Download Induction of apoptosis by Smad3 and down-regulation of

Oncogene (1998) 17, 1743 ± 1747
 1998 Stockton Press All rights reserved 0950 ± 9232/98 $12.00
http://www.stockton-press.co.uk/onc
SHORT REPORT
Induction of apoptosis by Smad3 and down-regulation of Smad3 expression
in response to TGF-b in human normal lung epithelial cells
Kiyoshi Yanagisawa1,5, Hirotaka Osada1,2, Akira Masuda1, Masashi Kondo1,5, Toshiko Saito1,
Yasushi Yatabe4, Kenzo Takagi5, Toshitada Takahashi3 and Takashi Takahashi1,2
1
Laboratory of Ultrastructure Research, 2Pathophysiology Unit and 3Laboratory of Immunology, Aichi Cancer Center Research
Institute, Kanokoden, Chikusa-ku, Nagoya 464; 4Department of Pathology and Clinical Laboratories, Aichi Cancer Center
Hospital, Kanokoden, Chikusa-ku, Nagoya 464; 5Internal Medicine 2, Nagoya University School of Medicine, Tsurumai-cho,
Showa-ku, Nagoya 466, Japan
Smad family members are essential intracellular signaling components of the transforming growth factor-beta
(TGF-b) superfamily involved in a range of biological
activities. Two highly homologous molecules, Smad2 and
Smad3, have so far been identi®ed as receptor-activated
Smads for TGF-b signaling and have become the focus
of intensive studies. However, no de®nite di€erences in
regulation or function have been established between
these TGF-b signaling molecules. In the present study,
we show that the expression of Smad3, but not its close
relative, Smad2, is down-regulated by TGF-b mediated
signals themselves in human lung epithelial cells. This
down-regulation of Smad3 by TGF-b treatment did not
appear to result from shortening of the half-life of
Smad3 mRNA. Constitutive expression of Smad3 in the
presence of TGF-b induced apoptotic cell death, with an
adverse e€ect on the cell growth of human lung epithelial
cells. Apoptotic cell death could also be induced by
forced expression of Smad2 in the presence of TGF-b,
but less eciently than by that of Smad3. These ®ndings
clearly de®ne the distinctions between Smad2 and Smad3
for the ®rst time in that a qualitative di€erence was
observed with regard to the regulation of their expression
in response to TGF-b, while Smad2 and Smad3 appeared
to have quantitatively di€erent capabilities regarding the
induction of apoptotic cell death in human lung epithelial
cells.
Keywords: Smad; TGF-b; apoptosis; down-regulation;
lung
The TGF-b superfamily members, including activin,
bone morphogenetic protein and TGF-b, play important regulatory roles in cell growth, morphogenesis, cell
di€erentiation, and apoptosis (Kingsley, 1994; Massague, 1996; Roberts and Sporn, 1993). Signaling
mediators downstream of the receptors for the TGFb superfamily were recently identi®ed in a variety of
organisms, and termed `Smad' in vertebrates (Derynck
et al., 1996). While seven Smad genes have been
reported in the literature thus far, they can be classi®ed
into three types according to their functions, i.e.,
receptor-activated Smads (Smad1 and Smad5 for BMP;
Smad2 and Smad3 for TGF-b and activin), co-Smad
Correspondence: Takashi Takahashi
Received 28 January 1998; revised 14 April 1998; accepted 15 April
1998
(Smad4) and anti-Smads (Smad6 and Smad7) (Baker et
al., 1996; Derynck et al., 1996; Eppert et al., 1996;
Gra€ et al., 1996; Hahn et al., 1996; Hayashi et al.,
1997; Hoodless et al., 1996; Lechleider et al., 1996; Liu
et al., 1996; Riggins et al., 1996; Thomsen, 1996;
Topper et al., 1997; Yingling et al., 1996; Zhang et al.,
1996). Smad2 and Smad3, the receptor-activated, TGFb signaling Smads, are known to be highly homologous
with regard to amino acid sequence and structural
characteristics, and both have been found to mediate
TGF-b and activin signals by forming heteromers with
Smad4 (Eppert et al., 1996; Gra€ et al., 1996; Riggins
et al., 1996; Wrana and Pawson, 1997; Zhang et al.,
1996). Despite extensive studies of the Smad gene
family, however, no de®nitive di€erences in the
regulations or functions of these receptor-activated
Smads have been identi®ed.
