[CANCER RESEARCH 50, 5245-5249. September 1, 1990]
DNA Fingerprinting of 7,12-Dimethylbenz[a]anthracene-induced and Spontaneous
CD-I Mouse Liver Tumors
Brian J. Ledwith, Richard D. Storer, Srinivasa Prahalada, Sujata Manam, Karen R. Leander,
Matthew J. van Zwieten, Warren W. Nichols, and Matthews O. Bradley1
Merck Sharp and Dohme Research Laboratories, West Point, Pennsylvania 19486
ABSTRACT
Determining to what degree chemicals and environmental agents con
tribute to the development of cancer would be materially enhanced by
the ability to distinguish chemically induced tumors from those that arise
spontaneously. Using DNA fingerprinting as an assay, we investigated
whether somatic DNA rearrangements are more frequent in chemically
induced mouse liver tumors than they are in spontaneous mouse liver
tumors. Tumors were induced by a single i.p. injection of 12-day old male
Crl:CD-l(ICR)BR (CD-I) mice with 20 nmol/g 7,12-dimethylbenz|a]anthracene and were harvested 9 to 12 months after injection. Sponta
neous tumors were obtained from 94- to 98-week old male CD-I mice.
We detected 8 rearrangements in 14 7,12-dimethylbenz|a]anthraceneinduced tumors, which corresponds to a high rearrangement frequency of
about 2% (of the minisatellite bands examined). Furthermore, 6 of these
rearrangements included complete band losses which must have occurred
early in tumor development. However, only 2 band changes were observed
in 15 spontaneous tumors, and both changes were intensity shifts which
may represent rearrangements that occurred later during tumor progres
sion. Histological examination showed that the higher frequency of
rearrangements in 7,12-dimethylbenz|a]anthracene-induced tumors ver
sus spontaneous tumors was not related to differences in the degree of
tumor progression or malignancy. Our results suggest that DNA finger
printing may be a valuable assay for differentiating certain chemically
induced tumors from spontaneous tumors.
INTRODUCTION
The carcinogenic potential of chemicals is generally assessed
in part by in vivo tumorigenicity assays in rodents. However,
several of the mouse strains or stocks commonly used in chem
ical tumorigenicity assays, including C57BL/6 x C3H FI (here
after called B6C3F,) (1) and Crl:CD-l(ICR)BR (referred to as
CD-I) mice (2), can exhibit high and variable incidences of
spontaneous tumors which can complicate the evaluation of
tumorigenicity data. For example, in some studies chemically
treated animals have had a tumor incidence that was higher
than that in concurrent control animals, but still within the
limits of historical controls (2). Our ability to evaluate the
carcinogenic effect of a chemical in such studies would be
greatly enhanced by the development of methods that could
molecularly distinguish spontaneous from chemically induced
tumors.
Previous studies in B6C3F| mice have suggested that some
chemically induced tumors could be differentiated from spon
taneous tumors by determining the specificity of the carcino
gens for inducing particular types of point mutations in the ras
genes (3-6). Several chemicals (3,4) have been shown to induce
liver tumors that have a specific distribution of ras point mu
tations different from those observed in spontaneous liver tu
mors (3). However, other chemically induced tumors are similar
to spontaneous tumors in their spectrum of ras point mutations
(4, 6), and not all tumors (spontaneous or chemically induced)
Received 1/5/90; revised 4/17/90.
The costs of publication of this article were defrayed in part by the payment
of page charges. This article must therefore be hereby marked advertisement in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1To whom requests for reprints should be addressed.
contain activated ras genes (3-7). These facts, together with the
generally accepted concept of tumor development as being a
multistep process, suggest that additional methods of analyzing
tumor DNA for the action of carcinogens will be needed for
optimum differentiation of chemically induced from sponta
neous tumors.
