Targeted point mutations of p53 lead to
dominant-negative inhibition of
wild-type p53 function
Annemieke de Vries*†‡, Elsa R. Flores*, Barbara Miranda*†, Harn-Mei Hsieh*, Conny Th. M. van Oostrom‡, Julien Sage*,
and Tyler Jacks*†§
*Department of Biology and Center for Cancer Research, and†Howard Hughes Medical Institute, Massachusetts Institute of Technology, Cambridge, MA
02139; and‡National Institute of Public Health, Laboratory of Health Effects Research, 3720 BA Bilthoven, The Netherlands
Communicated by Robert A. Weinberg, Whitehead Institute for Biomedical Research, Cambridge, MA, December 31, 2001 (received for review
December 17, 2001)
The p53 tumor suppressor gene is the most frequently mutated
gene in human cancers, and germ-line p53 mutations cause a
familial predisposition for cancer. Germ-line or sporadic p53 mu-
tations are usually missense and typically affect the central DNA-
binding domain of the protein. Because p53 functions as a tet-
rameric transcription factor, mutant p53 is thought to inhibit the
function of wild-type p53 protein. Here, we studied the possible
dominant-negative inhibition of wild-type p53 protein by two
different, frequently occurring point mutations. The R270H and
P275S mutations were targeted into the genome of mouse em-
bryonic stem cells to allow the analysis of the effects of the mutant
proteins expressed in normal cells at single-copy levels. In embry-
onic stem cells, the presence of a heterozygous point-mutated
allele resulted in delayed transcriptional activation of several p53
downstream target genes on exposure to ? irradiation. Doxoru-
bicin-induced apoptosis was severely affected in the mutant em-
bryonic stem cells compared with wild-type cells. Heterozygous
mutant thymocytes had a severe defect in p53-dependent apopto-
tic pathways after treatment with ? irradiation or doxorubicin,
whereas p53-independent apoptotic pathways were intact. To-
gether these data demonstrate that physiological expression of
point-mutated p53 can strongly limit overall cellular p53 function,
supporting the dominant-negative action of such mutants. Also,
cells heterozygous for such mutations may be compromised in
terms of tumor suppression and response to chemotherapeutic
sis, cell-cycle arrest, DNA repair, recombination, cellular dif-
ferentiation, and senescence (1). Among other upstream stimuli,
DNA damage is a potent activator of p53 function, and p53 is
required for DNA damage-induced G1arrest and apoptosis in
many cell types (1), in part, by activating the expression of
downstream target genes (1–4). Given these functions, mutation
of p53 during tumorigenesis would be expected to lead to
inappropriate S-phase entry or survival of damaged cells, pos-
sibly promoting genomic instability (1). In addition, in model
systems, p53 mutant tumors and cell lines have been relatively
In humans, p53 mutations have been detected in at least 52
different cancer types, and ?50% of all human tumors carry
point mutations in p53 (8, 9). Heterozygous germ-line mutations
in p53 predispose individuals to a wide range of tumor types
at an early age, a condition known as Li–Fraumeni syndrome
The majority of both sporadic and germ-line p53 mutations
are missense and occur in the conserved DNA-binding domain
in the central portion of the protein. Depending on the tumor
type, certain ‘‘hotspot’’ mutations are found, including at codons
273 and 248 (9). These residues make direct contact with the
DNA helix and accordingly seem important for the transcrip-
he p53 tumor suppressor protein has been proposed to
function in many, diverse cellular processes, such as apopto-
tional activation function of p53 (9). However, virtually all
mutations of p53 abolish its ability to bind specific DNA
sequences and activate the expression of its target genes (4, 8, 9).
Extensive data from studies in vitro and in cell culture suggest
that many missense mutations in p53 can inhibit the function of
the wild-type protein in a dominant-negative manner, which
would indicate that a heterozygous mutation in p53 could result
in functional inactivation of cellular p53. To regulate down-
stream target genes, p53 binds the DNA as a tetrameric protein
complex. Mutated protein within this complex is thought to
abolish the DNA-binding capacity of the entire complex. Ex-
periments with ectopic expression of wild-type and mutant p53
protein have demonstrated inhibition of DNA-binding activity
data on this point and general concern about the effects of
ectopic expression on the results exist (4 and references therein).
Finally, despite possible dominant-negative function of missense
p53 mutants, in approximately 50% of human tumors harboring
such mutations, the remaining wild-type allele is mutated or lost,
suggesting that complete loss of normal p53 can promote
tumorigenesis further (8, 9).
