[CANCER RESEARCH 55, 3883-3881). September 1, 1995]
Spontaneous and Ionizing Radiation-induced
/753-deficient Mice1
Chromosomal Abnormalities
in
Simon D. Bouffler,2 Christopher J. Kemp,3 Allan Balmain, and Roger Cox
Biomedicai Effects Depurimela, National Radiological Protection Board, Chillón, Didcot, Oxon, OXII ORQ [S. D. B., R. C.¡,Unilei/ Kingdom, and Cancer Research Campaign
Beatson Laboratories, Beatson Institute for Cancer Research, Carscube Estate, Switclihack Road, Bearsden, Glasgow, G61 IBD ¡C.J. K., A. B.], United Kingdom
ABSTRACT
Chromosomal
abnormalities
have been assessed in p53-deficient
mice.
The in vivo frequency of spontaneous stable aberrations in bone marrow
cells was elevated by approximately 20-fold inp53 nulls and 13-fold in />.'>.!
and cellular consequences such as death by apoptosis, cell cycle
arrest, and repair of damage (17). In broad agreement with this
argument, p53 has been found to play a role in genetic stability as
judged by permissiveness for drug-induced DNA amplification (15,
16) and control of chromosome number (15, 18, 19).
Cytogenetic analysis is one of the few methods available for screen
ing, albeit on a gross scale, alterations to the entire genome of
individual cells. We have used a number of cytogenetic techniques to
examine genome stability in pJJ-deficient mice in vivo. Through the
use of direct bone marrow preparations, the levels of spontaneous and
radiation-induced stable chromosomal aberrations (translocations, de
hétérozygotes
compared to wild-type. No excessive induction of stable
aberrations by -/-irradiation was observed, but p53 deficiency resulted in
excess radiation-induced hyperploidy (>10-fold wild-type frequency). No
influence of p53 genotype on sister chromatid exchange or G2 chromatid
damage was observed in mitogen-stimulated spleen cell cultures; however,
a p53 effect on postirradiation mitotic entry was seen. Abnormalities in
chromosome segregation and mitotic delay following irradiation in p53deficient mice suggest a G2-M checkpoint role for p53 and are broadly
consistent with data on tumorigenesis
letions, chromosome gains, and losses) in hemopoietic stem and
progenitor cells of adults were assessed. Hemopoietic tissue is of
particular relevance since acute leukaemia is common in Li-Fraumeni
patients (3) and p53-deficienl mice frequently develop lymphoid
tumors spontaneously and following irradiation (6, 7). The experi
ments reported here relied upon G-band analysis of stable aberrations;
in these animals.
INTRODUCTION
Mutation of the p53 gene is among the most frequent genetic
alterations noted in human tumors (1, 2). This frequent alteration
suggests that the function of p53 protein is important in the mainte
nance of normal cellular phenotype. This proposition is supported by
studies of the hereditary cancer-prone disorder Li-Fraumeni syn
drome, a disease linked to germline mutations in p53 (3, 4) where
patients are at increased risk for the early development of various
tumors. A dramatic demonstration of the role ofp53 in carcinogenesis
comes from studies with transgenic "knock-out" mice. Mice homozy-
the scoring of large numbers of cells was facilitated by the use of an
automated karyotype analysis system (20). The use of mitogen-stim
ulated spleen cell cultures allowed examination of shorter-term radi
ation effects such as chromatid breakage and sister chromatid ex
change induction in p5:?-deficient genotypes. These approaches allow
the following questions to be addressed: (a) does p53 status influence
the in vivo burden of damaged cells in the hemopoietic system of mice
spontaneously or following ionizing irradiation? (b) is the frequency
gous for a null p53 alÃ-eledevelop tumors at very high rates early in of sister chromatid exchanges influenced by p53 status? (c) are early
life (5-8). The latent period for spontaneous tumors in heterozygous
postirradiation cytogenetic responses affected by p53 status? and (d)
p53 knock-out (+/—) animals lies between that of nulls and wildcan cytogenetic features of hemopoietic cells from /?5.?-deficient mice
types (6-8). In addition, the tumor latent period mp53 hétérozygotesbe correlated with their abnormal spontaneous and radiation-induced
cancer incidence?
can be very significantly reduced by treatment with the chemical
carcinogen, dimethylnitrosamine (6), or by ionizing radiation expo
sure (7). As the effect of radiation on p53 nulls is a function of age,
MATERIALS AND METHODS
it has been suggested that complete loss of p53 function is rate
Wild-type (p53 +/+), pS3 hétérozygote
(+/-), and p53 null (-/-)
mice,
limiting in carcinogenesis (7). Irradiation of young p53 nulls reduces
originally
obtained
from
L.
Donehower
and
A.
Bradley
(Baylor
College
of
the already very short tumor latency period in these animals (7),
Medicine, Houston, TX), were maintained in a mixed 129-NIH/Ola genetic
indicating that alteration of additional genomic targets is required for
background as described (7). The genotype of animals was determined by
tumorigenesis.
Southern blot analysis of tail tip DNA samples (21). For the analysis of
The biological role of p53 is complex and not yet fully understood.
radiation-induced chromosomal changes in bone marrow, adult mice were
Cells lacking functional p53 are defective in entry into apoptosis
treated with a 3 Gy whole-body dose of cobalt-60 gamma radiation at 0.85 Gy
min"1. Bone marrow samples were taken 2 weeks or 4 weeks postirradiation;
following some, but not all, stimuli, and radiation-induced apoptosis
is p53-dependent (9-12). In cellular in vitro systems, p53 plays a role
all animals were between 12 and 16 weeks of age at the time of sampling.
Direct femoral bone marrow metaphase preparations were produced by pub
in G, cell cycle arrest caused by DNA damage and other cellular
lished methods (22). Briefly, animals were sacrificed by cervical dislocation,
stresses; functional loss of p53 reduces or abolishes G, arrest follow
femora were removed, and marrow from each bone was flushed into 10 ml
ing a number of treatments including ionizing radiation (13-16).
hypotonie solution (0.56% KC1) containing 100 ¡û
10X trypsin-EDTA solu
These and related observations led to the suggestion that p53 occupies
tion (GIBCO) and 0.02 fig/ml Colcemid (Ciba). Following incubation at 37°C
an important position in the signaling pathway between DNA damage
for 25-30 min, cells were fixed in three changes of 3:1 methanol:acetic acid.
