ã
Oncogene (2000) 19, 3126 ± 3130
2000 Macmillan Publishers Ltd All rights reserved 0950 ± 9232/00 $15.00
www.nature.com/onc
SHORT REPORT
p51A (TAp63g), a p53 homolog, accumulates in response to DNA damage
for cell regulation
Iyoko Katoh1, Ken-ichi Aisaki2, Shun-ichi Kurata3, Shuntaro Ikawa4 and Yoji Ikawa*,1,2
1
Department of Retroviral Regulation, Medical Research Division, Tokyo Medical and Dental University, Tokyo 113-8519, Japan;
Human Gene Sciences Center, Tokyo Medical and Dental University, Tokyo 113-8510, Japan; 3Medical Research Institute, Tokyo
Medical and Dental University, Tokyo 113-8510, Japan; 4Department of Cell Biology, Institute of Development, Aging and Cancer,
Tohoku University, Sendai 980-8575, Japan
2
p51A, or TAp63g, a translation product of gene p51, or
p63, was identi®ed as a homolog of p53 in its primary
structure and transactivating function. p53 plays a
decision-making role in inducing either cell cycle arrest
or apoptosis in response to DNA damage, and thereby
preserves genome integrity of living cells. To compare
the biological activities between p51A and p53, cell lines
with low-level, constitutive expression of each protein
were obtained by cDNA transfection of mouse erythroleukemic cells. Production of p51A with an apparent
molecular mass of 57-kilodalton (kD) accompanied
induction of p21waf1 and appearance of hemoglobinproducing cells. After DNA-damaging treatment either
with ultraviolet light (UV) irradiation or with actinomycin D, the p51A protein accumulated in time courses
corresponding to those of wild-type p53, and caused an
increase in the hemoglobin-positive cell count. In
contrast, p53-accumulated cells underwent apoptosis
without exhibiting the feature of erythroid di€erentiation. The mode of p21waf1 and Bax-a upregulations varied
between p51A- and p53-expressing cells and between the
types of DNA damage. These results suggest the
possibility that p51A induces di€erentiation under
genotoxic circumstances. There may be cellular factors
that control p51A protein stability and transactivating
ability. Oncogene (2000) 19, 3126 ± 3130.
Keywords: p51; p63; p53; p21waf1; Bax; DNA damage
p51, also termed p63, a member of the p53-family
genes, encodes p51A (TAp63g) with strong structural
and functional similarities to p53 (Osada et al., 1998;
Yang et al., 1998). However, p51 considerably di€ers
from p53 in view of (i) infrequent p51 gene mutations
detected in tumors (Sunahara et al., 1999; Hagiwara et
al., 1999; (ii) p51 expression in speci®c cell types and
stages in development (Osada et al., 1998; Mills et al.,
1999; Yang et al., 1999); and (iii) production of
multiple translates from p51 including one with a
dominant-negative activity (Yang et al., 1998). It is in
question whether or not p51 functions as a tumor
suppressor. Here we show that p51A also has a feature
to accumulate in response to DNA damage to induce
*Correspondence: Y Ikawa
Received 27 January 2000; revised 19 April 2000; accepted 20 April
2000
gene expression, proposing that it is another property
shared by the three classes of p53 family proteins
(White and Prives, 1999).
Cell line EL-Tg-gp55-1-2-3 (Xu et al., 1995; Kato et
al., 1997), here referred to as 1-2-3, was established
from an enlarged spleen of a mouse with the
polycythemic Friend virus gp55 transgene (Aizawa et
al., 1990) and has monoallelic expression of p53val135, a
temperature-sensitive mutant p53 (Michalovitz et al.,
1990; Martinez et al., 1991). We generated 1-2-3derived cell lines with p51A or wild-type p53 expression
driven by Rous sarcoma virus (RSV) promoter which
did not undergo growth suppression at 378C. Reportedly, the p51A entire open reading frame (ORF)
includes the region spanning exons 1 and 2 that
encodes the 39 amino acids at the extreme NH2terminus of the protein (Hagiwara et al., 1999). The
human p51A cDNA cloned into the expression vectors
in this study was found to have the intact ORF
encoding 487 amino acid residues. Furthermore, only
11 residue substitutions are present in mouse *TAp63g
consisting of 483 residues (Yang et al., 1993).
