IIIIIIIIIIIIIIIIIIIIIIIIIIIII

“le ’ lllllllllllllllllllllllllllllllllllllllllllllllllllIlllllllll

3 1293 01568 7621
LIDHAI‘"

Michigan State
University

 

This is to certify that the

thesis entitled

EFFECT OF LIPIDS AND IRON ON HETEROCYCLIC ARONATIC
AHINE FORMATION IN AQUEOUS MODEL SYSTEMS

presented by

LISA SCRANTON

' has been accepted towards fulfillment
of the requirements for

MASTERS . FOOD SCIENCE

degree 1n

 

Date APRIL 17, 1997

 

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

l__'

PLACE IN HEI’URN BOX to romovo this checkout from your record.
TO AVOID FINES return on or before date duo.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU In An Affirmative Action/Ema! Opportunity Institution
W1

EFFECT OF LIPIDS AND IRON ON HETEROCYCLIC AROMATIC AMINE
FORMATION IN AQUEOUS MODEL SYSTEMS

By

Lisa Scranton

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

MASTER OF SCIENCE

Department of Food Science and Human Nutrition

1 997

ABSTRACT

EFFECT OF LIPIDS AND IRON ON HETEROCYCLIC AROMATIC
AMINE FORMATION IN AQUEOUS MODEL SYSTEMS
By

Lisa Scranton

This study was designed to investigate the efi‘ect of lipids of varying
degrees of unsaturation on heterocyclic aromatic amine (HAA) formation in an
aqueous model system. Saturated and unsaturated lipids were added to model
systems utilizing creatinine, glucose, glycine or phenylalanine, and water. The
model systems were heated at 180°C in closed stainless steel test tubes for 30
minutes. HAAS were isolated by solid phase extraction and separated/quantitated
by high performance liquid chromatography. Corn oil, olive oil, and tristearin had
no effect on species or yield of HAAS formed. Trilinolein increased MeIQx (2-
amino-3,8-dimethylimidazo[4,5flquinoxaline) and IQx (2-amino-3-methyl-
imidazo[4,5-f]quinoxaline) formation in the glycine system. Ferrous sulfate and
myoglobin were added to some of the samples. Ferrous sulfate increased MeIQx
formation without lipid. Myoglobin decreased IQx formation with or without lipid
present. These results indicate that the chemical effect of lipids on HAA formation

is probably not significant under household cooking conditions.

To my husband, Alec, and my children Gregg, Audrey, and Elizabeth
for their encouragement, patience, and inspiration

iii

ACKNOWLEDGMENTS

I would like to express gratitude to my major professor, J. I. Gray, for his
encouragement and guidance in my degree program, and his willingness to
accommodate my atypical schedule during these last several years. He truly
provided an educational opportunity I otherwise would not have had.

Appreciation goes to Dr. Gale Strasburg and Dr. Alden Booren for serving
on my academic committee. Thank-you, Dr. Strasbourg, for answering my many
day-to-day questions. Thank-you, Dr. Booren, for your willingness to work
individually with me on HACCP, and for taking me along on all those plant trips!

I would also like to thank the students and staff in Dr. Gray’s and Dr.
Strasburg’s laboratories for help, encouragement, and camaraderie. Special thanks
is due to Dr. Enayat Gomaa, for all her helpful suggestions and all those hours
spent toiling over the malfunctioning HPLC for me!

Dr. Rob Tempelman’s assistance and advice on the statistical analysis of my
data is greatly appreciated, especially as this project hinged on the correct
application of statistical analysis.

I am very grateful to my mother-in-law, Margie Scranton, my mother,
Kathy Cook, and my grandmother, Helen Oltman, for their many hours of

babysitting and for keeping my household running the last several years.

TABLE OF CONTENTS

 

INTRODUCTION

 

LITERATURE REVIEW

 

History of Heterocyclic Aromatic Amines

 

Discovery

 

Categories

 

Mutagenicity/Carcinogenicity

 

Incidence and Consumption of HAAs

 

Mechanism of Formation

 

Chemical Formation

 

Factors Contributing to HAA Formation
Role of Fat in HAA Formation

 

MATERIALS AND METHODS

 

Materials

 

Methods

 

 

Weighing/Mixing of Ingredients

 

Heating
Extraction of HAAs

 

High Performance Liquid Chromatography
Mass Spectrometry

 

 

Quantitation of HAAs

 

Statistical Analysis

 

@MMLII

10
l4
l6
17
19
26
31
31
32
32
33
33
35
36
37
38

 

RESULTS AND DISCUSSION 41
Study #1 :Efi‘ect of Trilinolein on HAA Formation in Aqueous Phenyl-

 

 

 

 

 

 

alanine and Glycine Model Systems 40

Study #2: Effect of Com Oil and Olive Oil on HAA Formation in

an Aqueous Phenylalanine Model System 46

Study #3: Effect of Ferrous Sulfate and Myoglobin on HAA Form-

ation in an Aqueous Glycine Model System 48
SUMMARY AND CONCLUSIONS 52
FUTURE RESEARCH 53
APPENDIX 56

 

REFERENCES 58

LIST OF TABLES

LITERATURE REVIEW
Table

1. Heterocyclic aromatic amines produced in model systems
(Skog,l993)

 

2. Mutagenicity of heterocyclic aromatic amines and typical
carcinogens in Salmonella typhimurium (Sugimura and Sato,
1982; Sugimura et al., 1988)

3. Average values of HAAS formed in selected food products
(Skog,l993)

 

 

RESULTS AND DISCUSSION
Table

4. Effect of trilinolein on HAA formation in aqueous phenyl-

 

alanine and glycine model systems

5. Effect of corn oil and olive oil on HAA formation in a
phenylalanine model system

 

6. Effect of ferrous sulfate and myoglobin on HAA formation

 

in a glycine model system with and without lipid

vii

11

14

16

42

47

49

LIST OF FIGURES

LITERATURE REVIEW

Figge
1. Chemical structures of some HAAs in cooked foods
(Skog,l993)

 

2. Theoretical reaction pathway for formation of IQ and IQx
compounds. R, X, and Y may be H or CH3: Z may be CH

 

or N (Jagerstad et al., 1983a)

RESULTS AND DISCUSSION

Figge
3. Effect of tristearin versus trilinolein on HAA formation in

 

glycine and phenylalanine model systems

4. Effect of tristearin versus trilinolein on HAA formation in

 

a phenylalanine model system

5. Effect of ferrous sulfate (Fe) or myoglobin (Mb) on the
formation of HAAS in a glycine model system with and
without trilinolein (TL)

 

viii

20

44

45

50

INTRODUCTION

Heterocyclic aromatic amines (HAAS) are a class of compounds containing
multiple aromatic rings with at least one exocyclic amino group and often an
exocyclic methyl group. These compounds have been found in cooked meat and
some other protein-containing foods (Sugimura, 1982), and were first discovered
in 1977 by Sugimura and associates after observing the mutagenicity of smoke
condensate obtained during the broiling of fish (Sugimura et al., 1977).

The most common HAAS found in food products are IQ (2-amino-3-
methylimidazo[4,5f]quinoline), MeIQ (2-amino-3,4dimethylimidazo[4,5f]quino-
line), MeIQx (2-amino-3,8 dimethylimidazo[4,5f]quinoxaline), 4,8 DiMeIQx (2-
amino-3,4,8 trimethylimidazo[4,5flquinoxaline), and PhIP (2-amino-l-methyl-6-
phenylimidazo [4,5b]pyridine) (Skog, 1993). Other HAAs have subsequently been
isolated from and identified in foods and model systems, and the current number of
HAAS known to occur in foods is now 17 (Johansson et al., 1993).

Mutagenicity of heterocyclic amines is generally measured by the Ames assay
(Ames et al., 1975; Sugimura et al., 1988). They are not mutagenic as found, but
require activation by cytochrome p45 OIA2 enzyme to their hydroxyamino forms
(McManus et al., 1990). Esters formed with acetic or sulfuric acid probably are the
final DNA adducts (Sugimura, 1982). Mutagenicity varies widely between
individual heterocyclic amines, but is as high as 661,000 revertants/ug toward S.
typhimurium TA98. Aflatoxin B 1, a documented carcinogen, causes only 6,000
revertants/ug of this strain of Salmonella. Not all mutagens are carcinogens,
however, and carcinogenicity must be determined with animal or cell culture
studies. Rodent and primate assays have shown many HAAS to be multisite

carcinogens, including IQ, MeIQ, MeIQx, and PhIP, as well as many of the so-
1

2
called nonpolar HAAS (Ohgaki et al., 1991). Some of these neoplastic sites include
the liver, small and large intestines, blood vessels, mammary gland, and lungs.
While results from these animal studies are highly suggestive of HAAS acting as
carcinogens in humans, this point has been more difficult to prove. Davis et a1.
(1993) found that mutagenic activation of HAAS in the Ames assay was in some
cases greater using human microsomes than using those of rodents.

Human consumption of HAAS has been estimated to range from 4-16 ug/day
(Wakabayashi et al., 1992), based on analyses of the HAA levels in various meat
products by a number of investigators (F elton et al., 1984: Kikugawa and Kato,
1987). However, doses of HAAS used in animal feeding studies have generally
been at many thousands of times the level of human lifetime exposure (Stavric,
1994), so it is difficult to extrapolate these studies to human experience. Some
epidemiological evidence suggests a link between daily consumption of well-done
meat or fish and risk of colorectal cancer (Schiffinan and Felton, 1990;
Gerhardsson de Verdier et al., 1991). Currently these compounds are considered
possible or probable human carcinogens by the International Agency for Research
on Cancer, and therefore their occurrence in meats and other foods deserves the
close scrutiny of researchers (IARC, 1993).

Given the likelihood that many HAAS are probably human carcinogens, it
seems prudent to decrease intake of these compounds, ideally without
compromising the nutritional or aesthetic quality of the diet. In order to do so, it is
necessary to understand both the mechanism of HAA formation and the effect of
possible contributing factors such as cooking methods, time and temperature of
cooking, and composition and content of lipid. The exact mechanism of formation
of HAAS is unknown, although various postulates have been published for specific
compounds (J agerstad et al., 1983a; Nyhammer, 1986). It is currently accepted that
precursors to HAAS are specific amino acids and creatinine, with sugars enhancing

formation under some conditions (Skog and Jagerstad, 1990). The reaction

3

requires heat, and is assumed to proceed utilizing Maillard reaction intermediates
(Spingarn and Garvie, 1979; Sugimura, 1982; Jagerstad et al., 1983a; Shibamoto et
al., 1981). While contributing factors such as time, temperature, and method of
cooking have been investigated by a number of researchers who have published
similar findings (Bjeldanes et al., 1983; Knize et al., 1994b; Skog et al., 1992a),
there has been no such agreement on the contribution of lipids to HAA formation.
Bjeldanes et al. (1983) found that mutagenicity of various beef patties was
independent of fat content, while Knize et al. (1985) showed that mutagenicity of
beef patties increased with increasing fat content up to 15%, then decreased
slightly with higher fat content. Nilsson and colleagues (1986) found that the
addition of different types of flying fat increased mutagenicity of meat products.
Johansson et a1. (1993) utilized an aqueous model system of creatinine, sugar, and
an amino acid to Show an increase in MeIQx formation upon addition of corn or
olive oil to the system.

The studies that have investigated lipid contribution to HAA formation have
been hampered by several factors. Most of these studies measured mutagenicity
only, not individual HAA formation. Only the experiments done by Johansson et
al. (1993) in an aqueous model system separated physical (heat transfer) from
chemical (molecular contribution) effects, and this study along with others did not
employ adequate statistical analyses. Information which makes the question of
lipid contribution to HAA formation more pressing includes the large body of
work showing high-fat diets to be epidemiologically linked to cancer (Alavanja et
al., 1996; Zhang, et al., 1996) as well as literature demonstrating the ability of fat
to act as a cancer promoter (Reddy, 1978, Weisburger et al., 1983).

Based on these observations, the hypothesis for the current studies was as
follows: Increasing degrees of lipid unsaturation will lead to increasing amounts of

HAA formation in aqueous model systems containing HAA precursors and lipid.

4

Iron will increase HAA formation in these model systems when lipid is present
through accelerated lipid oxidation. Objectives for the studies were:

(1) To determine the effect of specific lipids on the species and quantities of HAAS
formed in this model system.

(2) To compare the effects of saturated versus unsaturated lipid on the species and
quantities of HAAS formed in this model system.

(3) To study the effect of both heme and nonheme iron on the Species and
quantities of HAAS formed in the model system, both with and without lipid

present.

LITERATURE REVIEW

History of Heterocyclic Aromatic Amines

Discovery
Mutagens and carcinogens have been known to exist in the environment for

many decades. Pollutants present in air and water, including pesticides and
industrial chemicals, have been studied extensively (Sugimura and Sato, 1983).
Mutagenic or carcinogenic fungal toxins such as aflatoxins have been known to
‘ contaminate food Since the 19503 (Harrison et al., 1993). Several food additives
such as AF -2, a nitrofuran derivative used as a preservative in Japan, and N—nitroso
compounds, formed in cured meats from the interaction of nitrite and secondary
amino compounds, have been shown to be mutagenic or carcinogenic (Kada, 1973;
Tricker and Preussmann, 1991).

Mutagens/carcinogens which are innately present in foods or formed during
cooking of these foods are more difficult to isolate, and their effect on humans is
more difficult to determine. HAAS fall under this final category, and have become
the subject of extensive research in the last twenty years. The large body of
information which now exists regarding HAAS is due in part to the development of
the Ames Salmonella typhimurium assay for mutagenicity (Ames et al., 1975). The
Ames assay allows rapid determination of mutagenic potential which did not exist
prior to 1975.

The formation of mutagenic/carcinogenic substances in food as a result of
cooking or processing first became the subject of research in the early 1960’s,

when benzo[or]pyrene, a polycyclic aromatic hydrocarbon, was identified in

5

6
barbecued meats (Lijinsky and Shubik, 1964). About ten years later, Sugimura and

associates began to study the mutagenicity of smoke condensates from the cooking
of meat and fish, as well as the mutagenicity of the charred meat surfaces (Nagao
et al., 1977; Sugimura et al., 1977). The mutagenic levels found were far higher
than could be explained by the benzo[a]pyrene content, so a search began for other
mutagens in meats. Commoner and associates reported mutagenic compounds in
beef and beef broth cooked at temperatures ranging from 190-300°C, and
confirmed the absence of mutagens in raw meat. They were the first to attempt to
isolate and separate these compounds using chromatography (Commoner et al.,
1978; Dolara et al., 1979). Almost twenty years later, at least 17 HAAS have been
isolated from meat and other proteinaceous foods, and researchers continue to find

new compounds (J ohansson et al., 1993).

Categories
The heterocyclic aromatic amines that have been isolated from food can be

divided into two categories based on the temperatures at which they form.
“Pyrolytic” mutagens are those which are formed above 300°C and are
characterized by a pyridine ring with an amino group attached (Skog, 1993).
Pyrolytic mutagens can be deaminated by treatment with nitrite under acidic
conditions, after which they are no longer mutagenic (Tsuda et al., 1985). As these
mutagens are formed at such high temperatures, they are not generally found in
foods cooked under normal household conditions. Exceptions to this rule include
the findings of two a-carbolines, 2-amino-9H-pyrido[2,3-b]indole (AorC), and 2-
amino-3-methyl-9H-pyrido[2,3-b]indole (MeAaC), in grilled fish (Gross and
Gruter, 1992).

“Thermic” mutagens are the second category of HAAS based on formation
temperature, and includes HAAS formed below 300°C. These compounds, also

known as aminoimidazoazaarenes can be further subdivided into quinolines,

7

quinoxalines, pyridines, and fur0pyridines (Skog, 1993). A brief review of the

most common HAAS in each category follows. Structures are shown in Figure 1.

 

 

 

 

 

 

 

 

 

 

 

 

 

N
N_CH3
/
\
N
IQ
NH;
N
N N—CH3
/
I;
N
IQx
NH;
N
H3C N N—CH3
/
I
H3C N
7,8-DiMeIQx
rHs
/ N
| >—NH2
PhIP

 

 

 

 

 

 

 

 

 

 

 

 

NH2
N =<
N —CH3
/
\
N CH3
MeIQ
NH2
N 4
H3C N N—CH3
/
I
N
MeIQx
NH;
N
H3C N N—CH3
/
I
N CH3
4, 8-DiMeIQX

 

 

 

Figure 1. Chemical structures of some HAAS in cooked foods (Skog, 1993).