Slides were prepared
Received 4/11/95; accepted 7/6/95.
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.
1This work was financially supported by the National Radiological Protection Board,
the Cancer Research Campaign, and the Commission of the European Communities
(contract F13PCT920011 to R. C.).
~ To whom requests for reprints should be addressed.
3 Present address: Fred Hutchinson Cancer Research Centre, 1124 Columbia Street,
Cl-015, Seattle, WA 98104.
and G-banded
as described
(20). Metaphases
were
analyzed using the automated karyotyping system of Piper and Breckon (20),
and full G-banded karyotypes were produced for each cell examined.
In vitro irradiation studies were performed on spleen cell cultures. Spleens
were removed from animals and disaggregated in RPMI 1640 (GIBCO)
containing 10% fetal calf serum, using scissors, followed by sequential passing
through 19 and 23 gauge needles. Following storage on ice for a maximum of
24 h, 5-ml cultures were set up in round-bottomed plastic universals, each with
1-5 X IO7 cells total in RPMI 1640 containing 20% fetal calf serum and 2
3883
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1995 American Association for Cancer Research.
CHROMOSOMAL ABNORMALITIES IN />55-DEFICIENTMICE
Table 1 Spontaneous and radiation-induced
chromosomal aberrations
aberrations"RadiationdoseGenotype
in bone marrow cells of wild-type and p53-deficient mice
Stable
aberrations*CDS
time(days)IS30Combined*1429Combined^1429Combined^No.
cellsanalyzed5135387348694311241494291RT065111134510515TT001104150448TD145943365224ID020223140437INV002200111101INS000010000011Ave
+CDG00001267131325CEX000001010000TAS000
(Gy)0++
30+-
303Postirradiationsample
" RT, reciprocal
translocation;
TT, terminal translocation;
TD, terminal deletion; ID, interstitial
deletion; INV, inversion;
INS, insertion. Average breakpoints/cell,
total
breakpoints/cell scored in all cells scored divided by the number of cells scored. TT, TD = 1 breakpoint; RT, ID, INV = 2 breakpoints; INS = 3 breakpoints.
h DM, double minute; CDG, chromatid gap; CDB, chromatid break; CEX, chromatid exchange; TAS, telomeric association.
' Combined data from the two postirradiation sampling times.
fig/ml concanavalin
A (Pharmacia);
cell cultures for SCE4 experiments
also
included bromodeoxyuridine (Sigma Chemical Co.) at 10 JAMfinal concentra
tion. Cultures were X-irradiated at either 280 kVp at 0.1 Gy/minute or 250 kVp
at 1 Gy/minute. Total culture times were 24 or 48 h, with the final two h in the
presence of 0.004 /j.g/ml Colcemid. Cultures for analysis of SCE were irradi
ated at 24 h and harvested at 48 h. Chromosome preparations were made using
a 15-min hypotonie treatment (0.56% KC1; room temperature) and three
the variance to mean ratios for the average breakpoints/cell data
indicate overdispersion of the aberrations. Highly damaged cells were
observed more frequently than expected; this was taken into account
in the statistical testing.
A notable finding was the presence of unstable aberrations in +/—
and -/- preparations. Unstable damage, e.g., chromatid breaks, gaps
changes of fixative (3:1 methanol:acetic acid). Slides were prepared and
stained with 5% Giemsa in pH 6.8 phosphate buffer for G2 damage experi
ments or, for SCE analysis, by established methods (23). In G2 damage
experiments, chromatid discontinuities of greater than a chromatid width were
scored as chromatid gaps; when lateral chromatid displacement was present,
the aberration was classified as a chromatid break.
and exchanges, has a very low probability of transmission unaltered
from one cell generation to the next, x2 tests indicate that unstable
aberrations overall were more frequent in +/- than in either +/+
(P < 0.001) or -/- (P < 0.005); the specific aberration type, chro
matid breakage, was more frequently observed in +/- than +/+
(P < 0.005) or -/- (P < 0.05). Therefore, elevated unstable aberra
tion frequency appeared to be a feature of +/— animals only. The
RESULTS
presence of this form of aberration can be taken as evidence for a
specific persistent instability, whereas the stable aberrations more
closely reflect the cumulative effects of all forms of chromosomal
damage and replication errors.
It has been reported previously that cultures of/^-deficient
mouse
fibroblasts and cells from Li-Fraumeni syndrome patients develop
aneuploidy, especially after several passages (15, 18, 19). Therefore,
the occurrence of aneuploidy in vivo was assessed in the mouse bone
marrow preparations (Table 2). A reduction in the number of euploid
(2n) cells with decreasing p53 copy number was noted. This is largely
accounted for by greater numbers of hypodiploid cells in hétérozy
gotes and nulls, both significantly different from controls (P < 0.005,
hétérozygotes;
P < 0.001, null; x2 test) but the difference between
Spontaneously Occurring Chromosomal Abnormalities in Bone
Marrow. Direct bone marrow chromosome preparations allow the
examination of metaphases from animals without extended culture
periods. Metaphase cells are collected within minutes of removal from
the animal and so reflect, as closely as possible, the chromosome
constitution of cells in vivo. Full G-banded karyotypes of 40-50 cells,
selected at random from three individuals each of wild-type, p53
hétérozygote,
and p53 null, were examined. Results are given in Table
1, and examples of aberration types are shown in Fig. 1. It should be
noted that in all experiments, slides were scored blind; also, no gross
abnormal pathology was seen in any of the unirradiated animals. In
the case of stable aberrations, i.e., those that can potentially be
transmitted through cell division, an elevated level was found in p53
hétérozygotes
and nulls by comparison with wild type. Overall levels
of stable aberrations can be compared using the figure for average
breakpoints/cell, calculated from the total breakpoints required to give
rise to the aberrations observed within the total number of cells
sampled (Table 1, footnote a). The variation with genotype in average
breakpoints/cell was statistically significant (P < 0.01, x2 test); both
deficient genotypes differ from wild-type: P < 0.05 and P < 0.005 for
hétérozygotes
and nulls, respectively. However, the difference be
tween hétérozygotes
and nulls was not significant. Terminal deletion
and reciprocal translocation events were the only aberration type
apparently overrepresented in p53-deficient animals. No evidence of
clonality was observed. Although the majority of damaged cells had
only one aberration, some cells with multiple aberrations were seen,
4 The abbreviation used is: SCE, sister chromatid exchange.
nulls and hétérozygotes
fails to reach statistical significance. Hyperdiploid cells were present in null animals at a low frequency. No
specific safeguards against chromosome loss during preparation were
used in this study. However, the low frequency of hypodiploidy in
control cells suggests that preparation artifacts were rare.
