Mutagenesis vol.14 no.1 pp.113–119, 1999
A comparison of the effects of diverse mutagens at the lacZ
transgene and Dlb-1 locus in vivo
Lidia Cosentino and John A.Heddle1
Department of Biology, York University, Toronto, Canada M3J 1P3
Transgenic assays permit the detection of mutations in any
tissue, whereas endogenous mutations can be measured in
very few. For this reason comparisons between these loci
when both can be measured in the same cells are of
considerable interest. Previous comparisons have been
inconsistent: usually these loci have responded alike, however, in some cases the endogenous locus has been more
sensitive and at other times the transgenic locus has been
more sensitive. Here we report a comparison of the lacZ
transgene of the Muta™Mouse and the endogenous Dlb-1
gene in the epithelium of the small intestine after acute
exposure to seven mutagens. Benzo[a]pyrene, 5-bromo-29deoxyuridine, methyl methane sulphonate, ethyl methane
sulphonate, N-ethyl-N-nitrosourea, mitomycin C and Nmethyl-N-nitrosourea were all given by gavage to F1
(MutaMouse3SWR) mice. Mutations were quantified 2
weeks after the end of treatment. The data shows that all
of the agents induced similar mutant frequencies at the
Dlb-1 locus and at the lacZ transgene. The acute treatments
generally produced only modest increases in mutant frequency at both loci. The higher background frequency
observed at the lacZ transgene reduces the ability of the
transgenic assay to detect the same absolute increase in
mutant frequency.
Introduction
An assumption in the development and use of transgenic
assays is that mutations at these loci accurately reflect mutations
at endogenous loci. Nevertheless, these targets differ in several
ways. First, their sequences and location in the genome differ.
Secondly, the prokaryotic DNA is heavily methylated, is
non-transcribed and is embedded in viral DNA. Third, the
transgenes are usually present in multiple tandem copies.
Although some differences observed between the loci would
not be surprising, as all endogenous genes are not identical,
comparisons of mutations in endogenous genes and transgenes
in the same tissue are valuable. Previous studies have demonstrated that the lacI transgene and the endogenous Dlb-1
locus respond similarly in vivo after acute i.p. treatment with
N-ethyl-N-nitrosourea (ENU) but respond differently to X-rays
in the mouse small intestine (Tao et al., 1993; Tao and Heddle,
1994). A significant increase in the mutant frequency was
detected at the Dlb-1 locus after X-ray treatment, but at lacI
the increase in mutant frequency was only barely detectable.
This difference is probably due to the fact that X-rays produce
mostly deletions and deletions that extend into the λ vector
will not produce viable phage and will not be recovered. This
indicates that some classes of mutations or some classes of
1To
mutagens may not produce detectable increases in mutant
frequency at this transgene. The plasmid mouse was developed,
in part, to overcome this difficulty (Gossen et al., 1995).
Comparisons between the endogenous hprt locus and the lacI
transgene in splenocytes after acute exposure have given mixed
results depending on the mutagenic agent. The induction of
mutations by ENU is very similar (Skopek et al., 1995),
whereas benzo[a]pyrene [B(a)P] produces many more lacI
mutations than hprt mutations (Skopek et al., 1996). The hprt
data is complicated by the fact that the mutant frequency
varies with time after treatment and this time response is
age dependent (Jones et al., 1987; Walker et al., personal
communication). It is not known if the lacI mutant frequency
reflects the changing pattern with time found for hprt in these
cells. In this respect the comparisons in the small intestine are
simpler, since Dlb-1, lacI and lacZ frequencies each reach a
stable plateau soon after induction (Tao et al., 1993; Cosentino
and Heddle, 1996).
Previous studies have suggested that the response of endogenous genes and transgenes to different mutagenic agents
differ, thus we investigated the effects of various agents at the
Dlb-1 locus and lacZ transgene. Here we report comparisons
of the mutant frequencies at these two loci after acute exposure,
by gavage, to seven agents. The agents were selected according
to the different DNA alterations they create; the presumptive
repair pathways involved and whether the compound is a
direct acting or a metabolically activated agent (Table I). With
the use of various compounds, each inducing a different
spectrum of adducts, it is possible to investigate the mutagenic
effects of the various adducts in the DNA. N-Methyl-Nnitrosourea (MNU) was chosen because it is a strong mutagen
and carcinogen and it reacts directly with DNA producing
methylated bases. MNU methylates DNA at the N7 position
of guanine and at the O6 position of guanine residues (IARC,
1978). These adducts, after two rounds of replication, produce
GC→AT transitions. The responses of Dlb-1 and lacI were
comparable after treatment with ENU, thus it was of interest
to observe the effect of another ethylating agent. Ethyl methane
sulphonate (EMS) is also a direct alkylating agent which
generates a high level of nitrogen adducts compared with ENU
(IARC, 1974). Thus, an effect specific to O6-ethylguanine
would be observed with ENU but not with EMS. Methyl
methane sulphonate (MMS) was selected because it produces
primarily methylation at N7 of guanine (81–85% of total DNA
methylation), producing GC→AT transitions (Lawley, 1976).