Several mouse models have been generated to study p53
develop tumors (mainly lymphomas) at high incidence and short
latency, clearly demonstrating the important role of p53 in tumor
suppression. Transgenic overexpression of p53 point mutations
(codon 135 Ala to Val or codon 193 His to Pro) resulted in lung
adenocarcinomas and osteosarcomas. On a background of germ-
line heterozygous p53 deletion, transgenic overexpression of
mutant A135V alleles can accelerate tumor development com-
pared with nontransgenic p53?/?mice. However, in p53?/?
knockout animals, no effect of the p53 A135V transgene was
observed, indicating that the mutant protein affected tumor
development by interfering with wild-type p53 (15). Liu et al.
(16) described a point mutation of the endogenous p53 gene in
mice. Animals heterozygous for a mutation in codon 172 (argi-
nine to histidine substitution) differed from p53?/?mice in
Moreover, loss of the wild-type p53 allele was rarely observed in
these tumors, again supporting a dominant-negative function for
the R172H allele. However, one drawback of this model is that
the mutant p53 allele also contains an altered splice acceptor
site, which leads to significantly reduced expression of the
Abbreviations: ES cell, embryonic stem cell; LFS, Li–Fraumeni syndrome; PI, propidium
§To whom reprint requests should be addressed. E-mail: email@example.com.
The publication costs of this article were defrayed in part by page charge payment. This
article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C.
§1734 solely to indicate this fact.
March 5, 2002 ?
vol. 99 ?
Given the importance of p53 mutation in tumor development
and therapy, we have constructed genomic p53 point mutations
in mouse embryonic stem (ES) cells by homologous recombi-
nation. The R270H mutation is the equivalent of human R273H,
a common mutation in many tumor types and also present in
some LFS patients. The P275S mutation (equivalent to human
P278S) is common in skin tumors but not in other tumor types
nor in LFS. This system has allowed us to examine the effects of
tumor-associated p53 point mutations expressed at single-copy
levels in otherwise genetically normal cells. The effects of these
mutations were examined on the biochemical and functional
properties of wild-type p53 in mutant ES cells and thymocytes
derived from them.
Materials and Methods
Cloning of p53 Mutant Targeting Vectors. The p53.R270H and
p53.P275S targeting vectors were constructed by cloning frag-
129sv) into the vector pBSK?. Both vectors contained an ?8-kb
genomic p53 sequence, extending from the XhoI site in exon 2,
to the EcoRI site at the extreme 3? end. First, a 1.3-kb BamHI?
HindIII fragment comprising exons 7–9 was subcloned into
pBSK?. Primers were generated harboring the desired mutation
(mutated nucleotide in bold):
These primers were used to introduce the mutations into the p53
sequence by site-directed mutagenesis with the QuikChange
Site-Directed Mutagenesis Kit (Stratagene). The presence of the
correct mutation was verified by sequencing. A selectable
marker cassette (kindly provided by S. Tonegawa, Massachusetts
Institute of Technology) was incorporated ?0.5 kb downstream
of the p53 gene. The cassette contains the neomycin-resistance
gene and the thymidine kinase (TK) gene, flanked by LoxP sites.
A pGK-DTA selection marker (encoding the diphtheria toxin A
Technology) was cloned at the 3? end of the vector for negative
selection (without drug treatment). The complete inserts of the
targeting vectors were sequenced, to exclude the presence of
additional (undesired) mutations in the p53 encoding sequences
and exon?intron boundaries.
Homologous Recombination Experiments in ES Cells. The targeting
vectors were linearized by digestion with NotI, and electropo-
rated into D3 ES cells by using standard procedures (17).
