a 70% solution of ethylamine (approxi-
mately 0.5 mole). The amine solution was chilled in an
ice bath, and the acidity adjusted to pH 3.0 with
GM HCl. A 100-ml aqueous solution containing 69 g
(1 mole) sodium nitrite was then added, followed imme-
diately by the drOpwise addition of 36 g (0.5 mole)
n-butyraldehyde. Stirring was continued for an hour
after all the aldehyde had been added. The mixture was
then extracted with three 50-ml aliquots of dichlorome-
thane (CH2C12). The extracts were transferred to a
500-ml Kuderna-Danish evaporative concentrator, and
the dichloromethane removed by warming over a steam
table. The concentrated product was tested with Griess

reagent and the infrared spectra recorded.

Ground Ham Model System

 

Preparation of Ham Samples

 

Fresh (uncured) ham muscle, trimmed free of
visible fat prior to grinding, was obtained from the

Michigan State University Food Stores. The previous

28

history of the meat was not known. The ham was divided
into lOO-g portions, and various additives were dis-
persed throughout the ham by a 60-sec blending in a
250-ml stainless steel Waring blender jar. Additives

incorporated into the ham were
(1) 20 ml water (control);
(2) 21 ml water containing 2.76 g NaNOZ;

(3) 20 ml water containing 2.76 g NaNO2 and 2.0 9

NaCl;

(4) 20 ml water containing 2.76 g NaNO2 and 2.0 g

hexanal;
(5) 20 ml water plus 2.0 g hexanal;

(6) 20 ml water plus 2.0 g hexanal; 2.76 g NaNO2

added just prior to extraction; and
(7) 20 ml water plus 2 m1 ethanol containing 1 mg

each dimethylnitrosamine and dibutylnitrosamine.

After blending, all samples were frozen at -26 C.

Two treatments were used in this model study:
(1) Frozen ham samples were transferred directly into
a forced air oven (Cenco No. 95396-16) and heated at
70 C for fifteen hours. (2) Ham samples were freeze-
dried (Stokes Freeze Dryer, Model 2003-F2) and then

heated as above.

29

Extraction

 

After the heat treatment, the ham samples were
broken into small pieces and placed in a one-quart
stainless steel Waring blender jar, covered with 200 ml
dichloromethane, and blended for five minutes. The
blender was powered by an explosion-proof motor
(Waring EP-l).

The extract was filtered through Whatman Sharkskin
filter paper into a 500-ml distillation flask. A few
boiling chips and several KOH pellets were added, and
the flask was arranged for distillation at atmospheric
pressure. Distillation was terminated when the flask
appeared dry and liquid no longer distilled.

The distillate was transferred to a 500-ml
Kuderna-Danish evaporative concentrator fitted with a
5-ml calibrated concentrator tube and a three-section
Snyder distilling column. Carborundum boiling chips
were placed in the concentrator tube, and the solvent
was carefully concentrated to about 4 ml in a hot
water bath. The apparatus was removed from the bath
and allowed to cool, draining any remaining solvent
from the distilling column. The column was then
removed and the solvent concentrated to 0.5 ml under

a slow stream of nitrogen.

30

Analysis of Ham Extract

The concentrated ham extract was analyzed for
nitroso content by spotting 40-ul aliquots on thin-
layer plates (pre-coated with 0.25 mm silica gel G),
developing the plates with hexane: diethyl ether:
dichloromethane (4:3:2), and visualizing the sample

with either Griess or ninhydrin reagents.

Gas Chromatography

 

A Beckman GC-S dual column gas chromatograph
equipped with flame ionization detectors was used.
Temperatures of the detector and injection port were
240 and 185 C, respectively. A 9 ft 8-in x l/8-in o.d.
stainless steel column containing 3% OV-210 on 100-120
mesh Supelcoport was used with a nitrogen flow of 26 cm3
per minute. The air and hydrogen flows were 300 and 18

cm3 per minute, respectively.

Gas Chromatography--Mass Spectrometry

 

Mass spectra were obtained using a combined
GLC-mass spectrometer LKB 9000 equipped with a glass
column (6 ft x 1/8 in) of 3% OV-210, with ionizing
electron energies of 22.5 or 70 eV; the flash heater
set 20 degrees above the GC column temperature,
molecular separator at 230 C, and the ion source at
290 C. The spectra were recorded as bar graphs by means

of an on-line data acquisition and processing program (73).

31

Infrared Spectrophotometry
A Beckman IR-12 double beam infrared spectro-
photometer was used to record the spectra of neat samples

(thin films on NaCl cells) with air as a reference.

BE
A Corning model 12 pH meter equipped with a
Sargent—Welch (S-30070-10) miniature combination
electrode was used to monitor the pH during the syn-

thesis of nitrosamines.

Thin-Layer Chromatography

Glass plates, coated with 0.25 mm silica gel G
or silica gel G-HR, were used for the separation and
colorimetric detection of nitrosamines in synthesis end-
products and in model system extracts. Large plates
(20 cm x 20 cm), coated with silica gel G-HR, were
activated for one hour at 100 C; smaller plates (5 cm x
20 cm), precoated with Silica gel G, were used without
activation. Standard nitrosamines were diluted to 0.1%
(l ug/ul) with dichloromethane, and S—ug aliquots were
spotted on each plate as a reference.

The plates were developed with one of two solvent
systems, n-hexane: diethyl ether: dichloromethane
(4:3:2) gr chloroform: methanol: water (65:25:4); then
visualized with Griess reagent and/or ninhydrin. To

visualize with Griess reagent, the developed plate

32

was Sprayed, observed briefly, then placed in a light-
proof drawer. After five minutes, the plate was
inspected for red spots and irradiated at 254 nm for
five minutes using a Mineralight R-52. Red Spots which
develOped prior to UV irradiation were considered false
positive. When ninhydrin was the visualizing agent,
the plate was sprayed with 30% acetic acid, irradiated
for five minutes, sprayed with ninhydrin reagent, and
then heated at 100 C. Red-purple spots which appeared
within 10—15 minutes indicated the presence of a ninhydrin—
positive compound. Occasionally, both colorimetric
reagents were used on a single plate. This was
achieved by covering one side of the plate while
spraying the other with Griess reagent. Then the
Griess-treated side was covered while 30% acetic acid
was sprayed on the side originally covered. The plate
was irradiated, checked for Griess-positive spots, then

sprayed with ninhydrin and heated as before.

Griess Reagent

 

Solutions of sulfanilic acid (1% in 30% acetic
acid) and a-naphthylamine (0.1% in 30% acetic acid) were

stored at 4 C and mixed 1:1 just before using.

Ninhydrin Reagent

 

A 0.3% ninhydrin solution was prepared using

ethanol containing 2% pyridine.

RESULTS AND DISCUSSION

Synthesis and Identification of
Standard NEBA

 

 

Since there is only a limited selection of
N-nitrosamines available through chemical supply com-
panies, consisting mainly of short-chained, symmetrical,
dialkyl-N-nitrosamines and a few cyclic compounds,
N-nitrosoethylbutylamine (NEBA) was prepared by
reacting ethylbutylamine with nitrous acid. Confirm-
ation of the product as NEBA was based on its boiling
point, Rf value, and IR and mass Spectra as compared
with values reported in the literature.