In contrast, B(a)P and mitomycin C (MMC) were selected
because, unlike the other agents, they require metabolic activation. B(a)P is a potent mutagen and carcinogen in animals. Its
carcinogenic effects depend on host metabolic activation to
produce chemically reactive products such as B(a)P diol
epoxide, which is capable of forming bulky DNA adducts and
ultimately inducing GC→TA transversions (IARC, 1973).
MMC, on the other hand, is a bifunctional alkylating agent
whom correspondence should be addressed. Tel: 11 416 736 2100, ext. 33053; Fax: 11 416 736 5698; Email: [email protected] and [email protected]
© UK Environmental Mutagen Society/Oxford University Press 1999
113
L.Cosentino and J.A.Heddle
Table I. Summary of agents
Mutagen
Occurrence
Mode of action
Benzo[a]pyrene
B(a)P
Widely distributed in Metabolically
the environment
activated
DNA lesion
DNA repair
pathway
Metabolized to B(a)P diol expoxide, to form bulky DNA adducts,
inducing GC→TA transversions
Nucleotide excision
repair
5-Bromo-2Laboratory use only
deoxyuridine (BrdU)
Thymine analog
Induces base changes as it incorporates into DNA as a thymine analog Base excision
repair
Ethyl
Laboratory use only
methanesulphonate
(EMS)
Direct acting DNA
ethylating agent
Major site of ethylating is N7 position of guanine and N3 position of
guanine, inducing GC→AT and AT→GC transition mutations
Base excision
repair
N-Ethyl-Nnitrosourea (ENU)
Direct acting DNA
ethylating agent
Ethylates the O2 and O4 position of thymine and the O6 position of
guanine, inducing AT→GC and GC→AT repair transition mutations
Alkyl transferasemediated
Methyl methane
Laboratory use only
sulphonate (MMS)
Direct acting
methylating agent
Methylates N7 of guanine producing GC→AT transitions
Base excision
repair
N-Methyl-NLaboratory use only
nitrosourea (MNU)
Direct acting
methylating agent
Methylates N7 and O6 position of guanine, inducing GC→AT
transitions
Alkyl transferasemediated repair
Mitomycin C
(MMC)
Requires activation
Induces DNA crosslinks
Nucleotide excision
repair
Laboratory use only
Toxic antitumor
antibiotic
that can crosslink DNA and is a potent clastogen (IARC,
1976). Interstrand crosslinks represent an important class
of chemical damage, as they block DNA replication and
transcription.
Materials and methods
Animals
All homozygous lacZ (Dlb-1b/Dlb-1b) Muta™Mouse were obtained from
Hazleton Research Products Inc. (Denver, PA) and bred with non-transgenic
SWR (Dlb-1a/Dlb-1a) mice, obtained from The Jackson Laboratory (Bar
Harbor, ME) to produce the animals used here. The F1 animals, therefore,
were hemizygous for the lacZ transgene and heterozygous at the Dlb-1 locus
(Dlb-1b/Dlb-1a). The F1 animals were ~3 months of age and weighed between
20 and 31 g. Two females and three males were randomly assigned to each
treatment group and four males and four females to each control group. Mice
were housed in standard plastic cages with wood chip bedding. Mouse chow
and water was supplied ad libitum. An independent Animal Care Committee
approved all experimental protocols in advance.
Chemical treatment
All animals were treated by gavage. One group of animals were treated with
10, 50 or 100 mg/kg B(a)P. The second group was treated with either 2500
or 5000 mg/kg bromodeoxyuridine (BrdU). A third group was treated with
50, 100 or 250 mg/kg EMS. Another group was treated with 100 mg/kg ENU.