Figure 1a shows analysis of the transfectants with
p51A, antisense p51A (Rp51A), p53, or antisense p53
(Rp53) for their gene expressions by semi-quantitative
reverse transcription-polymerase chain reaction (RT ±
PCR). RNA preparations from p51A-2C6, -3C2, -3D6
and -4E7, and Rp51A-2E8 (lanes 2 ± 6) were positive in
production of 629 base-pair (bp) DNA primed with
p51 transactivation domain-speci®c sequences. p53-3A6
and -4D5, and Rp53-2C3 (lanes 8 ± 10) were positive in
production of 699 bp p53-speci®c DNA. PCR without
RT failed to show those bands, assuring mRNA
syntheses from the genes introduced. RT ± PCR for
glyceraldehyde phosphate dehydrogenase (GAPDH)
mRNA indicated adequately performed RNA quantitation. Figure 1b shows protein analyses of the cell
lines by immunoblotting. Murine p53val135 detected by
PAb421, a universal antibody to p53, appeared as a
dense band in every lane. With 4A4, a p51 (p63)speci®c monoclonal antibody, a faint band at the
position of 57 kD marker was detected in the lanes for
p51A-3C2 and -3D4 (lanes 3 and 4), evidencing the
p51A protein which cells retained poorly as reported
(Yang et al., 1998). The apparent molecular mass
matched the size of a protein of 487 amino acid
residues. In an analysis with a human p53-speci®c
monoclonal antibody, PAb1801, p53 appeared as a
clear band in the lanes for p53-4A6 and -4D5 (lanes 7
and 8) re¯ecting relative quantities of their p53 mRNA
p51A responds to DNA damage
I Katoh et al
Figure 1 Detection of p51A and p53 expressions in 1-2-3
transfectants. (a) RNA analyses by gene ampli®cation. Identity
of the cell lines established is indicated above. Results of the
p51A- and p53-speci®c RT ± PCR and PCR without RT are
shown. Control RT ± PCR for GAPDH is also shown. Bands of
the 100 bp ladder (Life Technologies, Inc.) are indicated. P51A
and p53 cDNAs were cloned into pOPRSVI/MCS vector
(Stratagene) in sense and anti-sense orientations and introduced
into 1-2-3 cells using DMRIE-C reagent (Life Technologies Inc.).
G418 (600 mg/ml)-resistant cell clones were obtained. Cells were
cultured in normal growth medium (NGM) consisting of
RPMI1640, 10% fetal bovine serum and antibiotics for 2 ± 4
weeks before harvest. From total RNA (0.4 mg for a reaction),
target RNA was reverse transcribed by SuperScriptII RT (Life
Technologies Inc.) with PCR primers, and directly ampli®ed with
Taq DNA polymerase in a volume of 40 ml for 28 cycles. Primer
sequences are as follows: for detection of p51A, 5'-ATGTCCCAGAGCACACAG-3' and 5'-AGCTCATGGTTGGGGCAC-3'; for
p53, 5'-AGTGGATCCAGACTGCCTTCCGGGTCACTG-3' and
5'-GCGGATCCTAGGGCACCACCACACTAT-3'; for p21waf1,
5'-CGGTCCCGTGGACAGTGAGCAG-3' and 5'-GTCAGGCTGGTCTGCCTCCG-3'; for GAPDH, 5'-GGGTGGAGCCAAACGGGTC-3' and 5'-GGAGTTGCTGTTGAAGTCGCA-3'.
Conditions were: 948C for 30 s, 558C (or 608C) for 30 s, 728C
for 60 s. (b) Protein analyses by immunoblotting. Monoclonal
antibodies used are indicated on the left. Positions of biotinylated
size markers (New England BioLabs) of 46.5 and 57 kD are
marked by bars for each panel. Proteins detected are denoted with
arrows on the right. Results of the RT ± PCR for p21waf1 are also
shown at the bottom. For sample preparation, cells (106) were
washed on ice, lysed in the SDS ± PAGE sample bu€er (100 ml)
and boiled immediately. After sonication for DNA shearing, a
10 ml aliquot was analysed in each lane. Proteins were transferred
to polyvinylidene di¯uoride membrane, blocked, and incubated
with a primary antibody (0.1 mg/ml). Alkaline phosphataseconjugated secondary antibodies and CDP-Star chemiluminescent
substrate (New England BioLabs) were used for detection
(Figure 1a). When analysed by RT ± PCR, the cell lines
with obvious protein synthesis from transfected genes,
p51A-3C2, p53-4A56 and p53-4D5, had upregulation
in the expression of p21waf1, mouse p21waf1 originally
termed CAP20 (Gu et al., 1993) (lanes 3, 7 and 8). In
p51A-3D4 and -4E7 (lanes 4 and 5), it was less
apparent. Other cells did not have this feature (lanes 1,
6 and 9). Based on these data, we considered p51A-3C2
and p53-3A6 as the cell lines that grew producing
active p51A and p53, respectively, and used them in
the following experiments.