Quinolines

Two distinct imidazoquinoline compounds have been identified in foods. The
HAAS, 2-amino-3-methylimidazo[4,5-j]quinoline (IQ) and 2-amino-3,4-methyl-
imidazo[4,5-f]quinoline (MeIQ), were initially isolated from broiled sardines
(Kasai et al., 1980). They have subsequently been identified in fi'ied ground beef
(Barnes et al., 1983; Felton et al., 1984; Felton et al., 1986a), fried ground pork
(Vahl et al., 1988), and various fishes (Y amaizumi et al., 1986; Zhang et al., 1988).
IQ has also been found in beef extract (Taylor et al., 1985; Turesky et al., 1989),
and has been formed in model systems containing creatinine, glucose or fi'uctose,

and a variety of amino acids (Table l). MeIQ is generally not found in most model

systems.

Quinoxalines

The quinoxaline, 2-amino-3,8-dimethylimidazo—[4,5-f]quinoxaline (MeIQx),
was first isolated from fiied ground beef by Kasai et a1. (1981). MeIQx has
subsequently been found in fried ground pork, chicken, mutton, fish, and beef
extract (Gry et al., 1986; Wakabayashi et al., 1986; Sugimura et al., 1988; Turesky
et al., 1988, Vah] et al., 1988, Gross, 1990; Gross and Gruter, 1992). This HAA
has also been produced in a number of model systems containing creatinine, a
monosaccharide or disaccharide, and one of five amino acids (Table 1). MeIQx
can also be produced in model systems without a sugar, using only creatinine plus
threonine, serine, or alanine (Overvik et al., 1989).

Two other quinoxalines which have been isolated from meat are 2-amino-
3,4,8-trimethylimidazo[4,5-f]quinoxaline (4,8-DiMeIQx) and 2-amino-3,7,8-tri-
methylimidazo[4,5-f]quinoxaline (7,8-DiMeIQx). The latter compound has been
found in only a few meat products such as fried ground beef (Turesky et al., 1988)
and roasted eel (Lee and Tsai, 1991). However, 4,8-DiMeIQx has been isolated
from fiied ground beef (Grivas et al., 1985; Felton et al., 1986a, 1992; Turesky et

10

al., 1988; Murray et al., 1988; Sugimura et al., 1988), fiied ground pork (Gry et al.,
1986; Vahl et al., 1987), chicken (Sugimura et al., 1988), mutton (Kato et al.,
1986), fish (Zhang et al., 1988), and beef extract (Turesky et al., 1988; Gross,
1990). Both these mutagens have also been produced in model systems containing
creatinine, a monosaccharide or disaccharide, and an amino acid. As with MeIQx,
4,8-DiMeIQx is produced when one of five amino acids (threonine, alanine,
glycine, phenylalanine, or lysine) is used, but 7,8-DiMeIQx has only been isolated
from a system using glycine as the amino acid (Table 1).

A fourth quinoxaline, 2-amino-3-methylimidazo[4,5-j]quinoxaline (IQx) has
been isolated from model systems (Knize et al., 1988; Skog and Jagerstad, 1993;
Johansson et al., 1993), but only found once in a meat system (Becher et al., 1988).

Pyridines

The dominant HAA formed in cooked meat is a pyridine, 2-amino-l-methyl-
6-phenylimidazo[4,5-f]pyridine, abbreviated PhIP. This HAA was first isolated
from the crust of fried ground beef in 1986 (F elton et al., 1986b). Since then, many
other researchers have added to the body of knowledge regarding the occurrence,
chemistry, and mutagenicity/carcinogenicity of PhIP. According to findings from
model system studies, phenylalanine is the major amino acid that contributes to
PhIP formation (Shioya et al., 1987; Felton and Knize, 1990; Skog and Jagerstad,
1991; Manabe et al., 1992). Leucine may also be involved in PhIP formation
(Overvik et al., 1989). PhIP has been isolated from most of the same meat products
as the quinolines and quinoxalines. Fried beef, fiied ground pork, broiled chicken,
mutton, fish, and beef extract all contain PhIP (F elton et al., 1986a; Gry et al.,
1986; Vahl et al., 1987; Hayatsu et al., 1991; Gross and Gruter, 1992). The mass of
PhIP in beef has been estimated to be 10 times the weight of other known HAAS
combined (F elton et al., 1986b). Therefore, even though PhIP is less mutagenic

11
than the other HAAs (Table 2), it may play a larger role in the etiology of cancer
(Stavric, 1994).

Two other pyridines have been identified in meat, 2-amino-n,n,n-trimethyl-
irnidazopyridine (TMIP) and 2-amino-1,6-dimethylimidazopyridine (DMIP), but
occur in much smaller quantities than PhIP (Becher et al., 1988, 1989; Felton et
al., 1984).

New HAAs

Purification and isolation methods for HAAS continue to improve, and as a
result of these improvements, new HAAS are being identified. Many of these
HAAS contain oxygen. The structure of a furopyridine, first reported by Gry et al.
(1986), was elucidated by Knize et al. (1990) to be C10H10N4O. A pyridine similar
to PhIP, 4-OH-PhIP, was shown by Kurosaka et a1. (1992) to be mutagenic. Knize
et al. (1991) identified a mutagen fiom a model system with a molecular weight of
244 and composition of C12H12N402. In Japan, a new quinoxaline, 4-CHZOH-8-
MeIQx, was isolated from beef extract (Kim et al., 1994). Each new finding adds
to the accumulated knowledge of HAA formation in foods, and also illustrates the

need for continuing research in this area.

Mutagenicity/Carcmmcig
The development of the Ames assay for mutagenicity in 1975 made possible I
the rapid, consistent classification of various compounds as potentially mutagenic
or nonmutagenic. Following the lead of Sugimura et al. (1977), many research
groups began utilizing the Ames assay to determine mutagenicity in food products.
HAAS were discovered to be potent mutagens once activated by liver microsomes
(Sugimura et al., 1977; 1979). The initial activation step is thought to be N-
hydroxylation, mainly by the cytochrome p4501A2 mediated mixed-function

12

Table 1. Heterocyclic aromatic amines produced in model systems (Skog, 1993).

 

 

Compound Yield* Amino Su— Heating Reference
Acid gar Conditions
IQ 0.4 pro — Dry Yoshida et al. (1984)
1.0 gly fru DEG-Water Grivas et al. (1986)
3.0 phe — Dry Felton and Knize (1990)
13.5 phe glu Dry F elton and Knize (1990)
3.7 ser — Dry Knize et al. (1988)
MeIQ nd ala fru DEG-Water Grivas et al. (1985)
IQx 2.7 ser — Dry Knize et al. (1988)
nd gly glu Water Skog and J ohansson (unpublished,
1 993)
nd thr glu Water Skog and Jagerstad (1993)
MeIQx 4.4 gly glu DEG-Water Jagerstad et al. (1984)
0.9 ala glu DEG-Water Muramatsu and Matsushima (1985)
1.8 ala rib DEG-Water Muramatsu and Matsushima (1985)
4.2 lys rib DEG-Water Muramatsu and Matsushima (1985)
nd thr glu DEG-Water Negishi et al. (1985)
6—7 gly fru DEG-Water Grivas et al. (1986)
nd ser — Dry Overvik et al. (1989)
nd ala — Dry Overvik et al. (1989)
nd tyr — Dry Overvik et al. (1989)
4 gly glu DEG-Water Skog and Jagerstad (1990)
nd phe glu DEG-Water Skog and Jagerstad (1991)
10 ala, thr glu DEG-Water Skog et al. (1992b)
8.8- gly glu Water Johansson et al. (1993)
17.9
7-10 gly glu Water Skog and Jagerstad (1993)
9 thr glu Water Skog and Jagerstad ( 1993)
nd gly glu Water Johansson and J agerstad (1993)
4,8-DiMeIQx nd thr glu DEG-Water Negishi et al. (1985)
1.9-2.6 ala fru DEG-Water Grivas et al. (1985)
4.2 ala glu DEG-Water Muramatsu and Matsushima (1985)
1.5 ala rib DEG-Water Muramatsu and Matsushima (1985)
26.1 lys rib DEG-Water Muramatsu and Matsushima (1985)
nd gly glu DEG-Water Skog and Jagerstad (1990)
nd phe glu DEG-Water Skog and Jagerstad (1991)
36 ala, thr glu DEG-Water Skog et al. (1992b)
30 thr glu Water Skog and Jagerstad (1993)
nd gly glu Water J ohansson et a1. (1993)_
7,8-DiMeIQx 1.1 gly glu DEG-Water Negishi et al. (1984)
nd gly glu DEG-Water Skog and Jagerstad (1990)
nd gly glu Water J ohansson and Jagerstad (1993)

13

Table 1. @nt’d)

 

4,7,8- 6 ala, thr glu DEG-Water Skog et al. (1992b)
TriMeIQx
PhIP 3.6 phe glu DEG-Water Shioya et al. (1987)
735 phe — Dry Felton and Knize (1990)
560 phe glu Dry Felton and Knize (1990)
nd phe — Dry Overvik et al. (1989)
nd leu — Dry Overvik et a1. (1989)
20.9 phe glyu DEG-Water Skog and Jagerstad ( 1991)
6.4 phe — DEG-Water Skog and Jagerstad (1991)

<0.06 phe glu DEG-Water Manabe et al. (1992)

*Yield is in nmol/mmol creatin(in)e

Dry = dry heating at 180 or 200°C for 1 hour

DEG-Water = reflux boiling in diethylene glycol/water (5:1) or 14% water

Water = heated in water in closed metal tubes at 180°C for up to 30 minutes

Amino acids: pro = proline, gly = glycine, phe = phenylalanine, ser = serine, ala = alanine, thr =
threonine, lys = lysine, tyr = tyrosine, leu = leucine.

Sugars: fru = fi'uctose, glu = glucose, rib = ribose.

nd = not determined

oxidase system, although the p4501A1 system also may catalyze the reaction (Kato
and Yamazoe, 1987; McManus et al., 1990). After hydroxylation occurs,
esterification with acetic or sulfuric acid probably forms the final product which
can then bind to DNA. These adducts cause mainly fi'ameshift mutations in the
Ames assay using Salmonella typhimurium strains TA98 and TA 100 (Sugimura,
1982; Sugimura et al., 1988). In vivo, these frameshift mutations would then cause
abberrant proteins to be formed through the processes of translation and
transcription.

The mutagenicity of HAAS is well documented and is, in several cases,
greater than that of compounds which are known to be potent carcinogens, such as
aflatoxin B1 (Table 2). While many mutagens are also carcinogens, this is not
necessarily the case, and cell culture or whole-animal assays must be performed to
determine the carcinogenic potential of a compound. Long-term feeding

experiments with rodents and primates have demonstrated the carcinogenicity of a

14
number of the HAAS, including IQ, MeIQ, MeIQx, 4,8-DiMeIQx, and PhIP. The

primary sites of DNA-adducts caused by the IQ and IQx-type compounds appear
to be the liver, followed by kidney, heart, and bladder (Snyderwine et al., 1992).
Tumors have been observed in the liver, colon, small intestine, stomach, liver,
bladder, mammary gland, and pancreas of rodents fed IQ-type compounds.
(Takayama et al., 1984; Tanaka et al., 1985; Weisburger et al., 1986). PhIP, in
contrast to its weak mutagenic activity in the Ames assay, shows strong
mutagenicity in mammalian cell cultures (Thompson et al., 1987). Colon and
mammary gland carcinomas have been induced by PhIP in rodents, as well as
lymphomas (Esumi et al., 1989; Ito et al., 1991). In attempting to extrapolate
animal studies to humans, it should be noted that human microsomes Show a two-

fold greater ability than rat microsomes to activate PhIP (Davis et al., 1993).

Incidence and Consumption of HAAS

There remains little doubt as to the potent carcinogenic potential of HAAS in
many animal Species. Questions do exist regarding the interpretation of the results
of these studies (Stavric, 1994). Many of the studies used doses equivalent to
thousands of times the human lifetime exposure. Two studies using graduated
doses of MeIQx found a linear dose-response relationship between ingestion of
MeIQx and hepatic DNA adducts, even at doses too small to produce tumors in the
test animals (Y amashita et al., 1990; Ohgaki et al., 1991). Thus it is possible that
there exits a threshold level for HAA consumption below which cancer does not
form. It is more likely that the many HAAS consumed daily, along with exposure .
to other carcinogens and promoters, exert some type of additive effect on human
cancer risk. In order to examine human risk more closely, it is necessary to turn to

epidemiological evidence and our knowledge of the HAA content of various foods.

15

Table 2. Mutagenicity of heterocyclic aromatic amines and typical carcinogens in

Salmonella typhimurium (Sugimura and Sato, 1982; Sugimura et al., 1988)

 

 

Revertants/gg
Compound TA98 TA100
IQ 433,000 7,000
MeIQ 661,000 30,000
IQx 75,000 1,500
MeIQx 145,000 14,000
4,8-DiMeIQx 1 83,000 8,000
7,8-DiMeIQx 163,000 9,900
PhIP 1,800 120
Trp-P-l 39,000 1,700
Trp-P-2 104,200 1,800
Glu-P-l 49,000 3,200
Glu-P-2 1,900 1,200
Om-P-l 56,800 _
AaC 300 20
MeAaC 200 120
Aflatoxin B1 6,000 28,000
AF-2 6,500 42,000
4—Nitroquinolin l-oxide 970 9,900
Benzo[a]pyrene 320 660
N-Methyl-N’nitro-N- 0.00 870
nitrosoguanidine
N-nitrosodiethylamine 0.02 0. l 5
N-Nitrosodimethylamine 0.00 0.23

 

16

The average daily consumption of HAAS from food in the Western
hemisphere has been estimated to range from 0.1 ug/day to 16 ug/day (Turesky et
al., 1993; Wakabayashi et al., 1992). The actual intake is quite variable, based on
amount and type of meat eaten as well as cooking technique and degree of
doneness. Lifetime cancer risk for humans based on intake of 80 ng/kg/day is
significant (Felton et al., 1986a; Adamson, 1990). Actual epidemiological studies
considering HAA intake and cancer risk are sparse and contradictory. Lyon and
Mahoney (1988) reported no association between ingestion of fiied or broiled meat
and the risk of colon cancer. However, the data collection procedure which they
utilized has been criticized as being inaccurate (Schiffman and F elton, 1990). A
population study in Sweden found that total meat intake, consumption of brown
gravy, and heavily browned meat each increased risk of colon cancer
independently (Gerhardsson-de-Verdier et al., 1991). These same authors
concluded in a later review article that the epidemiological literature is insufficient
to justify any recommendations at this time (Steineck, et al., 1993).

Heterocyclic aromatic amines appear to be possible or probable human
carcinogens on the basis of animal studies and bacterial assays (IARC, 1993).
Clearly, more long term, in-depth human studies are needed to definitively classify
HAAs as human carcinogens. In order to elucidate an accurate relationship
between HAA consumption and carcinogenesis, it is also necessary to have precise
knowledge of the HAA content of foods and the factors which affect their
formation. Accurate values are not easy to obtain. The methods used to isolate and
quantitate HAAS in foods vary widely, as do the results (F elton et al., 1992). The
most accurate method currently in wide use is the solid-phase extraction procedure
developed by Gross (1990) and refined by Gross and Gruter (1992). Briefly, this
procedure utilizes tandem extraction with diatomaceous earth and an ion exchange
resin followed by further clean-up with a C18 column. Separation and

quantification of the HAAS is achieved with high performance liquid

17
chromatography (HPLC) using a photodiode array detector. Another accurate

approach which is not in general use involves spiking with heavy-isotope-labeled
standards, a clean-up procedure, then analysis by gas chromatography-electron-
capture negative ion chemical ionization mass spectrometry (Murray et al., 198 8).

Table 3 lists average values of HAAs in meat products (Skog, 1993).

Table 3. Average values of HAAs formed in selected food products (Skog, 1993)

 

Heterocyclic Aromatic Amines*

 

 

Food IQ MeIQ MeIQx 4,8-DiMeIQx PhIP

Fried ground beef 3.69 not 1.58 1.13 13.19
determined

Salmon 0.85 1.55 0.93 — —

Chicken — —— 2.22 0.81 38.1

Fried ground pork 0.04 0.02 1.4 0.6 4.5

Beef Extract 4.70 — 19.99 3.45 3.62

 

*Values are expressed in ng/g cooked or uncooked food, and are an average of

those quoted by Skog (1993).

Mechanism of Formation of HAAS

The next step beyond establishing mutagenic/carcinogenic potential of HAAS
and ascertaining levels formed in the food supply is to elucidate the mechanism of
HAA formation. Without this understanding, it is difficult to determine
contributing factors or formulate methods to control HAA formation.