Radiation-induced Chromosomal Aberrations in Bone Mar
row. Ionizing radiation is known to be carcinogenic and to cause
chromosomal aberrations (24, 25); furthermore, certain specific chro
mosomal aberrations are associated with specific tumors (26, 27). In
light of the increased sensitivity to radiation carcinogenesis in p53deficient mice (7), we examined the chromosome-damaging effects of
ionizing radiation in p53-deficient animals. Table 1 shows results of
stable aberration analysis in direct bone marrow preparations of 3
Gy-irradiated animals. Again, scoring was blind, and the gross pa
thology of most mice was normal. One of three 14-day/3 Gy-null
individuals had a slightly enlarged thymus, and one of two 29-day/3
3884
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1995 American Association for Cancer Research.
CHROMOSOMAL
ii
ABNORMALITIES
!! IK II
ÕÕ
|{ M il
IN pW-DEFICIENT
BH ii
19
lì M St M M
«i M
s2 M
51 M
9
MICE
7
If,
i
IO
3
s
II
M
IZ
Ie
M
ii
H ^
r
«fiÄÜ
IMili.:itiri
È?.II!M11MibII
IR
••
fi"
II
ili
i
II
K M
i!
io
19
C!
02
I
M134
«Õ
14
Su
SS
A» I*
?
¿i
«1i?HIli14
I
Fig. 1. Representative Ci-handed karyotypes from bone marrow cells. A, normal mouse karyotypc from an unirradiated wild-type animal; note the variation in the number 8
chromosomes in the centromeric region, which represents natural polymorphism between the 129 and NIH/Ola strains. B, karyotype from a 3-Gy irradiated, 29-day sample time p53
hétérozygote
showing a small inversion in the left hand chromosome 6 and monosomy of chromosome 10. C, karyotype from a 3-Gy irradiated, 14-day sample time p53 hétérozygote;
note the translocation between chromosomes 6 and 17, terminal deletion of material from chromosome 7, and monosomy of chromosome 15. D, karyotype from a 3-Gy irradiated,
29-day sample time p53 null animal; note the additional copies of chromosomes 1, 6, 8, 9, 12, 17, and Y, as well as monosomy of chromosome 2.
Gy-nulls had a somewhat enlarged spleen; the remaining sample
aberration types was observed in irradiated wild-type animals with no
one form predominating. Similarly, a range of aberration types was
noted in nulls and hétérozygotes,
although these genotypes seemed
more prone to terminal (telomeric) translocation induction. A related
observation here is the occurrence of telomeric association events in
irradiated nulls and hétérozygotes
(Table 1, far righthand column).
Telomeric associations represent end-to-end linkage of two entire
chromosomes, whereas terminal translocations appear as the fusion of
icant radiation effect in hétérozygotes
and nulls may be due in part to a chromosome fragment broken at an interstitial site with the end of
the elevated spontaneous frequencies of stable aberrations (Table 1) in another chromosome. Telomeric association has been noted as a
frequent event in high-passage-number Li Fraumeni syndrome fibrothat it is more difficult to demonstrate a statistically significant effect
in the presence of high background, although none of the aberration
blast cultures (18).
frequencies scored in this study is saturating.5 As with the data from
The unstable aberrations, chromatid gaps, breaks, and exchanges
were observed in irradiated p5J-deficient animals but not wild-type.
unirradiated animals, no evidence of clonality was observed. As with
the results from unirradiated animals, the aberration distribution
Radiation did not significantly elevate the frequency of chromatid
amongst cells was overdispersed (average variance:mean ratio, 1.8); gaps, breaks, or exchanges in any of the genotypes (Table 1). Double
again, this was taken into account for statistical testing. A range of minutes, very small fragments of chromosome, were present in irra
diated hétérozygotes
and more clearly in nulls but not in wild-type
5 S. D. Bouffler, unpublished data.
animals.
3885
groups were of three or four individuals.
For each genotype, the stable aberration frequencies observed at
14-15 days were not significantly different from those at 29-30 days
after irradiation; data for the two time points were, therefore, com
bined for statistical analysis. Comparing average breakpoints/cell,
3-Gy irradiation showed a significant effect on aberration induction in
only wild-types (P < 0.001, x2 test). The failure to observe a signif
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1995 American Association for Cancer Research.
CHROMOSOMAL
ABNORMALITIES
IN /.«-DEFICIENT
MICE
significant differences between genotypes in the extent of radiationinduced SCE were observed.
ofchromosomenumber/cell38-4031-4234-4035
Chromatid damage is one of the most rapidly expressed cytogenetic
effects of radiation. Such damage, e.g., chromatid gaps and breaks,
<2n23426424640%>2n02026630Range
Genotype+++-—Radialiondose(Gy)03II3II3%
2n986474324830%
probably represents a subpopulation of unrepaired DNA doublestrand breaks, whereas chromatid exchanges are the products of
misjoining events. By careful selection of irradiation and harvest
18036
—
times, it is possible to examine G2 chromatid damage specifically
(32). In the experiments reported here, concanavalin A-stimulated
12026--160
—
spleen cell cultures were irradiated with 0.25-1 Gy X-rays at time 0
" Each percentage is based on chromosome counts from 50 cells; irradiated samples
h, and mitotic cells were collected by 2 h exposure to Colcemid at 1-3
were taken at 29-30 days after 3 Gy y-radiation.
h or 3-5 h after irradiation. The majority of metaphase cells analyzed
would have been in G2 at the time of irradiation, with some contri
bution of late S phase cells, particularly at 3-5 h. Results are plotted
Table 3 Spontaneous and radiation-induced SCEs in wild-type and p53-deficient
in Fig. 2. All irradiated samples had significantly greater levels of
spleen cells
chromatid damage than unirradiated controls at both time points
induced
(P < 0.01; mean difference test). At 1-3 h after 1 Gy irradiation, more
dose
over
background25.21.64.82.44.4
Genotype+++-X-ray Gy00.25OJ011.250.500.250.5Mean
SCE/cell
SE6.0±
chromatid damage was seen in hétérozygotes
and nulls than wild type.