To examine whether p51A has the ability to respond
to DNA damage, cells were subjected to DNAdamaging treatment, UV-C (wavelength of 254 nm)
irradiation (20 J/m2) or incubation with actinomycin D
at 25 nM (Woo et al., 1998). More than 100 J/m2 of
UV irradiation was required for induction of 100% cell
death in the 1-2-3 culture. Twenty-®ve nM of
actinomycin D is well below the drug concentration
generally used for transcription inhibition (1 ± 2 mM),
and did not a€ect the RNA recovery within 24 h. The
mutant p53val135 protein amount detected by PAb421
was stable in 1-2-3 cells (Figure 2a) through the time
course to 24 h after UV irradiation as well as at the 4
and 24 h points of the actinomycin D treatment. In
contrast, PAb1801-detectable p53 band intensity increased signi®cantly after UV treatment (lanes 1 ± 4),
reaching maximum at 4 h in p53-3A6. It reduced to the
basal level in 24 h (lane 5), indicating transient
accumulation. p53 enhancement was also apparent in
actinomycin D-treated p53-3A6 (lanes 6 and 7). As
observed in various experiments in human and mouse
systems (Lu and Lane, 1993; Nelson and Kastan,
1994), human p53 stabilization by DNA damage was
achieved in the mouse cell line. Intriguingly, the p51A
protein content in p51A-3C2 cells increased after
exposure to UV or actinomycin D in time courses
similar to those of p53, indicating that p51A has the
ability to accumulate in response to DNA injury. RNA
analysis by ampli®cation (Figure 2b) showed that the
level of p51A mRNA relative to the control GAPDH
mRNA was not altered in p51A-3C2 cells, nor was the
level of p53 mRNA in p53-3A6 cells. Thus, p51A
accumulation occurred by a post-transcriptional mechanism as in the case of p53 (Levine, 1997; Ko and
Prives, 1996).
Neither p51A nor p53 expression at the steady state
level a€ected the protein amount of Bax-a, a 21 kD
active translate from the bax gene (Oltvai et al., 1993),
as found in the immunoblot analysis (Figure 2a, lane 1
of 1-2-3, p51-3C2 and p53-3A6). Moreover, in the
control 1-2-3 cells, the Bax-a content was unaltered
during the 24 h period of UV and actinomycin D
experiments. However, the band intensi®ed in the lanes
for UV- or actinomycin D-treated p53-4A6 samples
(lanes 2 ± 4, 6 and 7), coinciding with p53 accumulation. In p51A-3C2, the Bax-a protein composition
increased by 24 h incubation with actinomycin D,
whereas the relative protein amount did not signi®cantly alter after UV irradiation.
As described above, p21waf1 mRNA synthesis was
enhanced in p51A-3C2 and p53-3A6 under normal
culture conditions. It rose further in p51A-3C2 cells for
8 h (lanes 2 ± 4) after UV irradiation, and declined to
the steady-state level in 24 h (lane 5), suggesting
rapidly induced p21waf1 transcription by p51A accumu-
3127
Oncogene
p51A responds to DNA damage
I Katoh et al
3128
lation. In UV-irradiated p53-4A6 cells, increased p21waf1
expression was detected at 8 and 24 h, indicating
discordance between the p53 protein level and p21waf1
upregulation, which is consistent with earlier results
(Lu et al., 1996; HaapajaÈrvi et al., 1997). The p21waf1
mRNA level was reduced to the basal level during the
actinomycin D treatment in the p51A and p53
expressing cells (lanes 6 ± 8). In the control 1-2-3 cells,
neither of the treatments caused p21waf1 upregulation to
a detectable level. Thus, at least two of the known p53responsive genes seemed to undertake the regulation of
p51A in p51A-3C2 cells. However, p21waf1 and Bax-a
Figure 2 Protein and RNA alterations caused by DNA damage.