Much is not yet known about the mechanism by which HAAS are formed. It
is generally accepted that one route of HAA formation is through intermediates of

the Maillard or nonenzymatic browning reaction. This concept was first proposed

18
not long after HAAS were initially isolated (Spingarn and Garvie, 1979; Powrie et

al., 1981; Shibamoto et al., 1981; Wei et al., 1981). The precursors which have
been found to be essential for HAA formation are creatine or creatinine (Y oshida
and Okamoto, 1980a,b,c; Bjeldanes et al., 1982a,b) and amino acids (Taylor et al.,
1984, 1985). Sugars and water appear to be involved in the formation of some
HAAS (J agerstad et al., 1983b; Skog and Jagerstad, 1990). The contributions of the

various reactants are detailed later in the text.

Chemical Formation

The Maillard reaction is actually a series of reactions between reducing
sugars and proteins or amino acids, catalyzed by heat, that lead to the formation of
hundreds of compounds. Melanoidins are some of the color compounds formed
during the Maillard reaction that lead to the characteristic brown color of baked
breads and fried meats. Aldehydes, firranones, sulfur-containing heterocycles,
pyridines, pyrazines, and pyrroles all are volatile Maillard reaction products which
contribute to flavor and aroma (Rizzi, 1994).

The Maillard reaction begins when a reducing sugar reacts with the amino
group of an amino acid to form a glycosylamine. This compound rearranges to
give an N-substituted amino ketose, or Amadori compound. Amadori compounds
are thermodynamically unstable, and under alkaline conditions, these rings
fragment to 2- and 3-carbon compounds, which give rise to melanoidins (Hayashi
and Namiki, 1986). Slightly acidic conditions will cause the Amadori compounds
to lose their amine and give rise to l-deoxyosones (Pischetsrieder and Severin,
1994). Strongly acidic conditions will generate a 3-deoxyosone. Deoxyosones
undergo cyclization and dehydration, then fragment to low molecular weight 1,2-
dicarbonyls. These dicarbonyls react with amino acids (the Strecker degradation)
to form aldehydes, pyrazines, pyrroles, pyridines, and many other volatile
compounds (Rizzi, 1994).

19
The first hypothesis for the mechanism of HAA formation through Maillard

reaction pathways was proposed by Jagerstad et al. (1983a). This hypothesis
assumes the formation of pyridines and pyrazines through the Maillard reaction
pathway outlined above. One molecule of a pyridine or pyrazine then reacts with
an aldehyde (which has also been formed through the Maillard reaction) to form
the quinoline or quinoxaline portion of the HAA. Creatine, when heated,
dehydrates and cyclizes to form creatinine, and one molecule of creatinine adds to
the aldehyde to form an IQ-or IQx-type heterocyclic aromatic amine (Figure 2).

Confirmation of the reactants involved in HAA formation was offered by
Jagerstad et al. (1983b) in the way of a model system study that included refluxing
creatinine, amino acids, and glucose in a water/diethylene glycol mixture at 128°C.
This mixture exhibited high mutagenic potential in the Ames assay. Excluding any
one of the three reactants reduced the level of mutagenicity Significantly (Jagerstad
et al., 1983b). Addition of synthetic pyridines or pyrazines increased the mutagenic
potential by 50%. Grivas et al (1985) utilized a similar model system, employing
fructose as the sugar, and glycine as the amino acid. From this system, they
isolated MeIQx and IQ, offering further confirmation of the precursors needed to
form HAAS.

The theory of J agerstaad et al. (1983a) assumes that condensation first occurs
between pyridines or pyrazines and aldehydes, followed by ring closure with
creatinine. A variation of this theory was offered by Nyharnmar (1986), who
hypothesized that condensation first occurs between creatinine and an aldehyde,
with the addition of a pyrazine or pyridine as the last step. Support for the
condensation of aldehydes with creatinine was offered by Jones and Weisburger
(1988), who proposed a “concerted condensation” model of IQ-type HAA
formation after finding “IQ-like” mutagens resulting from the reaction of
aldehydes and creatinine. This model involves the condensation of two molecules

of acetaldehyde with creatinine in one step to form IQx-type compounds. They

20
found no mutagenic activity resulting flom the reflux of 2-vinylpyrazine with
creatinine, which may indicate that aldehydes are necessary for HAA formation. It
Should be emphasized that very few mechanistic studies have actually been
performed to verify reaction sequences for HAA formation.

An alternative explanation of early Maillard reaction chemistry raises
intriguing questions about HAA formation. A series of experiments performed by
Namiki and Hayashi (1981, 1983) challenges the long-held belief that Amadori
rearrangement follows aldol condensation in the Maillard reaction sequence
(Hodge, 195 3). Using electron spin resonance (ESR) to look for flee radicals, these
investigators followed the Maillard reaction of various sugars and amino acids. It
was found that N,N’-disubstituted pyrazine cation radicals were formed prior to
Amadori rearrangement. The reaction sequence was determined to first involve the
formation of glucosylamine, then flagrnentation to a 2-carbon glycolaldehyde
alkylimine and other products. The glycolaldehyde alkylirnines can then proceed
through a series of further reactions to form glyoxal, or condense to form a
dialkylpyrazine radical. According to Namiki and Hayashi (1981), the Amadori
rearrangement occurs after the formation of this radical.

Milic et al. (1993) performed ESR studies of HAA formation which
confirmed the pyrazine/pyridine flee radical cation findings. They were unable to
determine if these flee radicals participate in HAA formation. Therefore the

possibility of a flee radical mechanism for HAA formation does exist.

Eton Contributigg to HAA Formation
Creatinine

Creatine exists in vertebrate animals as creatine phosphate, which acts as an
energy reserve for muscles. Creatine can be cyclized to form creatinine through
dehydration at elevated temperatures (Laser-Reutersward et al., 1987a). Creatinine

has been clearly implicated, through both model system and meat

 

 

 

 

21

+

 

 

 

 

 

 

 

 

 

 

 

 

NHz

 

 

 

 

 

 

 

 

 

 

NH3 —/
/ HN
C6H1206 R—CQ HO O C N
Hexose . C90 V \CH3
Ammo Acrd Creatin e
l -H20
Y Z N /NH2
Ii + OHC—R + A/
N\
X N CHs Aldehyde 0 CH3
Pyridine
or Creatinine
Pyrazine l
N—CHs

 

 

IQ or IQx Compound

 

 

 

Figure 2. Theoretical reaction pathway for formation of IQ and IQx compounds. R,
X, and Y may be H or CH3: Z may be CH or N (Jagerstad et al., 1983a).

22

studies, as a precursor of HAAs. The first indication that muscle meats contain a
compound which causes mutagen formation was published by Bjeldanes et al.
(1982a,b). These researchers found low mutagenicity in cooked dairy products,
organ meats, and shrimp, compared to the high mutagenicity reported for cooked
muscle meats. Jagerstad et al. (1983a) proposed the involvement of creatinine in
HAA formation and demonstrated a 50% increase in mutagenicity of beef when
creatine was added before cooking. Other researchers subsequently verified the
increased mutagenicity of meat systems containing added creatine (Taylor et al.,
1985; Becher et al., 1988; Knize et al., 1988; Overvik et al., 1989). Laser-
Reutersward et al. (1987a) confirmed earlier findings of low mutagenicity in non-
muscle, high protein meat. They evaluated the levels of mutagenicity developed
during the cooking of a variety of bovine tissues which exhibited a wide range of
creatine concentrations. Kidney and liver were found to have very low levels of
creatine, and also low mutagenicity, while muscle portions, heart, and tongue
exhibited much higher mutagenic potential.

Model system studies provide the D most definitive proof of creatinine as a
HAA precursor, as they exclude other compounds contained in meat. Many model
system studies have involved the heating of an amino acid with creatine/creatinine
and a sugar to produce IQ-, IQx-, and pyridine-type HAAS (Table 1). Yoshido and
Okamoto (l980a,b) were the first to use creatine and amino acids in a model
system to produce mutagenic activity in the Ames assay. Knize et al. (1988) and
F elton and Knize (1990) were able to isolate IQ and PhIP flom dry-heating amino
acids with creatinine with no sugar present. Many other investigators have utilized
diethylene glycol as a reflux medium for model systems containing creatinine,
amino acids, and sometimes sugars (Skog and Jagerstad, 1990; Muramatsu and
Matsushima, 1985; Grivas et al., 1985). An aqueous system was employed by
Johansson et al. (1993). Despite the many differences in the model systems

employed, all these systems have in common amino acids and creatine/creatinine

23
as reactants. These are the only model systems flom which HAAs have been

isolated, a fact that provides significant evidence for creatinine as a necessary
precursor of HAAS. Felton and Knize (1990) have been able to Show incorporation
of labeled isotopes of creatine into the PhIP molecule, giving definitive evidence

of creatine’s role in the formation of PhIP.

Amino Acids
AS previously discussed, all model systems which have produced isolatable

HAAS have contained amino acids (see Table 1). Ever since Commoner et a1.
(1978) demonstrated mutagenicity in cooked beef, researchers have shown
repeatedly that HAA formation is limited to specific protein-rich foods (N agao et
al., 1977; Bjeldanes et al., 1982a,b). Mutagenicity is present at low levels in some
grain-based cooked foods, but none of the known heterocyclic aromatic amines
have been isolated flom these foods (Knize et al., 1994a). Some form of protein,
therefore, appears to be required for HAA formation. Several researchers have
attempted to determine the form of protein involved in this process. J agerstad et al.
(1983b) found that substituting proteins for amino acids in a model system
produced no mutagenic activity. Ashoor et al. (1980) reported that out of 19 amino
acids tested, only proline and hydroxyproline increased mutagenicity of flied beef
patties in the Ames assay. Taylor et al. (1985) demonstrated that HAA production
markedly increased in boiled beef extract when the amino acid tryptophan was
added to the reaction mixture. Overvik et al. (1989) recorded increases in the
mutagenicity of broiled pork up to 43-fold with the addition of 1% by weight of
various amino acids. The dipeptide, camosine (alanine plus histidine), has also
been found to contribute to mutagenic activity at approximately the same level as
single amino acids when utilized in a model system with creatinine and glucose

(Laser-Reutersward et al., 1987b).

24

The research to date clearly points to the flee amino acid content of meats
rather than their protein content as a determining factor in HAA formation (Skog,
1993). An interesting point that is made in several papers is that not all amino
acids contribute equally to mutagenicity (Ashoor et al., 1980; Jagerstad et al.,
1983b; Overvik et al., 1989). Threonine appears to have the greatest mutagenic
potential. Most HAAS can be formed flom more than one amino acid (Table 1).
Often, however, one amino acid causes greater amounts of a specific HAA to be
formed than does another amino acid. AS more research is conducted in this area,
findings may indicate that only certain amino acids are truly Significant

contributors to HAA formation.

Sugars

The role of sugars in the formation of HAAS is less clear than that of amino
acids and creatinine. Many HAAS have been isolated flom model systems using
only creatinine and amino acids as reactants (see Table 1). However, sugars have
been shown to increase HAA yield and change the relative ratio of HAAS formed
in these model systems (Muramatsu and Matsushima, 1985; Skog and Jagerstad,
1990; Knize et al., 1991). Skog and Jagerstad (1991) further proved this point in a
model system experiment where phenylalanine was heated with creatine at 180° to
yield only PhIP. When glucose was added to the reaction mixture under the same
conditions, the yield of PhIP increased three-fold, and MeIQx and 4,8 DiMeIQx
were also detected. Concentration of sugars appears to play a role in mutagen
formation as well. High concentrations of monosaccharides or disaccharides
(greater than half the molar amount of the creatinine) acted as inhibitors to the
formation of HAAS in a model system study employing increasing amounts of
these compounds. (Skog and Jagerstad, 1990).

The actual mechanism by which sugar enhances HAA formation was

unknown until Skog and Jagerstad (1993), using 14C-labeled glucose, produced

25

definitive evidence that monosaccharides actually contribute carbon atoms to the
HAAS IQx, MeIQx, and 4,8-DiMeIQx. Labeled carbon atoms flom the glucose
were incorporated into all three HAAS. Jagerstad et al. (1983a) suggested that the
pathway involving sugars is the preferred route of HAA formation in actual meat
systems. They tested mutagenicity in low-glucose beef, which was produced
through animal exhaustion and thus glycogen breakdown prior to slaughter. These
low-glucose beef patties produced very little mutagenic activity in the Ames assay,

but mutagenic potential increased two-to three-fold when glucose was added prior

to flying.

Other Factors

Several factors other than the immediate precursors have been found to exert
significant influence over HAA formation. Fat is one such element, and will be
considered separately in the next section. The factors with the best-defined
relationship to HAA production are time, temperature and method of cooking.

Methods of cooking that result in lower mutagen formation include
microwave cooking, roasting, and stewing (Dolara et al., 1979; Bjeldanes et al.,
1983; Miller and Buchanan, 1983). Microwave cooking of meat is of extremely
short duration, does not cause formation of a brown meat surface, and the pan
residue has been shown to contain the water-soluble precursors of HAAS. These
facts may explain the absence of mutagen formation in microwaved meats (F elton
et al., 1994). Contact flying and deep-fat flying are cooking methods which give
higher levels of mutagen production in meats (Nader et al., 1981).

Temperature has been demonstrated by many researchers to be an important
factor in HAA formation. In meat systems, mutagen formation has been
demonstrated to increase with increasing cooking temperature (Weisburger and
Spingarn, 1979; Bjeldanes et al., 1983; Knize et al., 1994b; Balogh, 1995).
Mutagenic activity has been reported at flying temperatures as low as 130°C

26

(Laser-Reutersward et al., 1989). The exact mathematical relationship between
cooking temperature and mutagen or HAA formation is unknown, although some
studies have suggested a roughly linear relationship within normal cooking
temperatures (Bjeldanes et al., 1983; Knize et al., 1994b).

Cooking time also affects HAA formation. Generally, longer cooking times
produce greater mutagenic potential and larger quantities of HAAS (Weisburger
and Spingarn, 1979; Miller and Buchanan, 1983; Knize et al., 1994b). A lag period
of 2 to 4 minutes during which the surface temperature of the meat is less than
100°C has been demonstrated. During this lag time, little mutagen formation is
found (Spingarn and Weisburger, 1979; Knize et al., 1985). More information on
time and temperature relationships to HAA formation in ground beef has been
detailed by Balogh (1995). Time and temperature relationships to HAA formation
are less predictable in model systems (Skog, 1993).

Due to the proven mutagenic and carcinogenic potential of HAAS, a number
of compounds have been tested to determine if they inhibit HAA formation. The
mechanisms of inhibition of these compounds vary, and much remains unknown as
to their modes of action. High concentrations of sugars were discovered
accidentally to inhibit mutagenic activity in beef stock (Taylor et al., 1986), and
this effect has Since been demonstrated consistently by other researchers (Skog and
Jagerstad, 1990; Skog et al., 1992a). Soy protein concentrate and defatted
glandless cottonseed flour, both of which contain naturally occurring antioxidants,
were shown to decrease mutagenicity in cooked beef (Wang et al., 1982; Rhee et
al., 1987). Polyphenolic compounds present in tea have been Shown to be effective
antioxidants (Sorata et al., 1984; Chen et al., 1990) and inhibitors of HAA
formation in model systems (Weisburger et al., 1994; Yen and Chen, 1995).
Vitamin E is another effective antioxidant that has been shown to inhibit HAA
formation by as much as 80% (Faulkner, 1994; Balogh, 1995).

27
Role of F at in HAA Formation

The role of lipids in HAA formation is an especially interesting topic due to
the epidemiological and laboratory evidence that high-fat diets in general, and
perhaps Specific lipids as well, may play a role in cancer promotion (Reddy et al.,
1978; Brennan-Craddock et al., 1990; Alldrick et al., 1987; Alavanja et al., 1996;
Ritskes et al., 1996). Oxidation products of lipids have been shown to produce
mutagenic activity in the Ames assay (Levin et al., 1982). A logical extension of
these data is to examine the possibility that fat increases the production in meats of
the probable human carcinogens, HAAS. Many researchers have examined the
effect of lipid source and content on HAA formation, mostly in meats but also in
model systems. Results have been variable.

Bjeldanes et al. (1983) examined the effect of meat composition on mutagen
formation in ground beef. Fat content was found to have minimal effect on the
mutagenicity of the samples, with defatted products actually displaying slightly
higher mutagenic potential. Spingarn et al. (1981) obtained opposite results when
examining the mutagenicity of flied ground beef patties. Mutagenic potential
increased with the fat content of the beef up to a maximum at 16.4% fat. Results
similar to those of Spingarn et al. (1981) were obtained by Knize et al. ( 1985), who
found maximum mutagenicity in 15% fat ground beef. Lower fat content produced
lower mutagenic potential, and 30% fat patties had Slightly decreased mutagenicity
compared to 15% fat patties. The authors attributed the decrease at 30% fat to
dilution of the reactants.