These excesses, while small (1.5-2-fold), were just statistically sig
0.48
±
Table 2 Spontaneous ami radiation-induced numerical chromosome aberrations
hone marrow cells of wild-type cinti p53-deßcient mice"
in
±0.811.2±
1.2"6.8
0.48±
±
111.
±1.3"5.4
2
0.47.6
±
±1.29.6
±1.1"SCE
" Significantly different from respective unirradiated control. P < 0.01. mean differ
ence tests.
Radiation effects on bone marrow cell ploidy were assessed by
chromosome counting in 50 randomly selected cells from the Gbanded preparations. Comparing irradiated and unirradiated cells (Ta
ble 2), it is evident that radiation treatment decreased the number of
euploid cells in all genotypes. Statistically significant induction of
hypodiploid cells was seen in wild-type animals (P < 0.001, x2 test).
Again, the elevated level of hypodiploidy seen in untreated p53deficient mice tends to mask radiation effects. Hyperdiploidy was also
caused by radiation with significant elevations in frequency observed
in hétérozygotes
and nulls (P < 0.001 and P < 0.05, respectively; x2
tests); no significant elevation in wild-type animals was noted. The
pattern of chromosome gains and losses appeared to be random, with
no particular chromosome(s) being strongly overinvolved.
Short-Term Cytogenetic End Points. The data presented thus far
in this report reflect as nearly as possible the in vivo situation. Any
differences in chromosomal aberration formation between genotypes
are also influenced by physiological factors such as clonal selection
and, possibly, elimination of very highly damaged cells. Short-term
cytogenetic assays using mitogen-stimulated spleen cells can begin to
address questions of the immediate effects of radiation exposure.
Selection of appropriate assays can also go some way to assessing
effects in specific phases of the cell cycle.
SCE formation is an S phase-dependent process (28, 29). Ionizing
radiation is, by comparison with many other agents, a modest inducer
of SCE (30, 31). By examining baseline and radiation-induced SCE in
wild-type and p5J-deficient cells, abnormalities in S phase or entry of
damaged cells into S phase should be apparent. SCE frequencies
observed in concanavalin A-stimulated spleen cells are shown in
Table 3; each average is calculated from a sample of 25-100 metaphases. Baseline SCE frequency in hétérozygotes
was modestly
higher than in wild-types, and in nulls, slightly lower; these differ
ences are not, however, statistically significant. In all genotypes, a
dose-dependent increase in SCE is observed following X-irradiation,
reaching statistical significance at 0.5 Gy (mean difference test). No
nificant (P < 0.05). This level of damage is significantly reduced by
3-5 h after 1 Gy in nulls and hétérozygotes
(P < 0.05), whereas in
wild-type cells, the amount of damage scored increased. The peak
levels of damage observed in each genotype following 1 Gy were not
significantly different. Comparison of the two postirradiation sam
pling times allows assessment of two major cellular damage re
sponses: the repair of damage with time, and the delay of entry of cells
into mitosis. The contribution of mitotic delay can be assessed by
scoring the mitotic index after irradiation. Table 4 gives quantitative
estimates of the radiation effect on mitotic cell frequency; the control
(unirradiated) mitotic frequencies in the three genotypes are not
statistically distinguishable. These data demonstrate a dose-dependent
decline in mitotic frequency in wild-type, whereas in hétérozygotes
and nulls, no such decline was seen. In fact, linear regression analysis
of the data suggests that it is statistically valid to suggest a dosedependent increase in mitotic yield in nulls. Data from hétérozygotes
did not produce a good fit.
DISCUSSION
The chromosome analysis presented here of hemopoietic stem cells
and progenitors from bone marrow of normal and /;5J-deficient mice
has shown that p53 plays an in vivo role in maintaining some aspects
of genome stability, even in the absence of DNA-damaging treat-
3-5
1-3
3-5
1-3
3-5
time post-irradiation, hours
0
Dose, Gy
D 0.25
D
0.5
01
Fig. 2. Radiation induction of chromatid gaps and breaks in normal and /¡5J-deficient
mouse spleen cells. Mitogen-stimulated cultures were irradiated with 0.25, 0.5, or 1 Gy as
indicated, and mitotic cells were collected by Colcemid treatment between 1 and 3 h or
3 and 5 h after irradiation. Chromatid lesions were counted in Giemsa-stained metaphase
preparations. Columns, the mean number of gaps and breaks/cell in a sample of 50 cells,
except the 1-Gy, 3-5 h wild-type sample, which was of 20 cells; bars, SEM. Frequencies
for control, unirradiated cells are as follows: + + , 0.02 ±0.01 ; + -, 0.08 ±0.03; - -, 0.
388ft
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1995 American Association for Cancer Research.
CHROMOSOMAL
ABNORMALITIES
Table 4 Frequence' of ntitolic cells in commi unti iiriulititt'Ã-lu1/lil-l\pe tiiul ¡>53tleßctt'iltspleen cell cultures
As expected, a 3-Gy dose of radiation significantly elevated stable
aberration frequencies in wild-type cells. The extent of aberration
induction in these animals was quantitatively similar to that observed
in an unrelated strain, CBA/H.'' The absence of any strong clones in
X-ray dose, Gy
0.5
0.25
0
Genotype
Mean"
Rei."
Mean"
++
10
1
8
0.8
+-
8
1
8
1
6.5
Rei."
Mean"
4.5
16
8
1
Rcl."
0.5
2
Mean"
Rei."
