Carcinogenesis vol.17 no.10 pp.2259-2265, 1996
Intestinal mutagenicity of two carcinogenic food mutagens in
transgenic mice: 2-amino-l-methyl-6-phenylimidazo[4,5-£]pyridin<
and amino(a)carboline
Xue Bin Zhang1, James S.Felton2, James D.Tucker2,
Cesare Urlando1 and John A.Heddle1^3
'Department of Biology, York University, Toronto, Canada M3J 1P3 and
Biology and Biotechnology Research Program, Lawrence Livennore
National Laboratory, Livennore, CA 94551-9900, USA
2
'To whom correspondence should be addressed
The heterocyclic amines produced during the cooking of
meat, including amino(a)carboline (AaC) and 2-amino-lmethyl-6-phenylimidazo[4,5-*]pyridine (PUP), are potent
bacterial mutagens and are carcinogenic in rodents. PhEP
is mutagenic in the small intestine, but its mutagenicity in
the colon, where most human intestinal cancers arise, has
not been reported, nor has the mutagenicity of AaC. In
this study, AaC (800 p.p.m.) was fed for 30 and 45 days
and PhIP (100 and 400 p.p.m.) was fed for 30, 60 and 90
days to groups of Fx (C57BL/6 X SWR) mice hemizygous
for multiple tandem copies of a lad transgene (the Big
Blue™ mouse) and heterozygous at the endogenous Dlb-1
locus. The mutant frequencies were assayed at Dlb-1 and
at lacI in the small intestine and at lad in the colon. PhIP
induced mutations at both loci in the small intestine
and induced slightly fewer mutations in the colon. The
accumulation of mutations at both loci appears to be linear
with both PhIP concentration and duration of exposure
and, thus, with dose (concentration x duration). The linear
increase with time is in agreement with predictions about
the effectiveness of chronic treatment protocols for tests of
in vivo mutagenicity. Unlike PhIP, AaC induced mutations
specifically in the colon and not in the small intestine,
thereby showing a dramatic tissue specificity. The rate
(mutations/p.p.m. day) was similar to PhIP.
Introduction
Of the mutagens identified by the use of the Salmonella/
mammalian microsome test, the heterocyclic amines (HAs*)
found in cooked food (1,2) have attracted the most interest.
These HAs are formed predominantly as pyrolysis products
during the cooking process and are most abundant in the welldone or charred parts of fried, broiled and barbecued fish,
chicken and meat. They are very mutagenic in the Ames
Salmonella strain TA98. Their carcinogenicity has also been
demonstrated in a variety of organs in both mice and rats,
including rat colon (3-5). HAs usually require metabolic
activation by cytochrome P450 for DNA adduct formation and
genotoxicity, key events in the initiation of carcinogenesis.
The concentration of HAs formed during the cooking process
depends on the temperature, duration and mode of cooking
(6,7). HAs are classified into two groups, according to their
•Abbreviations: HAs, heterocyclic amines; IQ, 2-amino-3-methyl
imidazo[4,5-/]quinoline;
PhIP,
2-amino-l-methyl-6-phenylimidazo[4,5fcjpyridine; AaC, amino(a)carboline; PBS, phosphate-buffered saline; X-gal,
5-bromo-4-chloro-3-indolyl fi-D-galactopyranoside.
© Oxford University Press
structures. One type, the 2-amino-3-methylimidazo[4,5-/|quinoline (IQ) type, is derived from amino acids, hexose and
creatinine. The other, non-IQ, type is formed via degradation
products of amino acids or proteins. The amounts of 2-aminol-methyl-6~-phenylimidazo[4,5-fc]pyridine (PhIP; IQ type) and
amino(a)carboline (AaC; non-IQ type) produced in cooked
meats are usually higher than other HAs (8).
PhIP is the most abundant of the mutagenic and carcinogenic
HAs produced in cooked meat and fish (9,10), so a role for
this compound in human carcinogenesis has been suspected.
PhIP is the most mutagenic HA in mammalian cells (10,11),
although its mutagenicity in Salmonella typhimurium is relatively low. PhlP-DNA adducts have been detected in white
blood cells, lung, small intestine, kidney, colon, liver and
stomach. The adduct level is particularly high in the colon
and lower in liver of male F344 rats (12). PhTP induces
sister chromatid exchanges in lymphocytes and micronuclei in
erythroid precursors, but does not induce detectable numbers
of translocations in bone marrow or lymphocytes (13). PhIP
induces lymphomas (14) and both colon and mammary
carcinomas (15) in mice, and intestinal adenocarcinomas in
rats (16). Although some studies have indicated that PhIP is
not a hepatocarcinogen in rats (17), it induced preneoplastic
liver foci in rats (18).
AaC is the second most abundant HA in very well-done
meat and fish. It has been identified in a pyrolysate of soybean
globulin (19), cigarette smoke condensate (20), beer and wine
(21), grilled bacon and cooked beef (22,23). Obviously, it is
important that mutations be studied in vivo in the appropriate
sex and species so that factors such as the uptake, distribution,
metabolic activation, detoxification, elimination and influence
on cell proliferation are comparable with those used for the
studies of carcinogenicity. The development of transgenic mice
with bacterial target genes that can be recovered and analyzed
for mutations in vivo makes this possible. Mutations that arose
in vivo can be assayed in any tissue from which sufficient
high molecular weight DNA can be obtained (24-26). In
addition, mutations of naturally occurring genes can be quantified in some cells (27-30).