Two studies have examined the effects of different types of flying fat on
mutagen formation in pork. Nilsson et al. (1986) investigated the mutagenicity of
pork flied in butter, lard, three types of margarine, and three types of vegetable oil
at two different temperatures. Mean mutagenicity was higher in samples flied in
fat versus samples fried without added fat at 250°C, but the investigators were

unable to correlate mutagenic potential and degree of unsaturation of the flying fat.

28
Overvik et al. (1987) performed a Similar study, flying pork in the same

margarines and oils, and examining mutagenicity of the pan residues as well as the
meat. Again, mean mutagenicity was greater in both pan residues and meat flied
with fat at 250°C versus samples flied without fat. No significant difference in
mutagenic potential between samples flied with fat and those flied without fat at
200°C was observed in either study. These investigators attributed the increased
mutagenicity formed with fat present to the enhancement of heat transfer caused by
cooking with fat. Barrington et a1. (1990) examined the effect of cooking methods
and type of flying fat on mutagen formation in lamb and beef. They demonstrated
that the class of flying fat influenced mutagen formation, with butter and
margarine producing higher mutagenic potential then oil, despite oil-flied meat
achieving the highest maximum cooking temperature. However, no statistical
treatment of the data was apparently performed for this part of the study.

One of the first studies to quantitate the formation of a Specific HAA in
response to two different levels of fat in meat was performed by Barnes et al.
(1983), who determined the amount of IQ formed in beef patties containing either
10.6% or 27.5% fat. F orty-fold higher quantities of IQ were formed in the high-fat
beef than in the low-fat beef. It was hypothesized that this effect was not due to
increased heat transfer, but could be due to a number of factors, including the
formation of flee radicals during lipid peroxidation. These flee radicals might then
enhance some Maillard reactions leading to HAA formation. Barnes et al. (1983)
also postulated that lipid oxidation will lead to increased formation of certain
Maillard reaction intermediates that are possibly HAA precursors such as
aldehydes and N-heterocyclic compounds.

In a more comprehensive study, Johansson and Jagerstad (1994) examined
HAA formation in cooked meats and their pan residues, with specific emphasis on
the role of fat content. Results indicated a significant positive correlation between

fat content and IQ formation. No relationship to fat content was reported for any

29
other HAA. Mutagenicity was not determined, so it is not known if total

mutagenicity increased with increasing IQ formation. It was noted that the total
HAAS formed were ten times greater in pan residue flom the beefburgers flied in
butter versus those flied in vegetable oil. The authors postulated that this
difference may be due to naturally occurring antioxidants in the vegetable oils.

The first researchers to investigate the effect of fat on mutagenicity in model
systems were Barnes and Weisburger (1983, 1984). Their results indicated that
several different types of fats increased mutagenic potential in the model system,
including beef suet, corn oil, and lard. Larger amounts of fat (20%) increased
mutagen formation even further. However, this study did not control for the
physical effect of fat, which is greater efficiency of heat transfer. A model system
was designed by Johansson et al. (1993) that permitted the use of water as a
solvent, and basically eliminated the heat transfer effect of added fat. Using this
model system, these investigators tested the formation of HAAs in a
glucose/glycine/creatinine mixture with addition of various fatty acids and edible
oils at approximately 20% by weight. In this system, MeIQx formation was found
to nearly double when com or olive oil was added to the system. Fatty acids had no
predictable effect. However, due to small sample Size and large standard
deviations, no statistical analysis was performed.

J ohansson and J agerstad (1993) utilized this same model system to study the
effects of oxidized fat, tocopherol content, and iron on HAA formation. The
oxidation status of fat seemed to make little difl‘erence in HAA formation.
Naturally occurring tocopherols did not affect the quantities of HAAS formed.
Addition of iron as ferrous sulfate significantly increased HAA formation, both in
the presence of fat and without fat. In samples containing lipid, HAA formation
was only significantly greater than in the control (no lipid) when iron was present.
MeIQx formation was greater in samples containing 40% fat than in samples

containing 20% fat. These results differ somewhat flom the study by J ohansson et

30

al. (1993 ), where addition of oil caused MeIQx formation to increase relative to the
control which contained no oil. Johansson and Jagerstad (1993) explained the
influence of iron as catalyzing flee radical reactions which lead to greater
formation of pyrazines and pyridines. They hypothesized that these pyrazine
compounds are immediate precursors of HAAS, and their increase leads to a higher
yield of HAAS.

Most of the aforementioned studies found some increase in mutagenicity or
HAA formation in samples with higher levels of intrinsic or added fat. Several
theories have been proposed to explain these observations. First, it is well known
that fat is an efficient conductor of heat. Large amounts of fat in or surrounding the
meat could effectively increase the cooking temperature at the meat surface, or
increase heat transfer into the meat (Knize et al., 1985; Nilsson et al., 1986). This
phenomenon is known as the physical effect of fat on HAA formation. It is
unlikely that this is the only explanation for the increase of HAAS formed in the
presence of fat. Barrington et al. (1990) observed higher mutagenicity in meat
products flied in butter or margarine than in those flied in oil, despite higher
surface temperatures in the oil-cooked products. Nilsson et al. (1986) and Overvik
et al. (1987) reported no greater mutagenicity in pork products flied in fat at 200°C
than in those flied without fat. Barnes et al. (1983) observed no differences in
surface temperature of meats cooked in fat versus those cooked without fat, despite
higher levels of IQ formation in the former. Johansson and Jagerstad (1993) and
Johansson et al. (1993) confirmed the possibility that a chemical effect of fat may
exist through their model system studies which eliminated the heat transfer effect.

The nature of the chemical effect of fat on HAA formation is still quite
speculative. Pyrazine and pyridine formation, which occurs during the Maillard
reaction, is known to be enhanced by the presence of lipids (Watanabe and Sato,
1971; Buttery et al., 1977; Parihar et al., 1981; Amoldi, 1990). If pyrazines and
pyridines are immediate precursors of HAAS, as proposed in the theory of

31
Jagerstad et al. (1983a), lipids could have a chemical role in increasing HAA

formation. Lipid oxidation, whichiis catalyzed by heat, can produce aldehydes and
carbonyl compounds (Farmer and Mottram, 1990), which may also be precursors
of HAAS (Barnes et al., 1983; Jagerstad et al., 1983a; Jones and Weisburger,
1988). Peroxidized lipids can also cause formation of flee radicals (Pokomy,
1981), and based on the effectiveness of phenolic antioxidants in controlling HAA
formation (Faulkner, 1994; Balogh, 1995), flee radicals may be involved in HAA

formation.

MATERIALS AND METHODS

Materials

Creatinine, L-phenylalanine, L-glycine, trilinolein (cis-cis 9,12), tristearin, D-
glucose, l-monostearoyl-rac-glycerol, ferrous sulfate heptahydrate, caffeine,
triethylarnine, and horse-heart myoglobin were purchased flom Sigma Chemical
Company (St. Louis, MO). All solvents, including water, were HPLC grade. The
heterocyclic amine standards (IQx, MeIQx, 4,8-DiMeIQx, TriMeIQx, and PhIP)
were obtained flom Toronto Research Chemicals (Toronto, Canada). The
heterocyclic aromatic amine standard FEMA (Flavour and Extracts Manufacturers’
Association) was a kind gift flom Dr. Mark Knize, Lawrence Livennore National
Laboratory, University of California, CA. The FEMA standard contained IQ,
MeIQ, Mqux, 4,8-DiMeIQx, and PhIP, each at 5.0 ng/ul.

Extrelut-20 columns and Extrelut refill diatomaceous earth were obtained
flom EM Separations (Gibbstown, NJ). Propyl-sulfonic acid (PRS) Bond-Elut
columns (500 mg), C18 cartridges (100 mg), and Hydromatrix diatomaceous earth
were purchased flom Varian, Inc. (Harbor City, CA). The heating module was a
Reacti-Therm III, model 18835, made by Pierce Co. (Rockford, IL). Stainless steel
test tubes, 2.3 ml. capacity, with threaded, self-sealing stainless steel caps were
manufactured by the Engineering Research Complex Machine Shop at Michigan
State University. A separate set of stainless steel test tubes was used for each
amino acid to avoid carryover of HAAS. A Waters 510 high performance liquid
chromatography machine (HPLC), equipped with a 991M photodiode array UV
detector and 420 Waters fluorescence detector (Millipore, Milford, MA) and a

32

33
Rheodyne 7125 injector with a 50 ul loop (Rheodyne, Inc., Cotati, CA) were used

for separation and quantitation of the HAAS.

Methods
Three model system studies were performed as described below. All

procedures mentioned apply to each study unless otherwise stated. Briefly, the

three studies were designed as follows:

Study #1: Eflect of T rilinolein versus T ristearin on HAA Formation in Aqueous
Phenylalanine and Glycine Model Systems

Study #1 examined the effects of a completely saturated lipid (tristearin) and
a highly unsaturated lipid (trilinolein) on HAA formation in a phenylalanine and a
glycine model system. Each amino acid system consisted of a control (no lipid), a
treatment with tristearin, and a treatment with trilinolein. The control and
treatments were replicated five times each. The control for the glycine system
consisted of 0.6 mmol creatinine, 0.3 mmol glucose, 0.6 mmol glycine, and 0.02 g
monostearin in 1.5 ml water. The phenylalanine control was composed of 0.6
mmol creatinine, 0.3 mmol glucose, 0.6 mmol phenylalanine, and 0.02g
monostearin in 1.5 ml water. Treatment #1 for each model system was the addition
of 0.3 g tristearin to the given reactants. Treatment #2 for each system was the

addition of 0.3 g trilinolein to the aforementioned reactants.

Study #2: Eflect of Com Oil and Olive Oil on HAA Formation in an Aqueous
Phenylalanine Model System

Study #2 examined the effect of two vegetable oils with different degrees of
unsaturation on HAA formation in the phenylalanine model system. This study

consisted of a control (tristearin), and two treatments (olive oil and corn oil).

34

Creatinine (0.6 mmol), glucose (0.3 mmol), phenylalanine (0.6 mmol),
monostearin (0.02 g), and tristearin (0.3 g) in 1.5 ml water comprised the control.
Treatment #1 substituted corn oil (0.3 g) for the tristearin. Treatment #2 substituted
olive oil (0.3 g) for the tristearin. Each treatment and the control were replicated
seven times. Three extra treatment #1 replications were performed and extracted
with Hydromatrix to compare extraction efl'rciency of Extrelut and Hydromatrix

diatomaceous earth.

Study #3 .' Efl'ect of Ferrous Sulfate and Myoglobin on HAA Formation in a
Glycine Model System with and without Lipid

Study #3 examined the effect of two sources of iron present in meat (ferrous
sulfate and myoglobin) on HAA formation in the glycine system, both in the
presence and absence of an unsaturated lipid. The five treatments entailed addition
to a glycine system of one of the following: ferrous sulfate, myoglobin, trilinolein,
ferrous sulfate plus trilinolein, or myoglobin plus trilinolein. The control contained
creatinine (0.6 mmol), glucose (0.3 mmol), and glycine (0.6 mmol) in 1.5 ml
water. A control for the lipid-containing samples consisted of the above reactants
plus 0.3 g trilinolein and 0.02 g monostearin. Treatment #1 added ferrous sulfate to
both the model system with trilinolein and the system without trilinolein in the
amount of 0.01 mmol/test tube. Treatment #2 added myoglobin to both model
systems also in the amount of 0.01 mmol. Each treatment plus the control were
replicated five times. Three extra replications of the glycine system control and
three replications of the glycine plus lipid control had nitrogen bubbled through the
reactants and the headspace filled with nitrogen immediately prior to capping the

test tubes. These extra replications were not included in the statistical analysis.

35

Safety of Designed System

Safety of the investigator and other laboratory workers was a major
consideration when designing the model system. Small, stainless steel test tubes
with tight-fitting caps were chosen rather than glass to minimize the hazard of
explosion during heating. The reaction was carried out in a closed hood separated
flom the rest of the laboratory. Test tubes were not opened until cooled below
20°C. Asbestos gloves were used to handle the hot test tubes and heating block.
All standard laboratory safety precautions such as goggles, lab coat, and latex

gloves for handling chemicals were followed as well.

Weighing/Mixing of Ingredients

Creatinine (0.6 mmol), glucose (0.3 mmol), and either phenylalanine or
glycine (0.6 mmol) were weighed accurately (i 0.5 mg) and poured into clean
glass test tubes. Monostearin (an emulsifier) was added at a level of 0.02 g to each
tube except those that were the control or control plus iron in study #3. Next, 0.3 g
lipid (tristearin, trilinolein, corn oil, or olive oil) was placed in each test tube.
HPLC-grade water (1.5 ml) was added to each tube. The tubes were then capped
and placed, one at a time, in a beaker of boiling water for 15 seconds to dissolve
the emulsifier and tristearin (both of which have melting points above 70°C). The
glass test tubes were shaken vigorously for 10 seconds upon removal flom the
water bath, and the contents immediately poured into the stainless steel tubes and
sealed with the threaded caps wrapped with Teflon tape. The only deviations flom
this procedure were for the controls which did not contain lipid or emulsifier.
These samples were mixed directly in the stainless steel test tubes rather than being
heated in a glass test tube first. Any samples scheduled to receive iron or
myoglobin had the ferrous sulfate or myoglobin dissolved in the water prior to
adding the water to the sample, at a concentration which delivered 0.01 mmol iron

to each test tube.

36

Heatin

The Reacti-Therm heating module was allowed to preheat for a minimum of
1.5 hours before heating of samples. It was discovered in preliminary experiments
that the temperature of the heating block dropped approximately 30°C when
loaded with 10 stainless steel test tubes. The heating temperature was 180° i 5°C,
so the heating block temperature reached prior to loading was set at 210°C. Silicon
oil (0.5 ml) was placed in each cavity in the heating block to facilitate heat transfer
flom the block to the test tubes. Samples were heated for exactly 30 minutes. Upon
completion of heating, the test tubes were plunged immediately into a beaker of

ice, where they were allowed to cool for 30 minutes before transfer of the product

to microvials for temporary storage at 5°C.

Extraction of HAAS

Sample preparation for HAA extraction flom the model system products
began with dissolution of the contents of two identical 1.5 ml microvials in sodium
hydroxide (NaOI-I). Samples containing tristearin were dissolved in 57 ml hot
(100°C), 10N NaOH. All others were dissolved in 57 ml 5N NaOH. The NaOH
and samples were well-mixed with a spatula. Aliquots (10 ml) were taken flom the
mixture and placed in four 250 ml beakers. The remainder of the mixture was
saved and used if any further extractions were required. Approximately 20 g of
Extrelut refill (experiments # 1 and 2) or Hydromatrix (experiment #3) was added
to each beaker. Two of the four beakers were spiked with a known amount of a
standard mixture of HAAS (Appendix I). This spiking procedure facilitated peak
identification during HPLC analysis, and allowed calculation of percent recovery
for each HAA. The rest of the extraction procedure followed the method developed
by Gross and Gruter (1992) and modified by Balogh (1995). Details of the
procedure can be obtained flom Balogh (1995).

37

Briefly, the extraction procedure was performed as follows. The
diatomaceous earth was mixed with the sample, and the mixture was packed into
the Extrelut 20 columns (20 g diatomaceous earth plus sample in each column).
Hydromatrix brand diatomaceous earth required the operator to apply pressure in
order to pack 20 g into a column; Extrelut brand did not. The columns were placed
under a ventilation hood, and the HAAS were eluted with dichloromethane/toluene
(95:5). PRS cartridges, which had been preconditioned with the same solvent
mixture and had needles attached, were coupled to the Extrelut columns. Solvent
was added periodically to keep the columns full until 45 ml had flowed through the
assembly. The PRS cartridges, which retained the HAAS, were removed and dried
under vacuum for 10-15 minutes. The PRS cartridges were connected to a
peristaltic pump and rinsed consecutively at 2 ml/minute with 6 ml of 0.1 M
hydrochloric acid, 15 ml methanol:0.l M hydrochloric acid (4:6), and 2-3 ml
water. C18 cartridges, which had been conditioned with 2 ml methanol and 2 ml
water, were connected to the PRS cartridges, and 20 ml 0.5 M ammonium acetate
buffer (pH=8.0) was pumped through the assembly to transfer the HAAS to the
C18 columns. The C18 cartridges were then rinsed and dried under vacuum for 30
minutes. The HAAs were eluted into microvials using 0.8 ml methanolzammonia

(9: 1), and evaporated under nitrogen for temporary storage at —20°C.