3
0.3
IO
1.3
IN /..«-DEFICIENT MICE
either irradiated or unirradiated individuals might be, in part, due to
the relatively small numbers of cells scored. Clones are only rarely
seen in irradiated CBA/H bone marrow, even with larger sample
sizes.7 The level of stable aberrations observed in irradiated null
1.3
15.5
2.6
" Mean frequency of mitotic cells for the specified genotype and radiation dose. The
frequency of mitotics per 5000 cells was counted in 1-3 and 3-5 h after irradiation
preparations, and the mean was calculated.
' Mean mitotic frequency per 5(KX)cells expressed as a proportion of the respective
control (0 Gy) mitotic frequency.
merits. An elevated frequency of spontaneously occurring stable ex
change-type aberrations was observed in both heterozygous and null
/^-deficient
animals compared to wild-type. This observation is
broadly similar to that of Bischoff et al. (18), who noted an increased
frequency of dicentric chromosomes in Li-Fraumeni syndrome fibroblast cultures. Dicentrics are the unstable equivalent of reciprocal
translocations that rarely pass through cell division due to problems in
chromosome separation presented by a chromosome with two func
tional centromeres. A persistent unstable phenotype in /^-deficient
mouse bone marrow was indicated in this study by the presence of
unstable chromatid type damage. This was most pronounced in p53
hétérozygotes,
for reasons which are not completely clear. It is,
however, possible that a p53 gene dosage effect upon chromatid
damage is being masked by the death of highly damaged null cells,
while relatively moderately damaged hétérozygote
cells more fre
quently survive to mitosis. Unstable breakage and fragmentation has
also been noted in Li-Fraumeni syndrome cell cultures (18). A pro
portion of this unstable damage is likely to be converted into stable
aberrations such as terminal deletions through cell division. This
proposal is supported by the higher incidence of terminal deletions
observed in p53 hétérozygotes
and nulls. Also, it might be expected
that a proportion of the chromatid damage be converted to chromo
some type damage in the subsequent cell cycle. However, acentric
fragments or dicentric chromosomes were not seen. This may be due
to a low frequency of conversion and the relatively small sample sizes.
This persistent instability suggests that cells of p5J-deficient animals
accumulate aberrations with time. The high frequencies of stable
aberrations observed in /»5.?-deficientcells are unlikely to be ac
counted for entirely by faulty processing of spontaneous damage
because the aberration frequencies are judged to be too high. As p53
protein can be located at sites of DNA replication in undamaged cells
(33), the occurrence of such aberrations is more probably due to an
abnormality in chromosomal replication. The apparent genome desta
bilizing effect of loss of one p53 alÃ-elemay account for the early
onset, multiple site cancer incidence seen in Li-Fraumeni patients (34,
35). In parallel and extending observations made in cell culture
systems (15, 18, 19), bone marrow cells of p53-deiicieM animals were
frequently found to be aneuploid.
With respect to all the aberration types discussed thus far, a single
wild-type p53 alÃ-eleappears to be acting genetically in a recessive or
codominant fashion; loss of one functional alÃ-eleappears to exert a
strong phenotypic effect. Since it was not possible to determine the
genotype of each individual cell scored, we cannot exclude the pos
sibility that all chromosomally abnormal cells have mutated to a
double p53 mutant genotype. Given the high frequency of such
chromosomally abnormal cells, this explanation seems unlikely, and
similar strong phenotypic effects in cells believed to be carrying a
single functional p53 alÃ-elehave been noted by others (15, 19).
animals was higher than in wild-types. However, the elevated spon
taneous frequency in nulls accounts for this, and the excess induced by
radiation was less than in wild-type. Little or no radiation induction
was observed in hétérozygotes.
Thus, p53 deficiency does not appear
to promote the aberrant processing of initial radiation damage leading
to these forms of chromosomal mutation. In the bone marrow system
used in these studies, factors such as clonal growth advantage and cell
death are superimposed upon the processing of initial damage to
genetic alterations. Although these factors might obscure some
damage-processing defects, they are, in our view, unlikely to mask a
major role of p53. Since such in vivo factors are likely to affect the
early phases of the neoplastic process, the statistically insignificant
differences in the observed frequency of induced stable aberrations
per se are unlikely to account for the differences between p53 geno
types in sensitivity to radiation carcinogenesis (7).
Despite the lack of an excessive overall chromosomal radiation
effect in /j5.?-deficient animals, some types of induced aberration
appeared more commonly. Telomeric association events were noted
only in irradiated nulls or hétérozygotes.
Similar events have been
recorded in unirradiated Li Fraumeni syndrome fibroblast cultures
(18). Terminal translocation events were also more frequent in hét
érozygotesand nulls than wild-types; in the case of nulls, this differ
ence reached statistical significance (P < 0.05, x2 test). This high
frequency of telomere-associated aberrations might be indicating a
role for p53 in the maintenance of normal chromosome end structure
or function and may relate to the activation of telomerase noted in
immortal cells and cancer cells (36). Double minute chromosomes are
another abnormality seen only in irradiated /»5_?-deficientanimals;
again, the excess seen in nulls is statistically significantly different
from wild type (P < 0.05, x2). These double minutes may be a
reflection of gene amplification, which is readily achieved in p53 null
cells (15, 16), but might also represent small interstitial deletion
events. An excess of other unstable aberration types was not observed
after irradiation. Given the long period after irradiation in the exper
iment, this is expected, because the cells scored have most probably
undergone several cycles since irradiation; therefore, radiationinduced unstable aberrations will have been eliminated. These obser
vations do, however, argue against the induction by radiation of
persistent genomic instability in /?5.i-deficient animals.
Cellular ploidy appeared to be strongly influenced by radiation
exposure. Chromosome losses were evident in wild-type cells,
whereas cells of p53-deficient animals suffered both gains and losses.
For this end point, there does seem to be a major difference in the
damage response of different p53 genotypes. Thus, p53 appears to
play a role in the control of ploidy, both spontaneously and in
response to ionizing radiation damage. Because gains and losses of
chromosomes were seen, the true level of whole chromosome changes
is probably being underestimated, since some apparently 2n cells
might be expected to have unbalanced karyotypes. The inter-genotype
variation seen in these data seems to reflect most closely the carci
nogenesis results of Kemp et al. (7). The significant radiation induc" S. D. Bouffler and R. Cox, unpublished observations.