In this study, we have used DNA fingerprinting to investigate
whether somatic DNA rearrangements are more frequent in
CD-I mouse liver tumors induced by DMBA2 than they are in
spontaneous liver tumors. DNA fingerprinting is a sensitive
assay for genomic rearrangements because a single probe can
simultaneously detect up to 40 different minisatellites (8, 9),
which are well dispersed throughout the genome (10) and
known to be hot spots for recombination (11, 12). DNA finger
printing has previously been used to assay for somatic DNA
rearrangements in human tumors (13). However, our study
represents the first comparison of chemically induced tumors
with spontaneous tumors and indicates that a carcinogen can
induce a much higher frequency of rearrangement than that
found in spontaneous tumors.
MATERIALS
AND METHODS
CD-I Mouse Liver Tumors. Male CD-I mice (Charles River Breeding
Laboratories, Wilmington, MA), 12 days old, were given a single i.p.
injection of DMBA (20 nmol/g body weight in 10% dimethyl sulfoxide
in trioctanoin). Liver tumors were harvested during two necropsies, 9
and 12 months after injection. Thirty-one of 38 DMBA-treated mice
survived until the 9- or 12-month necropsies, and 74% of the surviving
mice had a total of 226 hepatocellular tumors. In contrast, vehicletreated mice, used as concurrent controls, had a total of only 9 tumors
(in 3 of 38 mice). Therefore, nearly all of the tumors from the DMBAtreated animals were induced by the single dose of DMBA and were
not spontaneous. Since very few spontaneous tumors were present in
the vehicle-treated concurrent control animals, and since those present
were too small to analyze, we isolated spontaneous liver tumors from
an aging colony of male CD-I mice that were approximately 2 years
(94 to 98 weeks) old. The 9- to 12-month old DMBA-treated mice and
the 2-year-old aging mice thus provided us with essentially pure popu
lations of DMBA-induced tumors and spontaneous tumors, respec
tively. For this study, 14 of the DMBA-induced tumors and 15 of the
spontaneous tumors were analyzed by histology and DNA fingerprint
ing.
A section through each tumor and adjacent normal tissue was ex
amined histologically. A portion of the tumor was trimmed of surround
ing normal tissue and frozen in liquid nitrogen for subsequent DNA
preparation. In addition, normal liver tissue from each mouse was also
frozen for DNA preparation. Long-term storage of tissue samples was
at -70'C.
DNA Isolation. High molecular weight DNA was purified from
pulverized frozen tissue by lysis in 4 M guanidinium isothiocyanate
solution, CsCl gradient centrifugation, treatment with both proteinase
K and RNase A, phenol-chloroform extraction, and ethanol precipita
tion, essentially as described by others (14).
DNA Fingerprinting Analysis. Ten ¿igof each DNA sample was
2The abbreviations used are: DMBA, 7.l2-dimethylbenz|a]anthracene;
SSC.
standard saline-citrate or 150 mM NaCI/15 mvi sodium citrate: SDS. sodium
dodecyl sulfate; MHC, major histocompalibility complex.
5245
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1990 American Association for Cancer Research.
DNA FINGERPRINTING
OF CD-I MOUSE LIVER TUMORS
cleaved with //uri 11.separated in 0.8%-agarose gels (20 cm long) at 60
V for about 26 h, and then transferred to Gene Screen Plus (Du Pont,
Wilmington, DE) membranes in 0.4 M NaOH. The membranes were
prehybridized for 3 h at 60"C in 6x SSC, 5x Denhardt's solution (0.2%
Ficoll-0.2% polyvinylpyrrolidone-0.2% bovine serum albumin), and 1%
sodium dodecyl sulfate (SDS). Hybridization was for 16 h at 60°Cin
prehybridization mix containing 10% dextran sulfate and 5 x IO5cpm
(Cerenkov) of "P-labeled nick-translated probe (see below). Following
hybridization, the membranes were washed three times for 15 min each
at room temperature in Ix SSC/1% SDS, twice for 30 min each at
60"C in Ix SSC/1% SDS, and then three times for 15 min each at
room temperature in Ix SSC. Autoradiographic exposure of Kodak XOmat XAR-5 film was for 1 day at -70"C. All DNA samples were
analyzed in duplicate, and only reproducible band changes were scored
as positive changes.