Egh Performgnce Liquid Chromatgggphy
Separation of HAAs

The dry sample extracts were redissolved in 50 ul of methanol containing 50
ng/ul caffeine (experiment #1) or TriMeIQx (experiments #2 & 3). Samples were
injected into the HPLC in aliquots of 20 ul. The first mobile phase was filtered,
degassed 0.1 M triethylarnine, adjusted to pH 3.2 with either phosphoric acid or
hydrochloric acid. The second mobile phase was filtered, degassed acetonitrile.

One of two reversed phase silica columns was used to separate the HAAS. The first

38
experiment utilized a column flom Toso Haas (TSK-Gel ODS-80TM column with

5 pm particle size, 4.6 mm ID x 25 cm; Toso Haas, Montgomeryville, PA), and the
second and third experiments utilized a Waters Symmetry C8 column (3.9 mm ID
x 15 cm, Waters/Millipore, Milford, MA). A precolumn (Supelguard LC-8-DB,
Supelco, Bellefonte, PA) was attached between the injector port and column to
filter out unwanted compounds, and its cartridge was replaced approximately every

60 injections.

The flow rate of the mobile phase was 1 ml/min. For the first and second
experiments, the initial ratio of acetonitrile:buffer in the mobile phase was 8:92,
which increased to 17:83 during the first 10 minutes. Acetonitrile concentration
continued to increase until the ratio was 25:75 in another next 10 minutes, then
55:45 in the next 10 minutes. This allowed elution of all heterocyclic aromatic
amines flom the column. Over the next five minutes, the acetonitrile:buffer ratio
increased to 80:20 to facilitate elution of other compounds. After 35 minutes, the
ratio returned to its original 8:92 for 10 minutes to allow the column to re-
equilibrate before the next injection. The third experiment utilized a Slightly
different gradient based on observations of the Waters column during experiment
#2. The original ratio of acetonitrile:buffer was again 8:92, but changed to 20:80
over the first 15 minutes, then to 80:20 over the next three minutes. Cleaning of the
column was accomplished with an 80:20 ratio for 8 minutes, then the gradient
returned to its original 8:92 ratio for the last 9 minutes. This gradient allowed
adequate separation of HAAs while decreasing each run time to 35 minutes instead

of 45 minutes.

Detection of HAAs
Detection of the HAAS was performed by a photodiode array UV detector.
The IQ standards were detected at 254 nm, IQx compounds at 262 nm, and PhIP at

316 nm PhIP was also detected by a fluorescence detector with excitation set at

39

330 nm and emission at 375 nm. The software package used for peak observation
and integration was the Millennium 2000 Chromatography Manager (Millipore,
Milford, MA).

Mass Spectrometry

Fractions flom the glycine system were collected flom the HPLC for analysis
by mass spectrometry at the Michigan State University-NIH Mass Spectrometry
Facility. The samples were found to be contaminated by phosphate and potassium,
which blocked the spectrophotometric Signal. Multiple attempts were made to
clean the HPLC and collect flactions for analysis, but the contamination was still
present. An extracted sample that had not been flactionated on the HPLC was
analyzed at the Mass Spectrometry facility by direct probe fast atom bombardment
(FAB). This sample was shown to contain a compound of molecular weight 199,
corresponding to IQx, a compound of molecular weight 214, corresponding to
MeIQx , and a compound of molecular weight 228, corresponding to DiMeIQx. As
firrther analysis was not possible due to the impurity of the sample, it is not known
if this was 4,8-DiMeIQx or 7,8-DiMeIQx. A compound with molecular weight 242
was also found, corresponding to TriMeIQx. It is not known if this last compound
was actually TriMeIQx or another substance with the same molecular weight, as

TriMeIQx was not identified by retention time during I-IPLC of samples.

Qu_antitation of HAAS

Before HPLC separation of samples for each experiment, four aliquots of the
standard mixture of HAAS, four aliquots of internal standard, and four aliquots of
FEMA (10, 15, 20, and 25 ul) were injected. Linear regression (ng compound
versus peak area) was performed for each HAA in each mixture. A correlation

coefficient of 0.99 or greater was considered acceptable for FEMA or internal

40
standard, and 0.97 or greater was acceptable for the laboratory standard mixture of

HAAS.

Each peak area corresponding to an HAA was first corrected for expected
area of the internal standard based on the internal standard regression line. The
standard addition method of Gross and Gruter (1992) was then used for
determining extraction efficiency and for quantitation of HAAS flom the samples.
Each data point consisted of four subsamples; two spiked and two unspiked. The
average area of the Spiked samples minus the average of the unspiked samples
allowed comparison with the regression line for the standard mixture. Each data
point as determined below was then corrected for its individual extraction
efficiency, or percent yield.

Nanograms of each HAA formed were determined using the average of the
two unspiked subsamples. The linear regression slope for FEMA was used to
determine the exact amount of each HAA formed in each sample. FEMA was
employed for this calculation rather than the standard mixture, as the concentration
of each HAA in FEMA was known to 1/ 100 of a nanogram. The exception to this
rule was for the compound IQx. This HAA was not present in FEMA, so the

standard mixture regression line was used to calculate formation of IQx.

Statistical Analysis
Stuay #1: Eflect of T rilinolein versus T ristearin on HAA Formation in Aqueous

Phenylalanine and Glycine Model Systems

Statistical analysis was accomplished using the SPSS statistical software
package (student version 6.1 for Windows, 1994, Prentice Hall, New Jersey). One-
way analysis of variance (AN OVA) was performed for each heterocyclic amine.
Prior to this analysis, a log transformation was performed on the data. This

transformation was necessary due to a positive correlation between ng HAAS

41

formed and standard deviations. Student-Newman-Keuls post-hoe tests were

performed.

Stuay #2: Eflect of Com Oil and Olive Oil on HAA Formation in an Aqueous

Phenylalanine Model System
Statistical analysis was again performed with SPSS software. One way

ANOVAS were used to analyze differences in HAA formation between treatments
for each HAA. No post-hoe tests were necessary. Hydromatrix replications were
not included in any statistical treatment. Extraction efficiencies were virtually

identical in Hydromatrix and Extrelut samples.

Study #3 : Effect of Ferrous Sulfate and Myoglobin on HAA Formation in a
Glycine Model System with and without Lipid

Statistical analysis was again accomplished with one way AN OVAs using the
SPSS statistical software. Student-Newman-Keuls post-hoe tests were performed

with the same software.

RESULTS AND DISCUSSION

Study #1: Effect of Trilinolein on HAA Formation in Aqueous Phenylalanine and

Glycine Model Systems

Johansson et al. (1993) reported a large increase in MeIQx formation in an
aqueous model system of creatinine, glucose, and glycine with the addition of 20%
corn oil to the system. This observation is significant in light of the results of many
meat studies indicating increasing mutagenicity of meat with increasing fat content
up to approximately the same level of fat. The difficulty with investigations into
the effect of fat on HAA formation has been the inability to decouple the physical
effect of fat flom possible chemical effects. An aqueous model system heated in
very small test tubes made of material with excellent heat conductivity essentially
eliminates the physical effect of fat (increased heat transfer). The model system of
Johansson et al. (1993) appears to effectively decouple these two factors and
provide evidence of a large increase in HAA formation due to the chemical
interaction of fat within the Maillard reaction.

Many questions are raised by these results. It is not known whether HAAS
other than MeIQx increase in the presence of fat. There is no information on how
different types of fats of varying degrees of unsaturation affect HAA formation.
Perhaps the most important note of caution is that a rigorous statistical evaluation
of the data was lacking in the study of Johansson et al. (1993). As has been shown
by several investigators, data on HAA formation in meat and in model systems
have inherently large variability. This variability can be problematic if adequate

replications are not performed (Balogh, 1995).

42

43

Based in part on some of the questions discussed above, this study was
designed utilizing a model system similar to that developed by Johansson et al.
(1993). The objective of this study was to elucidate the chemical efl'ect of a
completely saturated fat (tristearin) and an unsaturated fat (trilinolein) on the
formation of several HAAS in the model system.

The ratios of reactants for this study and the time and temperature of heating
were selected to resemble those used by Johansson et al. (1993) while taking into
account the small Size of the test tubes. Glycine was chosen as the amino acid for
system #1 in order to compare results with those of Johansson et al. (1993).
Phenylalanine was selected for system #2 to maximize formation of PhIP, the
dominant HAA in meat systems. Lipid was added at 15% by weight rather than
20% because of literature evidence indicating that the greatest concentrations of
HAAS are formed at this fat level in ground beef patties, and because of the size
limitations of the test tubes. The amount of standard used to spike half the samples
(1000 ng of each HAA) was based on the expected amount of formation of HAAS.
The average amounts of each HAA formed for specific treatments with the
standard deviations, as well as the mean recovery are shown in Table 4.

The dominant HAAS formed in the glycine system were IQx and MeIQx.
PhIP, IQx, and MeIQx were the major HAAs formed in the phenylalanine system.
Small amounts of DiMeIQx were formed in both systems, but DiMeIQx was not
present in every sample, and therefore was not quantitated. Recoveries ranged
flom a low of 33% to 100% (Table 4). Larger amounts of HAAS were formed in
these systems than anticipated, which may have caused mean recoveries to be
slightly lower than actually calculated. This effect is due to difficulty in accurately
determining recovery when the amount of HAAS added to the spiked samples is
not greater than the amount of HAAS formed in these samples.

Statistical analysis of the data flom this study revealed a significant increase
(p<0.05) in MeIQx formation when trilinolein was added to the glycine system.

44

The amount of MeIQx formed when trilinolein was present was also significantly
greater than the amount formed with tristearin present. A similar profile was seen
with IQx formation in the glycine system. The addition of trilinolein Significantly
increased formation of IQx above that in the control and in the samples containing
tristearin. PhIP was not expected to be formed flom glycine in this system and was

not observed.

Table 4. Effect of trilinolein on HAA formation in aqueous phenylalanine and

 

 

 

 

glycine model systemsl’2’3’

System IQx MeIQx PhIP
Gly/Control 5.09:0.16’ 10.40:i:2.12° not formed
Gly/Tristearin 43310.52a 10.63i2.21" not formed
Gly/Trilinolein 6.30:0.32" 159042442b not formed
Phe/Control 2.59:0.55 4.67:1.55’ 44.90i4.76
Phe/Tristearin 27210.40 67211.45" 39.4:1757
Phe/Trilinolein 28010.13 60710.82” 24.6i11.90
Mean Recovery 87.8%* 87.8% 71.3%

 

 

IValues for each HAA are expressed in nmol/mmol creatinine used.

2Each value is the mean of five analyses i standard deviation.

3 Means in columns with different superscripts are significantly different flom each
other within each amino acid system at p<0.05.

*Recovery for IQx was calculated based on MeIQx recovery, as the standard did

not include IQx.

45
In contrast to the glycine system, the presence of lipids and their composition

did not change the amount of IQx formed in the phenylalanine system. MeIQx was
formed in somewhat greater quantity when tristearin was present compared to the
lipid-flee control. Trilinolein seemed to have no effect on MeIQx in this system.
PhIP formation appears to decrease dramatically in the presence of trilinolein
versus tristearin or no fat. However, this effect was statistically significant only at
the p<0.26 level, due to the large standard deviations for PhIP. Figures 3 and 4
depict the changes in HAA yields in both amino acid systems graphically.

The increase in formation of quinoxaline compounds in the glycine system
upon addition of an unsaturated lipid agrees with the data of Johansson ct al.
(1993). These results indicate a chemical role of fat in HAA formation under
specific conditions. The fact that tristearin did not increase HAA formation in this
system suggests that a certain degree of unsaturation is necessary for fat to
participate in reactions leading to HAAS. The role of unsaturated lipids may be to
increase the amount of intermediate pyrazine and pyridine compounds formed via
the Maillard reaction that lead to HAA formation. This theory was proposed by
Barnes et al. (1983) and Jagerstad et al. (1983a). Free radical reactions cause
acceleration of lipid oxidation. Thus, greater concentrations of lipid oxidation
products such as aldehydes may occur in the system, leading to increased HAA
formation (Parihar et al., 1981; Namiki and Hayashi, 1983).

The apparent decrease in PhIP formation in the phenylalanine system due to
the presence of trilinolein is difficult to explain. This decrease in PhIP formation,
while large, was not statistically significant and likely occurred simply by chance.
However, a similar result was achieved by Faulkner (1994) in a phenylalanine
model system when an unsaturated lipid was added. Assuming for the sake of
discussion that this trend was significant, explanations for this result are purely
speculative. There was no significant increase in any other HAA in this system

when trilinolein was added, so it is not the case that reactants were diverted to

46
production 'of other HAAS. It is possible that reactants in the presence of an

unsaturated lipid preferentially produced other compounds, such as melanoidins.
Another possibility is that, as mentioned above, lipid oxidation leads to increased
amounts of aldehydes and other compounds, but these compounds are not
precursors for PhIP. The effect of unsaturated lipids on PhIP formation Should be a

productive area for further research.

47

 

nmol/mmol CR

 

GTS

GTL

 

Model System/Lipid PTL

Figure 3. Effect of tristearin versus trilinolein on HAA formation in glycine and

phenylalanine model systems.

GMS = glycine control plus monostearin, GTS = glycine system plus tristearin,
GTL = glycine system plus trilinolein, PMS = phenylalanine system plus
monostearin, PTS = phenylalanine system plus tristearin, PTL = phenylalanine
system plus trilinolein.

48

 

 

nmollmmol CR

 

PMS PTS

PTL

Lipid

Figure 4. Effect of tristearin versus trilinolein on HAA formation in a

phenylalanine model system.

PMS = phenylalanine system plus monostearin, PTS = phenylalanine system plus
tristearin, PTL = phenylalanine system plus trilinolein.

49
Study #2: Effect of Com Oil am Olive Oil on HAA Formation in an Aqueous

Phenylalinine Model System

The results of the first study confirmed a chemical role for unsaturated fat in
the formation of HAAS in an aqueous Maillard reaction system. The unsaturated
fat in that study was trilinolein, which contains six double bonds per triglyceride
molecule. This is a greater degree of unsaturation than any cooking oil would
contain. It is unknown how differences in the degree of unsaturation of actual
cooking oils affect HAA formation. Based on information obtained flom Sigma
Chemical Company technical services, and the storage of the trilinolein under
argon gas in its container at -20°C, oxidation should have been less than that of
typical cooking oils. The trilinolein contained no antioxidants. Vegetable oils such
as com, soybean, and olive oil contain some naturally occurring antioxidants, and
generally are oxidized to some degree before use.

Therefore, a second study was designed to address practical aspects of how
the degree of unsaturation in vegetable oils affects HAA formation. The objective
of this study was to investigate the effect of corn oil and olive oil on the formation
of HAAS in the phenylalanine model system.

The major HAAs detected in the phenylalanine model system were IQx,
MeIQx, sometimes DiMeIQx, and PhIP. This observation is the same as for study
#1. Extraction efficiencies were similar to those of the previous study as well,
except for PhIP in the presence of tristearin. This atypically low average recovery
of PhIP was due to extraction efficiencies ranging flom 16-36% in one treatment.
It was proposed that the tristearin complexed with the large quantity of PhIP
present in the standard for this study (2500ng/sample) and made it unavailable for
extraction. Therefore the average recoveries flom the first study were used to

calculate this treatment (Table 4).

50

Table 5. Effect of corn oil and olive oil on HAA formation in a phenylalanine

 

 

 

model system.1’2‘3

System IQx MeIQx PhIP
Phe/Control 3.26i1.04 7.54i2.74 36.75i16.80
Phe/Olive Oil 2.80:0.42 6.06:1:1.99 40.93i12.01
Phe/Com Oil 2.94:0.49 7.34:1:1 .13 39.41il 3.89
Mean Recovery 81 .77% 93 . 10% 43 .87%

 

 

1Values are expressed in nmol HAA formed/mmol creatinine used.
2Each value is the mean of seven replications i standard deviation.

3No significant differences were found at p<0.05 level.

Statistical analysis revealed no Significant differences between any of the
treatments in this study. Absolute values and standard deviations were very similar
to those obtained flom the phe/control and phe/tristearin treatments in study #1.

Although the results of this study did not corroborate the results obtained
with pure unsaturated lipids in study #1, the lack of differences between treatments
gives some important information. First, the difference in the degree of
unsaturation of olive oil and corn oil is not adequate to influence the formation of
HAAS that require phenylalanine as a precursor. Second, it also appears that, at
least for PhIP, the degree of unsaturation in the range of zero double
bonds/triglyceride (tristearin) to 3.77 double bonds/triglyceride (corn oil) makes no
difference in the amount of HAA formation. In comparing data between studies #1
and 2, it also appears that the presence of 15% fat does not affect PhIP formation
as long as the degree of unsaturation is S that of corn oil. Here study #2 confirms
the statistical analysis of study #1, showing unsaturated oils to have no significant

effect on PhIP formation.