7 S. D. Bouffler, G. Breckon, and R. Cox. Chromosomal
regions in murine radiation acute myeloid leukaemogenesis,
mechanisms and telomeric
submitted for publication.
3887
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1995 American Association for Cancer Research.
CHROMOSOMAL
ABNORMALITIES
IN ^-DEFICIENT
MICT.
tion of hyperploidy in hétérozygotes
might account for the character
assays, again using flow cytometric techniques, little evidence of G2
istic high frequency loss of the normal p53 gene and duplication of blockage and G, emptying postirradiation can be seen in the published
mutant alÃ-elenoted in radiation-induced tumors in these animals (7).
data for any of the three p53 genotypes, although it is clear that
p53-dencienl cells are more likely to enter S phase following irradi
Whether this mechanism is involved in the full range of tumor types
that develop in the various mutant p53 backgrounds (5-8, 37) requires
ation. A clear demonstration of effective G2-M blockage in p53 null
further investigation. Gains and losses of chromosomes presumably
T cells 1 day following 1 Gy gamma irradiation has been reported
occur at a late stage of the cell cycle, and the data could, therefore, be recently (42); cell cycle stages were determined by flow cytometry.
indicating a role for p53 in the control of chromosome segregation in The apparent disagreement between the G2-M blockage data from
G2 or M of the cell cycle.
flow cytometric studies and those reported in this paper may be
To look more closely at the influence of p53 status on radiation
accounted for by the different sampling times used and by uncertain
damage processing in the absence of in vivo selective pressures, a ties on the viability of cells. Mitotic indices in the present study were
number of short-term cytogenetic assays using spleen cells were
scored up to 5 h after irradiation, whereas the flow cytometry data
carried out. Concanavalin A, used in these experiments, is a T-cellwere largely taken from 13-24 h after irradiation (13, 41). Recently,
specific mitogen and was chosen because a high proportion of spon
data has been published that suggests that the expression of very high
taneous and radiation-induced tumors in/^-deficient
mice are T-cell
levels of wild-type p53 in rat embryo fibroblasts can cause arrest of
lymphomas (5-8). Splenic T-cells of neither p53 nulls nor hétérozy cells in the G2-M phases of the cell cycle (44). Furthermore, Cross et
gotes had baseline SCE frequencies significantly different from wildal. (45) have demonstrated thatp5J null mouse embryo fibroblasts fail
type. Low doses of X-rays (0.25 and 0.5 Gy) induced similar numbers
to arrest following prolonged exposure (22 h) to the spindle inhibitors,
of SCEs in all genotypes. These experiments indicate that p53 has no
Colcemid or nocodozole; this failure to arrest leads to hyperploidy.
major role in the S phase recombinational events leading to SCE
To conclude, in the hemopoietic system of mice, the principal
formation.
effects of p53 deficiency appear to be a constitutive abnormality
giving rise to elevated frequencies of stable aberrations and a late cell
DNA damage processing at a later cell cycle stage was examined
cycle defect, leading to increased spontaneous and radiation-induced
using a G2 irradiation chromatid damage assay. Ionizing radiation was
very effective at causing this form of chromosome damage in all
aneuploidy together with faulty control of the entry of cells into
genotypes. The maximal levels of damage observed were comparable
mitosis. The intrinsic chromosomal instability that may be attributable
in all genotypes, suggesting that the conversion of radiation damage to to abnormalities in the chromosomal DNA replication machinery
chromatid lesions is independent of p53. However, peak levels of would be expected to contribute to the tumor susceptibility of p53deficient mice and, by implication, to Li-Fraumeni patients. We
chromosomal damage were observed earlier in p53 hétérozygotes
and
nulls than in wild-types following 1 Gy; therefore, differences in suggest that, in addition to the established role for p53 in G, cell cycle
kinetics might be of importance. Parshad et al. (38) observed a higher
arrest (13, 14, 17, 41), p53 may play an important role in a late cell
cycle stage (G2-M) damage-response pathway that regulates postirra
frequency of G-, chromatid lesions in lymphocytes of affected indi
viduals in a Li-Fraumeni syndrome pedigree at 0.5-1.5 h after irra
diation entry into mitosis. In the absence of excessive radiation
induction of stable aberrations in bone marrow cells and of SCEs in
diation. There was not, however, a simple relationship between high
G2 damage responders, cancer proneness, and p53 mutations in the
spleen cell cultures, the role for p53 in early phase cell cycle responses
pedigree examined (38). There are, nonetheless, data available that
to damage (13, 14, 17, 41) does not appear to be ubiquitous or
suggest that G2 chromatid damage may be associated with cancer
necessarily predominant. In the context of ionizing radiation, recent
observations point towards a distinct tumorigenic mechanism involv
predisposition (39, 40).
ing loss of wild-type and duplication of mutant p53 alÃ-elesin tumors
The G-, damage data reported here could be interpreted as damaged
cells of /?5.?-deficient genotype being more likely to proceed rapidly
induced in p53 hétérozygotes
(7). The data on aneuploidy reported
through to mitosis than similarly damaged wild-type cells. Two of the
here suggest that this may be associated with whole chromosome 11
possible explanations for this could be that wild-type cells are more
loss and gain; whereas chromosome 11 was not found here to be
likely to die prior to entry into mitosis, or wild-type cells are more
preferentially involved in gains or losses, numerical changes to chro
able to remain in a viable but G2 blocked state than p5J-deficient
mosome 11 were observed. This will be the subject of further inves
cells. An attempt to gain insight into this second possibility was made
tigation.
by assessing the numbers of mitotic cells in control and irradiated
cultures. In wild-type cultures, a radiation dose-dependent decrease in
ACKNOWLEDGMENTS
postirradiation mitotic index was observed. The data for p53 hétérozy
gotes were more variable, whereas the postirradiation mitotic yield in
We thank A. Edwards for advice on statistical analysis, D. Morris for
technical assistance, J. Moody and L. Gibbens for help with preparation of
nulls was higher than in wild type and increased with dose. These data
figures, and K. Brooks for typing the manuscript.
are contrary to the currently prevailing notion of p53 status having
little or no effect on G2 cell cycle blockage (13, 14, 41, 42). Support
for this latter hypothesis comes from flow cytometric determinations
REFERENCES
of cell cycle distributions after irradiation. The work of Kastan et al.