The probe used in this study is a synthetically generated tandemrepetitive DNA fragment1 based on a minisatellite that is present in the
mouse major histocompatibility complex (12). The tandem-repetitive
fragment, referred to as the M-core fragment, is 208 base pairs long
and consists of a tandem array of (AGGC)n repeats (n = 52). The
AGGC repeats and the mouse major histocompatability complex min
isatellite are 75% homologous to a consensus core sequence found in
human minisatellites (9). We constructed a plasmid, pGEM3Z-Mcore,
which consists of the M-core fragment cloned into the Smal site of the
pGEM3Z plasmid (Promega Corp., Madison, WI). For nick-transla
tion, M-core fragments were excised from pGEM3Z-Mcore plasmids
by codigestion with BamH\ and EcoRl and then purified by agarosegel electrophoresis. The M-core probe has been shown to produce
individual-specific DNA fingerprints in several species, including mice
and humans.3
Densitometry. Partial changes in the intensity of bands were quantitated using a Molecular Dynamics 300A Computing Densitometer
(Molecular Dynamics, Sunnyvale, CA). Total band intensity was meas
ured (by volume integration) after the adjacent background intensity
was subtracted, and then the band volume was normalized against
another band in the same lane that does not undergo an intensity
change.
RESULTS AND DISCUSSION
DNA Fingerprinting as an Assay for Somatic DNA Re
arrangements. Because minisatellites are highly polymorphic
among individuals, it is critical that normal tissue from the
same animal be used for comparison with the tumor material.
A rearrangement that alters the size of a minisatellite can be
detected by a change in the migration of its band in the tumor
DNA fingerprint relative to the normal tissue DNA fingerprint
(referred to as a band shift). It is possible that for a single
rearrangement, such as a deletion causing a decrease in band
size, two band changes could be detected, including the loss of
the original band and the gain of the smaller band. However,
sometimes the original band or the rearranged band may mi
grate in an unresolvable portion of the DNA fingerprint, such
that only one band change (a band gain or a band loss, respec
tively) is detected. Since DNA fingerprinting probes are not
loci specific, we cannot determine whether coincidental band
changes involving the gain of one band and loss of another band
represent a single rearrangement in one minisatellite or two
independent rearrangements in different minisatellites. Cloning
out the rearranged minisatellites for use as loci-specific probes
would be required to determine an identity between coincidentally changing bands. For consistency, we have chosen the
conservative approach of scoring only one rearrangement for
each coincident band loss and band gain, and we henceforth
3 B. J. Lcdwith. S. V. Manam, W. W. Nichols, and M. O. Bradley. Preparation
of synthetic, tandem-repetitive probes for DNA fingerprinting. Biotechniques. 9:
5-7, 1990.
refer to these possibly related changes as a "putative size shift."
Southern analysis is subject to several factors that can lead
to inconsistencies in the intensity of bands, including DNA
quality, membrane imperfections, unequal transfer throughout
the membrane, and variations in the concentration of blocking
agents bound to the membrane. DNA fingerprinting can be
very sensitive to such artifacts because many more bands are
examined per lane and because it is carried out under less
stringent hybridization conditions than in standard Southern
analysis. Therefore, we have chosen several conservative guide
lines to ensure that only real band changes are scored: (a) all
DNA samples are analyzed in duplicate, and only reproducible
band changes are scored as positive changes; (b) if differences
in DNA quality could possibly account for an observed band
alteration, then the band change is not scored; (c) densitometry
is performed on partial intensity changes and only those
changes representing a 2-fold or greater change in intensity (in
both duplicate fingerprints) are scored as positive changes.
DNA Fingerprinting of DMBA-induced Tumors. Neonatal
male CD-I mice were given a single i.p. injection of DMBA
and liver tumors were harvested 9 or 12 months later. At this
age, the number of spontaneous liver tumors observed in vehi
cle-treated, concurrent control mice was only about 3% of the
number of liver tumors observed in DMBA-treated animals.