51
Several investigators have speculated that naturally occurring antioxidants in

vegetable oils might inhibit HAA formation (Johansson et al., 1993). The corn oil
used in this experiment contained approximately 0.012% mixed tocopherols per
the manufacturer’s analysis. No information was available for the olive oil. The
presence of naturally occurring antioxidants did not appear to have any effect on
HAA formation in this model system. Because antioxidant concentration is so
crucial to its effectiveness (Balogh, 1995), it is difficult to say if these vegetable

oils would have any eflect in a meat system.

Study #3: Effect of Ferrous Sulfate and Mygglobin on HAA Formation in a
Glycine Model System.

The third study was designed to investigate the effect of iron on HAA
formation in the glycine model system. HAA yields in this system are known to be
affected by the addition of highly unsaturated lipids (Table 3; Johansson et al.,
1993). If oxidation of lipids plays a role in increasing intermediate compounds
necessary for HAA formation, addition of a prooxidant should increase the rate of
this oxidation and possibly further increase HAA formation. Iron, in the form of
heme compounds as well as flee iron, can facilitate the decomposition of lipid
hydroperoxides and speed the formation of hydroxyl radicals (Tappel, 1962;
Halliwell and Gutteridge, 1986). Johansson and Jagerstad (1993) found that the
addition of ferrous sulfate to an aqueous model system significantly increased
MeIQx formation without the presence of fat and also with the addition of 40% fat.

The objectives for this phase of the study were to determine the effect of flee
iron (ferrous sulfate) and myoglobin on HAA formation both in the absence and
presence of an unsaturated lipid. Ferrous sulfate was chosen to allow comparison
of data with the study of Johansson and J agerstad (1993). Myoglobin was selected
as a representative heme compound in meat systems. The latter group of

compounds are effective catalysts of lipid oxidation (Silberstein and Lillard, 1978).

 

52

Table 6. Effect of ferrous sulfate and myoglobin on HAA formation in a glycine

 

 

 

model system with and without trilinolein.1’2’3’4
System IQx MeIQx
Gly/Control 7.39:1 .23‘“b 15.48:1.34°
Gly/Fe 8.49:2.08° 26.29:2.9ob
Gly/Trilinolein 5.94:1 .85b 1579:039a
Gly/Trilinolein/Fe 5.71:1 .69b 19.98i6.83"
Gly/Mb 2.30:0.74“c 1518:245'
Gly/Trilinolein/Mb 1.83:t0.40b’° 16.27i4.49°
613713; """""" 4'. 67:85 ''''' REES-.1" s” ""
Gly/Trilinolein/Nz 5.04:1 . 17 17.07:2.93
Mean Recovery 63.50% 64.85%

 

1Values are expressed in nmol HAA formed/mmol creatinine used.

2All treatments above the dotted line were replicated five times. Treatments below
the dotted line were replicated three times and were not subjected to statistical
analysis.

3Means in the same column with different superscripts are significantly different
flom each other at p<0.05.

4Fe denotes ferrous sulfate, Mb denotes myoglobin, N2 denotes nitrogen gas.

53

nmollmmol CR

 

2‘
(D

Gly/Fe
Gly/T L
Gly/'1' L/Fe
Gly/Mb
Gly/1' UMb

System

Figure 5. Effect of ferrous sulfate (Fe) or myoglobin (Mb) on the formation of
HAAS in a glycine model system with and without trilinolein (TL).

Gly = glycine system control, Gly/Fe = glycine system plus ferrous sulfate, Gly/TL
= glycine system plus trilinolein, Gly/TL/Fe = glycine plus trilinolein plus ferrous
sulfate, Gly/Mb = glycine system plus myoglobin, Gly/TL/Mb = glycine system
plus trinlinolein, plus myoglobin.

54
A one-way AN OVA with Student-Newman-Keuls post-hoe testing was

performed for each HAA, except DiMeIQx, whose presence was not detected in all
samples. This analysis revealed that the glycine system with ferrous sulfate added
(no lipid) produced significantly greater (p<0.05) amounts of IQx and MeIQx than
did almost any other treatment (Table 6). When myoglobin was added to the
system, MeIQx production remained the same as the control, while IQx formation
actually decreased Significantly. The presence of trilinolein at 15% by weight made
no significant differences in any treatments. The system containing trilinolein plus
ferrous sulfate did form more MeIQx than the system containing trilinolein and no
iron, but the difference did not attain statistical significance. A notable difference
between the results of this study and the first study is that the control of this study
formed 50% more MeIQx than the control in study #1. The cause of this difference
is unknown, but a Similar variance was noted between the model system results
obtained by Johansson and Jagerstad (1993) and Johansson et al. (1993). Data
flom Table 6 are depicted in graph format in Figure 5.

The results of this study are consistent with those of Johansson and
Jagerstad (1993). They reported increases in MeIQx formation after 30 minutes of
heating when iron was added to the system without lipid, and when iron was added
to a system containing 40% lipid (not 20%). Several possible explanations exist for
these data. First, it appears that iron may have a role in the Maillard reaction as an
initiator of flee radical reactions that do not involve lipids. Results of this study,
along with the data flom antioxidant treatments of meat which suppress HAA
formation (Balogh, 1995), support the theory of a flee radical component of HAA
formation. Second, decomposition of low and moderate levels of lipids (IS-20%)
in the presence of iron may not produce a sufficient increase in HAA precursors
such as pyrazines or pyridines to make a difference in HAA yield.

The apparent lack of involvement of lipid oxidation in this study has several

implications. First, it may be that rate of HAA formation is crucial. Iron may be

 

55

hastening lipid breakdown and thus formation of HAA precursors, but the reaction
may reach steady state before the heating is finished, so the final yield is no
different than without iron. Second, it may be that the role of lipid is not due to
oxidation products at all, but rather that lipid is involved in the Maillard reaction in
another, as yet undetermined, mode.

The decrease in IQx in the myoglobin treatments is likely due to high
viscosity and the inability of the reactants to reach each other. Myoglobin
denatures at relatively low temperatures, and creates a network of thick strands.
When these test tubes were opened after heating, the contents were solidified flom
top to bottom, due to the high reaction temperature. MeIQx formation was
unaffected by the increase in viscosity. It is likely that MeIQx formation occurs
prior to or at a faster rate than that of IQx during heating, and therefore MeIQx was
produced before being affected by the increased viscosity. It is unknown if the
same effect would function in meat systems to keep the reactants physically
separated.

Although the data flom the test tubes heated in a nitrogen atmosphere were
not subjected to a statistical analysis, it appears that the lack of available oxygen
did not affect the HAA formation in these samples. The same species and
approximately the same yield of HAAS were produced as in many of the other
samples. Either enough oxygen was still present for any reactions requiring this

compound, or the reactions leading to HAA formation are not oxygen dependent.

Summary and Conclusions

Results of these three studies taken together Show an increase in formation
of specific quinoxaline-type HAAS formed flom one amino acid in the presence of
either trilinolein or ferrous sulfate. Through rigorous statistical treatment of the
data, several lipids were shown to have no effect on HAA formation in the two
model systems utilized (tristearin, corn oil, olive oil).

The three studies discussed here indicate that highly unsaturated lipids may
have a small chemical eflect on HAA formation, but it is likely that the major
result of fat on HAA production is the physical efl‘ect of increased heat transfer. It
appears that the chemical effect of lipids may differ depending upon the precursors
present, as HAA formation increased in the glycine system and apparently
decreased in the phenylalanine system. Iron in the form of ferrous sulfate (Fey)
increased HAA formation in the model system containing no lipid, but not in the
system containing lipid. This result points away flom lipid oxidation products
playing a major role in HAA formation during the Maillard reaction, and toward a
flee radical mechanism. Results indicating inhibition of the majority of HAA
formation by phenolic antioxidants in flied ground beef support the involvement of

a flee radical mechanism in their formation (Balogh, 1995).

56

 

Future Research

The research discussed here indicates that trilinolein has a chemical effect
on HAA formation when glycine or phenylalanine is the amino acid precursor.
Other model system studies need to be performed to determine whether actual
cooking lipids have an effect on HAA formation. If a chemical effect does exist for
any of these fats, it should be determined for as many amino acids as possible, as
the effect may vary greatly among the amino acid precursors.

Even more fundamental than the issue of a lipid effect on HAA formation is
the actual mechanism of formation itself. Each study examining the result of a
specific component on HAA formation adds more information regarding the
possible mechanism. However, a series of several studies could be designed to
give more definitive information. Continuous or intermittant monitoring of a
model system reaction containing lipid through electron spin resonance (ESR)
could indicate at what point flee radicals are present. Thermal gravidirnetric
analysis of the lipids could pinpoint the exact time/temperature of lipid oxidation,
and combined with ESR data, determine whether lipid oxidation products are
involved in HAA formation. Another study which could provide valuable
information regarding reaction intermediates would be detection and separation of
aldehydes, pyrazines, and pyridines by gas chromatography. Aliquots of the
reaction mixture could be analyzed at intervals to follow the reaction sequence as it
proceeded.

Other areas of research which can provide information about the reaction
mechanism of HAA formation are labeled-carbon experiments as performed by

Skog and Jagerstad (1993) and Felton et al. (1990). Labeling of probable reaction

57

58

intermediates such as aldehydes, pyrazines, and pyridines would contribute the
most knowledge to the field, as it is already known that creatinine, amino acids,
and in some instances, sugars are HAA precursors. The ongoing studies of
inhibition of HAA formation also help our understanding of the mechanism of
formation. If we know what substances inhibit or stop HAA formation, we can
make inferences regarding the pathways by which they are produced. Such
inhibition studies are best performed on actual meats rather than in model systems.
Most of the compounds which inhibit HAA formation are also antioxidants, such
as vitamin E, tea phenolics, and other plant extracts, and their efficacy is highly
concentration dependent. The concentrations of reactants in most model systems is
necessarily quite different flom those in meat due to lack of the meat matrix and
absence of the many other substances which do not contribute to HAA formation.
Inhibition data obtained using model systems would not be applicable to meat
products. A useful tool would be the availability of a probe, perhaps similar to the
enzyme-linked-immunosorbent assays (ELISA) now used for toxin detection,
which would allow rapid determination of the presence of HAAS. Such a tool
could quickly assess the need for or effectiveness of an inhibitory compound.

Finally, the continuing studies to quantitate the known HAAS in various
meat products and isolate new HAAS are very important. AS stated earlier, much of
the work on HAAS formed in foods was performed before an accurate quantitative
method was available (Gross and Gruter, 1992). Values reported in the literature
vary widely, and it is important to know where the HAAS are found in the food

supply if attempts are to be made to avoid them or decrease their production.

APPENDIX

APPENDIX

Study 1
HAA standard used for study #1 consisted of:

IQ (20 rig/141)

MeIQ (20 ng/ul)

MeIQx (20 ng/ul)

4,8-DiMeIQx (20 ng/ul)

PhIP (20 ng/ul)
Samples (two of every four) were spiked with 50 ul of this standard, which
delivered 1000 ng of each HAA to the sample.

Study 2
HAA standard used for study #2 consisted of:

IQx (20 ng/ul)

MeIQx (30 ng/ul)

4,8-DiMeIQx (20 ng/ul)

PhIP (50 ng/ul)
Two of every four samples were spiked with 50 ul of this standard, which
delivered 1000 ng IQx, 1000 ng DiMeIQx, 1500 ng MeIQx, and 2500 ng PhIP to

each sample.

59

 

60

Study 3 .
HAA standard used for study #3 consisted of:

IQx (20 ng/ul)

MeIQx (30 ng/ul)

4,8-DiMeIQx (20 ng/ 111)
Two of every four samples were spiked with 50 ul of this standard, which
delivered 1000 ng IQx, 1000 ng DiMeIQx, and 1500 ng MeIQx to each sample.

 

REFERENCES

 

REFERENCES

Adamson, R. H., Thorgeirsson, U. P., Snyderwine, E. G., Thorgeirsson, S. S.,
Reeves, J., Dalgard, D. W., Takayama, S., and Sugimura, T. 1990.
Carcinogenicity of 2-amino-3-methylimidazo[4,5j]quinoline in nonhuman
primates: induction of tumors in three macaques. Jpn. J. Cancer Res. 81:10-
14.

Alavanja, M. C., Brownson, R. C., and Benichou, J. 1996. Estimating the effect of
dietary fat on the risk of lung cancer in nonsmoking women. Lung Cancer

14(Suppl 1):63-74.

Alldrick, A. J ., Rowland, 1. R., and Wise, A. 1987. The hepatic conversion of some
heterocyclic amines to bacterial mutagens is modified by dietary fat and
cholesterol. Mutagenesis. 2(3):221-224.

Ames, B. N., McCann, J., and Yarnasaki, E. 1975. Methods for detecting
carcinogens and mutagens with Salmonella/mammalian-microsome
mutagenicity test. Mutat. Res. 31 :347-364.

Amoldi, A., Amoldi, C., Baldi, O., and Ghizzoni, C. 1990. Effect of lipids in the
Maillard reaction. In The Maillard Reaction. Advances in Life Sciences. Eds.,
P.A. Finot, H.U. Aeschbacher, R.F. Hurrell, and R. Liardon. Pp.133-138.
Birkhauser Verlag, Basel.

Ashoor, S. H., Dietrich, R. A., Chu, F. S., and Pariza, M. W. 1980. Proline
enhances mutagen formation in ground beef during flying. Life Sci. 26:1801-
1805.

Balogh, Z. 1995. Formation and inhibition of heterocyclic aromatic amines in flied
ground beef patties. MS. Thesis. Michigan State University, East Lansing,
MI, USA.

Barnes, W. S., Maher, J.C., and Weisburger, J. H. 1983. High-pressure liquid
chromatographic method for the analysis of 2-amino-3-methyl[4,5-f]imidazo-
quinoline, a mutagen formed during the cooking of food. J. Agric. Food Chem.
31:883-886.

61

 

62

Barnes, W. S., and Weisburger, J. H. 1983. Lipid content and mutagen formation
in the cooking of beef. Proceedings of AA CR, p. 95.

Barnes, W. S., and Weisburger, J. H. 1984. Formation and inhibition of mutagens
during flying of beef and relationship to fat content. Proceedings of AA CR, p.
102.

Barrington, P. J., Baker, R. S. U., Truswell, A. S., Bonin, A. M., Ryan, A. J., and
Paulin, A. P. 1990. Mutagenicity of basic flactions derived flom lamb and
beef cooked by common household methods. Food Chem. T oxicol. 28(3): 141-

146.

Becher, G., Knize, M. G., and Felton J. S. 1989. Identification and synthesis of
new mutagens flom a flied Norwegian product Var F oda 42(2):85-90.

Becher, G., Knize, M. G., Nes, I. F. and Felton, J. S. 1988. Isolation and
identification of mutagens flom a flied Norwegian meat product.
Carcinogenesis 9:247-253.

Bjeldanes, L. F ., Morris, M. M., Felton, J. S., Healy, S., Stuermer, D., Berry, P.,
Timourian, H., and Hatch, F. T. 1982a. Mutagens flom the cooking of food. II.
Survey by Ames/Salmonella test of mutagen formation in the major protein-
rich foods of the American diet. Food Chem. T oxicol. 20:357-363.

Bjeldanes, L. F., Morris, M. M., Felton, J. S., Healy, S., Stuermer, D., Berry, P.,
Timourian, H., and Hatch, F. T. 1982b. Mutagens flom the cooking of food. 11.
Survey by Ames/Salmonella test of mutagen formation in secondary sources
of cooked dietary protein. Food Chem. T oxicol. 20:365-369.

Bjeldanes, L. F ., Morris, M. M., Timourian, H., and Hatch, F. T. 1983. Effects of
meat composition and cooking conditions on mutagen formation in flied
ground beef. J. Agric. Food Chem. 31:18-21.

Brennan-Craddock, W. E., Coutts, T. M., Rowland, LR, and Alldrick, A. J. 1990.
Dietary fat modifies the in vivo mutagenicity of some food-bome carcinogens.
Mutat. Res. 230(1):49-54.

Buttery, R. G., Ling, L. C., Teranishi, R., and Mon, T. R. 1977. Roasted lamb fat:
basic volatile components. J. Agric. Food Chem. 25: 1227-1229.

63

Chen, Y., Zheng, R., J ia, Z., and Ju, T. 1990. Flavonoids as superoxide scavengers
and antioxidants. Free Radical Bio. Med. 9:19-21.