PhIP has been shown to be mutagenic at the Dlb-1 locus in
the small intestine by Brooks et al. (24). They noted that the
potency of PhIP at inducing mutations in the small intestine
contrasted with the relatively low rate of cancer induced at
this site. Since the Dlb-1 locus is used only in the small
intestine, they could not be certain whether or not the higher
rate of colon cancer was associated with a proportionately
higher mutagenicity of PhIP in the colon. Thus measurements
of the colonic mutagenicity of PhIP are of interest. There has
been no information about mutagenicity of AaC in vivo prior
to this study. The carcinogenic potency of AaC in the intestine
is still unknown.
The use of in vivo assays for mutation in mice is relatively
new and much remains to be learned about the methods and
the loci used. It is noteworthy that the lacl transgene, like its
2259
X.B.Zhang el al
lacZ counterpart, is a bacterial gene embedded in a long stretch
of reiterated prokaryotic DNA that is heavily methylated and
(presumably) unexpressed (32). In studies of the response of
the lacl transgene to acute mutation by ethylnitrosourea, it
behaved very similarly to an expressed host gene in the same
cells (33,34). Nevertheless, the transgenic complex shows
heterocyclic condensation behavior at meiosis (35) and is not
transmitted to progeny quite as often as would be expected
(36). Furthermore, in a study involving a chronic exposure
to ethylnitrosourea, the lacl transgene showed the linear
accumulation of mutations expected, whereas the expressed
host locus showed a non-linear accumulation of mutations,
indicative of preferential repair of the transcribed gene (37).
It was therefore of interest to determine whether or not this
difference between the loci would be found for another mutagen
when a similar chronic treatment regimen was used. We report
the results of this for PhIP and AaC here.
Materials and methods
The experiments reported here were approved by the Animal Care Committees
of York University and Lawrence Livermore National Laboratory.
Animals
Homozygous lacl C57BL/6 (Dlb-lb/Dlb-lb) transgenic mice were obtained
from Stratagene (La Jolla, CA). SWR (Dlb-l'/Dlb-l') mice were obtained
from the Jackson laboratories (Bar Harbor, ME). The two strains were crossed
to produce the Dlb-Ib/Dlb-1* mice hemizygous for multiple tandem copies of
lacl used in these experiments. The mice were bred at York University,
shipped by air to Livermore, where the experimental groups were exposed to
PhIP, and then returned by air to York University, where the mutation
measurements were made. The positive and negative controls were also
transported. The positive controls were treated with ethylnitrosourea after
their return to York University. More than 1 week elapsed after the end of
treatment before the animals were killed, which is adequate for expression
of mutations at both loci in this tissue (33).
Diet and mutagcns
PhIP and AaC were obtained from Toronto Research Chemicals. They were
milled dry into the standard diet and fed ad libitum. It was already known
that these concentrations are well tolerated for at least 90 days (13), the
maximum duration of the work described here. All of the exposures began at
the same time, except for the positive controls. The animals were killed
between 1 and 3 weeks after the end of the exposure, which is after the
minimum expression time. The frequency of these mutations does not change
once the 1 week expression time is over (33,38). Each of the experimental
groups contained at least three females and three males, although in some
cases mutation measurements could not be made on all of the animals and in
one case two extra animals were included. AaC was fed at 800 p.p.m. for 30
or 45 days. PhIP was fed at 100 or 400 p.p.m. for 30, 60 or 90 days. In
addition, PhIP was fed at 250 p.p.m. for 30 days. Ethylnitrosourea was freshly
dissolved in dimethylsulfoxide, diluted 20-fold with phosphate-buffered saline
(PBS), pH 7.4, and injected i.p. to give 100 mg/kg as the positive control.
All mice were housed in standard plastic cages with lids and wood chip
bedding at 22 ± 2°C, 70% humidity on a 12 h light/12 h dark cycle. All
animals were fed adlibitum and given unlimited access to water. The food
was replenished regularly and weighed when it was replenished.
lacl transgenic mouse mutagenesis assay
Genomic DNA was extracted from the epithelial cells of the small intestine
or colon and analyzed as.descnbed previously (25). Briefly, the cell suspension
was treated with proteinase K solution (2 mg/ml; Sigma), followed by phenol/
chloroform extraction and ethanol precipitation as described by Kohler el al.
(26). The A. shuttle vector was recovered by means of in vitro X packaging
extracts (Transpack™; Stratagene) as viable phage. These phage were placed
on Escherichia coli SCS-8, which produce an inactive C-tenninal portion of
the lacZ gene product. The E.coli were then plated on assay trays containing
agar with 5-bromo-4-chloro-3-indolyl P-D-galactopyranoside (X-gal). A mutation in the lacl repressor gene that allows transcription of the alacZ gene,
coding for the N-terminal portion of (J-galactosidase, leads to an active enzyme
by complementation, which cleaves X-gal and yields a blue plaque (26).
Plaques produced by phage with non-mutated lacl genes are colorless, since
the presence of an active repressor prevents the N-terminal portion of the
enzyme from being synthesized.
2260
Dlb-1 locus assay
The tissues of the small intestine were prepared as described (31) with a few
modifications (32). Briefly, the small intestine was flushed with cold PBS,
fixed in 10% formalin in saline, then cut along the mesenteric side. It was
then placed on a microscope slide, fixed in 10* formal saline for 1 h, rinsed
with PBS and incubated in 20 mM DL-dithiothreitol (Sigma) to remove mucus.