The product was a clear yellow liquid which
boiled at 92 C/15 mm. The Rf value obtained for this
compound using TLC plates precoated with silica gel G
and the chloroform: methanol: water solvent system was
0.95; Similar plates developed in the hexane: ether:
dichloromethane solvent system showed a positive spot
at an Rf of 0.51. The latter Rf increased to 0.65 when
plates which had been coated with silica gel G-HR and
activated at 100 C were used. The GC retention time

of this compound was recorded as 3.8 minutes when an

33

34

oven temperature of 110 C was used. The IR Spectrum,
Figure 2, Shows strong absorption bands at wavelengths
1450-1470 cm”1 and 1075 cm-1, corresponding with
stretching of the N=O and N-N groups, respectively (53,
55, 82). A mass spectrum for this compound is shown
in Figure 3.

Pensabene gt_al. reported a boiling point of
95 C/l4 mm for NEBA (53). A reported Rf value was not
found for NEBA; however, Sen gE_al. reported Rf values
of 0.57 and 0.77 for methyl-n-butylnitrosamine and
di-n-propylnitrosamine, reSpectively (65). The Rf

value of NEBA should fall within this range.

 

 

CHBCHZ \
I
CH3(CH2)3

3

TRANSMOTTANCE .%
a
O

 

 

 

N
O

 

AL

 

 

L A A J J L 4L L A J A A A 41 A L A A
3200 2800 2400 2000 neoo IGOO I400 1200 1000 000
wavsuuuaen. cu"

Figure 2.--Infrared spectrum of N-nitrosoethylbutylamine
(NEBA) standard prepared in this laboratory

35

huoumuonma menu SH comma

 

o):
omfl OOH om
. _ . _ _ L). _ L C . _ t _.L _). Cf. _ . _.. c . LlL _ . x
_ _ . . .3: _ .1 _ __
on. m: m»
/
\hm
mm

 

 

 

 

 

 

 

mm

 

. . m. Nv
on. :2 \nlmxe Io

ozlz
xmzomzo

 

 

 

 

10m

10¢

tom

tom

 

 

02

some uncommon Admmzv mcwEmHmuanaaumOmonuflclz mo Ednuoomm mmm211.m oudmwm

BONVCINHBV 3Al1V138

36

Saxby has investigated the fragmentation pattern
of twenty-four dialkyl-N-nitrosamines subjected to 75 eV
electron impact, and tabulated the relative occurrence
of eight important ions (60). All of these ions were
present in the mass spectrum of the NEBA synthesis end-
product, and their occurrence relative to the total
ionization is compared in Table 3 with the values
reported by Saxby. A reading of £29 2.8% for the ion
at m/e 130 is interpreted as meaning that ion m/e 130
contributed 2.8% of the total ionization, excluding ions

smaller than m/e 29.

TABLE 3.--Relative contribution of eight ions to the
total ionization produced by electron impact
fragmentation in synthesized NEBA as compared
with values previously reported (60)

 

 

 

 

Fragmentation ions % Total ioniZation (£29)
m/e NEBA NEBA
(60)
130 2.8 3.1
113 1.7 1.7
87 4.5 5.4
56 9.8 9.1
42 9.4 9.6
88 5.5 7.9
75 0.6 0.6
30 7.2 5.8

 

37

AS an additional check on the procedure used for
the synthesis of NEBA, N-nitrosodimethylamine (NDMA) was
prepared from dimethylamine and compared to NDMA pur-
chased from Eastman Kodak Company, Rochester, New York,
14650. The IR and mass spectra of NDMA from both
sources are in the Appendix (Figures 16, 17, 18),
along with spectra of seven other nitrosamine standards
(Figures 19-26), N-nitroso-diethylamine, -dipropylamine,
-dibutylamine, -methylpropylamine, -methy1butylamine,
-pyrrolidine, and -piperidine. The mass spectra of
these standards are almost identical to the spectra
published by Pensabene, gt_gl. (53); however, the IR
spectra obtained from a thin film of these standards
do not contain the large solvent peaks present in the
IR spectra reported by Pensabene, §E_al. (53).

Synthesis and_Identification
of’Imines

 

 

Imines synthesized from aldehydes and primary
amines were fractionally distilled, and their identity
confirmed by infrared and mass spectra prior to their
use as substrates in subsequent nitrosamine syntheses.

The boiling points of the two imines synthesized
were 26-30 C/20 mm for N-hexylidenemethylamine (HMI),
and 98 C/735 mm or 20 C/lS mm for N-butylidene—ethylamine
(BEI). A boiling point of 100-108 C/760 mm has been

reported for BEI (5, 40).

38

The IR spectra of HMI, Figure 4, and BEI,

Figure 5, each contain a strong absorption band at
1675-1678 cm‘l. Absorption between 1666-1680 cm"1 is
typical of C=N stretching (8). The absence of a C=O
stretch in the 1700-1800 cm’1 range indicates that any
excess aldehyde from the reaction mixture had been
removed from the imines during fractional distillation.

A mass spectrometer equipped with a gas chromato-
graphic inlet was used for the final identification of
the distillation fractions which had tentatively been
identified as imines on the basis of boiling point and
IR spectrum. Both HMI and BEI were positively identified
by inspection of their fragmentation patterns. The mass
spectrum of HMI (CSHll

molecular ion at m/e 113. A base peak at m/e 57 is the

CH:NCH3), Figure 6, shows the

result of a McLafferty rearrangement (Figure 7) in which

.4.
a C4H8 molecule is lost and the m/e 57 ion, CH2:CHNHCH3,

is produced (4). The series of peaks separated by 14 m/e
units, 42, 56, 70, 84, 98, represent the ions produced
by the simple carbon-carbon fission at each of the five
bonds of this type. The ion at m/e 112 results from
loss of a hydrogen atom from the methyl group attached
to nitrogen.

The molecular ion (M+) for BEI (C3H7CH:NC2H5) is

not present in the mass spectrum, Figure 8. Ions are

(m/e 84), M+-c H

present for M+-H (m/e 98), M+-CH 2 5

3

39

 

80*

TRANSMITTANCE.%

20>

 

3

&
O

    
 

CH31CH2’4CH - N- CH3

 

 

xbo A

2600 A

A A A A A A A A A A A A A 4 A 4
2400 2000 I800 I600 I 400 I 200 I000 800
WAVENUMBER,CM"

Figure 4.——Infrared spectrum of N—

hexylidenemethylamine (HMI)

 

80-

$

TRANSMITTANCE.%
b
O

N
O

 

 

   
 

 

 

 

 

A A
3200

2600

A A A A A A g A A A A A A 1 A I.
2400 2000 I800 I600 I400 I200 I000 800
WAVENUMBER.CM"

Figure 5.-—Infrared Spectrum of N—

butylidene—ethylamine (BEI)

40

3:5 mcwfimamcuocacoogmwooIz m0 eBHuoomm mmmZII6 ousmam

o\E

 

1

 

 

m:s?2

mzoIz zocluzocmzo

lax

 

Oh

so

 

 

tom

.3

Iom

Iomw

 

OOH

 

BONVONOBV 3A|1V138

41

cow ucmfiomum hm o\E am Show on Ham mo ucofimmcmuummu monommmquzII.h musmwm

 

n: .2.
so 35 «\1o
mzolzalzoflmzo :m\ filo
, A
+ z../_._.._o
«Io Hzonzmo mzo o/zmo

42

“Home ooseaasauoIocooHHSoooIz no sosuooon onczII.m mucosa

 

 

 

 

 

 

 