G418-resistant ES cell clones were analyzed for homologous
integration of the targeting vector by Southern blot analysis. The
Southern blot probe I consists of a 0.5-kb HindIII to KpnI
fragment of intron 1 sequence. Clones with homologous inte-
gration of the targeting vector were checked for the presence of
the mutation by a PCR?digestion-based assay. The following
primers were used to amplify the p53 alleles:
(situated in intron 7)
p53in9#1: 5?-ATGCGACTCTCCAGCCTTGGTA-3? (situat-
ed in intron 9)
The resulting PCR product (486 bp) was digested with either
MslI or BstNI. The R270H mutation results in a new MslI site in
the PCR product, and the P275S mutation results in loss of a
BstNI site compared with the wild-type p53 sequence. In this
way, the p53 mutant alleles can be discriminated from the
wild-type allele and from each other. Correctly targeted ES cell
clones with either the R270H or P275S heterozygous mutation
(p53?/R270Hor p53?/P275S) were expanded, and electroporated
with circular pMC-CreN plasmid (kindly provided by F. Alt,
Harvard Medical School, Boston). Clones that had lost the
selectable marker cassette were selected for by adding ganciclo-
vir to the culture medium. Loss of the cassette was confirmed by
Southern blot analysis by using probe II (BamHI–HindIII frag-
ment). All correctly targeted clones that were used for blastocyst
injection procedures were analyzed by reverse transcription–
PCR followed by sequencing to check for additional (undesired)
mutations. For all assays described here, multiple independent
mutant clones were used.
ES Cell Exposure to DNA-Damaging Agents. For the analysis of
transcriptional activation of p53 downstream target genes, ES
cells of all genotypes [wild-type (D3 parental cell line), p53?/?
(17), p53?/?(kindly provided by R. Jaenisch, Whitehead Insti-
tute for Biomedical Research, Cambridge, MA), p53?/R270Hand
p53?/P275S] were plated on day 1 at a density of 2.5 ? 106per
10-cm dish. At day 3, cells were exposed to a single dose of 500
cGy of ? irradiation (?-cell irradiator with a Cs source). At 1, 3,
and 6 hr after the treatment, cells were isolated by trypsinization
and centrifugation, and pellets were frozen for subsequent RNA
For the analysis of doxorubicin-induced apoptosis, ES cells of
all genotypes were plated on day 1 in a six-well tissue culture dish
and grown to subconfluence. Cells were treated with 1 or 2
?g?ml doxorubicin (Sigma) for 24 hr. At the indicated time
points, both floating and adherent cells were harvested, and
pelleted by centrifugation. Cells were washed once with 1? PBS,
and stained with annexin V conjugated to FITC and propidium
iodide (PI) by following the manufacturer’s protocol (Phar-
Mingen). Stained cells were analyzed on a Becton Dickinson
FACScan machine. The percentage of apoptotic cells was de-
termined by using CELLQUEST software.
RNA Preparation and Northern Blot Analysis. Total RNA was pre-
pared from ES cells pelleted and frozen in liquid nitrogen by
using Ultraspec RNA (Biotecx Laboratories, Houston), accord-
ing to the manufacturer’s instructions. Total RNA was prepared
from thymocytes by using Trizol reagent (GIBCO?BRL). Thy-
mocytes were pelleted, immediately resuspended into 0.8 ml of
Trizol, and stored at ?80°C. Further isolation of RNA was
according to the manufacturer’s procedure. RNasin (Promega)
was added to the RNA samples to prevent degradation.
Northern blotting was performed by using standard proce-
dures. cDNAs corresponding to bax, p21, mdm-2, cyclinG, and
gapdh were used as probes (19, 20). Radioactivity was quantified
by a PhosphorImager (Molecular Dynamics).
Rag2?/?Blastocyst Complementation Assay. Rag2?/?blastocysts
were isolated from matings between RAG2?/?mice. To obtain
Contribution of ES cells was determined by coat color.
Thymocytes, splenocytes, and cells isolated from the bone
marrow were stained with several monoclonal antibodies to
analyze reconstitution of the lymphoid lineages by the ES cells.
For this procedure, staining with FITC- or phycoerythrin-
conjugated monoclonal antibodies directed against CD4, CD8,
B220, IgM, and CD3 (PharMingen) were performed by proce-
dures described (21, 22). Flow cytometric analysis was per-
formed by using a FACScan (Becton Dickinson). Animals with
thymocyte populations containing less than 70% CD4??CD8?
double-positive cells were excluded from further analysis.
For the analysis of apoptosis, thymocytes were isolated from
wild-type, p53?/?, p53?/?, p53?/R270H?RAG2?/?, and
p53?/P275S?RAG2?/?chimeras (age 4–9 weeks) and kept in a
de Vries et al.PNAS ?
March 5, 2002 ?
vol. 99 ?
no. 5 ?