Thus, the vast majority of tumors in the DMBA-treated mice
were induced by the single dose of the carcinogen. Fourteen
DMBA-induced liver tumors, isolated from 5 different mice,
were assayed for somatic DNA rearrangements by comparing
their DNA fingerprints with that of their normal tissue coun
terpart. Using our conservative approach of scoring a coincident
band loss and band gain as one putative size shift, we detected
a total of 8 rearrangements in 14 tumors (Table 1).
Fig. IA shows the DNA fingerprints of the DMBA-induced
tumors that exhibited rearrangements, with the band changes
indicated by numbers within the figure. There were four cases
of putative size shifts that could represent size decreases by
internal deletions. Tumor D-b, had a band gain (band 4) as well
as the loss of a lower molecular weight band (band 5). Tumors
D-b, and D-b2, isolated from the same animal, both appeared
to have had discernible deletions in the same minisatellite locus
(bands 3 and 7, respectively). Also, tumor D-as had the gain of
a new band (band 2) together with a 7-fold loss in the intensity
of a larger band (band 1) (scored as a partial intensity shift in
Table 1). Conversely, there were two cases of putative size shifts
that could represent size increases by additions. Although faint,
band 8 represents a new band in tumor D-CI not found in the
adjacent normal DNA fingerprint. A second band of identical
intensity, just below the position of band 8, is found in the
normal DNA fingerprint but is lost in the tumor DNA finger
print. The gain of band 8 and the loss of the band just below it
that exhibits the same intensity were reproducible in duplicate
DNA fingerprints and were therefore scored as a putative size
shift. Likewise, band 10 shifts to a slightly higher molecular
Table 1 Minisatellite rearrangements in spontaneous and DMBA-induced CD-I
mouse liver tumors
Partial
Tumor type
No. of
tumors
examined
New
bands"
Complete
band losses
or shifts
intensity
losses
or shifts
Total band
changes/no.
of tumors
Spontaneous
15
1
2/15
KD
DMBA-induced
1(6)
14
8/14
' New bands include intensity increases. New bands or intensity gains that
appear coincidentally with a band loss are listed in parentheses but are not
counted, since they are scored under "Complete band losses or shifts" or "Partial
intensity losses or shifts" (depending on the degree of band loss).
5246
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1990 American Association for Cancer Research.
DNA FINGERPRINTING
OF CD-I MOUSE LIVER TUMORS
B.
A.
D-a
N
1
2
3
4
S
D-b
D-b
D-c D-d D-e
N
N
N l
l
2
N l
N
l
S-a
S-b
S-c
S-d
\
NI
NI
NI
l
S-e
NI
S-f
NI
Szi
&ÃœL O
NI
NI
N
Kb
Kb
—
23.1-
23.I-
9.4-
HIHI'i
-
-
-
-
~
V
^*
• *•
^—
a
-- —
—
•:
«•••IM«.
4.2.
illlllÜ
«•••§•
b
•HHI"
Fig. 1. DNA fingerprints of CD-I mouse liver tumors (numbered) and their corresponding normal tissue (/V).A, DMBA-induced tumors; B. spontaneous tumors.
Letters above the lanes identify the individual mouse from which the tumors were isolated. Band changes are indicated by numbers adjacent to the bands. All DNA
samples were analyzed in duplicate, and only reproducible band changes were scored as positive changes. The intensity increases in bands a and b in tumor D-b¡
represent two cases of apparent band changes which were not scored because they were probably caused by artifacts (see text). The position and size in kilobases (Kb)
of ///m/III-cleaved XDNA markers are denoted next to each set of lanes.
weight in tumor D-CI such that it forms a poorly resolved
doublet with the band just above it. This doublet was more
clearly resolved in a repeat experiment (not shown). In all of
these cases of putative size shifts, only one rearrangement was
scored in Table 1 for each pair of coincidental band changes.
In addition, two tumors had band changes that had no coinci
dent counterparts. Tumor D-b2 had the gain of a new band
(band 6), and tumor D-d, had a band loss (band 9).