QE
omfi OOH om
_ . _ . _ p r . _ . _ e .4 _ LL # ._ r b _ L. _
mm _ —_ _—_
cm ION
N¢
qu
Iom
so .222
Z. 0%
mxomxolznzomzomzomxo
on

 

OOH

 

BONVONDBV HAIIVWBH

43

(m/e 70), and M+-C3H7 (m/e 56). The large ion at m/e 71
is accounted for by a McLafferty rearrangement which
produces CH2:CHNHC2H5.
Aldehyde molecules are prone to interact by
aldol condensation to form B-hydroxyaldehydes. These
B-hydroxyaldehydes easily dehydrate to form stable
aldehydes in which a carbon—carbon double bond between
the a- and B-carbon atoms is conjugated with the
carbonyl group (51). Imines with an a-methylene group
can also undergo an aldol-type carbon-carbon conden-
sation (72). Analysis of the various fractions
obtained from the distillation of HMI and BEI confirmed
the presence of aldol-imine compounds in the higher
boiling fractions. For example, the aldol "dimer" of-

HMI (C4H -C-CH:NCH

9 H 3
CHC5H11

fraction collected at 100-115 C/20 mm. The mass spectrum

M.W. 195) was found in a distillation

of this imine, Figure 9, has a molecular ion at m/e 195.
The large peaks at m/e 42 and 152 represent the favored
carbon-carbon fission at the alpha and gamma bonds,
respectively (4). The series of peaks at m/e 110, 124,
138, 152, 166, and 180 represent ions formed by the
successive elimination of methylene units from the
alkyl chains. A similar fragmentation pattern was
obtained for the dimer of BEI, Figure 10. These

imine dimers could have resulted from (1) the

44

snofieoe Hzm mo Edupommm mmozrl.m ouomflm

 

 

 

“<5
0 O N O m OOH
PI_+I_+I_ _I__.efl rt .tI_s _
mo. 0:
mm. fix: me.
mm. ¢N_
om
nxocxuonoz
813.2 m __
:oIzuzoIoI

 

mm.

mm

om

mm mm

mlmxovmzo

N¢

 

.P

 

 

 

0N

ION

-oc

row

IOmw

 

OOH

 

BONVONHGV HAHN—lat!

seesaw: Hmm mo abuuoomm mmoer.OH mesmem

 

 

45

 

 

 

 

 

BONVONOBV 3A I .LV'IBH

 

 

 

$5
on. 00. on _
.p. _._.._.¢_.._: .
a _ _._ —__ E1 _
mm. o: oo ION
mm
3
on -O¢
no.3: -oo
n N N I
:o :o :o m: - om
mxomonzuzoIoImzomxo

46

condensation of an aldehyde "dimer" with an amine group
or (2) an aldol-type condensation of two imine monomers.

The presence of aldol-imine dimers and/or poly-
mers in the imine to be used in subsequent nitrosamine
syntheses was deemed undesirable, as this would compli-
cate identification of the end-products. Fortunately,
the presence of aldol-imines may be easily detected by
a shift in the C=N absorption. The peak at 1675 cm-1
in the IR spectrum of the HMI monomer (M.W. 113),

l in the HMI dimer

Figure 4, is shifted to 1650 cm-
(M.W. 195), Figure 11. Conjugation with an ethylenic
double bond has been reported to cause such a shift in

absorbance (8).

 

 

 

ryyw—rr v—1—v—erww— ~

       

3

CH3(CH2)3-fi-CH-N-CN3

5

runs» Iruucc .1:
3

3

 

 

o ,_____A L A A 1 A A A J A A A L L A
£00 2830 2400 2000 lm IGOO Tm 12” I000 000
WAVENuuata. cu"

 

Figure ll.--Infrared spectrum of HMI "dimer"

47

Attempted Syntheses of Nitrosamines

 

HMI and Sodium Nitrite

 

The end-products of the reaction between HMI
and sodium nitrite were collected over four temperature
ranges during vacuum distillation at 17 mm: less than
30 C, 30-80 C, 80-115 C, and 115-150 C. Each of these
four fractions and the distillation residue were tested
with Griess reagent for nitrosamine content. Negative
results were obtained in every instance.

‘IR spectra for each distillation fraction above
30 C and the spectrum of HMI prior to addition of sodium
nitrite are presented in Figure 12. Recalling that IR
spectra of aliphatic N-nitrosamines are distinguished
by absorption bands at 1425-1460 cm"1 for N=O and at

1030-1150 cm’1

for N-N (83), it is interesting to note
that all of the spectra in Figure 12 show absorption
between 1450-1470 cm-l, but none at 1075 cm-1. Since

the HMI absorbs at 1460 cm-1

, it is difficult to place
much significance on a similar absorption by these
"nitrosated" imines. The IR spectra of the standard
dialkyl-N-nitrosamines in the Appendix all Show a strong
absorbance at or near 1075 cm-1. Since none of the
nitrosated HMI distillation fractions demonstrated this

absorbance, it would appear that nitrosamines were not

formed by reacting HMI with sodium nitrite. Of further

48

ouwnuwc Ebaoom can
Hzm mo ousuxwfi m Scum ooHHAUch ~>H0>wuooemon .Q can .0 .m .4V
ncossocsm as eaxo mma cos .mHH .om one can Ham mo snoooon cosssmcHII.~H osooem

.ISD .mwo::zw><3 .ISD .cuo:=zw><3
DOD ODD. DON. DOW. DOD 08. . DD.D~ DOWN . DOWN . DD.“ 0 08 ODD. DON. 000. 95.

41]!

DD.DN DOWN.

4

 
  

8
8

é
é

 

L

t
I r
I

 

x 30MV11InsuveIi
z' annulment”.

 

 

 

y-
p.
. A
r
In I
P .
I
~ I . I

P»—»L>»L_._ ..» .p. I s s ,_.IK>»LILIP>>

 

 

OD.

 

7.20 . ¢w0332w><’ .130 . ¢u0332w><3
000 000. CON. 00V _ 00m. 08. OOON OO¢N OOON 8mm 08 COO. CON. 00¢ _ 8m. 9!. OOON §N OOON 8g
4 4 4 4 4 4 4 4 J14 4 4 4 4 I4 4 4 4 I] o 4 4 4 4 4 4 4 4 4 I 4 4 4 4 4 1|!!- 0

in

a M A r a U

s
é a

a
'
:I. ' aonuu usual

%' BONVLLINSNVUL

   

s
s

 

n:oIzl:oc.~:o.nxo

s
s

 

 

 

 

 

 

LI. . , . r . . ril 00. .F 0°—

 

49

interest concerning the IR spectra in Figure 12 are the
changes which occur between 1600 and 1700 cm-1. The HMI
has a strong peak at 1675 cm.-1 which is replaced by peaks

at 1645 and 1695 cm‘1

in the nitrosated fractions. The
peak at 1695 cm"1 predominates in the lower boiling
fractions, but decreases and almost disappears in the
higher boiling fractions.

Mass spectra of nitrosated HMI fractions which
boiled at 113 C and 125 C/20 mm are presented in Figures
13A and 13B. The spectra are similar, but the 125 C-
fraction apparently has a higher molecular weight. The

compounds responsible for these spectra were not

identified.