The loss of sensitivity to TGF-b has been shown to
be common in human cancers (Roberts and Sporn,
1993). While Smad4 was ®rst isolated as a pancreatic
tumor suppressor gene (Hahn et al., 1996), mutations
in the Smad2 and Smad4 genes have also been
reported in other common cancers of adults (Eppert
et al., 1996; Nagatake et al., 1996; Riggins et al., 1996;
Uchida et al., 1996). Our ®ndings that Smad2 and
Smad4 both at 18q21 were somatically mutated in a
fraction of human lung cancers (Nagatake et al., 1996;
Uchida et al., 1996) suggested that members of the
Smad gene family may play a role in the pathogenesis
of lung cancer, since normal lung epithelial cells are
known to exhibit a growth-inhibitory response to
TGF-b (Masuda et al., 1997; Masui et al., 1986).
The present study was undertaken to gain further
insights into the regulation and roles of the TGF-b
signaling pathway in lung epithelial cells. Using
BEAS2B (Reddel et al., 1988) and HPL1A lung
epithelial cell lines (Masuda et al., 1997), we ®rst
examined whether various external stimuli can alter
the expression levels of TGF-b signaling Smads, i.e.,
Smad2, Smad3 and Smad4. Northern blot analysis
showed readily detectable expression of all three
Smads in both BEAS2B and HPL1A (Figure 1a).
Addition of 1 ng/ml of TGF-b to the culture media
resulted in a signi®cant reduction of Smad3 mRNA
expression in both cell lines, whereas expression levels
of Smad2 and Smad4 remained virtually unaltered. In
contrast, treatment with 10 ng/ml of epidermal growth
factor (EGF) or use of various concentrations (0.1, 1,
10%) of fetal calf serum had no e€ect on either the
expression of Smad3 or that of Smad2 and Smad4
TGF-û signaling Smads in lung epithelial cells
K Yanagisawa et al
1744
signal down-regulates Smad3 mRNA expression, we
examined the half-lives of Smad3 mRNA in the
presence or absence of TGF-b in BEAS2B. BEAS2B
cells were cultured for 8 h in the presence or absence
of 1 ng/ml of TGF-b and then chased in the presence
of 5 mg/ml of actinomycin D together with 1 ng/ml of
TGF-b for up to 12 h. Northern blot analysis using
RNAs harvested at various time points revealed that
the half-life of Smad3 mRNA was 4.7 h with, and
5.1 h without TGF-b treatment, indicating that TGFb treatment did not signi®cantly alter the stability of
Smad3 mRNA (Figure 2). These results suggested that
the reduction of Smad3 mRNA levels may be mainly
caused by the decrease in the Smad3 gene transcription. Smad2 mRNA was found to have a considerably
longer half-life (12.5 h), and did not exhibit any
change in its half-life after the addition of TGF-b.
The next question was what the biological outcome
would be of constitutive expression of Smad3 in
BEAS2B cells, in which TGF-b was found to rapidly
down-regulate Smad3 expression. A tetracycline-o€
gene expression system, in which addition of
tetracycline to the culture media arrests gene
transcription (Ookawa et al., 1997), was employed
for this purpose. BEAS2B-tet-Smad3 stable transfecFigure 1 Northern blot analysis of Smad2, Smad3 and Smad4 in
response to external stimuli in two immortalized human normal
lung epithelial cell lines, BEAS2B and HPL1A. (a) Marked
reduction of Smad3 mRNA is observed only in the presence of
TGF-b, whereas no noticeable changes are detected as a result of
the addition of EGF or of either low (0.1%) or high (10%) serum
concentrations (data not shown). In contrast Smad2 and Smad4
expressions remain virtually unchanged under all experimental
conditions. (b) Reduction of Smad3 is detectable as early as 4 h
after addition of TGF-b. (a) and (b) BEAS2B and HPL1A cells
were maintained in Ham's F12 medium bu€ered with 15 mM
HEPES (pH 7.3) and supplemented with bovine insulin (5 mg/ml),
human transferrin (5 mg/ml), 1077 M hydrocortisone, 2610710 M
triiode thyronine, penicillin (100 IU/ml), streptomycin (100 mg/ml)
and 1% FCS. To examine the possible e€ects of external stimuli
on Smad2, Smad3 and Smad4 mRNA expression, BEAS2B and
HPL1A cells were cultured for 6 days in the presence of 1 ng/ml
TGF-b (Wako, Osaka, Japan) or 10 ng/ml EGF (Austral
Biologicals, San Ramon, CA). In addition, the e€ects of lower
and higher serum concentrations (0.1 and 10% FCS) were also
examined. Extraction of total cellular RNA and Northern blot
analysis was performed with the aid of probes generated by the
following oligonucleotide primers. Smad2S (sense), 5'-AGCTCTAGATGGCTTGCTGCCTTTGGTAAG; Smad2AS (antisense),
5'-AGCGAATTCAAGTCTTTTCCATGGGACTTG;
Smad3S
(sense), 5'-AGCTCTAGAACTCAAGAAGACGGGGCAGCT;
Smad3AS (antisense), 5'-AGCGAATTCCTACTTGGTGTCGTACCTGCG; Smad4S (sense), 5'-AGCAAGCTTGCTTCAGAAATTGGAGAC; Smad4AS (antisense), 5'-AGCGGATCCCCATCCTGATAAGGTTAAGGGC. Ethidium bromide staining of the gels is shown as a control for RNA loading
(Figure 1a and data not shown for serum). Northern
blot analysis using RNAs harvested at various time
points revealed that down-regulation of Smad3
expression can be observed as early as 4 h after
addition of 1 ng/ml of TGF-b, with a 50% reduction
at 9.3 h (Figure 1b). These ®ndings clearly indicated
for the ®rst time that expression of a TGF-b signaling
molecule, Smad3, but not its close relative, Smad2, is
regulated by a TGF-b mediated signal in human lung
epithelial cells.