PBS?FCS (2%) solution. The nonchimeric mice used were in the
same 129sv background as the ES cells. Cells were plated at a
density of 1 ? 106per well in 24-well plates in medium
[DMEM?Hepes (25 mM, pH 7.2), 5% FCS, penicillin?
streptomycin, glutamine]. To induce apoptosis, cells were ex-
posed to (i) a single dose of 500 cGy of ? irradiation with a ?-cell
irradiator with a Cs source, (ii) 0.2 ?g?ml doxorubicin (Adria-
mycin), (iii) 1 ?M dexamethasone, or (iv) a combination of 10
nM phorbol ester (phorbol 12-myristate 13-acetate) and 500 nM
calcium ionophore A23187 (chemicals all obtained from Sigma).
Cells were incubated at 37°C, and were, at the indicated time
points, analyzed for the amount of apoptosis. For this procedure,
cells were washed with cold PBS, and stained with FITC-labeled
annexin V antibody (PharMingen) and PI (Sigma). The relative
amounts of apoptotic cells were determined at various times by
binding of annexin V and subsequent fluorescence-activated cell
For analysis of expression levels in thymocytes of p53 target
genes, thymocytes were isolated and brought into culture as
described above at a density of 10 ? 106cells per 10-cm dish. At
time 0, cultures were exposed to a single dose of 500 cGy of ?
irradiation. Cells were pelleted at 2 or 5 hr after the treatment
and resuspended in 0.8 ml of Trizol. RNA was subsequently
isolated according to the manufacturer’s procedure.
Results and Discussion
Generation of ES Cells and Chimeras with Point Mutations in p53. To
study the possible dominant-negative effect of point mutations
in the tumor suppressor gene p53 in a physiological setting, we
generated ES cells carrying a mutation in the DNA-binding
domain of the p53 protein in a heterozygous state by gene
targeting. This approach allows for the analysis of the effects of
the mutated alleles when expressed from the p53 promoter at the
endogenous locus. With use of gene targeting, we generated ES
cells containing a mutation at codon 270 [arginine to histidine
(R270H)] or codon 275 [proline to serine (P275S)]. Both mu-
tations have been found frequently in both human and mouse
tumors, and the human equivalent of the codon 270 mutation
(R273H) is associated with LFS.
Targeting constructs containing either the R270H or P275S
mutation and a neomycin?TK-selectable marker cassette
(flanked by LoxP sites) were introduced into the mouse genome
by homologous recombination in ES cells (Fig. 1). Homologous
recombinant clones were identified by Southern blot analysis of
ES cell DNA digested with EcoRI and by using probe I, which
is situated outside the targeting vector (Fig. 1 and data not
shown). The presence of the mutation was subsequently deter-
mined by PCR. Homologous recombination frequencies were
29% and 50% for the R270H and P275S vectors, respectively. Of
these homologous clones, 26% (R270H) and 45% (P275S) of
the ES cell clones contained the mutation, indicating that in the
other clones recombination between the targeting vector and the
and the mutation in exon 8.
To exclude the possible influence by the selectable marker
of ES cell transfections was performed. For each mutation, at least
three independent and correctly targeted ES cell clones were
retargeted with a Cre-recombinase-expressing plasmid. As a result,
the neomycin?TK cassette should be removed from the targeted
p53 allele, because it is flanked by LoxP sites (Fig. 1). Excision of
the marker cassette was selected for by acquired resistance to
ganciclovir, and verified by Southern blot analysis with BamHI-
both mutations several independent clones were obtained lacking
the neomycin?TK marker cassette. These clones exclusively har-
bored the desired point mutation in p53, as detected by direct
sequencing of reverse transcription–PCR products spanning the
wild-type p53 allele seemed to be comparable in these cells, as
judged by the reverse transcription–PCR sequence signal (not
present in the mutant allele does not seem to affect expression. For
all assays described below, multiple independent mutant ES cell
clones were used.
Analysis of ES Cells. Point-mutated p53 proteins have been re-
ported to have dominant-negative properties against wild-type
p53. To test this possibility in this well-controlled cell system, we
analyzed the heterozygous p53 mutant ES cells for known p53
either the R270H or the P275S mutation (asterisk), and a neo-TK selectable marker cassette flanked by LoxP sites (first selection round). Homologous integration
In the homologous recombinant clones excision of the neo-TK selectable marker cassette was accomplished by transfection with circular pMC-CreN plasmid
(second selection round). The resulting allele differs from the wild-type allele, besides the R270H or P275S mutation, only in the presence of one LoxP site
downstream of the coding sequences.
www.pnas.org?cgi?doi?10.1073?pnas.052713099de Vries et al.
functions. As a control, homozygous and heterozygous p53
knockout ES cells were included in the experiments (17).