BEI and Sodium Nitrite

 

A second imine, N-n-butylidene-ethylamine (BEI),
was allowed to react with aqueous solutions of sodium
nitrite, fractionally distilled, and the distillate
tested for nitrosamine content. Two fractions were
collected, the first boiling at 62-72 C/12 mm, and the
second at 108-112 C/12 mm. Both fractions gave negative
Griess tests.

The IR spectrum of BEI prior to the addition of
sodium nitrite and the spectra of both distillation
fractions of the "nitrosated" BEI are shown in Figure 14.

As with HMI, the large C=N peak at 1675 cm".1 is replaced

50

a sons ooHHsonso

OON

Ire ._ 1_-r_.sul
d — 1—.

 

mm.

 

th

 

ND.

 

openneo ESHcOm com Hzm mo ousuxfla
coeuomnm BE om\u maa on» no Eduuoomm mmszI.4ma ousoem

omfi

 

 

3:.

mm.

 

OOH

mm

_N

 

mm

om

 

 

mm

_¢

 

 

 

 

mm

ION

Ios

Iow

Iom

 

 

OOH

SONVONHBV BAIIV'IBH

51

ouHHpHG ESHUOm one Hzm mo whopxwfi

m Scum coaawumflc cowuomum BE o~\o mNH on» mo Eduuoomm mmszI.mma ouomflm

 

 

 

 

 

 

 

 

 

«\E
oom omm com 09 02 om!
own .— I .1 I
mom 3. mm we
no -om
Cm o»
Dm— ¢N_
.94
no. .8
.8
SN com mm. 02

 

 

 

BONVCINHBV BAIIV'IHH

52

 

set

TRANSUITTANCE.%

3

 

 

7' *7t 7. Y ._

 

CH3CI-I2cu.‘,cu-I«I-crI.¢.cII3

 

 

 

 

2200

2800 I 2400 #2000 I I600 ‘ I600 ‘ I400 ‘ I200
wAVENuuaen. cu"

 

TRANSMITTANCE.‘

 

A 4

 

 

 

 

 

Ie‘oo‘ Ie‘oo‘Iioo‘Iz‘ooL

2000 ‘
wevsnuusen. cu"

2500 A 2400 ‘

 

3

9.5

Q
0

N
O

 

O-——L*e‘

TRANSMITTANCE.%

3RD

 

 

A

2000 ‘ 2400 72000 I000” A I600 ‘ I400 I200 I000 900
WAVENUMBER. cm"

Figure 14.9eInfrared spectra of BEI (Al and

the 62—72 (B) and 108—112 (C)
C/12 mm fractions distilled from
a mixture of BEI and sodium
nitrite

53

l in the "nitrosated"

with peaks at 1695 and 1645 cm“
samples; and, again, the peak at 1695 cm.1 is very

large in the lower boiling fraction, but small in the
high boiling fraction. All three spectra in Figure 14

Show absorbance in the 1450-1470 cm"l

N=O stretching
region, but only the low-boiling ”nitrosated" BEI
fraction has appreciable absorbance between 1030 and

1150 cm‘1

, the region of N-N stretching. The latter
Spectrum contains the peaks expected for a nitrosamine;
however, their presence does not dominate the spectrum,
as in the case of the standard nitrosamines.

The mass spectrum of the low-boiling fraction is

shown in Figure 15. This compound was not identified.

Imine and Nitric Oxide

 

An attempt was made to synthesize nitrosamines
by bubbling nitric oxide gas through the imine. This
system was simple, and since no water came in contact
with the reactants, the possibility of hydrolytic decom-
position of the intermediates and/or end-products was
eliminated (72).

DeSpite these advantages, this approach was
unsuccessful with both imines used, BEI and N-benzyli-
denemethylamine. The products always gave a false-
positive Griess test, i.e., Spontaneous color develop-
ment on a TLC plate without UV irradiation. Spontaneous

color development also occurred in samples which had

54

 

oneness season can Hem mo
chooses a some coeeeonsc cosuossm as ~H\o NAING no» mo sosoooon nnczII.mH choose

o\E
com and ooh om
om w.
1
V
I.—
oe m
3
V
cm W
N
0
V
om w
3

02

55

been extensively purged with nitrogen prior to analysis.
These false-positive results were probably due to
residual oxides of nitrogen. A comparison was made

of the IR spectra of BEI before and after exposure to
nitric oxide, and the spectrum of the nitric oxide;
treated imine following a ten-minute irradiation with
UV light. No Significant differences were noted in

the three spectra.

Aldehyde, Primary Amine,
and Sodium Nitrite

 

 

An aldehyde, n-butyraldehyde, was added to a
mixture of ethylamine and sodium nitrite. The purpose
of this procedure was to determine if the presence of
sodium nitrite during the synthesis of imine would
result in nitrosamine formation, where addition of
nitrite to the previously synthesized imine had failed.
In other words, would an intermediate in the formation
of imines be easier to nitrosate than the imine itself?

Concentrated ether extracts consistently gave
false-positive Griess tests. IR spectra from these
products were almost identical to the spectra pre-
viously shown in Figure 14 for BEI following the
addition of sodium nitrite. Apparently, the same
product is obtained regardless of whether nitrite is
added before or after the imine is formed. The shift

of the major absorption peak from 1675 cm.1 to

56

1695 cm"1 could be due to a nitrosated intermediate
/NO
\R I
has been reported to absorb at 1690 cm-

+
such as [R-CH=N J. A positively charged imine group

1 (72).

Ground Ham Model System

 

Although aldehydes and sodium nitrite were not
shown to form nitrosamines in the preceding experimental
syntheses, an attempt was made to form nitrosamines by
combining these reactants in a model dry food system.
Rather than use a relatively inert material such as
carboxy-methyl-cellulose (CMC), fresh (uncured) ground
ham muscle was chosen as the matrix upon which the
aldehyde and sodium nitrite were to be distributed.

Ham was chosen because it is a common food normally
consumed after being cured with sodium nitrite; and
because ham muscle, unlike CMC, contains proteins which
would supply free amino groups for reaction with aldehyde
added to the model system. Aldehydes, from the oxidative
decomposition of fatty acids, have been Shown to react
with the amino group of phospholipids to form imines (14).

Various combinations of sodium nitrite, sodium
chloride, and hexanal were included in both the "wet"
(non-freeze-dried) and "dry" (freeze-dried) systems.
Freeze-drying the meat emulsions prior to heating was
intended to reduce the possibility of decompositiOn
of nitrosamines or nitrosamine—intermediates as a

result of hydrolysis (Dr. W. H. Reusch, Michigan State

57

University, personal communication). The amounts of
aldehyde and nitrite added to the model system were
based on an estimate of the millimoles of free amino
groups in 100 g ham muscle. Lean trimmed ham.muscle
contains 20% protein (75); and twenty grams of ham
protein contain approximately 3.2 g of the amino acids
lysine, arginine, asparagine, and glutamine (62).
Based on these values, the free amino groups in 100 g
ham muscle were estimated as 20 mM. Enough hexanal for
a 1:1 stoichiometric reaction with the amino groups,
a two-fold excess of sodium nitrite, and 2% sodium
chloride were added to the appropriate model system
variables. To insure that nitrosamines were not form-
ing during the extraction procedure, sodium nitrite was
incorporated into the baked emulsion of some of the
samples at the beginning of the extraction procedure.
The negative result obtained for this variable (see
Table 4) indicates that nitrosamine formation did not
occur during the extraction procedure. Recovery con-
trol samples were spiked with dimethyl- and dibutylni-
trosamine standards at a concentration of 10 ppm (1 mg
nitrosamine/100 9 meat).