(a) Analysis of UV or actinomycin D-treated cells for protein
composition. Three cell lines used are indicated on the left of the
panels. Proteins detected are noted on the right. Time courses (in
hours) after UV irradiation and by incubation with actinomycin D
(act. D) are shown at the top. Untreated control sample is
indicated by (7). Positions of size markers are also indicated in
kD. For detection of b-actin, 4A4- or PAb1801-probed membranes were reprobed with an anti-b-actin antibody (Santa Cruz).
For DNA damage experiments, cells were plated (56105/ml) in 6cm dishes with NGM (5 ml). UV irradiation was performed by
exposing the cultures to a UV lamp so that it delivered 20 J/m2 on
the surface of the dishes. Actinomycin D solution (0.5 mM) made
in ethanol was diluted to 25 nM with NGM prior to incubation.
Cells for protein analysis were washed and lysed by adding 100 ml
of the SDS ± PAGE sample bu€er to 106 cells found intact under
microscopy. (b) RNA analysis by RT ± PCR. Cell lines, RNA
template ampli®ed, positions of size markers (in bp) and time
points are indicated. For p51A and p53 RNA detection, Taq
polymerase reaction was performed for 26 cycles
Oncogene
responses varied depending on whether p51A or p53
accumulated, as well as with the methods of DNA
damage.
Nuclear staining was performed at 24 h in the UV
and actinomycin D experiments (Figure 3a ± c) to
detect apoptosis induction. In 1-2-3 and Rp51A-2E8
cultures, 9.5 ± 13% of the cells contained a condensed,
fragmented nucleus, indicating occurrence of p53independent apoptosis at a lower incidence. In
contrast, p53-3A6 cultures had a 68% fraction of
Figure 3 Cellular responses to DNA damage. (a) Induction of
apoptosis. The ratio (in per cent) of [count of cells with
fragmented nucleus] to [total cell count (N4500)] was determined
after nuclear staining. Cell lines and type of treatment are
indicated. Mean of two separate experiments is shown. None of
the raw values deviated more than +20% from the mean. (b)
Detection of nuclear condensation and fragmentation. Cells
stained with Hoechst 33342 are shown. For nuclear stain, cells
were ®xed in glutaraldehyde (1.25% in phosphate-bu€ered saline,
PBS) for 16 h at 48C, and then the glutaraldehyde washed away
with PBS. Cell suspension was combined with Hoechst 33342
solution (6 mM) at 5 : 1 and observed under ¯uorescence
microscopy. (c) Detection of hemoglobin-producing cells by
DAF-staining. Live cells were directly incubated with DAF
(0.005%) and 0.15% H2O2 in 25 mM Tris-HCl, pH 7.2, and
observed microscopically. Cells in dark blue color were counted
as DAF-stain-positive cells
p51A responds to DNA damage
I Katoh et al
apoptotic cells after UV irradiation, and a 30%
fraction after incubation with actinomycin D, indicating that p53-dependent apoptosis was eciently
induced. However, the population of dying cells in
DNA-damaged p51A-3C2 cultures, 18% in UVirradiated samples and 8.6% in actinomycin D-treated
sampled, did not vary from those in 1-2-3 and Rp51A2E8 treated in the same way.
We next examined these cells for hemoglobin
production as a characteristic of erythroid di€erentiation by staining with 2,7-diamino¯uorene (DAF).
When cultured under normal growth conditions, a
0.5% cell population in p51A-3C2 cultures was found
DAF-stain positive. One day after exposure to UV or
actinomycin D, the DAF-stain-positive cell count
increased to 3.6% or to 6.0%, respectively. In contrast,
less than 0.1% population in the 1-2-3, Rp51A-2E8
and p53-3A6 cultures were stained with DAF, and the
ratio did not increase after UV or actinomycin D
treatment. Taken together, while p53 accumulation
resulted in induction of apoptosis, p51A accumulation
led to hemoglobin production.
The most intriguing result in this study was that
p51A molecules also accumulated and elicited their
transactivating function in response to DNA damage.