Commoner, B., Vithayathil, A., and Dolara, P. 1978. Mutagenic analysis as a
means of detecting carcinogens in foods. J. Food Protect. 41(12):996-1003.

Davis, C. D., Schut, H. A. J., Adamson, R. H., Thorgeirsson, U.P., Thorgeirsson,
SS, and Snyderwine, E. G. 1993. Mutagenic activation of IQ, PhIP and
MeIQx by hepatic microsomes flom rat, monkey and man:low mutagenic
activation of MeIQx in cynomolgus monkeys in vitro reflects low DNA adduct
levels in vivo. Carcinogenesis l4 (1):61-65.

Dolara, P., Commoner, B., Vithayathil, A., Cuca, G., Tuley, B., Madyastha, P.,
Nair, S., and Kriebel, D. 1979. The effect of temperature on the formation of
mutagens in heated beef stock and cooked ground beef. Mutat. Res. 60:231-

237.

Esumi, H., Ohgaki, H., Kohzen, B., Takayama, S., and Sugimura, T. 1989.
Induction of lymphoma in CDFl mice by the food mutagen, 2-arnino-l-
methyl-6-phenyl-imidazo[4,5-b]pyridine. Jpn. J. Cancer Res. 80:1 176-1 178.

Farmer, L. J., and Mottram, D. S. 1990. Interaction of lipid in the Maillard reaction
between cysteine and ribose: the effect of a triglyceride and three
phospholipids on the volatile products. J. Sci. Food Agric. 53:505-525.

Faulkner, J. 1994. Mechanism of heterocyclic amine formation in flied ground
beef-the role of oxidized lipid and the Maillard reaction. Ph.D Thesis.
Michigan State University, East Lansing, MI, USA.

Felton, J. S., Fultz, E., Dolbeare, F. A., and Knize, M. G. 1994. Reduction of
heterocyclic aromatic amine mutagens/carcinogens in flied patties by
microwave pretreatment. Food Chem. T oxicol. 30(10):897-903.

Felton, J. S., and Knize, M. G. 1990. Heterocyclic amine mutagens/carcinogens in
foods. In Handbook of Experimental Pharmacology, 94/1. Eds. C.S. Cooper
and PL. Grover. pp.471-502. Springer-Verlag, Berlin.

Felton, J. S., Knize, M. G., Roper, M., Fulz, E., Shen, N. H. and Turteltaub, K. W.
1992. Chemical analysis, prevention, and low-level dosirnetry of heterocyclic
amines flom cooked food. Cancer Res. (Suppl) 52:2103S-2107s.

64

Felton, J S., Knize, M. G., Shen, N. H., Lewis, P. R., Andresen, V. D., Happe, J.,
and Hatch, F. T. 1986b. The isolation and identification of a new mutagen
flom flied ground beef: 2-amino-l-methyl-6-phenylimidazo[4,5-b]pyridine
(PhIP). Carcinogenesis 7: 1081 -1086.

Felton, J. S., Knize, M. G., Shen, N. H., Andresen, B. D., Bjeldanes, L. F., and
Hatch, F. T. 1986a. Identification of the mutagens in cooked beef. Environ.
Health Perspect. 67: 1 7-24.

Felton, J. S., Knize, M. G., Wood, C., Wuebbles, B.J., Healy, S. K., Stuermer,
D.H., Bjeldanes, L. F., Kimble, B. J. and Hatch, F. T. 1984. Isolation and
characterization of new mutagens flom flied ground beef. Carcinogenesis
5(1):95-102.

Gerhardsson-de-Verdier, M. Hagman, U., Peters, R. K., Steineck, G., and Overvik,
E. 1991. Meat, cooking methods and colorectal cancer: a case-referent study in
Stockholm. Int. J. Cancer. 49(4):520-525.

Grivas, S., Nyharnmar, T., Olsson, K., and Jagerstad, M. 1985. Formation of a new
mutagenic DiMeIQx compound in a model system by heating creatinine,
alanine and fluctose. Mutat. Res. 151:177-183.

Grivas, S., Nyharnmar, T., Olsson, K., and Jagerstad, M. 1986. Isolation and
identification of the food mutagens IQ and MeIQx flom a heated model
system of creatinine, glycine, and glucose. Food Chem. 20:127-136.

Gross, G. A. 1990. Simple methods for quantifying mutagenic heterocyclic
aromatic amine in food products. Carcinogenesis 11:1597-1603.

Gross, G. A., and Gruter, A. 1992. Quantitation of mutagenic/carcinogenic
heterocyclic amines in food products. J. Chromat. 592:271-278.

Gry, J ., Vahl, M., and Nielsen, P. A. 1986. Mutagens in Fried Meat Pub]. No. 139,
National Food Administration, Copenhagen, Denmark.

Halliwell, B., and Gutteridge, J. M. C. 1986. Iron and flee radical reactions: two
aspects of antioxidant protection. T [BS 1 1 :372-375.

Harrison, J. C., Carvajal, M., and Garner, R. C. 1993. Does aflatoxin exposure in
the United Kingdom constitute a cancer risk? Environ. Health Perspect.
99:99-105.

65

Hayasahi, T., and Namiki, M. 1986. Role of sugar flagrnentation in an early stage
browning of amino-carbonyl reaction of sugar with amino acid. Agric. Biol.
Chem. 50:1965-1970.

Hayatsu, H., Arimotao, S., and Wakabayashi, K. 1991. Methods for separation and
detection of heterocyclic amine. In Mutagens in Food: Detection and
Prevention. Eds., H. Hayatsu. pp.101-112. CRC Press, Boca Raton, FL.

Hodge, J. E. 1953. Chemistry of browning reactions in model systems. J. Agric.
Food Chem. 1:928-943.

IARC. 1993. Monographs on the Evaluation of Carcinogenic Risk to Humans:
Vol. 56. Some Naturally Occurring Substances: Food Items and Constituents
Heterocyclic Aromatic Amines and Mycotoxins. pp. 163-242. International
Agency for Research on Cancer, Lyon.

Ito, N., Hasegawa, R., Sano, M., Tamano, S., Esumi, H., Takayama, S., and
Sugimura, T. 1991. A new colon and mammary carcinogen in cooked food, 2-
amino-l-methyl-6-phenyl-imidazo[4,5-b]pyridine (PhIP). Carcinogenesis.
12:1503-1506.

Jagerstad, M., Laser-Reutersward, A., Oste, R, Dahlqvist, A., Grivas, S., Olsson,
K., and Nyharnmar, T. 1983a. Creatine and Maillard reaction products as
precursors of mutagenic compounds formed in flied beef. In The Maillard
Reaction in Foods and Nutrition ACS Symposium Series 215. Eds. G.R.
Waller, Feather, M.S. pp. 507-519. American Chemical Society, Washington,
DC

Jagerstad, M., Laser-Reutersward, A., Olsson, R., Grivas, S., Nyharnmar, T.,
Olsson,, K., and Dahlqvist, A. 1983b. Creatin(in)e and Maillard reaction

products as precursors of mutagenic compounds: effects of various amino
acids. Food Chem. 12:255-264.

Jagerstad, M., Olsson,. K., Grivas, S., Negishi, C., Wakabayashi, K., Tsuda, M.,
Sato, S., and Sugimura, T. 1984. Formation of 2-amino-3,8-dimethylimidazo-
[4,5-f]quinoxaline in a model system by heating creatinine, glycine, and
glucose. Mutat. Res. 126:239-244.

Johansson, M., and Jagerstad, M. 1993. Influence of oxidized deep-flying fat and
iron on the formation of food mutagens in a model system. Food Chem.
T oxicol. 31(12):971-979.

66

Johansson, M., and Jagerstad, M. 1994. Occurrence of mutagenic/carcinogenic
heterocyclic amines in meat and fish products, including pan residues,
prepared under domestic conditions. Carcinogenesis 15(8): 151 1-15 1 8.

Johansson, M., Skog, K., and Jagerstad, M. 1993. Effects of edible oils and fatty
acids on the formation of mutagenic heterocyclic amines in a model system.
Carcinogenesis 14:89-94.

Jones, R. C., and Weisburger, J. H. 1988. Inhibition of aminoimidazoquinoxaline-
type and aminoimidazol-4-one-type mutagen formation in liquid reflux models
by L-tryptophan and other selected indoles. Jpn. J. Cancer Res. 79:222-230.

Kada, T. 1973. Escherichia coli mutagenicity of furylfuramide. Jpn J Genet.
48:302-305.

Kasai, H., Yamaizumi, Z., Wakabayashi, K., Nagao, M., Sugimura, T., Yokoyama,
S., Miyazawa, T., Spingarn, N. E., Weisburger, J. H., and Nishirnura, S. 1980.
Potent novel mutagens produced by broiling fish under normal conditions.
Proceedings of the Japan Academy 56(B): 278-283.

Kasai, H., Yamaizumi, Z., Shiomi, T., Yokoyama, S., Miyazawa, T., Wakabayashi,
K., Nagao, M., Sugirnua, T., and Nishimura, S. 1981. Structure of a potent
mutagen isolated flom flied beef. Chem. Lett. 4:485-488.

Kato, T., Kikugawa, K., and Hayatsu, H. 1986. Occurrence of the mutagens 2-
amino-3,4,8-trimethylimidazo[4,5-f]quinoxaline (4,8 DiMeIQx) in some
Japanese smoked, dried fish products. J. Agric. Food Chem. 34:810-814.

Kato, R., and Yamazoe, Y. 1987. Metabolic activation and covalent binding to
nucleic acids of carcinogenic heterocyclic amines flom cooked foods and
amino acid pyrolysates. Gann. 78:297-31 1.

Kikugawa, K., and Kato, T. 1987. Formation of mutagens, 2-amino-3,8-
dimethylimidazo[4,5-f]quinoxaline (MeIQx) and 2-amino-3,4,8-trimethy-
limidazo[4,5-f]quinoxaline (4,5-DiMeIQx) in heated fish meats. Mutat. Res.
179:5-14.

Kim, I., Wakabayashi, K., Kurosaka, R., Yamaizumi, Z., Jinno, F., Koyota, S.,
Tada, A., Nukaya, H., Takahashi, M., Sugimura, T., and Nagao, M. 1994.
Isolation and identification of a new mutagen, 2-amino—4-hydroxy-methyl-3,8-
dimethylimidazo[4,5-f]quinoxaline (4-CH20H-8-MeIQx), flom beef extract.
Carcinogenesis l 5( 1 ):2 1 -26.

 

67

Knize, M. G., Andresen, B. D., Healy, S. K., She, H. N., Lewis, PR, Bjeldanes, L.
F ., Hatch, F. T. and Felton, J. S. 1985. Effects of temperature, patty thickness,
and fat content on the production of mutagens in flied ground beef. Food
Chem. T oxicol. 23(12):]035-1040.

Knize, M. G., Cunningham, P. L., Griffin, E. A. Jr., Jones, A. L., and Felton, J. S.
1994a. Characterization of mutagenic activity in cooked-grain-food products.
Food Chem. T oxicol. 32(1):]5-21.

Knize, M G., Dolbeare, F. A., Carroll, K. L., Moore, D. H., and Felton, J. S.
1994b. Effects of cooking time and temperature on the heterocyclic amine
content of flied beef patties. Food Chem. T oxicol. 32:595-603.

Knize, M. G., Hopmans, B., and Happe, J. A. 1991. The identification of a new
heterocyclic amine mutagen flom a heated mixture of creatine, glutamic acid
and glucose. Mutat. Res. 260:313-319.

Knize, M. G., Roper, M., Shen, N. H., and Felton, J. S. 1990. Proposed structures
for an amino-dimethylimidazoflrropyridine mutagen in cooked meats.
Carcinogenesis 11(12):2259-2262.

Knize, M. G., Shen, N. H., and Felton, J. S. 1988. The production of mutagens in
cooked foods. In Proceedings of the Air Pollution Control Association, April
1988, pp. 88-130.

Kurosaka, R., Wakabayashi, K., Ushiyama, I-I., Nukaya, H., Nobuhiko, A.,
Sugirnua, T., and Nagao, M. 1992. Detection of 2-amino-1-methyl-6-(4-
hydroxyphenyl)imidazo[4,5-b]pyridine in broiled beef. Jpn. J. Cancer Res.
83:919-922.

Laser-Reutersward, A., Agerhem, H., Skog, K., Jagerstad, M., Lidstrom, A., and
Sundemar, A. 1989. The relationship between mutagenic activity and flavour
development in flied beef steak. Var F oda 42(suppl.2):65-70.

Laser-Reutersward, A., Skog, K., and Jagerstad, M. 1987a. Mutagenicity of pan-
flied bovine tissues in relation to their content of creatine, creatinine,
monosaccharides, and flee amino acids. Food Chem. T oxicol. 25(10):755-762.

Laser-Reutersward, A., Skog, K., and Jagerstad, M. 1987b. Effects of creatine and
creatinine content on the mutagenic activity of meat extracts, bouillons and
gravies flom different sources. Food Chem. T oxicol. 25(10):747-754.

68

Lee, H., and Tsai, SJ. 1991. Detection of IQ-type mutagens in canned roasted eel.
Fd. Chem. T oxicol. 29:517-529.

Levin, D. E., Hollstein, M., Christman, M. F ., Schwiers, E. A., and Ames, B. N.
1982. A new Salmonella tester strain (TA102) with A:T base pairs at the Site
of mutation detects oxidative mutagens. Proc. Natl. Acad. Sci. (U .S.A.),
79:7445-7449.

Linjinsky, W., and Shubik, P. 1964. Benzo[a]pyrene and other polynuclear
hydrocarbons in charcoal broiled meat. Science 145:53-54.

Lyon, J. L., and Mahoney, A. W. 1988. Fried foods and the risk of colon cancer.
Am. J. Epidemiol. 128:1000-1006.

Manabe, S., Kurihara, N., Wada, O., Tohyama, K., and Aramaki, T. 1992.
Formation of PhIP in a mixture of creatine, phenylalanine and sugar or
aldehyde by aqueous heating. Carcinogenesis 13:827-830.

McManus, M. B., Burgess, W. M., Veronese, M. E., Huggett, A., Quattrochi, L.
C., and Tukey, R. H. 1990. Metabolism of 2-Acetylaminofluorene and
benzo[a]pyrene and activation of food-derived heterocyclic amine mutagens
by human cytochromes P-450. Cancer Res. 50:3367-3376.

Milic, B. L., Djilas, S. M., and Canadanovic-Brunet, J. M. 1993. Synthesis of some
heterocyclic amino-imidazoazarenes. Food Chem. 46:273-276.

Miller, A. J ., and Buchanan, R. L. 1983. Detection of genotoxicity in flied bacon
by the salmonella/mammalian microsome mutagenicity assay. Food Chem.
T oxicol. 21(3):319-323.

Muramatsu, M., and Matsushima, T. 1985. Formation of MeIQx and 4,8-DiMeIQx
by heating mixtures of creatinine, amino acids, and monosaccharides. Mutat.
Res. (Abstract) 147:266-267.

Murray, S., Gooderham, N. J., Boobis, A. R., and Davies, D. S. 1988.
Measurement of MeIQx and DiMeIQx in flied beef by capillary column gas
chromatography electron capture negative ion chemical ionization mass
spectrometry. Carcinogenesis 9:321-325.

69

Nader, C. J ., Spencer, L. K., and Weller, R. A. 1981. Mutagen production during
pan-broiling compared with microwave irradiation of beef. Cancer Lett.

13:147-151.

Nagao, M., Honda, M., Seino, Y., Yahagi, T., and Sugimura, T. 1977.
Mutagenicities of smoke condensates and the charred surface of fish and meat.
Cancer Lett. 2:221-226.

Namiki, M., and Hayashi, T. 1981. Formation of novel flee radical products in an
early stage of Maillard reaction. Progr. Food Nutr. Sci. 5:81-91.

Namiki, M., and Hayashi, T. 1983. A new mechanism of the Maillard reaction
involving sugar flagrnentation and flee radical formation. In The Maillard
Reaction in Foods and Nutrition Eds., G.R. Waller and MS. Feather. pp.21-
46. American Chemical Society.

Negishi, C., Wakabayashi, K., Tsuda, M., Sato, S., Sugimura, T., Saito, H., Maeda,
M., and Jagerstad, M. 1984. Formation of 2-amino-3,7,8-trimethyl[4,5-
flquinoxaline, a new _ mutagen, by heating mixture of creatine, glucose, and
glycine. Mutat. Res. 140:55-59.

Negishi, C., Wakabayashi, K., Yamaizumi, J., Saito, H., Sato, S., and Jagerstad, M.
1985. Identification of 4,8-DiMeIQx, a new mutagen. (Abstract). Mutat. Res.
147:267-268.