The MMC group received 1, 2 or 4 mg/kg and the MMS group was given an
acute treatment of 50, 100 or 150 mg/kg. The last group of animals was
treated with 50, 75 or 100 mg/kg MNU. Animals were sacrificed 2 weeks
after treatment.
Chemicals
B(a)P (CAS no. 50-32-8), ENU (CAS no. 759-73-9) and MNU (CAS no.
000684935) were purchased from Sigma Chemical Co. (St Louis, MO). BrdU
(CAS no. 59-14-3) and MMC (CAS no. 50-07-7) were purchased from
Boehringer Mannheim (Laval, Quebec, Canada). EMS (CAS no. 000062500)
was purchased from Aldrich (Milwaukee, WI) and MMS (CAS no. 000063273)
was purchased from Eastman Kodak Co. (Rochester, NY). After test solutions
were freshly prepared, all animals were treated with the appropriate dose of
the agent. Control animals were treated with a phosphate-buffered saline
(PBS) solution, also by gavage.
Tissue collection
All animals were sacrificed by cervical dislocation. The jejunal section of the
small intestine was reserved for the Dlb-1 assay, while the remainder of the
tissue was used for the lacZ assay. After the intestine was flushed with PBS
and inverted, it was placed in 3 ml KCl/EDTA solution and forced in and out
of a 5 ml needleless syringe. The cell suspension was stored at –70°C for
future use.
DNA isolation
Genomic DNA was purified from the cell suspension with a proteinase K
solution (2 mg/ml) for 3 h at 55°C, followed by phenol–chloroform (1:1)
114
extraction and precipitation with ethanol as described by Kohler et al. (1990).
The precipitated DNA was spooled onto a hooked glass Pasteur pipette, air
dried and dissolved in Tris–EDTA buffer. The concentration of DNA was
determined spectrophotometrically at 260 nm.
DNA packaging of lacZ mutations
The λ phage shuttle vector, which contains the entire lacZ target gene, was
recovered by in vitro packaging with Transpack™ packaging extract (Strategene,
La Jolla, CA), under conditions recommended by Hazleton Research Products
Inc. (Denver, PA). Briefly, 8 µl genomic DNA (1–2 mg/ml) was used in each
packaging reaction and incubated for a total of 3 h at 30°C. The reaction was
terminated by dilution with 470 µl phage buffer. Approximately 500 µl
packaged phage were incubated in 2 ml bacterial suspension at room
temperature for 20–30 min. A 5 µl aliquot was diluted in 100 µl LB broth
containing 10 mM MgSO4 for a concurrent titer on non-selective agar plates.
The remaining phage/bacteria mixture was mixed with ~8 ml freshly prepared
top selection agar supplemented with 0.3% phenyl-β-D-galactopyranoside
(P-gal) (Sigma). P-gal is used as a selective agent for bacteria infected with
phage containing non-functional lacZ genes. P-gal in the selection plates
prevents the formation of wild-type plaques and thus the plaques observed
represent lacZ mutants (Gossen et al., 1992). After incubation overnight at
37°C, mutant and non-mutant plaques were scored. The number of plaques
recovered from each animal varied, with a mean of ~200 000 plaques and a
range of 47 000–1 575 000 plaques.
Dlb-1 assay
Whole mounts of the small intestine were prepared as described by Winton
et al. (1988). Briefly, the small intestine was divided into its three sections
(duodenum, jejunum and ileum) and flushed clean with PBS. The jejunum
was used for this assay while the remaining sections of the intestine were
used for the transgenic assay. After the jejunum was flushed with 10% formal
saline, one end was sealed between two microscope slides and clipped. The
intestine was inflated with 10% formal saline using a blunt end needle and
fixed for ~3 min. It was cut along the mesenteric side, placed on a microscope
slide, villi side up, stretched and held in place by plastic-coated paper clips
under which a small piece of coverslip was placed. The slides were placed in
10% formal saline to fix for at least 1 h, at which time they were then rinsed
with PBS and incubated overnight in 20 mM DL-dithiothreitol (Sigma)
dissolved in 20% ethanol, 150 mM 80% Tris (pH 8.2). Mucus was removed
by pipetting the solution over the intestinal tissue. Before staining, slides
were rinsed three times with PBS and incubated in 0.1% phenylhydrazine
hydrocholoride (Sigma) in PBS for 30 min to block endogenous peroxidases.