There are several bands that appear to have undergone size
or intensity shifts which we did not score because they could
have been caused by artifacts. Tumors D-b2 and D-CI both
appeared to have size increases in their highest molecular
weight band (about 25 kilobases), such that these bands now
comigrate with band 9 in tumor D-d,. However, ethidium
bromide staining of the agarose gels showed that the normal
DNA samples for both mice D-b and D-c (D-bN and D-cN,
respectively) were partially degraded such that the highest mo
lecular weight fragments in their Hae\\\ digests ran below the
highest molecular weight fragments in their respective tumor
samples. In fact, the bands in question on the autoradiograph
comigrate exactly with the highest fragments in each lane as
seen in the ethidium bromide-stained gel. Thus, it is likely that
these apparent band shifts were caused by differences in the
quality of DNA and therefore they were not scored. In contrast,
the reproducible loss of band 9 in tumor D-d, was not due to
DNA quality differences since both tumor D-di and its adjacent
normal tissue had identical ethidium bromide staining patterns
that were representative of high molecular weight DNA di
gested with Haelli.
The quality of D-bN DNA is also likely to be responsible for
the apparent gains in the intensity of bands a and b in tumor
D-b2. The D-bN fingerprint (adjacent to tumor D-b2) has a
higher lane background and an overall poorer band resolution
as compared to the D-b2 lane, particularly in the regions sur
rounding bands a and b, which probably has obscured these
bands in the D-bN fingerprint. In support of this, examination
of tumor D-bi and its adjacent D-bN fingerprint reveals that
bands a and b are also found in tumor D-b,, and furthermore,
while band a is still obscured in this D-bN fingerprint, band b is
not obscured. Thus, the apparent intensity changes in bands a
and b in tumor D-b2 are most likely due to artifacts in DNA
quality and band resolution in the D-bN fingerprint, and there
fore these changes were not scored.
In addition to rearrangements, point mutations could also
potentially cause a change in a DNA fingerprint by either
destroying or creating a restriction site. However, observed
point mutation frequencies, relative to the number of potential
or existing restriction sites in each minisatellite restriction
fragment, are generally too low to permit detection in this assay
(15). Also, DNA sequencing of minisatellites has shown that
size polymorphisms in minisatellite restriction fragments are
almost exclusively due to rearrangements (8, 9, 11), and band
shifts observed in human tumor DNA fingerprints were found
to be independent of the restriction enzyme used and thus could
not be due to point mutations in restriction enzyme recognition
sites (13). Therefore, the ability to detect size changes in minisatellite restriction fragments by DNA fingerprinting is prob
ably due to minisatellites being hot spots for recombination
(11, 12). To demonstrate that the band changes we observed
were due to rearrangements and not point mutations, we re-
5247
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1990 American Association for Cancer Research.
DNA FINGERPRINTING OF CD-I MOUSE LIVER TUMORS
peated the DNA fingerprinting analysis of tumors D-b2 and DCi using 5fl«3Ainstead of Haelll for restriction digestion. We
observed the same changes in bands 6, 7, and 10 in the SauiA.
DNA fingerprints (data not shown) as we found in Hae\\\ DNA
fingerprints (Fig. \A)\ thus these band shifts could not be due
to point mutations affecting Haelll recognition sites.
Since about 30 minisatellite bands are resolved in each tumor
DNA fingerprint, the 8 rearrangements we observed in 14
DMBA-induced tumors correspond to a minisatellite re
arrangement frequency of about 2% (8 rearrangements per
about 420 bands examined), which is relatively high considering
that the average germ line recombination frequency of minisatellites is approximately 0.4%, and the frequency of somatic
rearrangements in minisatellites is thought to be much lower
(8, 9, 11). Furthermore, the rearrangements were well distrib
uted among the DMBA-treated animals. A band shift was
detected in at least one tumor from each of the DMBA-treated
mice that were examined (Table 2).