The minimum level of nitrosamines detectable
with the Griess and ninhydrin reagents was 1 ug;

therefore, a negative result in Table 4 means that

58

TABLE 4.--Presence of nitrosamine-positive compounds
in the extract from "dry” and "wet“ model
food systems as indicated by Griess and
ninhydrin reactions on thin-layer plates

 

 

 

Treatments
Additives
. Non-freeze
Freeze-dried Dried
Control - -
Nitrite - +
Nitrite + NaCl - +
Nitrite + aldehyde - +
Aldehyde - -
Aldehyde +
nitrite added just
before extraction - -
Nitrosamine + +

 

59

less than 1 ug of nitrosamine was present in the 40 ul
concentrated extract used for each determination.

All of the "wet" treatment samples which con-
tained nitrite, alone or in combination with other
variables, gave positive Griess and ninhydrin tests.
Some of the nitrite-containing "dry"-system samples
also gave positive Griess 25 ninhydrin tests, but
negative results were recorded in Table 4 unless both
tests were positive. In both wet and dry systems, the
Rf values of the positive spots were the same, 0.30
and 0.65. Samples which gave a positive Griess test
were analyzed by GC-MS. The spectra obtained were
compared with the Spectra of 25 standard nitrosamines
(53) and 101 pyrazines (44), but could not be identified.

It seems probable that the same nitroso-positive
compounds were formed in each system, but were present
in larger quantities in or were more easily extracted
from the "wet” (non-freeze-dried) system. Also, the
surface of the "wet" samples developed extreme case-
hardening during the lS-hour heating in circulating
dry air, and this may have helped retain the nitroso-
positive compounds during the heat treatment. Formation
of nitrosamines in the non-freeze-dried ham muscle would
seem reasonable in view of the excessive nitrite level

used, l38-times the legal maximum. Fiddler et a1. (26)

60

have shown nitrosamine formation will occur in meat
emulsions, provided an excessive nitrite concentration
is present.

The low levels of nitrosamine-positive compounds
found in the freeze-dried ham emulsion could represent
compounds which formed prior to freeze-drying, or low
residual levels resulting from losses of volatile
nitrosamines during the freeze-drying and/or heating

steps.

SUMMARY AND CONCLUSIONS

Attempts were made to synthesize N-nitrosamines
by the reaction of imines with sodium nitrite or with
nitric oxide gas, and by the reaction of nitrite with
aldehyde and primary amines. Under the reaction con—
ditions imposed, nitrosamines either were not formed,
were non-volatile, or were not identified by the
detection methods used. Acid hydrolysis of imines
and/or "nitrosated" imines probably accounts for the
failure to form nitrosamines from carbonyl-amine pre-
cursors in aqueous systems. Any nitrosamines which
may have been formed in the "dry" imine-nitric oxide
system were masked by the spontaneous red color which
formed when these samples were tested with Griess
reagent.

In an attempt to react sodium nitrite with
carbonyl-amine compounds in a dry food system, various
combinations of sodium nitrite, sodium chloride, and
hexanal were distributed on a ground ham muscle matrix
prior to freeze-drying. The freeze-dried samples were

exposed to heat over a prolonged time in an attempt

61

62

to enhance formation of carbonyl—amine condensation
products. Freeze-dried meat, to which 138-times the
maximum legal concentration of sodium nitrite had been
added, contained traces of Griess-positive compounds.
However, non-freeze-dried controls contained considerably
more of these apparent nitrosamines. There was no evi-
dence that the addition to the meat of components which
would enhance the probability of non-enzymatic browning
in the system contributed to the presence of nitroso-
positive reactants in the extracts. Samples containing
the apparent nitrosamines were analyzed by gas chroma-
tography-mass spectrometry, but were not identified
when compared to mass spectra of known nitrosamines and

pyrazines.

APP END IX

MHOHMHOQMH mHSu SH coNHmoSuSMm “mo oSm mSomEou xmcoM
SmEumom Eoum commoousm ASL mSHEmHSSHoEHoomOHuHSIz mo mnuomom commumSHIl.mH ouomHm

.IID .mwmzazw><3 .IZD .mwmi:zw>(3

 

 

 

 

 

000 000. OON. 00v. 00w. 009 . 00.0N 4 DOWN . DOWN . OOjNfl 0 ODD 000. CON. 00¢. 00w— OOQ. . 00.0N . DOWN J 00.0qu Gown 0
m <

i ON .. ON
1 1
A U i U
v V
. co m . 04 w.
_ 2 W W
A 1. k r A II.
1 1
v v
/ i 8 N l i 8 N
, m 3 n 3
\ ID 3 \ ID 3
, ozIz/n . .x 02.2; .x

/ xu nxu
. {A 00 A 8
DO. I 00.

 

 

 

63

64

hSomEou Mmoom Sofiummm

Eoum commoondm mSHEmHmSuoEHUOmOHHHSIz mo Euuuoomm mmozII.hH ousmHm

 

 

 

 

 

 

$8
0 ma 0 OH o m
. _ ___. . __. A.
A I O N
on
I E as I 014
I n. I
mm o m
S. .222 mzo
I ozIzn I om
mzo
, a. 02

 

 

 

 

BONVGNOBV 3A I 111138

65

huoumuoaoH
mHSu SH cmnHmoSuSMw oSHEdHMSHoEHUOmouuHSIz mo ESHpoomm mmszI.mH musmHm

 

 

 

 

 

BONVONDGV BALM/138

 

 

 

 

3E
Om; .oofi om.
. I . _ I I I _ . . =1 _
m m. fiON
I we Io¢
cm
I Iow
m E
I $63.2 \ ID 1
ozsz om
mzo
a. 02

AHV oSHoHHonnwmI oSm

.Amv oSHcHHomHmI .flov oSHEmawuooHMSuoI .Amv oSHEMkuSoHNSuoEI

.nmu USHEMHSQOHQH>SuoEI .AQV wSHEdHhuSnHoI .AUV wSHEMHmmouchI
.Amg mSHESHSSuoHoI .A4V oSHEMHNSuoEHcOmouuHSIz mo muuooom commumSHII.mH oHSSHm

. cl
.Io lggil .IDU III->3 7.0 I!!!
. , .

. o HIFIJE _ EIEIIBIIJe . _
u I A e

 

 

 

I 8

    

:;:E3
rfi::s:.....§::.s
‘ ‘ ' )3.‘ A: K

,1
5 i

 

 

- 1—A—_§.A. EA.
I'nuuuuuvn
I‘ ”Inn-lulu“

 

66
I5; I

 

l'lleuIIlleA
I' Ia'l AIIII'IA

‘5
‘ 4' If“ T‘T

Y
(1
C
C
II‘
[’71—‘1

 

 

SSMQEOU 0300M Smfimom
Scum .commoonnm oSHfimHMSuoHcomonuHSIz mo Samuelson mmozII.o~ ousmHm

 

67

 

 

 

 

 

 

 

 

 

o):
OWL OOH om
»_L_L_p_.—L—._.I_._._r.tp._.._1._t
_ a . :1 ___. ._
N9
-om w.
_I
w
on .014 m
3
V
.8 m
No. .225. \Nzomxo m
oznz
firomzo cc Iom .No
R 3
NV mm

 

 

OOH

68

MSMSEOU xmoom Smfimmm
Scum comgousm oSHfimammouoHGOmonuHSIz mo Suuommm mmmZII.HN ousmHm

 

 

 

 

 

 

 

 

 

 

 

 

 

35
on; OOH Om
._._._F_t .H.-_t_._L_. L..._. I_l.
I .I A _____ ._. 2 H
m:
.0. 70W
mm
Om.
Ioe
On
ea Iom
on. .222 \Nzomzomzo
oZIzz Iom
«zomxomzo
Oh me.