After three washes with PBS and a 10 min incubation in PBS containing
0.5% albumin (fraction V; Boehringer), the slides were stained with the
Dolichos biflorus agglutinin–peroxidase conjugate (Sigma) at 5 µg/ml in PBS/
albumin (fraction V). The peroxidase was developed using 3,39-diaminobenzidine (Sigma) solution for 45 min. The slides were rinsed twice with PBS and
stored in 10% formal saline until analyzed. The slides were scored with a
dissecting microscope at 503 magnification. The Dlb-1b/Dlb-1a epithelial
cells stain dark brown; mutant cells, which have no lectin-binding ability,
appear as unstained white vertical ribbons on the villus. The number of villi
A comparison of the lacZ transgene and Dlb-1 locus in vivo
Fig. 1. Mutant frequency observed after acute treatment by gavage with (a) B(a)P, (b) BrdU, (c) EMS, (d) MMC, (e) MMS and (f) MNU at (u) the Dlb-1
locus (–SEM) and (s) lacZ transgene (1SEM).
Fig. 2. Mutant frequencies detected at the Dlb-1 locus and the lacZ
transgene after acute treatment are highly correlated for all mutagens tested.
scored per animal was estimated from duplicate counts of the number of villi
in the first and last field. Each field contained ~200 villi, to yield an average
of 9000 villi/animal. Since there are ~10 stem cells/villus (Cosentino et al.,
1996), ~90 000 stem cells were analyzed per animal.
Results
Individual animal data and mutant frequencies are presented
in Table II. The majority of agents produced only a marginal
increase in mutant frequencies at either locus. MMC proved
to be the weakest mutagen, whereas MNU was the most
mutagenic agent, inducing a 200- to 400-fold increase in both
Dlb-1– and lacZ– mutations.
The dose–response data were tested for linearity using the
SAS statistical test for regression analysis. Only treatment
with B(a)P, MMS and MNU resulted in a dose-related increase
in mutant frequency for both loci (Table III). Treatment with
BrdU resulted in a dose-related increase in Dlb-1– mutants but
not lacZ– mutants (Table III).
Because not all dose–response curves are linear, the mutant
frequencies of the highest dose was tested for significance
against controls by one-way analysis of variance (ANOVA)
using Microsoft® Excel 97. EMS (F 5 64.41, P , 0.05) and
MMC (F 5 8.74, P 5 0.02) induced a significant increase in
Dlb-1 mutations at the highest dose, whereas a significant
increase in lacZ mutations was not detected. The higher
background mutant frequency observed at the lacZ locus (3.1
6 1.2) compared with that of the Dlb-1 locus (1.2 6 0.5)
reduces the ability of the transgenic assay to detect small
increases, as discussed later. Despite the different spontaneous
mutation frequencies, the induced mutant frequencies at the
Dlb-1 locus and lacZ transgene were not significantly different
and were nearly identical after treatment with B(a)P (P 5
0.93), BrdU (P 5 0.97), EMS (P 5 0.85), MNU (P 5 0.87),
MMC (P 5 0.86) and MMS (P 5 0.74) (Figure 1). It is
interesting to note that although each mutagen produces a
distinct spectrum of mutations, resulting from the specificity
of DNA binding and the type of DNA repair involved, all of
the agents induced similar mutant frequencies at both loci.
115
L.Cosentino and J.A.Heddle
Table II. Individual animal data for the endogenous Dlb-1 locus and the lacZ transgene
Treatment
group
Dose
(mg/kg)
No. of
villi
Dlb-1
No. of
mutants
Control
B(a)P
BrdU
EMS
Mut. freq.a
(105)
8213
6956
7301
9020
9453
11 038
8220
8369
2
0
2
0
1
1
1
1
2
0
3
0
1
1
1
1
6490
10 440
8155
9486
7420
7
6
2
2
2
11
6
3
2
3
50
9826
10 863
7290
15
8
4
15
7
6
100
9984
8568
9025
11 224
9366
3
10
16
16
15
3
12
18
14
16
2500
9072
8910
8094
8959
8283
2
1
7
3
1
2
1
9
3
1
5000
8280
8085
8008
9000
7224
3
8
5
6
7
4
10
6
7
10
50
10 373
10 229
10 763
9698
10 010
8
1
1
3
1
8
1
1
3
1
10
7508
10 773
3981
1
2
3
1
2
10
250
8856
8401
10 868
5720
8282
5
4
6
3
4
6
5
6
5
5
10
aMutant frequency per mutable locus.
bNumber of plaque-forming units
lacZ
No. of
mutants
Mut. freq.