Expression Levels of p53-Responsive Genes upon DNA Damage. To
test what the effect of specific mutations will be on the tran-
scriptional activation of known p53 target genes, we treated ES
cells heterozygous for either the R270H or the P275S mutation
with a single dose of ? irradiation (500 cGy), an agent known to
induce p53 in many cell types. RNA was isolated at several time
points after the treatment, and expression levels of p53 targets
bax, mdm2, and cyclinG were determined. As is shown in Fig. 2,
expression levels of bax, cyclinG, and mdm2 rapidly increase on
exposure to ? irradiation in wild-type ES cells. Induction in
and decrease again afterward. p53?/?ES cells were indistin-
guishable for wild-type cells in this assay. Cells lacking p53
function (p53?/?) do not show induction of the three target
genes, demonstrating the p53 dependence of the effect. ES cells
harboring either the R270H or P275S point mutation showed a
clearly diminished activation of p53 targets response after
exposure to ? irradiation (Fig. 2). Although induction of bax,
cyclinG, and mdm2 RNA was observed, the response was
delayed and peak levels were reduced compared with controls.
Thus, we can conclude that single-copy expression of point-
mutant p53 is capable of dominant inhibition of p53 function, at
least at the level of transcriptional activity.
Apoptosis. To examine the apoptotic response of p53 mutant ES
cells, we treated them with doxorubicin, a chemotherapeutic
compound that induces double-strand breaks. Cells were treated
with two different doses of doxorubicin (1 or 2 ?g?ml), and
examined for the percentage of apoptotic cells 24 hr later by
using annexin V staining. As shown in Fig. 3, both doses of
doxorubicin caused a high percentage of wild-type cells to
undergo apoptosis (57% at 1 ?g?ml and 79% at 2 ?g?ml). In
contrast, cells completely lacking p53 were completely resistant
to 1 ?g?ml, and only 29% of these cells were dead after exposure
to 2 ?g?ml doxorubicin. The p53?/?cells had a partial response,
demonstrating an effect of gene dosage in this pathway. The
apoptotic response of the p53?/R270Hand p53?/P275SES cells was
highly similar to that of p53?/?cells. One day after treatment
with 1 ?g?ml doxorubicin, only 10–15% of the mutant cells had
undergone apoptosis, and at a dose of 2 ?g?ml 39% (p53?/R270H)
and 29% (p53?/P275S) ES cells had died. Thus, in this assay for the
cellular effects of p53 function, the presence of a single mutated
allele potently inhibited wild-type p53.
Analysis of Thymocytes by the Rag2?/?Blastocyst Complementation
Assay. The transcriptional and cell death data from ES cells were
consistent with a dominant-negative function for these point-
mutated alleles of p53. We next sought to examine the effect in
another cell system. In thymocytes, the apoptotic response on
exposure to various agents has been particularly well character-
ized, and both p53-dependent and independent pathways have
been identified (23, 24). To obtain thymocytes heterozygous for
the p53 point mutations, we injected heterozygous p53 mutant
ES cells (several independent ES cell clones of each mutation)
into Rag2?/?blastocysts. Because Rag2?/?mice do not form
mature B and T cells (22, 25), lymphocytes are derived exclu-
sively from the ES cells in the resulting chimeras.
? irradiation. (A) Wild-type D3 (???); p53?/?(???); p53?/?(???); p53?/R270H
(??R270H); and p53?/P275S(??P275S) ES cells were grown for 2 days, and
exposed to a single dose of ? irradiation (500 cGy). At 1, 3, and 6 hr after the
treatment, RNA was isolated. Northern blots were probed with bax, cyclinG,
mdm2, and gapdh cDNA probes. The 0-hr time points represent untreated
cells, isolated at the same time as the 3-hr time point after ? irradiation. (B)
Quantitation of Northern blot signals normalized for expression levels of
gapdh (used as a loading control). Note that in the bax graph, the scale of the
y axis is different from the other axes. Œ, Wild-type D3; F, p53?/?; I, p53?/?;
E, p53?/R270H; ?, p53?/P275S.
Expression levels of p53 target genes in ES cells upon treatment with
(F), p53?/?(I), p53?/R270H(E), and p53?/P275S(?) ES cells were grown to
subconfluency, and exposed to either 1 or 2 ?g?ml doxorubicin in the culture
medium. After 24 hr, the cells were stained with PI and annexin V antibody,
and the numbers of viable cells were determined by fluorescence-activated
cell sorter analysis. Data are averages of four independent experiments.