DNA Fingerprinting of Spontaneous Liver Tumors. In contrast
to DMBA-induced tumors, two intensity shifts but no complete
band shifts were observed in the 15 spontaneous liver tumors
analyzed (Table 1). In tumor S-f, (Fig. IB), there was a 2.5-fold
decrease in the intensity of band 1 together with a reciprocal
increase (2.3-fold) in the intensity of the band just below it (the
total intensity of the two bands was conserved between the
normal and tumor DNA fingerprints). In tumor S-h,, there was
a 2.5-fold increase in the intensity of band 2. It is possible that
the intensity increase was caused by an amplification of the
minisatellite represented by band 2 or, alternatively, by the
appearance of a new band that comigrates with band 2 in the
tumor S-h, fingerprint.
When tabulating the data in Table 1, we separated complete
band losses from partial band losses (intensity losses) because
the degree of a band loss can sometimes give an indication of
the percentage of cells in the tumor sample that contain the
rearrangement. New bands (and intensity increases) are listed
separately, because without a knowledge of their original inten
sity, they cannot provide an indication of the percentage of
tumor cells with the change. The incomplete nature of an
intensity shift, such as that observed in spontaneous tumor Sfi, could represent heterogeneity within the tumor and thus a
change that occurred later during tumor progression. Alterna
tively, some of the normal band could remain in the tumor
sample because of contamination with normal tissue (16, 17)
or because a homozygous band was made heterozygous by the
rearrangement. However, in this study care was taken to collect
Table 3 Histological examination of spontaneous tumors
Tumor"S-a,S-b,S-c,S-d,S-e,S-f,S-g,S-h,S-i,s-j,S-k,S-l,S-m,S-n,S-o,Size»
(cm)1.91.21.01.52.31.92.01.82.01.51.52.02.01.00.4Histological
diagnosisCarcinomaAdenomaAdenomaAdenomaAdenoma
shift_—_——+-+————
" Letters denote the individual mouse from which the tumors were isolated.
* Sizes represent the greatest dimension of the tumor.
' + or —,whether or not. respectively, rearrangements (band shifts) were
observed.
only tumor tissue for DNA preparation, and individuals are not
generally homozygous for minisatellite loci (8, 9). In DMBAinduced tumors, 6 of the 7 band changes that involved the loss
or shift of a band were complete losses of the normal band (Fig.
l.-l). and the only partial shift was still a 7-fold reduction in
intensity. Thus the DMBA-induced rearrangements must have
occurred early in or before the development of the tumors,
consistent with the single-dose protocol of DMBA treatment
used to initiate these tumors, and possibly in contrast to the
intensity changes seen in spontaneous tumors which may have
occurred later during tumor progression.
Histopathology. Histological examination showed that the
DMBA-induced (Table 2) and spontaneous (Table 3) tumor
groups contained similar proportions of adenomas and carci
nomas and that, in general, the spontaneous tumors were larger
than the DMBA-induced tumors (spontaneous tumors were
obtained from older mice). Also, band shifts were found in both
adenomas and carcinomas (of various sizes) within the DMBAinduced group. Thus, the far greater frequency of minisatellite
rearrangements observed in DMBA-induced tumors relative to
spontaneous tumors cannot be attributed to differences in their
malignancies or in their degree of tumor progression.
Choice of the DNA Fingerprinting Probe. The synthetic mouse
minisatellite probe (M-core) we used to obtain the data pre
sented above was far more effective in detecting minisatellites
that underwent somatic rearrangement than was a similar probe
that we generated' using a human minisatellite consensus core
sequence (9) (data not shown). Wide differences have been
noted in the germline recombination frequency of different
minisatellite loci (11) and it is likely that somatic recombination
Table 2 Histological examination of DMBA-induced tumors
frequencies also vary between minisatellites. Since each miniTumor"D-a,D-a2D-a,D-34D-a,D-b,D-b2D-c,D-d,D-d¡D-d,D-d4D-e,D-e2Size*
satellite probe recognize a different family of minisatellites
shifts-———+++++———+(cm)1.31.10.90.90.60.71.21.51.41.70.71.31.00.8Histological
diagnosisCarcinomaAdenomaCarcinomaAdenomaAdenomaAdenomaCarcinomaCarcinomaCarcinomaCarcinomaAdenomaAdenomaAdenomaAdenomaBa
(related by the degree of sequence homology) (8,9), it is possible
that some probes may be more effective than others in detecting
minisatellites that are highly prone to rearrangement. The Mcore probe may be particularly effective because it is based on
a minisatellite in the mouse major histocompatibility complex
that is known to be a "hot spot" for recombination (12).