(105)
Mean 6 SE
155 000
112 500
52 500
1 575 000
145,000
125 500
75 000
190 000
3
10
2
23
3
3
2
3
2
9
4
2
2
2
3
2
4.8 K 0.5
390 000
57 500
127 500
65 000
105 000
18
2
7
2
3
5
4
6
3
3
3.9 K 0.5
9.4 K 2.3
42 000
157 500
62 500
4
13
7
10
8
11
9.7 K 0.8
1.2 K 0.5
12.5 K 2.6
3.3 K 1.4
7.2 K 1.6
2.8 K 1.3
167 500
165 000
87 500
62 500
120 000
185 000
132 500
147 500
450 000
17
10
10
6
17
19
15
24
no data available
2
8
9
7
19
2
4
7
5
4
80 000
505 000
47 500
280 000
7
9
16
3
4
8
10
4
no data available
337 500
135 000
65 000
152 000
25
7
3
2
1
2
3
2.0
no data available
4.4 K 2.8
332 500
270 000
100 000
5.2 K 0.2
457 500
80 000
117 500
172 500
15
14
3
3.1 K 1.2
14.9 K 4.1
4.4 K 1.1
6.0 K 1.5
3.3 K 1.4
5
5
3
4.2 K 0.6
29
6
5
6
6
5
10
6
no data available
5.9 K 0.3
There are 10 stem cells/villus (Cosentino et al., 1996).
The large error bars seen for the 100 mg/kg EMS treatment
group are partly a result of a small sample size, as two males
died, leaving only one male and two females for that treatment
group (Figure 1c).
Discussion
The similarity in the response of the lacZ transgene and the
Dlb-1 gene is surprising (Figure 1). The induced mutant
frequencies detected at the Dlb-1 locus and the lacZ transgene
116
Mean 6 SE
No. of
pfub
for all the mutagens tested are highly correlated (Figure 2).
The correlation derived here implies that for an increase in
lacZ– mutants, the number of Dlb-1– mutants also increases.
The almost perfect quantitative agreement must be regarded
as a matter of chance, for other endogenous loci and other
transgenes can show different mutant frequencies. That there
are exceptions is shown by the differential response of lacI
and Dlb-1 to X-rays, with Dlb-1 being the more sensitive
locus (Tao et al., 1993). The sensitivity of any locus to
mutation depends on a number of factors. Clearly the larger
A comparison of the lacZ transgene and Dlb-1 locus in vivo
Table II. Continued
Treatment
group
Dose
(mg/kg)
No. of
villi
Dlb-1
No. of
mutants
MMC
1
8708
5409
8740
8878
7533
5
1
2
3
1
1
2
2
3
1
9065
8752
7227
3738
9377
1
4
1
1
6
1
1
1
3
6
6859
5145
11 155
5318
7271
1
2
5
2
5
2
4
5
4
7
8316
11 532
9647
8022
7021
3
1
4
1
1
4
1
4
1
1
6128
9084
9500
7680
8895
3
2
0
2
2
5
2
0
3
2
8966
8976
7592
9366
7668
2
1
4
10
5
2
1
5
11
7
50
4450
6851
3973
4200
73
151
124
93
164
220
312
221
75
3973
10 894
6650
6660
124
240
163
178
312
220
245
267
100
3575
4606
4272
5606
196
191
227
155
548
415
531
277
2
4
MMS
50
100
150
MNU
Mut. freq.a
(105)
aMutant frequency per mutable locus.
bNumber of plaque-forming units
Mean 6 SE
No. of
pfub
lacZ
No. of
mutants
Mut. freq.
(105)
Mean 6 SE
145 000
310 000
212 500
3
2
7
2
4
2
no data available
no data available
2.1 K 0.1
3.2 K 1.0
87 500
177 500
210 000
132 500
3
3
7
4
6
3
3
2
no data available
3.1 K 0.3
4.1 K 0.7
212 500
182 500
122 500
160 000
9
4
8
4
5
4
6
4
no data available
4.1 K 0.1
2.3 K 0.7
97 000
132 500
50 000
52 500
2
5
1
1
no data avalable
2.5 K 0.5
2.4 K 0.8
165 000
327 500
95 000
127 500
7
4
15
5
5
5
3
2
no data available
4.1 K 0.6
5.2 K 1.7
77 500
327 500
217 500
177 500
5
6
17
5
13
6
10
6
no data available
5.8 K 0.3
50 000
87 500
30 000
110 000
111
171
74
259
222
195
247
235
302 500
340 000
20 000
100 000
859
990
61
232
284
291
305
232
155 000
200 000
52 500
296 000
669
854
298
876
432
427
568
297
2.9 K 0.8
229.5 K 30.6
302.5 K 25.5
442.7 K 62.9
2
4
2.0
2
224.9 K 11.0
278.0 K 16.0
430.8 K 55.3
There are 10 stem cells/villus (Cosentino et al., 1996).