Endogenous peroxidases were blocked with 0.1% phenylhydrazine HC1
(Sigma). The preparations were washed and stained with DBA-peroxidase
(Sigma) solution. Slides were scored as described by Winton et al. (31).
Results
The mutagenicity of PhIP at the Dlb-1 locus in the small
intestine is given in detail in Tables I-HI, but is more easily
seen in Figure 1. The mutant frequency increased more or less
linearly with time (Figure la). This is what would be expected
if the mutation rate is constant with time at any one concentration and the mutations are neutral (42). The mutant frequency
also increased approximately linearly with concentration of
PUP in the diet for a given duration of exposure (Figure
lb). Consequently, there is a linear relationship with dose
(concentration X time) in Figure lc. A very similar pattern is
evident at the lacl locus in the colon, as can be seen in Figure
2. The detailed data are also in Tables I-IH, because the same
animals were used for both measurements. Again the mutant
frequency appears to increase linearly with time, concentration
and dose. The frequency of lacl mutants is remarkably similar
to the frequency of Dlb-1 mutants, given the large differences
in mutant frequency often observed for different loci.
Since two different tissues are being compared as well as
two different loci, it is of interest to know whether the same
locus has been affected equally in the two tissues. This
comparison between the colon and the small intestine at lacl
is shown in Figure 3 for animals exposed for 30 days. The
mutant frequencies are quite similar, although the mutant
frequencies in the colon are somewhat lower. Clearly PhIP is
mutagenic in both tissues, but is more mutagenic in the small
intestine than in the colon.
In contrast, AaC is not mutagenic at the Dlb-1 locus in the
small intestine under these conditions (Figure 4a), but is
mutagenic at the lacl locus in the colon (Figure 4b). The
detailed data are given in Table IV. Obviously it is of
interest to know whether this difference is a difference between
the Dlb-1 locus and the lacl locus or between the small
intestine and the colon. As shown in Figure 4b and the detailed
data in Table V, AaC was not detectably mutagenic at lacl in
the small intestine. Thus AaC is a colon-specific mutagen in
the intestine.
Discussion
The results are of interest with respect not only to the biological
effects and specificity of the chemicals tested but also to the
characteristics of the assays used. Quantitative measurements
of mutagenicity have previously been made at the hprt locus
in fibroblasts and lymphocytes (28,30,31,39), but frequently
these are not the target cells for carcinogens. More recently
the Dlb-1 assay has been developed for the small intestine and
has proved to be an easy, reliable and sensitive assay. Mutations
at this locus seem to be neutral (i.e. they provide no selective
advantage or disadvantage to the cells containing them), so
that the mutant frequency observed is the integral of the
mutation rates over the life of the animal (33,40). A similar
situation is seen for the lacl transgene in the Big Blue™
mouse (38). The latter gene can be analyzed for mutation in
Mutagenlclty of PUP and amlno(a)cart>oUrie
Table L Mutant frequencies in mice fed PhIP for 30 days
Treatment
(p.p.m.)
100
250
400
ENU
Animal
(sex)
8
8
9
9
8
8
8
9
9
9
8
8
8
9
9
6
8
8
8
9
9
9
9
8
9
Dlb-r Mutants/105 stem cells'
lacl- Mutants/105 plaques'1
Villi
Mutants
MJ.*
Mean±SE
10 400
13 200
15 900
14 800
11 900
13 500
s.l.c
13 600
16 500
1
3
3
6
23
23
_
19
36
32
41
71
54
45
85
81
_
38
_
39
80
47
19
26
1.0
2.3
1.9
4.1
19.3
17.0
2.3±0.7
S.I.
8000
11 700
13 600
12 400
13 300
11 200
12 100
S.I.
12 200
S.I.
8500
12 700
13 300
13 600
8400
18.0±1.7
14.0
22.0
40.0
35.0
52.0
44.0
34.0
76.0
67.0
41.0±3.3
53.0+17
31.0
Plaques
Mutants
S.I.
S.I.
_
49 800
10 000
24 500
11 500
15 300
20 900
29 000
14 100
34 200
37 400
2400
12 500
13 000
13 600
21 000
13 300
2
1
1
0
0
5
1
2
7
3
0
1
4
6
10
3
_
8
_
8
11
10
7
S.I.
34 100
46.0
63.0
35.0
14.0
31.0
S.I.
23.0±8 5
24
21
20
21
500
900
900
000
M.F.b
Mean±SE
7.0±3.0
4.0
10.0
4.0
0.0
0.0
24.0
3.0
14.0
20.0
8.0
0.0
8.0
30.0
44.0
48.0
23.0
7J±3.9
13±5.3
37±5.0
24.0
33.0
50.0
48.0
33.0
41±7.5
*In the small intestine, there being 10 mutable stem cells/villus.
b
In DNA isolated from the colon.
c
s.l., indicates that the sample was lost in processing.
Table n . Mutant frequencies in mice fed PhIP for 60 days
Treatment
(p.p.m.)
0
100
400
ENU
Animal
(sex)
6
8
9
9
8
8
8
9
9
9
8
8
8
9
9
9
8
9
UJCI' Mutants/105 plaquesb
Dlb-r Mutants/105 stem cells'
Villi
Mutants
M.F.1
Mean±SE
Plaques
15 200
12 900
12 900
14 400
12 600
11 900
12 700
14 600
12 100
14 900
13 200
11 900
13 100
12 700
12 300
16 400
4800
7700
3
6
1
3
30
45
42
70
61
43
143
117
150
145
123
113
9
12
2.0
4.7
0.8
2.1
24.0
38.0
33.0
48.0
50.0
29.0
108.0
98.0
115.0
114.0
97.0
69.0
19.0
16.0
2.4±0.8
31 500
s.l.c
22 400
68 800
19 300
37.0±4.2
100.0±7.0
18.0±1.5
Mutants
M.F.b
Mean±SE
9.5
5.6±2.9
3
-
0
5
4
S.I.