Nilsson, L., Overvik, E., Fredholm. L., Levin, O., Nord, CE, and Gustafsson, J.
A. 1986. Influence of flying fat on mutagenic activity in lean pork meat.
Mutat. Res. 171:115-121.

Nyhammar, T. 1986. Studies on the Maillard reaction and its role in the formation
of food mutagens. Doctoral Thesis, Swedish University of Agricultural
Sciences, Uppsala, Sweden. ISBN 91-576-265 8-8.

Ohgaki, H., Takayama, S., and Sugirnua, T. 1991. Carcinogenicities of
heterocyclic amines in cooked food. Mutat. Res. 259:399-410.

Overvik, E., Kleman, M., Berg, 1., and Gustavsson, J. A. 1989. Influence of
creatine, amino acids and water on the formation of the mutagenic
heterocyclic amines found in cooked meat. Carcinogenesis 10:2293-2301.

 

7O

Overvik, E., Nilsson, L., Redholm L., Levin, O., Nord, C. E., and Gustafsson, J. A.
1987. Mutagenicity of pan residues and gravy flom flied meat. Mutat. Res.
187 :47-53.

Parihar, D. B., Vasundhara, T. S., and Vijayayaghavan, P. K. 1981. Effect of metal
ions such as iron and copper on the formation of nitrogen heterocyclic
compounds in sesame oil-a-amino acids model systems under flying
conditions. J. Food Tech. 16:313-319.

Pischetsrieder, M., and Severin, T. 1994. The Maillard reaction of disaccharides.
In Maillard Reactions in Chemistry, Food, and Health. Eds., T.P. Labuza,
G.A. Reineccius, V.M. Monnier, J. O’Brien, and J.W. Baynes. pp.37-42. The
Royal Society of Chemistry, Thomas Graham House, Science Park,
Cambridge.

Pokomy, J. 1981. Browning flom lipid-protein interactions. Prog. Food Nutr. Sci.
5:421-428.

Powrie, W. D., Wu, C. H., Rosin, M. P., and Stich, H. F. 1981. Clastogenic and
mutagenic activities of Maillard reaction model systems. J. Food Sci. 46: 1433-
1438.

Reddy, B. S., Hedges, A. R., Laakso, K., and Wynder, E. L. 1978. Metabolic
epidemiology of large bowel cancer: fecal bulk and constituents of high-risk
North American and low-risk Finnish population. Cancer. 42(6):2832-2838.

Rhee, K. S., Donnelly, KC, and Ziprin, Y. A. 1987. Reduction of mutagen
formation in flied ground beef by glandless cottonseed flour. J. Food
Protection. 50(9):753-755.

Ritskes-Hoitinga, J ., Meijers, M., Tirnmer, W. G., Wiersma, A., Meijer, G. W., and
Weststrate, J. A. 1996. Effects of two dietary fat levels and four dietary
linoleic acid levels on mammary tumor development in Balb/c-MMTV mice
under ad libitum feeding conditions. Nutr. Cancer. 25(2):161-172.

Rizzi, G. 1994. The Maillard reaction in foods. In Maillard Reactions in
Chemistry, Food, and Health, Eds., T.P. Labuza, G.A. Reineccius, V.M.
Monnier, J. O’Brein, and J.W. Baynes. pp.11-19. The Royal Society of
Chemistry, Thomas Graham House, Science Park, Cambridge.

Schifflnan, M. H., and Felton, J. S. 1990. Fried foods and risk of colon cancer.
Letter to the Editor. Am. J. Epidemiol. 131 :376-378.

71

Shibamoto, T., Nishimura, O., and Mihara, S. 1981. Mutagenicity of products
obtained flom a maltol-ammonia browning model system. J. Agric. Food
Chem. 29:643-646.

Shioya, M., Wakabayashi, K., Sato, S., Nagao, M., and Sugimura, T. 1987.
F oramtion of a mutagen, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine
(PhIP) in cooked beef, by heating a mixture containing creatinine,
phenylalanine and glucose. Mutat. Res. 191:133-138.

Silberstein, D. A., and Lillard, D. A. 1978. Factors affecting the autoxidation of
lipids in mechanically deboned fish. J. Food Sci. 43(3):764-766.

Skog, K. 1993. Cooking procedures and food mutagens: A literature review. Food
Chem. T oxicol. 31(9):655-675.

Skog, K., and Jagerstad, M. 1990. Effects of monosaccharides and disaccharides
on the formation of food mutagens in model systems. Mutat Res. 25:263-272.

Skog, K, and Jagerstad, M. 1991. Effect of glucose on the formation of PhIP in a
model system. Carcinogenesis. 12:2297-2300.

Skog, K., and Jagerstad, M. 1993. Incorporation of carbon atoms flom glucose into
the food mutagens MeIQx and 4,8-DiMeIQx using "C-labelled glucose in a
model system. Carcinogenesis 14:2027-2031.

Skog, K., Jagerstad, M. and Laser-Reutersward, A. 1992a. Inhibitory effect of
carbohydrates on the formation of mutagens in flied beef patties. Food Chem.
T oxicol. 30(8):681-688.

Skog, K., Knize, M. G., Jagerstad, M., and Felton, J. S. 1992b. Formation of new
heterocyclic amine mutagens by heating creatinine, alanine, threonine, and
glucose. Mutat. Res. 268: 191 -1 97.

Snyderwine, E. G., Schut, H. A. J., Adamson, R. H., Thorgeirsson, U. P., and
Thorgeirsson, S. S. 1992. Metabolic activation and genotoxicity of
heterocyclic arylamines. Cancer Res. (Suppl.) 52:2099s-21023.

Sorata, Y., Takahama, U., and Kirnura, M. 1984. Protective effect of quercetin and
rutin on photosensitized lysis of human erythrocytes in the presence of
hematoporphyrin. Biochem. Biophys. Acta. 799:313-317.

 

72

Spingarn, N. E., and Garvie, C. T. 1979. Formation of mutagens in sugar-ammonia
model systems. J. Agric. Food Chem. 27(6):]319-1321.

Spingarn, N. E., Garvie-Gould, C., Vuolo, L. L., and Weisburger, J. H. 1981.
Formation of mutagens in cooked foods. IV. Effect of fat content in flied beef
patties. Cancer Lett. 12:93-97.

Spingarn, N. B., and Weisburger, J. H. 1979. Formation of mutagens in cooked
foods. I. Beef. Cancer Lett. 7:259—264.

Stavric, B. 1994. Biological significance of trace levels of mutagenic heterocyclic
aromatic amines in human diet: a critical review. Food Chem. T oxicol.

32(10):977-994.

Steineck, G., Gerhardsson-de-Verdier, M., and Overvik, E. 1993. The
epidemiological evidence concerning intake of mutagenic activity flom the
flied surface and the risk of cancer cannot justify preventive measures. Eur. J.
Cancer Prev. 2(4):293-300.

Sugimura, T. 1979. Naturally occurring genotoxic carcinogens. In Naturally
Occurring Carcinogens-Mutagen and Modulators of Carcinogenesis. Eds.,
E.C. Miller, J .A. Miller, I. Hirono, T. Sugimura, T., S. Takayama. pp.241-261.
Japan Scientific Societies Press, Tokyo.

Sugimura, T. 1982. Mutagens, carcinogens, and tumor promoters in our daily food.
Cancer 49: 1970-1984.

Sugimura, T., Nagoa, M., Kawachi, M., Honda, T., Yahagi, T., Seino, Y., Sato, S.
1977. Mutagen-carcinogens in foods, with special reference to highly
mutagenic pyrolytic products in broiled foods. In Origins of Human Cancer.
Eds., H.H. Hiatt, J.D. Watson, J.A. Winsten. pp. 1561-1576. Cold Spring
Harbor Laboratory, New York.

Sugimura, T., and Sato, S. 1982. Mutagens, carcinogens, and tumor promoters in
our daily food. Cancer 49:1970-1984.

Sugimura, T., and Sato, S. 1983. Mutagens-carcinogens in foods. Cancer Res.
(Suppl) 43:24153-24215.

73

Sugimura, T., Sato, S., and Wakabayashi, K. 1988. Mutagens/carcinogens in
pyrolysates of amino acids and proteins and in cooked foods: heterocyclic
aromatic amines. In Chemical Induction of Cancer, Structural Basis and
Biological Mechanism, Eds., V. T. Woo, D. V. Lai, J.C. Arcos and M. F.
Argus. pp.681-710. Academic Press, New York.

Takayama, S., Masuda, M., Mogami, M., Ohgaki, H. Sato., S., and Sugimura, T.
1984. Induction of cancers in the intestine, liver and various other organs of
rats by feeding mutagens flom glutamic acid pyrolyzate. Gann 75:207-213.

Tanaka, T., Barnes, W. S., Williams, G. M. and Weisburger, J. H. 1985.
Multipotential carcinogenicity of the flied food mutagen 2-amino-3-methyl-
imidazo[4,5-f]quinoline in rats. Jpn. J. Cancer Res. 76:570-576.

Tappel, A. L. 1962. Heme compounds and lipoxidase as biocatalysts. In
Symposium on Foods: Lipids and their Oxidation. Eds., H.W. Schultz, E.A.
Day, and Sinnhuber, R.O. Avi Publishing, Westport, Conneticut.

Taylor, R. T., Fultz, E., and Knize, M. G. 1985. Mutagen formation in a model
beef boiling system. IH. Purification and identification of three heterocyclic
amine mutagens-carcinogens. .1. Environ. Sci. Health A20(2):l35-l48.

Taylor, R. T., Fultz, B., and Knize, M. 1986. Mutagen formation in a model beef
supernatant flaction. IV. Properties of the system. Environ. Health Perspec.
67:59-74.

Taylor, R. T., Shore, V., and Fultz, E. 1984. Mutagen formation in a model beef
boiling system. II. Effects of proteolysis and comparison of soluble flactions
flom several protein sources. J. Environ. Sci. Health A19(7):8l9-845.

Thompson, L. H., Tucker, J. D., Stewart, S. A., Christensen, M. L., Salazar, E. P.,
Carrano, A.V., and Felton, J. S. 1987. Genotoxicity of compounds flom
cooked beef in repair deficient CHO cells versus Salmonella mutagenicity.
Mutagenesis 2:483-487.

Tricker, A.R., and Preussmann, R. 1991. Carcinogenic N-nitrosamines in the diet:
occurrence, formation, mechanisms and carcinogenic potential. Mutat. Res.
259:277-289.

 

74

Tsuda, M., Negishi, C., Makino, R., Sato, S., Yamaizumi, Z., Hirayarna, T., and
Sugimura, T. 1985. Use of nitrite hypochlorite treatments in the determination
of the contributions of IQ-type and non-IQ-type heterocyclic amines to the
mutagenicities in crude pyrolyzed materials. Mutat. Res. 147:335-341.

Turesky, R. J., Bur, H., Huynh-Ba, T., Aeschbacher, H. U., and Milon, H. 1988.
Analysis of mutagenic heterocyclic amines in cooked beef products by high-
performance liquid chromatography in combination with mass Spectrometry.
Food Chem. T oxicol. 26:501-509.

Turesky, R. J., Forster, C. M., Aeshbacher, H. U., Wurzner, H. P., Skipper, P. L.,
Trudel, L. J ., and Tannenbaum, S. R. 1989. Purification of the food-bome
carcinogens 2-amino-3-methylimidazo[4,5-j]quinoline and 2-amino-3,8-
dimethyl-imidazo[4,5-f]quinoxaline in heated meat products by
irnmunoaffrnity chromatography. Carcinogenesis 10: 15 1-156.

Turesky, R. J., Stillwell, S. R. J., Skipper, P. L. and Tannenbaum, S. R. 1993.
Metabolism of the food-bome carcinogens 2-amino-3-methylimidazo[4,5-f]-
quinoline and 2-arnino-3,8-dimethylimidazo[4,5f]-quinoxaline in the rat as a
model for human biomonitoring. Environ. Health Perspec. 99: 123-128.

Vahl, M., Gry, J., and Nielsen, P. A. 1987. Mutagens in flied pork and the
influence of the flying temperature. XVI Annual Meeting of the European
Environmental Mutagen Society, July pp. 19-23. Zurich, Switzerland.

Vahl, M., Gry, J., and Nielsen, P. A. 1988. Mutagens in flied pork and the
influence of flying temperature. (Abstract). Mutat. Res. 203:239.

Wakabayashi, K., Nagao, M., Esumi, H., and Sugimura, T. 1992. Food-derived
mutagens and carcinogens. Cancer Res. (Suppl) 52:20923-20983.

Wakabayashi, K., Takahasi, M., Nagao, M., Sato, S., Kinae, N., Tomita, I., and
Sugimura, T. 1986. Quantification of mutagenic and carcinogenic heterocyclic
amines in cooked foods. In Develop. Food Sci. .363-371.

Wang, Y. Y., Vuolo, L. L., Spingarn, N. B., and Weisburger, J. H. 1982.
Formation of mutagens in cooked foods vs. the mutagen reducing effect of soy
protein concentrates and antioxidants during flying of beef. Cancer Lett.
16:179-189.

75

Watanabe, K., and Sato, Y. 1971. Some alkylsubstituted pyrazines and pyridines in
the flavor components of shallow flied beef. J. Agric. Food Chem. 19:1017-

1019.

Weisburger, J. H., Barnes, W. S., Lovelette, C. A., Tong, C., Tanaka, T., and
Williams, G. M. 1986. Genotoxicity, carcinogenicity and mode of action of
the flied food mutagen 2-amino-3-methyl-imidazo[4,5-j]quinoline (IQ).
Environ. Health Perspec. 67: 12 l - 127.

Weisburger, J. H., and Spingarn, N. E. 1979. Mutagens as a function of mode of
cooking of meats. In Naturally Occurring Carcinogens-Mutagens and
Modulators of Carcinogenesis. Ed., E.C.Miller. pp. 177-184. University Park
Press, Baltimore, MD.

Weisburger, J. H., Nagao, M., Wakabayashi, K., and Oguri, A. 1994. Prevention of
heterocyclic amine formation by tea and tea polyphenols. Cancer Lett. 83: 143-
147.

Weisburger, J. H., Reddy, B. S., Barnes, W. S., and Wynder, E. L. 1983. Bile
acids, but not neutral sterols, are tumor promoters in the colon in man and in
rodents. Environ. Health Perspec. 50: 101-107.

Wei, C. I., Kitamura, K., and Shibamoto, T. 1981. Mutagenicity of Maillard
browning products obtained flom a starch-glycine model system Food
Cosmet. T oxicol. 19:749-751.

Yamaizumi, Z., Kasai, H., Nishimura, S., Edmonds, C. G., and McCloskey, J. A.
1986. Stable isotope dilution quantification of mutagens in cooked foods by
combined liquid chromatography-themospray mass Spectrometry. Mutat. Res.
173:1-7.

Yarnashita, K., Adachi, M., Kato, S., Nakayama, H., Ochiai, M., Wakabayashi, K.,
(Sato, S., Nagao, M., and Sugimura, T. 1990. DNA adducts formed by 2-
amino-3,8-dimethyl-imidazo[4,5-f]quinoxaline in rat liver: Dose-response on
chronic administration. Jpn. J. Cancer Res. 81:470-476.

Yen, G. C., and Chen, H. Y. 1995. Antioxidant activity of various tea extracts in
relation to their antirnutagenicity. J. Agric. Food Chem. 43:27-32.

Yoshida, D., and Okamoto, H. 1980a. Formation of mutagens by heating creatine
or amino acids with addition of fatty acids. Agric. Biol. Chem. 44(12):3025-
3027.

76

Yoshida, D, and Okamoto, H. 1980b. Formation of mutagens by heating the
aqueous solution of amino acids and some nitrogenous compounds with
addition of glucose. Agric. Biol. Chem. 44(10):2521-22522.

Yoshida, D., and Okamoto, H. 1980c. Formation of mutagens by heating creatine
and glucose. Biochem. Biophys. Res. Comm. 96(2):844-847.

Yoshida, D., Sato, Y., and Mizusake, S. 1984. Isolation of 2-amino-3-methyl-
imidazo[4,5-f]quinoline as a mutagen flom the heated product of a mixture of
creatine and proline. Agric. Biol. Chem. 48:241-243.

Zhang, X. B., Tao, K., Urlando, C., Shaver-Walder, P., and Heddle, J. A. 1996.
Mutagenicity of high fat diets in the colon and small intestine of transgenic
mice. Mutagenesis 11(1):43-48.

Zhang, X. M., Wakabayashi, K., Liu, Z. O, Sugimura, T., and Nagao, M. 1988.
Mutagenic and carcinogenic heterocyclic amines in Chinese cooked foods.
Mutat. Res. 201:181-188.

 

"11111111111111‘3