Doxorubicin-induced apoptosis in ES cells. Wild-type D3 (Œ), p53?/?
de Vries et al. PNAS ?
March 5, 2002 ?
vol. 99 ?
no. 5 ?
Reconstitution of the Lymphoid System. First, we analyzed the
reconstitution of the thymus in the chimeras. The total number
of thymocytes isolated from the thymus was on average the same
between the chimeras and, more importantly, did not differ
substantially from age-matched, nonchimeric wild-type, p53?/?
and p53?/?animals (not shown). Flow cytometric analysis of the
thymocytes isolated from mice of the different genotypes re-
vealed normal percentages of CD4?CD8?double-positive
(DP) cells (86%, 82%, and 88% for wild-type, p53?/R270H, and
p53?/P275Sanimals, respectively; data not shown). In addition,
the profile of CD4?CD8?double-negative (DN), CD4?and
CD8?single-positive cells was equal to that of wild-type, p53?/?
and p53?/?mice. For the cell death experiments, chimeras with
low thymocyte numbers or low (?70%) DP numbers were not
used for the apoptotic analyses. Finally, spleens and bone
marrow of the chimeras consisted of normally differentiated T
and B cells, as analyzed by staining the cells with CD4 and CD8;
B220 and CD3 and IgM antibodies (not shown). In conclusion,
ES cells with a heterozygous point mutation in p53 reconstitute
the T and B cell compartments normally in a Rag2?/?blastocyst
p53-Dependent Apoptosis in Thymocytes. Immature wild-type DP
thymocytes undergo apoptosis upon exposure to certain DNA-
damaging agents. For ? irradiation and doxorubicin, efficient
apoptosis requires a functional p53 pathway (23, 24). To deter-
mine whether a heterozygous point mutation in p53 can affect
this process, we exposed wild-type, p53?/?, p53?/?, p53?/R270H,
and p53?/P275Sthymocytes to both types of DNA-damaging
agents and measured percentages of apoptotic cells over time
with annexin V and PI staining. As shown in Fig. 4A, as expected
p53?/?thymocytes were highly resistant to DNA-damage-
induced apoptosis. Wild-type thymocytes died rapidly, and
p53?/?thymocytes were somewhat resistant to these agents.
Significantly, after ? irradiation, thymocytes heterozygous for
either the R270H or P275S mutation were more resistant to cell
death than either wild-type or p53?/?cells, but did display a
slightly higher degree of apoptosis than p53?/?cells. A similar
pattern of responses was observed with thymocytes of all five
genotypes after several doses of ? irradiation (not shown). After
exposure to doxorubicin the apoptotic response of the heterozy-
gous mutants was even more similar to that of p53?/?thymo-
cytes, and again, p53?/?thymocytes showed an intermediate
response (Fig. 4A). These results show clearly that thymocytes
heterozygous for a point mutation in p53 have a defect in their
apoptotic response after the induction of DNA damage, pro-
viding perhaps the best evidence to date for the dominant-
negative effects of certain tumor-associated p53 mutations.
Expression of p53-Responsive Genes in Thymocytes upon DNA Dam-
age. We next examined the induction of p53 target genes in
treated thymocytes. As shown in Fig. 4B, thymocytes show a
(0.2 ?g?ml), or dexamethasone (1 ?M). At the time points indicated, thymocytes were stained with annexin V and PI. The relative percentage of viable cells
(negative for both PI and annexin V) for each sample is shown. All values are normalized to the number of cells remaining viable in untreated cultures derived
from the same animal stained simultaneously. Data are representatives of ?2 independent experiments (i.e., mice). (B) Northern blot of p53 target genes in
thymocytes after ? irradiation (500 cGy). Two or 5 hr after the treatment, RNA was isolated, and Northern blots were probed with bax, cyclinG, and gapdh cDNA
probes. (C) Quantitation of Northern blot signals normalized for expression levels of gapdh (used as a loading control). Œ, Wild type; F, p53?/?; I, p53?/?; E,
p53?/R27OH; and ?, p53?/P275S.