1. Levine, A. J., Momand, J.. and Finlay, C. A. The p53 tumour suppressor gene. Nature
(13) shows that human cells with mutant p53 accumulate in the G2-M
(Lond.), 351: 453-456, 1991.
phases at 13-16 h after irradiation. These cells were also found to
2. Hollstein, M., Sidransky. D., Vogelstein, B., and Harris, C. C. p53 mutations in
human cancers. Science (Washington DC), 253: 49-53, 1991.
continue to enter S phase from G, after irradiation, with the overall
3. Malkin, D., Li, F., Strong, L. C., Fraumeni, J. F., Nelson, C. E., Kim, D. H., Gryka,
radiation effect being an emptying of the G, compartment and cells in
G., Bischoff, F. Z., Tainsky, M. A., and Friend, S. H. Germline p53 mutations in a
familial syndrome of sarcomas, breast cancer and other neoplasms. Science (Wash
G2 increasing. The data as presented did not distinguish between G2
ington DC), 250: 1233-1238, 1990.
and M cells, nor whether they are viable or dead. The radiation doses
4. Srivastava, S., Zou, Z.. Pirollo, K., Blattner, W., and Chang, E. H. Germline
used in those experiments were fairly high (~2 Gy), where one would
transmission of a mutated p53 gene in a cancer prone family with Li-Fraumeni
syndrome. Nature (Lond.), 348: 747-749. 1990.
expect lethality to be at or above 50-60%, even for cells with mutant
5. Donehower, L. A., Harvey, M.. Slagle, B. L., McArthur, M. J., Montgomery, C. A.,
p53 (43). Data for radiation-induced cell cycle blockage in embryo
Butel, J. S., and Bradley. A. Mice deficient for p53 are developmentally normal but
are susceptible to spontaneous tumours. Nature (Lond.), 356: 215-221, 1992.
fibroblasts from /?5J-deficient mice are also available (41). In these
3888
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1995 American Association for Cancer Research.
CHROMOSOMAL
ABNORMALITIES
6. Harvey, M., McArthur, M. J., Montgomery, C. A.. Bulel, J. S., Bradley, A., and
Donehower, L. A. Spontaneous and carcinogen-induced tumorigenesis in pS3 defi
cient mice. Nat. Genet., 5: 225-229, 1993.
7. Kemp, C. J., Wheldon, T., and Balmain, A. p53 deficient mice are extremely
susceptible to radiation induced tumorigenesis. Nat. Genet., 8: 66-69, 1994.
8. Purdie, C. A., Harrison, D. J., Peter, A., Dobbie, L., White, S., Howie, S. E. M.,
Salter, D. M., Bird, C. C., Wyllie, A. H., Hooper, M. L., and Clarke, A. R. Tumour
incidence, spectrum and ploidy in mice with a large deletion in the p53 gene.
Oncogene, 9: 603-609, 1994.
9. Lowe, S. W., Schmitt, E. M., Smith, S. W., Osborne, B. A., and Jacks, T. p53 is
required for radiation induced apoptosis in mouse thymocytes. Nature (Lond.), 362:
847-849, 1993.
10. Clarke, A. R., Purdie, C. A., Harrison, D. J., Morris, M. G., Bird, C. C., Hooper,
M. L., and Wyllie, A. H. Thymocyte apoptosis induced by p53 dependent and
independent pathways. Nature (Lond.). 362: 849-852, 1993.
11. Merritt, A. J., Potten, C. S., Kemp, C. J., Hickman, J. A., Lane, D. P., and Hall, P. A.
The role of spontaneous and radiation induced apoptosis in the gastrointestinal tract
of normal and p53 deficient mice. Cancer Res., 54: 614-617, 1994.
12. Lotem, J., and Sachs, L. Haemopoietic cells from mice deficient in wild type p53 are
more resistant to induction of apoptosis by some agents. Blood, 82: 1092-1096, 1993.
13. Kastan, M. B., Onyekwere, O., Sidransky, D., Vogelstein, B., and Craig, R. W.
Participation of p53 protein in the cellular response to DNA damage. Cancer Res., 5/:
6304-6311, 1991.
14. Kuerbitz, S. J., Plunkett, B. S., Walsh, W. V., and Kastan, M. B. Wild type p53 is a
cell cycle checkpoint determinant following radiation. Proc. Nati. Acad. Sci. USA,
89: 7491-7495, 1992.
15. Livingstone, L. R., White, A.. Sprouse, J., Livanos, E., Jacks, T., and Tlsty, T. Altered
cell cycle arrest and gene amplification potential accompanying loss of wild type p53.
Cell, 70: 923-935, 1992.
16. Yin, Y., Tainsky, M. A., Bischoff, F. Z., Strong, L. C., and Wahl, G. M. Wild type
p53 restores cell cycle control and inhibits gene amplification in cells with mutant p53
alÃ-eles.Cell, 70: 937-948, 1992.
17. Lane, D. P. p53, guardian of the genome. Nature (Lond.), 358: 15-17, 1992.
18. Bischoff, F. Z., Yim, S. O., Pathak, S., Grant, G., Siciliano, M. J., Giovanella, B. C.,
Strong, L. C., and Tainsky, M. A. Spontaneous abnormalities in fibroblasts from
patients with Li-Fraumeni cancer syndrome: aneuploidy and immortalisation. Cancer
Res., 50: 7979-7984, 1990.
19. Harvey, M., Sands, A. T., Weiss, R. S., Hegi, M. K., Wiseman, R. W., Panlazis, P.,
Giovanella, B. C., Tainsky, M. A., Bradley, A., Donehower, L. A. In vilro growth
characteristics of embryo fibroblasts isolated from p53 deficient mice. Oncogene, 8:
2457-2467, 1993.
20. Piper, J., and Breckon, G. An automated system for karyotyping mouse chromo
somes. Cytogenet. Cell Genet., 50: 111-115, 1989.
21. Kemp, C. J., Donehower, L. A., Bradley, A., and Balmain, A. Reduction of p53 gene
dosage does not increase initiation or promotion but enhances malignant progression
of chemically induced skin tumours. Cell, 74: 813-822, 1993.