Concluding Remarks. DMBA, a powerful initiator of carcinogenesis (18, 19), is known to cause point mutations (20) and
chromosomal aberrations (21) in DNA. Our DNA fingerprint
ing results indicate that a single-dose treatment of DMBA can
also induce a large number of minisatellite rearrangements that
" Letters denote the individual mouse from which the tumors were isolated.
are clonally selected during tumor outgrowth. Since only two
* Sizes represent the greatest dimension of the tumor.
rearrangements were observed in spontaneous tumors and they
f + or -. whether or not. respectively, rearrangements (band shifts) were
may have occurred later during tumor progression, our results
observed.
5248
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1990 American Association for Cancer Research.
DNA FINGERPRINTING
OF CD-I MOUSE LIVER TUMORS
suggest that DNA fingerprinting using informative probes may
be a valuable assay for distinguishing certain chemically induced
tumors from spontaneous tumors.
Presently, the DNA fingerprinting assay is not sensitive
enough to distinguish spontaneous from induced tumors in
each individual tumor on the basis of minisatellite re
arrangement frequency, since only about 30 minisatellite bands
are examined per tumor and we found approximately 1 re
arrangement per 50 bands in DMBA-induced tumors. There
fore, several tumors from a given group (e.g., DMBA-induced
tumors) must be analyzed in order to adequately determine the
minisatellite rearrangement frequency. For example, in a group
of 15 tumors (each tumor analyzed individually), a total of
about 450 bands can be examined, which allows a good esti
mation of rearrangement frequency within the tumor group.
The sensitivity of the assay could potentially be increased to
the point of being informative on a single tumor basis by using
additional probes and thus examining more minisatellites per
tumor sample. However, the additional probe must, like the Mcore probe, detect families of minisatellites that are highly prone
to rearrangement. Cloning minisatellites that have already ex
hibited rearrangements in the DMBA-induced tumors could be
an effective means of generating such probes.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
ACKNOWLEDGMENTS
We thank Jeff Hudson of Molecular Dynamics for the loan of the
300A computing densitometer and for his assistance in using it.
15.
REFERENCES
17.
1. Chu, K. C., Cueto, C., and Ward, J. M. Factors in the evaluation of 200
National Cancer Institute carcinogen bioassays. J. Toxicol. Environ. Health,
«.•251-280,1981.
2. van Zwieten, M. J., Majka, J. A., Peter, C. P.. and Burek, J. D. The value of
historical control data. In: H. C. Grice, and J. L. Ciminera. (eds.). Carcinogenicity, pp. 39-51. New York: Springer-Verlag, 1988.
3. Reynolds, S. H., Stowers. S. J., Patterson, R. M.. Maronpot, R. R., Aaronson,
S. A., and Anderson, M. W. Activated oncogenes in B6C3F, mouse liver
16.
18.
19.
20.
21.
tumors: implications for risk assessment. Science (Washington DC), 237:
1309-1316, 1987.
Wiseman. R. W.. Stowers, S. J. Miller, E. C, Anderson. M. W., and Miller,
J. A. Activating mutations of the c-Ha-ras protooncogene in chemically
induced hepatomas of the male B6C3F1 mouse. Proc. Nati. Acad. Sci. USA,
«J:5825-5829, 1986.