-
14 100
10 300
14 200
20 100
51 800
79 100
46 100
65 500
77 900
24 000
17 700
10400
4
3
2
10
32
40
26
57
67
17
12
5
0.0
7.3
21.0
28.0
29.0
14.0
50.0
62.0
51.0
56.0
87.0
86.0
71.0
68.0
48.0
28.0±6.0
69.0±6.2
58.0±10.0
'In the small intestine, there being 10 mutable stem cells/villus.
'in DNA isolated from the colon.
c
s.l., indicates that the sample was lost in processing.
any tissue from which sufficient high molecular weight DNA
can be isolated. The neutrality of these mutations, at least in
the small intestine, is indicated by the fact that the mutant
frequency is constant after the cessation of exposure and that
weekly doses give an additive response (33,38). This provides
support for the contentions of Tao et al. (38), Shephard et al.
(41) and Heddle et al. (42) that chronic exposures would
increase the sensitivity of the transgenic assays. This is
important because the assays are expensive and the spontaneous
mutant frequency is high (34). The PhIP data presented here
confirm the results obtained for lacl in the small intestine over
a similar time period (37). Although PhIP, like ethylnitrosourea,
is sufficiently mutagenic that it can be detected with a subacute
treatment protocol, clearly chronic exposure, by increasing the
2261
X.B.Zhang et aL
Table i n . Mutant frequencies in mice fed PhIP for 90 days
Treatment
(p.p.m.)
Animal
(sex)
ViUi
0
100
400
ENU
6
6
9
9
6
S
6
9
9
9
6
6
6
9
9
9
6
9
9
lacl' Mutants/105 plaquesb
Dlb-1- Mutants/105 stem cells*
13 100
9600
15 600
s.l.c
12 700
17 900
13 000
11 600
12 000
10 600
15 000
8300
6500
12 700
16 100
9200
12 300
12 000
11 400
M.F.'
Mean±SE
Plaques
Mutants
M.F.b
Mean±SE
2
3
3
1.5
3.1
1.9
2.2±0.5
_
23.0
39.0
22.0
39.0
16.0
44.0
143.0
178.0
134.0
123.0
127.0
120.0
9.0
8.3
20.0
0
0
2
0
6
6
2
6
2
4
3
12
3
49
16
6
9
0.5
0.0
8.6
0.0
31.0
26.0
65.0
35.0
16.0
34.0
107.0
111.0
73.0
209.0
165.0
34.0
43.0
2.2±2.2
29
69
28
45
19
46
214
148
87
156
205
110
11
10
23
27 300
2000
23 300
6100
19 400
23 000
3100
17 200
12 800
11 600
2800
10 800
4100
23 400
Mutants
_
31.0±4.7
138.0±8.8
S.I.
12.0±3.8
9 700
17 500
S.I.
20 900
35.0+6.7
133.0±24.0
39.4±4.0
'In the small intestine, there being 10 mutable stem cells/villus.
b
In DNA isolated from the colon.
s.l., indicates that the sample was lost in processing.
c
Table IV. Mutant frequencies in mice fed 800 p.p.m. amino(a)carboline
Treatment
(days)
Animal
(sex)
Dlb-r Mutants/105
Villi
lacl' Mutants/105 plaquesb
stem cells*
Mutants
M.F.°
Untreated controls (see PhIP experiments done simultaneously)
30
6
12 900
3
2.3
6
9700
3
3.1
6
12 700
6
4.7
9
13 200
1
0.8
9
12 900
3
2.3
9
12 700
6
4.7
45
6
10 800
10
9.2
6
11 000
6
5.5
6
12 000
4
3.3
9
11 600
3
2.6
2
13 700
7
5.1
9
12 200
6
4.9
ENU (from PhIP experiments done simultaneously)
Mean±SE
3.1
3.0±0.6
5.2+0.9
29.0
Plaques
Mutants
M.F.b
21 200
27 200
16 100
19 200
20 100
20 600
19 000
45 400
60 900
41 700
42 500
67 200
20
29
15
10
17
54
18
28
42
46
73
37
94.0
107.0
93.0
52.0
85.0
262.0
94.0
62.0
69.0
110.0
Mean±SE
3.5
116.0±30.0
94.0±18.0
172.0
55.0
50.0
*In the small intestine, there being 10 mutable stem cells/villus.
•"In DNA isolated from the colon.
mutant frequency, increased the sensitivity of the assay. The
fact that the mutant frequency increased linearly with dose
(time X concentration) indicates that a lower dose can be
detected if the exposure time is increased. This is true also for
the Dlb-1 locus in response to PhIP.
The results reported here are of considerable interest with
respect to the concept of dose. In acute experiments, the
duration of exposure is determined by pharmacokinetic considerations and may differ with the treatment. This makes it
difficult to determine the nature of the dose-response curve.