 

00L

HONVONDBV BAIIV'IBH

69

. hSomEOD MmooM Shannon
Scum commnonsm mSHEonuSQHoOmOHuHSIz Ho Efiuuoomm mmozII.N~ whomHm

 

 

 

 

 

 

 

 

 

 

 

 

o\E
om OOH om
..pb.—._p—._a._..L4I.~..._t_._._—“"u_._
_ _ _ _ ___ ___ _
mm. .2 2.
I o: mm ION M
_I
V
IL
I Iolw A
:2
w
1 low A.”
N
@9222 0
emxuzornzo W
I 0.2-2 mm Iomw O
fixmonmxo so _ c 3
so
OOH

 

 

7O

EfimoSommm .mH musnfiom m .oHEmSOOHm
USS oHEoSU oSomHSmmuo Ham usqumSH .Hmmom .m .Ho

Scum omSHmuao mSHaoahmoumHmSumEOmOHHHSIZ mo ESHuommm mmMZII.mN ouomHm

 

 

 

 

 

m\c:
O m 0 O H . o m
. c . _ . FI. _ . _ . _II _ . HI. _ . _ . g.. c .._ i _ _ .
_ _ _ .:: _. :__
No. no en
T.
E
No. .222 «zomzonzo a.
\
oZIz/
mzo
we

 

 

ION

Ioc

Iom

womw

 

 

OOH

30Nv0anv aAlivjaa

71

MD .oHHSmcnomcom .xoounSHoSm .omoom SHHOBHOU
.moHHoumnoomq Sonommom H0>0HHSD .oSHonmo .Q .Ho
Scum ooSHmuno oSHEHHMHSQHmSumEOmOHDHSIZ mo Efluuommm mmmZII.vN oHSmHm

 

 

 

 

 

 

 

 

 

 

o\E
om; 00L om
I. _ . _ . _ . _ . _ .IL fl 1 Le_. 1.... . “.I uL=E+I “1. cm __I.
.. _. .
I -om u
1
w.
2. SN I
A I014 A
2a
Hv
- Iom ann.
mm
9.2;. masonic m
\ N
I ozlz Iom 0
/m:o 3
a. 02

 

 

72

MSSQEOU Mooom Sofiummm
Scum ommmSoHSm mSHoHHOHHNmomOHuHSIz mo Ebuuommm mmmer.mN mndem

 

 

 

 

 

 

 

 

 

. oxE
0 ma 0 OH O m
I _ . _ r _ r e . _ I _ _I__. _ _ . +7. . A.+__:.. _fl 2;.
.. .om
- 0.4
I O Q
Do. .>>.2 NIDIINID
0212A .Iom
NIOINIO :4 mN
CO;

 

 

 

OOH

BONVONHBV 3AllV'138

73

MSomfioo_Mmcom Smfiummm
Scum commoousm MSHoHHomHSOmOHuHSIz mo Efluuoomm mmoer.mN mesmHm

 

 

 

 

 

 

 

 

 

 

 

oxE
Om; OOL om. ,
. _ . rI. _ . _ . r . _ l _ . _ I _ 4:. . r:. u ?L Iqw1.. fl _ A
¢m
v:

I . .om m...
1
av
I;

I no Iota M

mm 3
w
I Iow n
m
3.222 \Nonmzo w
I ozlz/ vow: Iom .20.
Nzouuzo
a. 02

 

REFERENCES CITED

REFERENCES CITED

Archer, M. C., Clark, S. D., Thilly, J. E., and
Tannenbaum, S. R. 1971. Environmental nitroso
compounds: reaction of nitrite with creatine
and creatinine. Science 114:1341-1343.

Ashton, M. R. 1970. The occurrence of nitrates and
nitrites in foods. British Food Manufacturing
Industries Research Association, Literature
Survey No. 7.

Barnes, J. M. and Magee, P. N. 1954. Some toxic
properties of dimethylnitrosamine. Brit. J.
Industrial Med. 11:167-174.

Budzikiewicz, H., Djerassi, C., and Williams, D. H.
1967. Mass Spectrometry of Organic Compounds.
Holden-Day, Inc., San rancisco. pp. 329-330,
388-390.

 

 

Campbell, K. N., Sommers, A. H., and Campbell,
B. K. 1944. The preparation of unsymmetrical
secondary amines. J. Amer. Chem. Soc. 66:82-84.

Christiansen, L. N., Johnson, R. W., Kautter, D. A.,
Howard, J. W., and Aunan, W. J. 1973. Effect
of nitrite and nitrate on toxin production by
Clostridium botulinum and on nitrosamine for-
mation in periShable canned comminuted cured
meat. Appl. Microbiol. 25:357-362.

 

Crosby, N. T., Foreman, J. K., Palframan, J. F., and
Sawyer, R. 1972. Estimation of steam-volatile
N-nitrosamines in foods at the 1 ug/kg level.
Nature 238:342-343.

Curran, D. J. and Siggia, S. 1970. Analysis of
azomethenes. In. S. Patai (ed.), The Chemistry
of the Carbon-NItrogen Double Bond. Interscience
Publishers, Ihb., New York. pp. 149-180.

74

10.

11.

12.

13.

14.

15.

16.

17.

Daiber, D. and Preussmann, R.

75

1964. Quantitative

determination of organic N-nitroso compounds
with the photochemical degradation of the nitro-

samine linkage (in German).

344-352.

Devik, O. G.

the Maillard reaction.

2302-2303.

Druckrey, H., Schildbach, A., Schmahl, D., Preussmann,
R., and Ivankovic, S. 1963. Quantitative analyse

Z.

Anal. Chem. 206:

1967. Formation of N-nitrosamines by

Acta Chem. Scand. 21:

der carcinogenen wirkung von diathylnitrosamin

(in German).

Druckrey, H., Preussmann, R.,

Schmahl, D. 1967.

Z. Krebsforsch.

£2:

Arzneimittel-Forsch. 13:841-851.

Ivankovic, S., and

Organotrope carcinogene
wirkungen bei 65 verschiedenen N-nitroso-
verbindungen an BD-Ratten (in German).

103-201.

Druckrey, H., Landschutz, C., and Ivankovic, S.
1970. Transplacentare erzeugung maligner
tumoren des nervensystems. II:
soharnstoff an 10 genetisch definierten ratten-
stammen (in German, English summary). Z.

Krebsforsch.

Dugan, L. R.,

13:371-386.

Jr. and Rao, G. V.
of the flavor contribution of products of the
Maillard reaction.
Army Natick Laboratories, Natick, Massachusetts.

Tech. Rep.

Athyl-nitro-

1971. Evaluation

72-27-FL’ UoSo

Du Plessis, L. S. Nunn, J. R., and Roach, W. A.
1969. Carcinogen in a Transkeian Bantu food
Nature 222:1198-1199.

additive.

Ender, F. 1966.

N-nitrosodimethylamine, the active

principle in cases of herring meal poisoning.

Publication of the 4th.

Congress,

Zurich.