The effect of heterozygous point mutations in p53 on apoptosis in thymocytes. (A) p53-dependent and -independent apoptosis. Thymocytes of mice
www.pnas.org?cgi?doi?10.1073?pnas.052713099de Vries et al.
p53-dependent induction of bax and cyclinG RNA upon ?
irradiation, with levels rising at 2- and 5-hr time points. In p53?/?
thymocytes, the induction of these genes was delayed and
reduced. The induction of bax and cyclinG RNA in thymocytes
heterozygous for the R270H or P275S mutation was comparable
to p53?/?cells, despite the clear difference in DNA-damage-
induced apoptosis between cells of these genotypes (Fig. 4 B and
C). Thus, the kinetics or extent of induction of at least these
target genes cannot fully account for the dominant-negative
affects of these p53 point mutations in inhibition of p53-
dependent apoptosis. However, given the uncertainty about the
specific target genes required for p53-dependent apoptosis, it
remains possible that critical target genes are underexpressed in
the mutant cells, accounting for the phenotype. Indeed, the
analysis of the transcriptional profile of these cells might help
define p53 targets required for apoptosis. In addition, other
functions of p53, including transcriptional repression (1), might
be affected by the presence of the point mutant protein.
p53-Independent Apoptosis. To establish whether the inhibition of
DNA damage-induced apoptosis by the p53 point mutations was
specific for the p53 pathway or indicative of a general apoptotic
defect, we treated the thymocytes of different genotypes with
stimuli that induce p53-independent apoptosis (23, 24). Isolated
thymocytes were exposed in vitro to dexamethasone (1 ?M; Fig.
4A) or a combination of phorbol 12-myristate 13-acetate?
ionomycin (10 nM?500 nM, data not shown). Wild-type, p53?/?,
and p53?/?thymocytes all died rapidly after these treatments,
harboring either the R270H or the P275S mutation were com-
parable to the thymocytes of the other three genotypes, indi-
cating that the p53-independent apoptotic pathways tested are
intact in these cells.
The analysis of mouse ES cells and thymocytes heterozygous for
tumor-associated p53 point mutations demonstrates that single-
copy expression of altered p53 protein can substantially inhibit
wild-type p53 function, which is consistent with a dominant-
negative function of the mutant alleles. These data support the
model in which tumor cells progressively lose p53 function
through the acquisition of such p53 mutations, often followed by
loss of the wild-type p53 allele. These studies also reveal
assay-specific differences in the effects of these mutations, which
may help account for some of the previous conflicting data on
this point. As discussed above, a disparity occurred between the
induction of p53 target genes in mutant thymocytes treated with
DNA-damaging agents and the inhibition of apoptosis in these
cells. In addition, as judged by standard in vitro p53 DNA-
binding assays using extracts from unstimulated wild-type and
mutant ES cells, the presence of point mutant p53 did not
significantly affect overall p53 DNA-binding activity (data not
shown). Given that these assays are performed in the presence
of p53 antibody to stabilize the protein–DNA complex, it is
possible that the effects of the mutations were masked. It will be
necessary to assay p53 DNA binding in vivo under different
conditions (and possibly at different promoters) to assess accu-
rately the effects of these p53 mutations.
The ability of these point-mutant p53 alleles to affect p53-
dependent apoptosis potently has important implications for
understanding the course of tumorigenesis in sporadic cancers
with p53 mutations and in patients with LFS. Specifically, these
data show clearly that emerging tumor cells that acquire this sort
of p53 point mutation will have a selective advantage under
conditions that induce p53-dependent apoptosis, including after
endogenous DNA damage events or DNA damage induced by
chemotherapy or radiation. Indeed, the data underscore the
functional difference between the acquisition of a p53 missense
versus loss-of-function mutation in the gene in the course of
tumor development. These results also raise the interesting
possibility that LFS patients carrying certain types of p53
mutations may have less severe side effects of treatment with
some forms of chemotherapy or radiation, given that many of
these effects are secondary to p53-dependent apoptosis in
normal cell types. It will be interesting to examine the effects of
these and other p53 point mutations on tumorigenesis in the
mouse, and possible gain-of-function properties of certain p53
We thank Laura Attardi, Timo Breit, and Harry van Steeg for helpful
discussions and critical reading of the manuscript. We thank Denise
Crowley and Roderick Bronson for histological and pathological analysis
of mice, Jan de Wit for performing the ?-induced cell death assays in ES
cells, and Tara Schmidt for breeding and maintenance of the RAG2?/?