Reynolds, S. H., Stowers, S. J.. Maronpot. R. R., Anderson, M. W., and
Aaronson. S. A. Detection and identification of activated oncogenes in
spontaneously occurring benign and malignant hepatocellular tumors of the
B6C3F, mouse. Proc. Nati. Acad. Sci. USA, 83: 33-37, 1986.
Stowers. S. J., Wiseman, R. W., Ward, J. M.. Miller, E. C, Miller, J. A.,
Anderson. M. W., and Eva, A. Detection of activated proto-oncogenes in Nnitrosodiethylamine-induced liver tumors: a comparison between B6C3F,
mice and Ficher 344 rats. Carcinogenesis (Lond.), 9: 271-276, 1988.
Fox, T. R., and Watanabe. P. G. Detection of a cellular oncogene in
spontaneous liver tumors of B6C3F, mice. Science (Washington DC). 238:
596-597. 1985.
Jeffreys, A. J., Wilson, V., and Thein, S. L. Hypervariable minisatellite
regions in human DNA. Nature (Lond.), 314:61-73. 1985.
Jeffreys, A. J., Wilson. V., and Thein, S. L. Individual-specific (fingerprints'
of human DNA. Nature (Lond.), 316: 76-79. 1985.
Wong, Z., Wilson, V.. Patel, I., Povey, S.. and Jeffreys. A. J. Characterization
of a panel of highly variable minisatellites cloned from human DNA. Ann.
Hum. Genet.. 51: 269-288, 1987.
Jeffreys, A., Royle. N. J.. Wilson. V., and Wong, Z. Spontaneous mutation
rates to new length alÃ-elesat tandem-repetitive hypervariable loci in human
DNA. Nature (Lond.), 332: 278-281, 1988.
Kobori, J. A., Strauss, E., Minard, K.. and Hood, L. Molecular analysis of
the hotspot for recombination in the murine major histocompatibility com
plex. Science (Washington DC), 234: 173-179, 1986.
Thein, S. L., Jeffreys. A. J.. Gooi, H. C, Cotter, F., Flint, J., O'Connor, N.
T. J., Weatherall, D. J.. and Wainscoat, J. S. Detection of somatic changes
in human cancer DNA by DNA fingerprint analysis. Br. J. Cancer, 55: 353356, 1987.
Méese,E., and Blin. N. Simultaneous isolation of high molecular weight
RNA and DNA from limited amounts of tissues and cells. Gene Anal.
Techn.. 4: 45-49, 1987.
Skolnick, M. H. An approach to monitoring mutation rates using restriction
fragment lengths. IARC Sci. Pubi. (France), 78: 253-266, 1986.
Fearon. E. R., Feinberg. A. P., Hamilton, S. H., and Vogelstein, B. Loss of
genes on the short arm of chromosome 11 in bladder cancer. Nature (Lond.),
318: 377-380, 1985.
Vogelstein. B.. Fearon, E. R.. Kern, S. E., Hamilton, S. R.. Preisinger, A.
C., Nakamura, Y., and White, R. Allelotype of colorectal carcinomas. Science
(Washington DC), 244: 207-211. 1989.
Barbacid, M. ras genes. Annu. Rev. Biochem., 56: 779-827, 1987.
Yuspa, S. H., and Poirier, M. C. Chemical carcinogenesis: from animal
models to molecular models in one decade. Adv. Cancer Res., 50: 25-70,
1988.
Singer. B., and Kusmierek. J. T. Chemical mutagencsis. Annu. Rev.
Biochem.. 52: 655-693. 1982.
Mitelman, F., and Lavan, G. The chromosomes of primary 7,12-dimethylbenz(a)anthracene-induced rat sarcomas. Hereditas. 71: 325-334, 1972.
5249
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1990 American Association for Cancer Research.
DNA Fingerprinting of 7,12-Dimethylbenz[a]anthracene-induced
and Spontaneous CD-1 Mouse Liver Tumors
Brian J. Ledwith, Richard D. Storer, Srinivasa Prahalada, et al.
Cancer Res 1990;50:5245-5249.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/50/17/5245
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1990 American Association for Cancer Research.