The mechanism that controls p53 stability and activity
involves a p53-binding protein, MDM2 (Shieh et al.,
1997; Haupt et al., 1997; Kubbutat et al., 1997), and
kinases represented by ATM (Banin et al., 1998;
Canman et al., 1998). MDM2 directly binds to the
NH2-terminus of p53. Molecular association between
p51A and MDM2 has not been proved. Another p51
translate p51B (TAp63a) is well retained in the same
expression system. Sequences determining the p51A
stability are, therefore, speculated to reside in the
COOH-terminus of p51A. DNA-damage-induced p51A
phosphorylation is expected.
The result of hemoglobin production observed in
p51A-expressing cells should not be taken to indicate
that p51A functions in erythroid di€erentiation in vivo,
since p63 gene expression is con®ned to the sites of
epidermal-mesenchymal interactions in mouse embryogenesis, and p63-mutated humans and animals are
de®cient in limb and skin development (Mills et al.,
1999; Yang et al., 1999; Celli et al., 1999). Our result in
cell culture is interpreted to indicate that p51A, when it
is expressed in an active form, could lead to cell
di€erentiation. Neither p21waf1 nor Bax-a seemed to
send a direct signal for di€erentiation.
In spite of the well conserved primary and putative
higher order structures between p51A and p53, and
comparable transactivating ability found in reporter
gene expression assays with common sequence motif,
the two proteins induced di€erent biological responses
to DNA damage. This may not be attributable to the
quantitative di€erence in their transactivational activity, if any (Shimada et al., 1999), under the normal
cellular circumstances. Induction of p21waf1, a CDK
inhibitor responsible for p53-caused G1-S transition
arrest in cell cycle (El-Deiry et al., 1993), did not occur
in parallel with that of Bax-a, a mitochondrial protein
implicated in p53-induced apoptosis (Miyashita and
Reed, 1995). More selective activation may occur
among p53-responsive genes in the nuclear events
following DNA scission resulting from UV irradiation
or DNA intercalation by actinomycin D, which may
further vary between p51A- and p53-expressing cells
and depending on cell types.
In our earlier experiment, cytomegalovirus promoter-driven, 5 ± 10-fold more ecient p51A gene expression resulted in apoptosis induction (Osada et al., 1998)
as well as hemoglobin production. When overexpressed, p51A, as well as p53, possibly dominate the
cellular mechanism that controls stability and activity
of the molecule.
Another p53-homologue, p73a was originally found
to be insensitive to UV or actinomycin D (Kaghad et
al., 1997). In recent reports, however, p73a also
accumulates to induce apoptosis in response to ionizing
irradiation and cisplatin, where c-Abl tyrosine kinase is
required (Gong et al., 1999; Agami et al., 1999; Yuan
et al., 1999). Thus, p53 family proteins, p53, p73a, and
p51A (TAp63g) are indeed closely related not only
structurally but also functionally. An elaborate cellular
signaling system may be present for each protein
regulation.
3129
References
Agami R, Blandino G, Oren M and Shaul Y. (1999). Nature,
399, 809 ± 813.
Aizawa S, Suda Y, Furuta Y, Yagi T, Takeda N, Watanabe
N, Nagayoshi M and Ikawa Y. (1990). EMBO J., 9, 2107 ±
2116.
Banin S, Moyal L, Shieh S-Y, Taya Y, Anderson CW,
Chessa L, Smorodinsky NI, Prives C, Reiss Y, Shiloh Y
and Ziv Y. (1998). Science, 281, 1674 ± 1677.
Canman CE, Lim D-S, Cimprich KA, Taya Y, Tamai K,
Sakaguchi K, Appella E, Kastan MB and Siliciano JD.
(1998). Science, 281, 1677 ± 1679.
Celli J, Duijf P, Hamel BCJ, Bamshad M, Kramer B, Smits
APT, Newbury-Ecob R, Hennekam RCM, Buggenhout
GV, Haeringen AV, Woods CG, Essen AJV, Waal RD,
Vriend G, Haver DA, Yang A, McKeon F, Brunner HG
and Bokhoven HV. (1999). Cell, 99, 143 ± 153.
El-Deiry WS, Tokino T, Velculescu VE, Levy DB, Parsons
R, Trent JM, Lin D, Mercer WE, Kinzler KW and
Vogelstein B. (1993). Cell, 75, 817 ± 825.