22. Lee, J. J., Warburton, D., and Robertson, E. J. Cytogenetic methods for the mouse:
preparation of chromosomes, karyotyping and in situ hybridization. Anal. Biochem.,
189: 1-17, 1990.
23. Perry, P., and Wolff, S. A new Giemsa method for the differential staining of sister
chromatids. Nature (Lond.), 257: 156-158, 1974.
24. UNSCEAR, United Nations Scientific Committee on the Effects of Atomic Radiation.
UNSCEAR 1993 Report to the General Assembly, Annex E, Mechanisms of radiation
oncogenesis. New York: United Nations, 1993.
25. UNSCEAR, United Nations Scientific Committee on the Effects of Atomic Radiation.
UNSCEAR 1994 Report to the General Assembly, Annex A, Epidemiological studies
IN p5J-DEFICIENT
MICE
of radiation carcinogenesis. New York: United Nations, 1994.
26. Solomon, E., Borrow, J., and Goddard, A. D. Chromosome aberrations and cancer.
Science (Washington DC), 254: 1153-1160, 1991.
27. Rabbitts, T. H. Chromosomal translocations in human cancer. Nature (Lond.), 372:
143-149, 1994.
28. Wolff, S., Bodycote, J., and Painter, R. B. Sister chromatid exchanges induced in
Chinese hamster cells by UV irradiation at different stages of the cell cycle: the
necessity of cells to pass through S. Mutât.Res., 25: 73-81, 1974.
29. Schvartzman, J. B., and Gutierrez, C. The relationship between the cell time available
for repair and the effectiveness of a damaging treatment in provoking the formation
of sister chromatid exchanges. Mutât.Res., 72: 483-489, 1980.
30. Perry, P., and Evans, H. J. Cytological detection of mutagen-carcinogen exposure by
sister chromatid exchange. Nature (Lond.), 267: 156-158, 1975.
31. Solomon, E., and Bobrow, M. Sister chromatid exchanges: a sensitive assay of agents
damaging human chromosomes. Mutât.Res., 30: 273—278,1975.
32. Sanford, K. K., Parshad, R., Gantt, R., Tarane, R. E., Jones, G. M., and Price, F. M.
Factors affecting and significance of G2 chromatin radiosensitivity in predisposition
to cancer. Int. J. Radiât.Biol., 55: 963-981, 1989.
33. Wilcock, D., and Lane, D. P. Localization of p53, retinoblastoma and host replication
proteins at sites of viral replication in herpes infected cells. Nature (Lond.), 349:
429-431, 1991.
34. Li, F. F., and Fraumeni, J. F., Jr. Rhabdomyosarcoma in children: epidemiological
study and identification of a familial cancer syndrome. J. Nati. Cancer Inst., 43:
1365-1373, 1969.
35. Li, F. P., Fraumeni, J. F., Jr., Mulvihill, J. J., Blattner, W. A., Dreyfus, M. G., Tucker,
M. A., and Miller, R. W. A cancer family syndrome in twenty four kindreds. Cancer
Res., 48: 5358-5362, 1988.
36. Kim, N. W., Piatyszek, M. A., Prowse, K. R., Harley, C. B., West, M. D., Ho, P. L.
C, Coviello, G. M., Wright, W. E., Weinrich, S. L., and Shay, J. W. Specific
association of human telomerase activity with immortal cells and cancer. Science
(Washington DC), 266: 2011-2015, 1994.
37. Harvey, M., Vogel, H., Morris, D., Bradley, A., Bernstein, A., and Donehower, L. A.
A mutant p53 transgene accelerates tumour development in heterozygous but not
nullizygous p53 deficient mice. Nat. Genet., 9: 305-311, 1995.
38. Parshad, R., Price, F. M., Pirollo, K. F., Chang, E. H., and Sandford, K. K. Cytoge
netic responses to G2 phase X-irradiation in relation to DNA repair and radiosensi
tivity in a cancer prone family with Li-Fraumeni syndrome. Radial. Res., 136:
236-240, 1993.
39. Sanford, K. K., and Parshad, R. Detection of cancer prone individuals using cytogenetic response to X-rays. In: C. Obe and A. T. Natarajan (eds. 0, Chromosomal
Aberrations: Basic and Applied Aspects, pp. 113-120. Berlin: Springer-Verlag, 1990.
40. Scott, D., Spreadborough, A., Levine, E., and Roberts, S. A. Genetic predisposition
in breast cancer. Lancet, 344: 1444, 1994.
41. Kaslan, M. B., Zhan, Q., El-Deiry, W. S., Carrier, F., Jacks, T., Walsh, W. V.,
Plunkett, B. S., Vogelstein, B., and Fornace, A. J., Jr. A mammalian cell cycle
checkpoint pathway utilizing p53 and GADD45 is defective in ataxia telangiectasia.
Cell, 71: 587-597, 1992.
42. Strasser, A., Harris, A. W., Jacks, T., and Cory, S. DNA damage can induce apoptosis
via p53 independent mechanisms inhibilable by bcl2. Cell, 79. 329-339, 1994.
43. Lee, J. M., and Bernstein, A. p53 mutations increase resistance to ionizing radiation.
Proc. Nati. Acad. Sci. USA, 90: 5742-5746, 1993.
44. Stewart, N., Hicks, G. G., Paraskevas, F., and Mowat, M. Evidence for a second cell
cycle block at G2/M by p53. Oncogene, IO: 109-115, 1995.
45. Cross, S. M., Sanchez, C. A., Morgan, C. A., Schimke, M. K., Ramel, S., Idzerda,
R. L., Raskind, W. H., and Reid, B. J. A p53 dependent mouse spindle checkpoint.
Science (Washington DC), 207: 1353-1356, 1995.
3889
Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1995 American Association for Cancer Research.
Spontaneous and Ionizing Radiation-induced Chromosomal
Abnormalities in p53-deficient Mice
Simon D. Bouffler, Christopher J. Kemp, Allan Balmain, et al.
Cancer Res 1995;55:3883-3889.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/55/17/3883
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 16, 2017. © 1995 American Association for Cancer Research.