International Buiatric

Ender, F., Havre, G;, Helgebostad, A., Koppang, N.,
Madsen, R., and Ceh, L. 1964.
identification of a hepatoxic factor in herring
meal produced from sodium nitrite preserved

herring.

Naturwiss.

51:637.

Isolation and

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

76

Ender, F., Havre, G., Madsen, R., Ceh, L., and
Helgebostad, A. 1967. Studies on conditions
under which N-nitroso-dimethylamine is formed in
herring meal produced from nitrite-preserved
herring. The risk of using nitrite uncritically
as a preservative agent (in English). Z.
Tierphysiol. Tierernahr. Futtermittelk 22:181-189.

Ender, F. and Ceh, L. 1968. Occurrence of nitro-
samines in foodstuffs for human and animal con-
sumption. Food Cosmet. Toxicol. 6:569-571.

Fan, T. Y. and Tannenbaum, S. R. 1972. Stability
of N-nitroso compounds. J. Food Sci. 31:274-276.

Fazio, T., Damico, J. N., Howard, J. W., White, R. H.,
and Watts, J. O. 1971. Gas chromatographic
determination and mass spectrometric confirmation
of N-nitrosodimethylamine in smoke-processed
marine fish. J. Agr. Food Chem. 12:250-253.

Fazio, T., White, R. H., and Howard, J. W. 1971.
Analysis of nitrite- and/or nitrate-processed
meats for N-nitrosodimethylamine. J. Assoc.
Offic. Anal. Chem. 54:1157-1159. -

Federal Register. 1971. Reg. 121.1230, Sodium
nitrite used in processing smoked chub. Title
21, Food and Drugs, p. 272.

Feigl, F., Anger, V., and Oesper, R. E. 1966.
Spot Tests in Organic Analysis. Elsevier Pub-
liShihg Co., New York. pp. 290-294.

 

Fiddler, W., Doerr, R. C., Ertel, J. R., and
Wasserman, A. E. 1971. Gas-liquid chromato-
graphic determination of N-nitrosodimethylamine
in ham. J. Assoc. Offic. Anal. Chem. 54:1160-
1163.

Fiddler, W., Piotrowski, E. G., Pensabene, J. W.,
and wasserman, A. E. 1972. Some current
observations on the occurrence and formation of
N-nitrosamines. Proceedings of the 18th.
Meeting, Meat Research WOrkers, Guelph, Ont.

Fiddler, W., Pensabene, J. W., Doerr, R. C., and
Wasserman, A. E. 1972. Formation of N-nitro-
sodimethylamine from naturally occurring quater-
nary ammonium compounds and tertiary amines.
Nature 236:307.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

77

Fishbein, L. and Falk, H. L. 1969. Chromatography
of alkylating agents. II. Nitrosamines,
epoxides, lactones, methanesulfonates and mis-
cellaneous derivatives. Chromatogr. Rev.
11:365-455.

Foreman, J. K. 1972. N-nitrosamines as possible
food carcinogens. Proceedings of the 18th Meet-
ing, Meat Research WOrkers, Guelph, Ont.

Freimuth, U. and Glaser, E. 1970. Zum auftreten
von nitrosaminen in lebensmitteln (in German,
English summary). Die Nahrung 14:357-361.

Hatt, H. H. 1943. Pages 211-213. In. A. H.
Blatt (ed.), Organic Syntheses, C5Ilective
Vol. 2. John Wiley and Sons, Inc., New York.

 

Havre, G. N. and Ender, F. 1971. The formation
of methyl substituted pyrazines in the Maillard
reaction, and the interference by such compounds
in the determination of N-nitrosamines in food-
stuffs (in English). Z. Lebens.-Unters. und-
Forsch. 116:74-79.

Hedler, L. and Marquardt, P. 1968. Occurrence of
diethylnitrosamine in some samples of food.
Food Cosmet. Toxicol. 6:341-348.

Howard, J. W., Fazio, T., and Watts, J. O. 1970.
Extraction and gas chromatographic determination
of N-nitrosodimethylamine in smoked fish:
application to smoked nitrite-treated chub.

J. Assoc. Offic. Anal. Chem. 53:269-274.

JOhnSOn, D. E. and RIIAOdeS' Jo W. 1972. N-
nitrosamines in smoke condensate from several

varieties of tobacco. J. National Cancer Inst.
48:1845-1847.

Kadar, R. and Devik, 0. G. 1970. Pyrazines as
interfering substances in the determination of
nitrosamines in roasted foods. Acta Chem.
Scand. 24:2943-2948.

Koppang, N. 1964. An outbreak of toxic liver
injury in ruminants. Nord. Vet.-Med. 16:305-
322.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

78

Koppang, N., Slagsvold, P.,.Hansen, M. Aas, Sfignen,
E., and Svenkerud, R. 19642 Feeding experiments
with meal produced from herring preserved with
sodium nitrite and formalin. Nord. Vet.-Med.
16:343-362.

Koprowski, J. A.. 1968. Studies of the detection
of dimethylnitrosamine in Lake Michigan chub
(Leucichth 3 ho i). M.S. Thesis, Michigan
State Un1versIty, East Lansing, Michigan.

Layer, R. W. 1963. The chemistry of imines.
Chem. Rev. 63:489-510.

Lijinsky, W., Keefer, L., and Loo, J. 1970. The
preparation and properties of some nitrosamino
acids. Tetrahedron 26:5137-5153.

Lijinski, W., Conrad, E., and Bogart, R. van de.
1972. Carcinogenic nitrosamines formed by
drug/nitrite interactions. .Nature 239:165-167.

Logten, M. J. van, Tonkelaar, E. M. den, Kroes, R.,
Berkvens, J. M., and Esch, G. J. van. 1972.
Long-term experiment with canned meat treated
with sodium nitrite and glucono-d-lactone in
rats. Food Cosmet. Toxicol. 12:475-488.

Maga, J. A. and Sizer, C. E. 1973. Pyrazines in
foods. In. J. E. Furia (ed.), Critical
Reviews IE Food TechnolOQy. CRC Press, Inc.,
Cleveland. pp. 39-115.

Magee, P. N. 1968. Possible mechanisms of car-
cinogenesis by N-nitroso compounds and alkylat-
ing agents. Food Cosmet. Toxicol. 6:572-573.

Magee, P. N. 1971. Toxicity of nitrosamines:
their possible human health hazards. Food
Cosmet. Toxicol. 2:207-218. .

Magee, P. N. 1971. Liver carcinogens in the human
environment. Liver Cancer 1971:110-120.

Magee, P. N. and Barnes, J. M. 1956. The pro-
duction of malignant primary hepatic tumours in
the rat by feeding dimethylnitrosamine. Brit.
J. Cancer 12:114-122.

Magee, P. N. and Barnes, J. M. 1967. Carcinogenic
nitroso compounds. Adv. Cancer Res. 12:163-247.

50.

51.

52.

53.

54.

55.

56.

57.

58.

79

Mirvish, S. S., wallcave, L., Eagen, M., and Shubik,
P. 1972. Ascorbate-nitrite reaction: possible
means of blocking the formation of carcinogenic
N-nitroso compounds. Science lZZ:65-68.

Morrison, R. T. and Boyd, R. N. 1966. Organic
Chemistry, 2nd. Ed. Allyn and Bacon, Inc.,
Boston. pp. 743 and 758.