Gong JG, Costanzo C, Yang H-Q, Melino G, Kaelin Jr WG,
Levrero M and Wang JY. (1999). Nature, 399, 806 ± 809.
Gu Y, Turk CW and Morgan DO. (1993). Nature, 366, 707 ±
710.
HaapajaÈrvi T, PitkaÈnen K, Tsubari M and Laiho M. (1997).
Mol. Cell. Biol., 17, 3074 ± 3080.
Hagiwara K, McMenamin MG, Miura K and Harris CC.
(1999). Cancer Res., 59, 4165 ± 4169.
Haupt Y, Maya R, Kazaz A and Oren M. (1997). Nature,
387, 296 ± 299.
Kaghad M, Bonnet H, Yang A, Creancier L, Biscan JC,
Valent A, Minty A, Chalon P, Ferrara P, McKeon F and
Caput D. (1997). Cell, 90, 809 ± 819.
Kato MV, Sato H, Nagayoshi M and Ikawa Y. (1997). Blood,
90, 1373 ± 1378.
Ko LJ and Prives C. (1996). Genes Dev., 10, 1054 ± 1072.
Kubbutat MHG, Jones SN and Vousden KH. (1997).
Nature, 387, 299 ± 308.
Levine AJ. (1997). Cell, 88, 323 ± 331.
Lu X and Lane DP. (1993). Cell, 75, 765 ± 778.
Lu X, Burbidge SA, Grin S and Smith HM. (1996).
Oncogene, 13, 413 ± 418.
Oncogene
p51A responds to DNA damage
I Katoh et al
3130
Oncogene
Martinez J, Georgo€ I, Martinez J and Levine AJ. (1991).
Genes Dev., 5, 151 ± 159.
Michalovitz D, Halevy O and Oren M. (1990). Cell, 62, 671 ±
680.
Mills AA, Zheng B, Wang X-J, Voge H, Roop DR and
Bradley A. (1999). Nature, 398, 708 ± 713.
Miyashita T and Reed JC. (1995). Cell, 80, 293 ± 299.
Nelson WG and Kastan MB. (1994). Mol. Cell. Biol., 14,
1815 ± 1823.
Oltvai ZN, Milliman C and Korsmeyer SJ. (1993). Cell, 74,
609 ± 619.
Osada M, Ohba M, Kawahara C, Ishioka C, Kanamaru R,
Katoh I, Ikawa Y, Nimura Y, Nakagawara A, Obinata M
and Ikawa S. (1998). Nature Med., 4, 839 ± 843.
Shieh S-Y, Ikeda M, Taya T and Prives C. (1997). Cell, 91,
325 ± 334.
Shimada A, Kato S, Enjo K, Osada M, Ikawa Y, Obinata M,
Ikawa S and Ishioka C. (1999). Cancer Res., 59, 2781 ±
2786.
Sunahara M, Shishikura T, Takahashi M, Todo S,
Yamamoto N, Kimura H, Kato S, Ishioka C, Ikawa S,
Ikawa Y and Nakagawara A. (1999). Oncogene, 18, 3761 ±
3765.
White E and Prives C. (1999). Nature, 399, 734 ± 737.
Woo RA, McLure KG, Lees-Miller SP, Rancourt DE and
Lee PWK. (1998). Nature, 394, 700 ± 704.
Xu X, Yamamura Y, Tsukada T, Yoshida MA, Senda H,
Nagayoshi M, Ikeuchi T and Ikawa Y. (1995). Jpn. J.
Cancer Res., 86, 284 ± 291.
Yang A, Kaghad M, Wang Y, Gillet E, Flemming MD,
DoÈtsch V, Andrews NC, Caput D and McKeon F. (1998).
Mol. Cell., 2, 305 ± 316.
Yang A, Schweizer R, Sun D, Kaghad M, Walker N,
Bronson RT, Tabin C, Sharpe A, Caput D, Crum C and
McKeon F. (1999). Nature, 398, 714 ± 719.
Yuan Z-M, Shioya H, Ishiko T, Sun X, Gu J, Huang YY, Lu
H, Kharbanda S, Weichselbaum R and Kufe D. (1999).
Nature, 399, 814 ± 817.