Obviously pharmacokinetic factors influence the results
obtained under chronic exposure conditions too, but if the
animals reach a more or less steady-state situation, then the
mutant frequency should increase linearly with time. This
appears to be the case for PhIP, for which we have the most
2262
data. The results from AaC are equivocal. The linearity of the
dose-response curve, when dose is defined as the product of
the exposure concentration and the duration of exposure, is
particularly interesting and provides a basis for extrapolation
to lower doses. Human exposure is at much lower doses
and for much longer times. These experiments indicate that
measurements could be made at concentrations much closer
to the human exposure conditions by increasing the duration
of the exposures.
Recently an important difference between the response of
the Dlb-1 gene and the lacl gene has been discovered (37).
During chronic treatment with ethylnitrosourea, the mutant
frequency at lad increased linearly with time, as it does for
PhIP. At the Dlb-1 locus, however, there was an initial deficit
of Dlb-1 mutations relative to lacl, followed by an exponential
Mutageniclty of PhlP and amino(a)carbollne
140 0
a
b
/
100.0
80 0
/
/ /
/ /
/
1200'
rC
/
'
/
400'
20.0-
7
SO
100
DAYS ON DIET
0
200
4000
CONCENTRATION Of P « P
20000
40000
DOSE (ppm-cteym)
Fig. 1. Mutagenicity of PhlP at the Dlb-1 locus in the small intestine. The
point at 0 is the average of all of the controls in each case, (a) Duration of
exposure at: O, untreated control; • , 100 p.p.m.; A, 250 p.p.m.;
O, 400 p.p.m. (b) Concentration in diet for various periods: O, untreated
control; • , 30 days; • , 60 days; • , 90 days, (c) Data from all treatments
are shown as a function of dose (= concentration X duration of exposure).
20
60 0
40
20
DAYS ON DIET
40
DAYS ON DIET
Fig. 4. Mutagenicity of AaC. The error bars represent the standard error of
the mean, (a) Dlb-1 locus in the small intestine, (b) • , lacl locus in the
small intestine; • • lad locus in the colon.
TC
Table V. Mutant frequencies at lad in the small intestine
90
100
DAYS ON DIET
0
200
400
CONCENTRATION OF Pt«>
20000
60T
100
200
300
Animal
(sex)
Plaques
Mutants
M.F."
Mean±SE
30
6
6
9
9
9
$
6
9
9
9
21 100
22 900
18 200
6400
17 400
22 900
20 800
21 800
24 500
23 750
2
3
0
0
1
3
2
2
4
3
9.0
18.0
0.0
0.0
6.0
13.0
10.0
9.0
13.0
13.0
7.0±7.0
40000
DO3E (ppro-dayi)
Fig. 2. Mutagenicity of PhlP at the lad locus in the colon. The point at 0 is
the average of the controls, (a) Duration of exposure at: O, untreated
control; Q, 100 p.p.m.; A, 250 p.p.m.; O, 400 p.p.m. (b) Concentration in
diet for various periods: O, untreated control; • , 30 days; • , 60 days;
• , 90 days, (c) Data from all treatments are shown as a function of dose
(= concentration X duration of exposure).
0
A(a)C treatment
(days)
400
CONCENTRATION OF PHIP
Fig. 3. Mutant frequency observed at the lacl locus after 30 days of feeding
PhlP at the concentration shown. X colon, • small intestine.
increase in Dlb-1 mutations with continued exposure. This is
not the result of selection, for the mutant frequency is stable
after treatment ceases. The hypothetical mechanism involves
preferential repair of the Dlb-1 gene, which is transcribed, at
low doses, such that many lesions are repaired before DNA
replication and so do not lead to mutations. As exposure
continues, however, the stem cells are forced to proliferate
faster to compensate for cell death, thus putting themselves at
45
12.0±2.0
'lad' Mutants/105 plaques.
increased risk of both further cell death and mutation. If this
hypothesis is correct, then the fact that this is not seen with
PhlP indicates: (i) that PhlP is not an S phase-dependent
mutagen (although it does induce sister chromatid exchanges;
13); (ii) that PhlP does not kill S phase cells preferentially;
(iii) that the PUP lesions are not repaired by a transcriptionally
linked repair system. Clearly further studies of PhlP are
warranted.
Brooks et al. (24) noted that PhlP, although strongly mutagenic at the Dlb-1 locus in the small intestine, is carcinogenic
primarily in the colon, like many other agents, and suggested
that this poses a biological problem. What factor protects the
small intestine from cancer? In fact, Brooks et al. (24) could
not be certain that PhlP would not be a more potent mutagen
in the colon than the small intestine, i.e. that differential
mutation rates could not explain the tissue specificity. Our
results confirm their findings at the Dlb-1 locus and show that
PhlP is mutagenic at the lacl transgene to a similar extent not
only in the small intestine, but also in the colon. The problem
posed is thus real: the higher rate of carcinogenicity of PhlP
in the small intestine is not caused by a higher rate of mutation
in the colon. There must be some other important factor. A
similar result has been obtained for the induction of apoptosis
by the intestinal carcinogen 1,2-dimethylhydrazine (43).
Although there is a very strong gradient of carcinogenicity in
the intestine, increasing from the stomach to the rectum, the
greatest induction of apoptosis is in the small intestine, where
cancers are induced much less frequently than in the colon. It
2263
X.B.Zhang et al
may be that the mutation spectra in the two tissues differ and
that some kinds of mutation are much more frequent in the
colon than in the small intestine, although this seems unlikely.