the mutable target, the more likely mutations are to arise,
although the mutable target may not be in proportion to the
size of the locus. The φX174 transgenic mouse, for example,
has a mutable target of 1 bp within the 5 kb gene sequence
(Burkhart et al., 1993). Secondly, the nature of the mutation
screen influences the size of the mutable target and thus the
mutation frequency. If the selection is very rigorous, then only
the more extreme mutants will be detected. lacI mutants, for
example, vary in the intensity of the blue color on selection
plates with X-gal. The nature of the agar and the amount of
X-gal in the plates will influence the fraction of lacI mutations
detectable and thus the mutation frequency. This is as true of
endogenous loci as transgenic ones. Ouabain resistance in
mammalian cells is much less frequent than thioguanine
resistance (Baker et al., 1979). In the case of ouabain, the
enzyme is essential and the mutation is dominant. Probably
the mutational target is a few base pairs at most and only base
pair substitutions are recoverable. Thioguanine resistance can
arise by deletion of the whole gene, by single base pair
substitutions and by frameshift mutations at numerous sites in
the hprt gene. This provides a satisfactory explanation for the
~100-fold difference in mutation rates at these loci.
Differences between loci are much more likely than similarities. Nevertheless, the lacI transgene (1 kb), about one third
117
L.Cosentino and J.A.Heddle
Table III. Summary of statistical analysis
Treatment group
Loci
F
df
P value1
Outcome
B(a)P
Dlb-1
lacZ
22.35
8.22
3
3
,0.001
0.002
dose-related in MF2
dose-related in MF
BrdU
Dlb-1
lacZ
7.26
2.22
2
2
0.009
0.148
dose-related in MF
no dose-related increase in MF
EMS
Dlb-1
lacZ
2.32
1.75
3
3
0.123
0.199
no dose-related increase in MF
no dose-related incrfease in MF
MMC
Dlb-1
lacZ
2.39
0.81
3
3
0.109
0.507
no dose-related increase in MF
no dose-related increase in MF
MMS
Dlb-1
lacZ
3.46
4.59
3
3
0.043
0.014
dose-related increase in MF
dose-related increase in MF
MNU
Dlb-1
lacZ
24.96
65.37
3
3
,0.001
,0.001
dose-related increase in MF
dose-related increase in MF
1Probability
2MF,
due to chance alone. P values were determined by linear regression analysis using SAS.
mutant frequency.
the length of the lacZ transgene (3 kb), the hprt endogenous
gene (~1 kb of coding sequence) and Dlb-1 (of unknown size)
all show very similar mutant frequencies after ENU treatment
(Tao et al., 1993; Skopek et al., 1995; Cosentino and Heddle,
1996). Other comparisons are few and involve lacI and hprt.
One chemical, B(a)P, produces many more lacI mutations than
hprt mutations (Skopek et al., 1996). It is noteworthy that,
unlike Dlb-1, the hprt mutant frequency changes with time,
rising to a maximum and falling thereafter. This makes
comparisons difficult, especially as the lacI time course has not
always been determined and might differ. These comparisons
indicate that age, manifestation time and the kinetics of the
lymphoid system, in addition to treatment effects of various
chemical agents, influence the frequency of hprt– mutations.
Clearly, more studies are needed, but the Dlb-1 comparisons
suggest that the transgene does not usually behave anomalously.
The data reported here shows that MNU, B(a)P and MMS
induced a significant dose-related increase in Dlb-1– and lacZ–
mutant frequencies (Table II). In addition to EMS (250 mg/kg),
MMC (4 mg/kg) and BrdU (5000 mg/kg) induced a significant
increase in Dlb-1– mutations (P , 0.05, P 5 0.02 and P 5
0.003, respectively) but not in lacZ– mutations. The lack of
significant increases in lacZ– mutations is probably due to a
higher background mutant frequency observed at the lacZ
locus (3.1 6 1.2) compared with that at the Dlb-1 locus (1.2 6
0.5). Thus, the induced mutant frequency (fold increase)
was significantly lower for lacZ than for Dlb-1. The lower
spontaneous mutant frequency in the Dlb-1 assay allowed this
test to detect smaller absolute numbers of mutations.