 

Panalaks, T., Iyengar, J. R., and Sen, N. P. 1973.
Nitrate, nitrite, and dimethylnitrosamine in
cured meat products. J. Assoc. Offic. Anal.
Chem. 66:621-625.

Pensabene, J. W., Fiddler, W., Dooley, C. J.,
Doerr, R. C., and wassermann, A. E. 1972.
Spectral and gas chromatographic characteristics
of some N-nitrosamines. J. Agr. Food Chem.
36:274-275.

Preussmann, R., Neurath, G., Wfilf-Lorentzen, G.,
Daiber, D., and Hengy, H. 1964. Anffirbemethoden
und dfinnschichtchromatographie von organischen
N-nitrosoverbindungen (in German). Z. Anal.
Chem. 393:187-192.

Rao, C. N. R. and Bhasker, K. R. 1969. Spectro-
scopy of the nitroso group. In. H. Feuer (ed. ),
The Chemistry of the Nitrot and—Nitroso Groups.
Inter-science PubliShers, Inc., New York.
pp. 137- 163.

Ray, J. D. and Ogg, R. A., Jr. 1956. Preparation
of nitric oxide. J. Amer. Chem. Soc. 16:5993.

Roach, W. A. 1972. The possible presence of
nitrosamines in Transkeian Bantu foodstuffs.
In. P. Bogovski, R. Preussmann, E. A. Walker,
and W. Davis (eds. ), N-nitroso Compounds
Analysis and Formation. International Agency
for ResearCh on Cancer, world Health Organi-
zation, Lyon, France. IARC Scientific Publ.
No. 3. pp. 74-78.

 

Sakshaug, J., Sfignen, E., Hansen, M. Aas, and
Koppang, N. 1965. Dimethylnitrosamine; its
hepatoxic effect in sheep and its occurrence in
toxic batches of herring meal. Nature 366:
1261-1262.

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

80

Saxby, M. J. 1970. The separation of volatile
N-nitrosamines and pyrazines. Analytical
Letters 6:397-399.

Saxby, M. J. 1972. Mass spectrometry of volatile
derivatives. V. N-dialkylnitrosamines.
J. Assoc. Offic. Anal. Chem. ‘66:9-12.

Scanlan, R. A. and Libbey, L. M. 1971. N-nitro-
samines not identified from heat induced d-
glucose/L-alanine reactions. J. Agr. Food Chem.
12:570-571.

Schweigert, B. S. and Payne, B. J. 1956. A
summary of the nutrient content of meat. Amer.
Meat Inst. Found., Bull. No. 30.

Sen, N. P. 1970. Gas-liquid chromatographic
determination of dimethylnitrosamines as
dimethylnitramine at picogram levels. J.
Chromatogr. 66:301-304.

Sen, N. P. 1972. The evidence for the presence
of dimethylnitrosamine in meat products. Food
Cosmet. Toxicol. 66:219-223.

Sen, N. P., Smith, D. C., Schwinghamer, L., and
Marleau, J. J. 1969. Diethylnitrosamine and
other N-nitrosamines in foods. J. Assoc.
Offic. Anal. Chem. 63:47-52.

Sen, N. P., Smith, D. C., Schwinghamer, L., and
Howsam, B. 1970. Formation of nitrosamines in
nitrite-treated fish. Can. Inst. Food Technol.
J. 6:66-69.

Sen, N. P. and Dalpe, C. 1972. A simple thin-
layer chromatographic technique for the semi-
quantitative determination of volatile nitro-
samines in alcoholic beverages. Analyst 97:
216-220. ‘ "

Sen, N. P., Donaldson, B., Iyengar, J. R., and
Panalaks, T. 1973. NitrOSOpyrrolidine and
dimethylnitrosamine in bacon. Nature 241:
473-474. "‘ '

Sen, N. P., Miles, W. F., Donaldson, B., Panalaks,
T., and Iyengar, J. R. 1973. Formation of
nitrosamines in a meat curing mixture. Nature
266:104-105.

70.

71.

72.

73.

74.

75.

76.

77.

78.

81

Serfontein, W. J. and Smit, J. H. 1967. Evidence
for the occurrence of N-nitrosamines in tobacco.
Nature 214:169-170.

Sidgwick, N. V. 1966. Nitrosamines, nitramines,
and nitrosohydroxylamines. £23 I. T. Miller
and H. D. Springall, The Organic Chemistry of
Nitrogen, 3rd. Ed. Clarendon Press, Ox or .
pp. -601.

 

Smith, P. A. S. 1965. The Chemistry of Open-Chain
Organic NiErogen Compounds. Volume I. 7W. A.
Benjamin, Inc., New York. pp. 297-300.

Sweeley, C. C., Ray, B. D., WOOd, W. I., Holland,
J. F., and Kritchevksy, M. 1970. On-line
digital computer system for high-speed single
focusing mass spectrometry. Anal. Chem. 63:
1505-1516.

Telling, G. M., Bryce, T. A., and Althorpe, J.
1971. Use of vacuum distillation and gas
chromatography-mass spectrometry for determi-
nation of low levels of volatile nitrosamines
in meat products. J. Agr. Food Chem. 19:
937-940. ‘-

U.S. Department of Agriculture. 1963. Composition
of Foods, by B. K. watt and A. L. MerrilI.
HandBooE No. 8. Government Printing Office,
Washington, D.C. p. 48.

 

Walters, C. L. 1971. The detection and estimation
of trace amounts of N-nitrosamines in a food
matrix. Laboratory Practice 26:574-578.

Walters, C. L., Johnson, E. M., and Ray, N. 1970.
Separation and detection of volatile and non-
volatile N-nitrosamines. Analyst 26:485-489.

Walters, C. L., Johnson, E. M., Ray, N., and
Woolford, G. 1972. Nonvolatile nitrosamines
from biological precursors and their detection.
£6: P. Bogovski, R. Preussmann, E. A. walker,
and W. Davis (eds.), N-nitroso Compounds Analysis
and Formation. InternationaIAgencyforResearch
on Cancer, World Health Organization, Lyon,
France. IARC Scientific Publ. No. 3. pp. 79-80.

 

 

79.

80.

81.

82.

83.

84.

82

Wasserman, A. E. 1972. A survey of analytical
procedures for traces of N-nitrosamines in
foods. Int. Union Pure and Appl. Chem., Inf.
Bull. Tech. Rep. No. 5.

Wechsler, W., Kleihues, P., Matsumoto, S., Zulch,
K. J., Ivankovic, S., Preussman, R., and
Druckrey, H. 1969. Pathology of experimental
neurogenic tumors chemically induced during
prenatal and postnatal life. Ann. N. Y. Acad.
Sci. 2§2:360-408.

Weisburger, J. H. and Weisburger, E. K. 1966.
Chemicals as causes of cancer. Chem. and Eng.
News 1966:124-142.

Williams, A. A., Timberlake, C. F., Tucknott,
O. G., and Patterson, R. L. S. 1971. Obser-
vations on the detection of dimethyl- and
diethyl-nitrosamines in foods with particular
reference to alcoholic beverages. J. Sci. Food
Agr. 22:431-434.

Williams, R. L., Pace, R. J., and Jeacocke, G. J.
1964. Applications of solvent effects. I.
The spectra of secondary nitrosamines. Spec-
trochimica Acta 29:225-236.

WOlff, I. A. and Wasserman, A. E. 1972. Nitrates,
nitrites, and nitrosamines. Science 177:
15-19 0