Possibly, fewer mutations are needed to induce cancer in the
colon than the small intestine because of the number of
loci involved in the control of cellular proliferation. 1,2Dimethylhydrazine poses an additional problem in that it is
very weakly mutagenic, as judged by these assays (38), and
would not be expected to be such a potent carcinogen if
mutagenic potency were the critical factor. There are, of course,
two mutation spectra to consider: that produced by the chemical
at the sentinel locus and that which is carcinogenic in the
target tissue. A mismatch between these two could lead to an
incorrect assessment of carcinogenic potency (48). PhIP
induced colon tumors in rats (21) and induced 5'-GGGA3'—»5'-GGA-3' Ape gene mutations in colonic tumors of F344
rats (46). More than 95% of Ape mutations are frameshift or
nonsense mutations that result in a truncated protein, but no
mutations in the Ki-ras or p53 genes were found in colon
tumors induced by PhIP (46,47). Therefore, though inactivation of the APC protein may play a major role in PhIP
colon tumors, only Ape mutations were found in four of eight
colon tumors. Even if the mutation spectrum is the same in
the colon and the small intestine, the nature of the mutations
involved in carcinogenesis in the two tissues may differ.
Alternatively, there may be other, tissue-specific factors that
are important. Nevertheless, the results are surprising.
AaC, unlike PhIP, showed a clear tissue specificity for
mutation, with the colon being the responsive tissue. AaC is
second only to PhIP in abundance in well-done cooked food.
The mutagenicity of AaC in S.typhimurium TA100 and TA98
is lower than PhIP (6). The explanation for the tissue specificity
of AaC is not known. Possibly the intestinal flora, which are
far more abundant in the colon than the small intestine,
metabolize AaC to a direct mutagen to induce mutations
locally, i.e. in the colon but not in the small intestine. Possibly
there is a difference in the distribution of metabolites produced
by the mouse. Unlike PhIP, which is known to be metabolized
in the liver (44,45), transported via the blood as N-hydroxyPhlP and yV-hydroxy-PhIP A'-glucuronides and deconjugated
in the colon, the metabolism and transport of AaC in vivo are
still unknown. Possibly there are specific microsomal P450s
or other enzymes that metabolize AaC in the colon but not in
the small intestine.
Our results indicate that chronic exposure protocols will
prove to be preferable for in vivo mutagenicity studies with
the transgenic assays such as the Muta™ mouse, the Big
Blue™ rat or the Big Blue™ mouse (although we have studied
only the latter). The tissue specificity of carcinogens is not
explicable entirely by the mutant frequencies induced in the
various tissues: PhIP is somewhat more mutagenic in the small
intestine than the colon, but most intestinal cancers are induced
in the colon. AaC, in contrast, shows a dramatic tissue
specificity in the mutational response.
Acknowledgements
We thank Marilyn Ramsey, Ke Sheng Tao, Pamela Shaver-Walker, Germaine
Dawod, Lidia Cosentino and Yolanda Paashuis-Lew for their help with the
experiments reported here. This work was supported by grants from the
National Cancer Institute of Canada, the Natural Sciences and Engineering
Research Council of Canada and US National Cancer Institute grant 55861
and was performed under the auspices of the US Department of Energy at
the Lawrence Livermore National Laboratory under contract no. W-7405ENG48.
2264
References
l.Sugimura,T. and Sato.S. (1983) Mutagens-carcinogens in foods. Cancer
Res., 43, 2415s-2421s.
2.0hgaki,H, Takayama,S. and Sugimura,T. (1991) Carcinogenicities of
heterocyclic amines in cooked food. Mutat. Res., 259, 399-410.
3. MatsukuxaJJ., Kawachi,T., Morino.K., Ohgaki,H., Sugimura,T. and
Takayama,S. (1981) Carcinogenicity in mice of mutagenic compounds
from a tryptophan pyrolysate. Science, 213, 346-347.
4.0hgaki,H., Hasegawa,H., Kato.T., Suenaga,M., Ubukata,M., Sato.S.,
Takayama,S. and Sugimura,T. (1986) Carcinogenicity in mice and rats of
heterocyclic amines in cooked foods. Environ. Health Perspect., 67,
129-134.
5.Wakabayashi,K., Nagao.M., Esumi.H. and Sugimura,T. (1992) Foodderived mutagens and carcinogens. Cancer Res., 52, 2092s-2098s.
6.FeltonJ.S. and Knize,M.G. (1991) Occurrence, identification, and potential
mutagenicity of heterocyclic amines in cooked food. Mutat. Res., 259,
205-217.
7.Wakabayashi,K., Ushijima,H., Takanashi.M., Nukaya,H., Kim,S.-B.,
Hirose,M., Ochiai,M., Sugimura,T. and Nagao.M. (1993) Exposure to
heterocyclic amines. Environ. Health Perspect., 99, 129-133.
8.Layton,D.W., Bogen,K.T., Knize,M.G., Hatch.F.T, Johnson,M. and
FeltonJ.S. (1995) Cancer risk of heterocyclic amines in cooked foods: an
analysis and implications for research. Carcinogenesis, 16, 30-52.
9.Felton4.S., Knize.M.G., Shen.N.H., Lewis.P.R., Andresen,B.D., HappeJ.
and Hatch,F.T. (1986) The isolation and identification of a new mutagen
from fried ground beef: 2-amino-l-methyl-6-phenylimidazo[4,5-i]
pyridine (PhIP). Carcinogenesis, 7, 1081-1086.
lO.FeltonJ.S. and Knize.M.G. (1990) Heterocychc-amine mutagens/
carcinogens in foods. In Cooper.C.S. and Grover.P.L. (eds), Handbook of
Experimental Pharmacology. Spring-Verlag, Berlin, Germany, vol. 94/1,
pp. 471-502.