The spontaneous frequency influences the sensitivity of
these assays, as pointed out by Skopek et al. (1995). Induced
mutations arise as an absolute increase in mutant frequency,
so the lower the spontaneous frequency, the more sensitive
the assay is to the same increase. This is apparent in our
studies with these mutagens, most of which produced rather
small increases in mutant frequency. The increases are remarkably similar at the lacZ and the Dlb-1 loci, but the Dlb-1 data
are much more significant because of the lower spontaneous
mutant frequency at this locus. This was also observed in a
study when animals were treated with cyclophosamide (CP).
CP-induced mutations were detected at the hprt locus but not
in the lacI transgene in splenic T cells (N.J.Gorlick, personal
118
communication). This fully confirms the prediction of Skopek
et al. (1995). Thus the high spontaneous frequency reduces
the sensitivity of the transgenic assays. The solution that has
been proposed is to increase the exposure by using subacute
and chronic protocols as recommended by Shephard et al.
(1994), Tao et al. (1994) and Heddle et al. (1995). Evidence
supporting this solution is accumulating rapidly. Suzuki et al.
(1993), for example, have shown that multiple treatments with
MMC resulted in a 2-fold increase in lacZ mutations in the
bone marrow, whereas a single acute dose did not induce a
significant increase in lacZ mutations.
It should be mentioned that the Dlb-1 locus and the lacI
transgene respond differently to chronic treatment, i.e. when
animals are treated chronically with ENU there is an initial
deficiency of Dlb-1 mutations relative to lacI mutations. With
continued treatment, lacI mutations accumulate linearly for 90
days. In contrast, the Dlb-1 mutants accumulate more slowly
at first and then at an accelerated rate (Shaver-Walker et al.,
1995). This is also true for the lacZ transgene (Cosentino and
Heddle, in preparation). In addition to treatment protocols, we
have observed that the induced mutant frequency is greatly
influenced by the route of mutagen administration. When
animals are treated with ENU (100 mg/kg) by gavage, a lower
mutant frequency is induced at the Dlb-1 locus when compared
with animals treated i.p. (503105 versus 803105, respectively).
It is evident that there are many factors that can affect the
induction of mutations in animals. Although Dlb-1 and lacZ
respond similarly to acute doses, the exposure given may not
accurately reflect all of the biological phenomena occurring
in vivo, as observed by the differential mutation frequencies
obtained during a chronic dosing regime. Like Shephard et al.
(1994), we believe that a subacute or chronic treatment protocol
would likely maximize the sensitivity of the assays (Heddle
et al., 1995). It is evident from this study that in addition to
treatment protocol, background mutant frequency, mode of
mutagen administration, selection of test substances and dose
level influence the detection limit of these mutational assays.
Although the majority of agents induced similar mutant frequencies at the two loci, they were not very mutagenic in the
small intestine. It is possible that each agent may be tissuespecific, species-specific or strain-specific. Obviously, one must
A comparison of the lacZ transgene and Dlb-1 locus in vivo
consider all influential factors when designing experiments in
order to maximize the power of the assay.
Winton,D.J., Blount,M.A. and Ponder,B.A.J. (1988) A clonal marker induced
by mutation in mouse intestinal epithelium. Nature, 333, 463–466.
Accepted on July 29, 1998
Acknowledgements
This work was supported by a grant from the National Cancer Institute of
Canada. We thank HRP Inc. for permission to cross the Muta™Mouse to
produce the F1 necessary for this study. We also thank Cesare Urlando and
Jennifer Moody who helped with the experiments. We are also grateful to
Jason Belas for his helpful comments and suggestions.
References
Baker,R.M., Van Voorhis,W.C. and Spencer,L.A., (1979) HeLa cell variants
that differ in sensitivity to monofunctional alkylating agents, with
independence of cytotoxic and mutagenic responses. Proc. Natl Acad. Sci.
USA, 76, 5249–5253.
Burkhart,J.G., Burkhart,B.A., Sampson,K.S. and Malling,H.V. (1993) ENUinduced mutagenesis at a single A:T base pair in transgenic mice containing
φX174. Mutat. Res., 292, 69–81.