mouse colony. We are grateful to Gigi Lozano for the mdm-2 probe, to
Jackie Lees for the cyclinG probe, to Frank Gertler for the pGK-DTA
plasmid, to Susumu Tonegawa for the loxP-neo-TK-loxP selectable
marker cassette, and to Fred Alt for the pMC-CreN plasmid. This work
was supported in part by funding from the Dutch Cancer Society (to
A.d.V.). T.J. is an Associate Investigator at the Howard Hughes Medical
1. Levine, A. J. (1997) Cell 88, 323–331.
2. Cox, L. S. & Lane, D. P. (1995) BioEssays 17, 501–508.
3. Gottlieb, T. M. & Oren, M. (1996) Biochim. Biophys. Acta 1287, 77–102.
4. Ko, L. J. & Prives, C. (1996) Genes Dev. 10, 1054–1072.
5. Lowe, S. W., Bodis, S., McClatchey, A., Remington, L., Ruley, H. E., Fisher,
D. E., Housman, D. E. & Jacks, T. (1994) Science 266, 807–810.
6. Aas, T., Borresen, A. L., Geisler, S., Smith-Sorensen, B., Johnsen, H., Varhaug,
J. E., Akslen, L. A. & Lonning, P. E. (1996) Nat. Med. 2, 811–814.
7. Delia, D., Goi, K., Mizutani, S., Yamada, T., Aiello, A., Fontanella, E.,
Lamorte, G., Iwata, S., Ishioka, C., Krajewski, S., et al. (1997) Oncogene 14,
8. Greenblatt, M. S., Bennett, W. P., Hollstein, M. & Harris, C. C. (1994) Cancer
Res. 54, 4855–4878.
9. Hollstein, M., Sidransky, D., Vogelstein, B. & Harris, C. C. (1991) Science 253,
10. Malkin, D. (1994) Biochim. Biophys. Acta 1198, 197–213.
11. Kern, S. E., Pietenpol, J. A., Thiagalingam, S., Seymour, A., Kinzler, K. W. &
Vogelstein, B. (1992) Science 256, 827–830.
12. Milner, J., Medcalf, E. A. & Cook, A. C. (1991) Mol. Cell. Biol. 11, 12–19.
13. Milner, J. & Medcalf, E. A. (1991) Cell 65, 765–774.
14. Attardi, L. D. A. & Jacks, T. (1999) Cell. Mol. Life Sci. 55, 48–63.
15. Harvey, M., Vogel, H., Morris, D., Bradley, A., Bernstein, A. & Donehower,
L. A. (1995) Nat. Genet. 9, 305–311.
16. Liu, G., McDonnell, T. J., Montes de Oca Luna, R., Kapoor, M., Mims, B.,
El-Naggar, A. K. & Lozano, G. (2000) Proc. Natl. Acad. Sci. USA 97, 4174–4179.
17. Jacks, T., Remington, L., Williams, B. O., Schmitt, E. M., Halachmi, S.,
Bronson, R. T. & Weinberg, R. A. (1994) Curr. Biol. 4, 1–7.
18. Macleod, K., Hu, Y. & Jacks, T. (1996) EMBO J. 15, 6178–6188.
19. Macleod, K., Sherry, N., Hannon, G., Beach, D., Tokino, T., Kinzler, K.,
Vogelstein, B. & Jacks, T. (1995) Genes Dev. 9, 935–944.
20. Hurford, R. K., Cobrinik, D., Lee, M. & Dyson, N. (1997) Genes Dev. 11,
& Jacks, T. (1996) Genes Dev. 10, 2055–2066.
22. Chen, J., Lansford, R., Stewart, V., Young, F. & Alt, F. W. (1993) Proc. Natl.
Acad. Sci. USA 90, 4528–4532.
23. Lowe, S. W., Schmitt, E. M., Smith, S. W., Osborne, B. A. & Jacks, T. (1993)
Nature (London) 362, 847–849.
24. Clarke, A. R., Purdie, C. A., Harrison, D. J., Morris, R. G., Bird, C. C., Hooper,
M. L. & Wyllie, A. H. (1993) Nature (London) 362, 849–852.
25. Shinkai, Y., Rathbun, G., Lam, K. P., Oltz, E. M., Stewart, V., Mendelsohn, M.,
Charron, J., Datta, M., Young, F., Stall, A. M., et al. (1992) Cell 68, 855–867.
de Vries et al.PNAS ?
March 5, 2002 ?
vol. 99 ?
no. 5 ?