11. ThompsonJ^.H , TuckerJ.D., Stewart,S.A., Christensen,M L., Salazar,E.P.,
Carrano,A.V. and FeltonJ.S. (1987) Genotoxicity of compounds from
cooked beef in repair-deficient CHO cells versus Salmonella mutagenicity.
Mutagenesis, 2, 483-487.
12.Schut,H.AJ. and Herzog.C.R. (1992) Formation of DNA adducts of 2amino-l-methyl-6-phenylimidazo[4,5-fc]pyridine (PhIP) in male F344 rats.
Cancer Lett., 67, 117-124.
13.Director,A., NathJ., Rarasey.MJ., Swiger.R.R. and TuckerJ.D. (1996)
Cytogenetic analysis of mice chronically fed the food mutagen 2-aminol-methyl-6-phenylimidazo[4,5i] pyridine (PhIP). Mutat. Res., 359, 53-61.
14.Esumi,H., Ohgaki.H., Kohzen.E., Takayama.S. and Sugimura,T. (1989)
Induction of lymphoma in CDF1 mice by the food mutagen, 2-amino-lmethyl-6-phenylimidazo[4,5-i]pyridine. Jpn J. Cancer Res., 80, 11761178.
15.Ito,N., Hasegawa,R., Sano,M., Tamano.S., Esumi.H., Takayama,S. and
SugimuraX (1991) A new colon and mammary carcinogen in cooked
food,
2-amino-l-methyl-6-phenylimidazo[4,5-i]pyridine
(PhIP).
Carcinogenesis, 12, 1503—1506.
16. Ochiai,M., Ogawa.K., Wakabayashi.K., Sugimura,T., Nagase.S., Esumi,H.
and Nagao.M. (1991) Induction of intestinal adenocarcinoma by 2-aminol-methyl-6-phenylimidazo[4,5-6]pyridine in Nagase analbuminemic rats.
Jpn J. Cancer Res., 82, 363-366.
17.Takayama,S., Yamashita.K., Wakayabashi,K.t Sugimura,T. and Nagao^M.
(1989) DNA modification by 2-amino-1 -methyl-6-phenylimidazo-[4,5fclpyridine in rats. Jpn J. Cancer Res., 80, 1146-1148.
18.Kleman,N., Overvik,E., Blanck,A. and GustafssonJ.A. (1989) The food
mutagens 2-amino- l-methyl-6-phenylimidazo-[4,5-b]pyridine (PhIP) and
2-amino-3,8-dimethylimidazo[4,5-/lquinoxaline (MelQx) initiate enzyme
altered foci in the 'resistant hepatocyte' model. Carcinogenesis, 9,
1696-1700.
19.Yoshida,D., Matsumoto.T, Yoshimura,R. and Matsuzaki.T. (1978)
Mutagenicity of amino-a-carbolines in pyrolysis products of soybean
globulin. Biochem. Biophys. Res. Commun., 83, 915-920.
20. Yoshida,D. and Matsumoto.T. (1980) Amino-a-carbolines as mutagenic
agents in cigarette smoke condensate. Cancer Lett., 10, 14-140.
21.Manabe,S., Suzuki,H., Wada,O. and Ueki,A. (1993) Detection of the
carcinogen 2-amino- l-methyl-6-phenylimidazo[4,5-fc]pyridine (PhIP) in
beer and wine. Carcinogenesis, 14, 899-901.
22. Knize.M.G., Dolbeare^.A., Carroll.K.L., Moore.D.H. II and FeltonJ.S.
(1994) Effect of cooking time and temperature on the heterocyclic amine
content of fried beef. Food Chem. Toxicol., 7, 595-603.
23.Gross,G.A., Turesky,RJ., Fay,L.B., Stillwell.W.G., Skipper,P.L. and
Tannenbaum,S.R. (1993) Heterocyclic aromatic amine formation in grilled
bacon, beef and fish and in grill scrapings. Carcinogenesis, 14, 2313-2318.
Mutagenktty of PhIP and amlno(a)carboUne
24.GossenJ.A., de Leeuw.WJ.R, Tan.C.H.T., Lohman,P.H.M., Berends,F.,
Knook,D.L., Zwarthoff,E.C. and VijgJ. (1989) Efficient rescue of
integrated shuttle vectors from transgenic mice: a model for studying
mutations in vivo. Proc. Nail Acad. Sci. USA, 86, 7981-7985.
25.Kohler,S.W., Provost,G.S., Heck,K., KretzJU-., BuUock,W.O., SorgeJ.A.,
PutmanJ)!. and ShortJM- (1991) Spectra of spontaneous and mutageninduced mutations in the lacl gene in transgenic mice. Proc. Natl Acad.
Sci USA, 88, 7958-7962.
26.BurkhardJ.G., Winn,R.N., VanBeneden.R. and Malling,H.V. (1992) The
$X174 bacteriophage shuttle vector for the study of mutagenesis. Environ.
Mol. Mutagen., 19 (Suppl. 20). 8.
27.Brooks,R.A., GooderhamJvIJ., Zhao.K., Edwards.RJ., HowanU..A.,
BoobisAJ"and
WintonJDJ.
(1994)
2-Amino-l-methyl-6phenylimidazo[4,5-6]pyridine is a potent mutagen in the mouse small
intestine. Cancer Res., 54, 1665-1671.