Cosentino,L. and Heddle,J.A. (1996) A test for neutrality of mutations of the
lacZ transgene. Environ. Mol. Mutagen., 28, 313–316.
Cosentino,L., Shaver-Walker,P. and Heddle,J.A. (1996) The relationship among
stem cells, crypts and villi in the small intestine of mice as determined by
mutation-tagging. Dev. Dyn., 207, 420–428.
Gossen,J.A., Molijn,A.C., Douglas,G.R and Vijg,J. (1992) Application of
galactose-sensitive E.coli strains as selective hosts for lacZ– plasmids.
Nucleic Acids Res., 20, 3254.
Gossen,J.A., Martus,H.J., Wei,J.Y. and Vijg,J. (1995) Spontaneous and X-ray
induced deletions in a lacZ plasmid-based transgenic mouse model. Mutat.
Res., 331, 89–97.
Heddle,J.A., Shaver-Walker,P., Tao.,K.S. and Zhang,X.B. (1995) Treatment
protocols for transgenic mutation assays in vivo. Mutagenesis, 10, 467–470.
IARC (1973) Monographs on the Evaluation of the Carcinogenic Risk of
Chemicals to Humans. Benzo(a)pyrene. IARC Scientific Publication no. 3.
IARC, Lyon, France, pp. 91–158.
IARC (1974) Monographs on the Evaluation of the Carcinogenic Risk of
Chemicals to Humans. Ethyl Methanesulphonate. IARC Scientific
Publication no. 7. IARC, Lyon, France, pp. 245–252.
IARC (1976) Monographs on the Evaluation of the Carcinogenic Risk of
Chemicals to Humans. Mitomycin C. IARC Scientific Publication no. 10.
IARC, Lyon, France, pp. 171–179.
IARC (1978) Monographs on the Evaluation of the Carcinogenic Risk of
Chemicals to Humans. N-Nitroso-N-mehylurea. IARC Scientific Publication
no. 17. IARC, Lyon, France, pp. 227–255.
Jones,I.M., Burkhart-Schultz,K., Strout,C.L. and Crippen,T.L. (1987) Factors
that affect the frequency of thioguanine-resistant lymphocytes in mice
following exposure to ethylnitrosourea. Environ. Mutagen., 9, 317–329.
Kohler,S.W., Provost,S., Kretz,P.L., Dycaico,M., Sorge,J.A and Short,J.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.
Lawley,P.D. (1976) Carcinogenesis by alkylating agents. In Searle,C.E. (ed.),
Chemical Carcingenesis, ACS Monograph 173. American Chemical Society,
Washington, DC, pp. 83–244.
Shaver-Walker,P.M., Urlando,C., Tao,K.S., Zhang,X.B. and Heddle,J.A. (1995)
Enhanced somatic mutation rates induced in stem cells by low chronic
exposures to ethylnitrosourea. Proc. Natl Acad. Sci. USA, 92, 11470–11474.
Shepard,S.E., Lutz,W.K. and Schlatter,C. (1994) The lacI transgenic mouse
mutagenicity assay: quantitative evaluation in comparison to tests for
carcinogenicity and cytogenetic damage in vivo. Mutat. Res., 306, 119–128.
Skopek,T.R., Kort,K.L. and Marino,D.R. (1995) Relative sensitivity of the
endogenous hprt gene and the lacI transgene in ENU-treated Big Blue
B6C3F1 mice. Environ. Mol. Mutagen., 26, 9–15.
Skopek,T.R., Kort,K.L., Marino,D.R., Mittal,L.V. Umbenhauer,D.R.,
Laws,G.M and Adams,S.P. (1996) Mutagenic response of the endogenous
hprt gene and lacI transgene in benzo(a)pyrene-treated Big Blue B6C3F1
mice. Environ. Mol. Mutagen., 28, 376–384.
Suzuki,T., Hayashi,M., Sofuni,T. and Myhr,B.C. (1993) The concomitant
detection of gene mutation and micronuclus induction by mitomycin C
in vivo using lacZ transgenic mice. Mutat. Res., 285, 219–224.
Tao,K.S. and Heddle,J.A. (1994) The accumulation and persistence of somatic
mutations in vivo. Mutagenesis, 9, 187–191.
Tao,K.S., Urlando,C. and Heddle,J.A. (1993) Comparison of somatic mutation
in a transgenic versus host locus. Proc. Natl Acad. Sci. USA, 90, 10681–
10685.
119