28.JonesJ.M., Burkhart-Shultz,K. and Carrano,A.V. (1985) A method to
quantify spontaneous and in vivo induced thioguanine-resistant mouse
lymphocytes. Mutat. Res., 147, 97-105.
29.Dean3_J. and Senner.K.R. (1977) Detection of chemically induced somatic
mutation in Chinese hamsters. Mutat. Res., 46, 403-407.
3O.Maier,P. (1983) The granuloma pouch assay. In de SerresJU. (ed.),
Chemical Mutagens: Principles and Methods for their Detection. Plenum
Press, New York, vol. 8, pp. 233-260.
31.Winton.DJ., BlounUM.A. and PonderJ3.AJ. (1988) A clonal marker
induced by mutation in mouse intestinal epithelium. Nature, 333, 463-466.
32. Kohler.S.W., Provost,G.S., Kretz,P.L., Dycaico,MJ., SorgeJ.A. and
ShortJ.M. (1990) Development of a short-term, in vivo mutagenesis assay:
the effects of methylation on the recovery of a lambda phage shuttle
vector from transgenic mice. Nucleic Acids Res., 18, 3007-3013.
33.Tao,K.S., Urlando.C. and HeddleJ.A. (1993) Comparison of somatic
mutation in a transgenic versus host locus. Proc. Natl Acad. Sci. USA, 90,
10681-10685.
34.Skopek,T.R., Kort,KX. and MarinoJXR. (1995) Relative sensitivity of the
endogenous hprt gene and the lacl transgene in ENU-treated Big Blue™
B6C3F1 mice. Environ. Mol. Mutagen., 26, 9-15.
35.Heng,H.H.Q., Tsui,L.-C. and Moens,P.B. (1994) Organization of
heterologous DNA inserts on the mouse meiotic chromosome core.
Chromosoma, 103, 401-407.
36. HeddleJ.A., Tao.K., Swiger,R.R. and TuckerJ.D. (1995) The transmission
rate of the lacl transgene from the Big Blue™ mouse. Mutat. Res., 348,
63-66.
37. Shaver-Walker,P.M., Urlando.C., Tao.ICS., ZhangXB. and HeddleJ.A.
(1995) Enhanced somatic mutation rates induced in stem cells of mice by
low chronic exposures to ethylnitrosourea. Proc. Natl Acad. Sci. USA, 92,
11470-11474.
38.Tao,K,S. and HeddleJ.A. (1994) The accumulation and persistence of
somatic mutations. Mutagenesis, 9, 187-191.
39. HeddleJ.A., BoucnA, Khan,M.A. and GingerichJ.D. (1990) Concurrent
detection of gene mutations and chromosomal aberrations induced in vivo
in somatic cells. Mutagenesis, 5, 179-184.
4O.Zhang,X.B., Urlando.C., Tao,K.S. and HeddleJ.A. (1995) Factors affecting
somatic mutation frequencies in vivo. Mutat. Res., 338, 189—201.
41.Shephard,S.E., Lutz,W.K. and Schlatter.C. (1994) The lacl transgenic
mouse mutagenicity assay: quantitative evaluation in comparison to tests
for carcinogenicity and cytogenetic damage in vivo. Mutat. Res., 306,
119-128.
42. HeddleJ.A., Shaver-Walker,P, Tao.K.S. and Zhang,X.B. (1995) Treatment
protocols for transgenic mutation assays in vivo. Mutagenesis, 10,467-480.
43.Ronen,A. and HeddleJ.A. (1984) Site-specific induction of nuclear
anomalies (apoptotic bodies and micronuclei) by carcinogens. Cancer
Res., 44, 1536-1540.
44.AlexanderJ., Walling., Rossland.OJ., Solberg,K.E., Holme J.A.,
Becher.G., Anderssonjt. and Grivas.S. (1991) Formation of a glutathione
conjugate and a semistable transportable glucuronide conjugate of N2oxidized species of 2-amino-l-methyl-6-phenylimidazo[4^-ft]pyridine
(PhIP) in rat liver. Carcinogenesis, 12, 2239-2246.
45. Kaderlik,K.R., Mulder.GJ., Turesky.RJ., Lang,N.P. and Kadluber,F.F.
(1994) Glucuronidation of A/-hydroxy heterocyclic amines by human and
rat liver microsomes and formation of two distinct yV-glucuronides of NOH-PhlP in vivo and in vitro in the rat and dog. Carcinogenesis, 15,
1695-1701.
46. Kakiuchi.H.,
Watenabe,M.,
UsnijimaX,
Toyota>l.,
Imai.K.,
WeisburgerJ.H., Sugimura,T. and Nagao^i. (1995) Specific 5'-GGGA3'-»5'-GGA-3' mutation of the Ape gene in rat colon tumours induced
by 2-armno-l-methyl-6-phenylimidazo[4,5-i]pyridine. Proc. Natl Acad.
Sci. USA, 92, 910-914.
47. Kakiuchi M-, Ushijima,T., Ochiai,M., Imai JC, Ito ,N., Yachi A , Sugirnura,T.
and Nagao,M. (1993) Rare frequency of activation of the Ki-rai gene in
rat colon tumours induced by heterocyclic amines: possible alternative
mechanisms of human colon carcinogenesis. Mol Carcinogen., 8, 44-48.
48. HeddleJ.A. and Swiger,R.R. (1996) Risk estimation from somatic mutation
assays. Mutat. Res., in press.
Received on November 9, 1995; revised on June 13, 1996; accepted on June
17, 1996
2265