As an initial step towards elucidation of the
possible mechanism by which the TGF-b elicited
Figure 2 Northern blot analysis for determining of the half-lives
of Smad2 and Smad3 mRNA. (a) and (b) Stability of neither
Smad2 nor Smad3 mRNAs is a€ected by the addition of TGF-b.
(a) and (b) BEAS2B cells were cultured for 8 h in the presence or
absence of 1 ng/ml of TGF-b and chased in the presence of 5 mg/
ml of actinomycin D (Sigma Chemical Co., St. Louis, MO)
together with 1 ng/ml of TGF-b for up to 12 h. RNAs were
extracted at various time points after the addition of actinomycin
D and subjected to Northern blot analysis using the cDNA
probes described above. The amounts of Smad2 and Smad3
mRNAs were measured with a BAS2000 image analyser (Fuji
Photo Film Co., Ltd., Toyko, Japan). Ethidium bromide staining
of the gels is shown as a control for RNA loading
TGF-û signaling Smads in lung epithelial cells
K Yanagisawa et al
tants (clones 3 and 10) were generated by transfection
of pTA-hyg, which contained a fusion gene (tTA)
between the tetracycline responsive repressor and the
activation domain of herpes simplex virus VP16
protein, followed by transfection of pT2GN-Smad3
harboring wild-type Smad3 cDNA in a tTA responsive
pT2GN (Zhang et al., 1996). As shown in Figure 3a,
the induction of Smad3 in BEAS2B-tet-Smad3 clone 3
by removal of tetracycline resulted in a moderately
inhibitory e€ect on cell growth in the absence of
TGF-b (tet7, TGF-b7), while TGF-b treatment
alone also caused modest growth inhibition (tet+,
Figure 3 The e€ect of constitutive expression of Smad3 in the
presence of TGF-b on cell proliferation and viability. (a)
Constitutive Smad3 expression in the presence of TGF-b results
in a decrease in the number of cells. (b) Apoptotic cell deaths are
observed only under the Smad3-induced and TGF-b-added
conditions. (a) A tetracycline-o€ gene expression system, in
which the addition of tetracycline (Wako, Osaka, Japan) to the
culture media arrests gene transcription, was used to regulate
expression of Smad3. BEAS2B-tet-Smad3 stable transfectants
(clones 3 and 10) were generated by sequential transfection of
1.25 mg of pTA-hyg and then of 2 mg of pT2GN-Smad3 using the
DMRIE-C method (Life Technologies, Inc., Gaithersburg, MD).
The transfected cells were selected using 100 mg/ml hygromycin B
(Sigma Chemical Co.) and 200 mg/ml G418, respectively.
BEAS2B-tet-Smad3 cells were seeded in 96-well cell culture
plates at 5000 cells/well in 100 ml of medium containing 1 mg/ml
of tetracycline and preincubated at 378C for 24 h. Smad3 was
then induced by removal of tetracycline and cells were exposed to
1 ng/ml of TGF-b. MTS assay was performed at various time
points using the TetraColor One cell proliferation assay system
(Seikagaku Corp., Tokyo, Japan). Each data point represents the
mean+s.d. of triplicate samples. Three independent experiments
gave similar results. (b) Apoptotic cell deaths were visualized by
the TUNEL method using the in situ cell death detection kit
(Boehringer Mannheim Gmbh, Mannheim, Germany)
TGF-b+). Notably distinct results were obtained
when cells were placed under constitutive Smad3
expression in the presence of TGF-b (tet7, TGFb+). In contrast to the gradual increase in the
number of cells with tet+, TGF-b+ or tet7, TGFb7 treatment, the number of cells decreased steadily
under tet7, TGF-b+ condition and did not recover,
showing the signi®cantly adverse e€ect of constitutive
Smad3 expression in the presence of TGF-b. The
terminal deoxynucleotidyl transferase-mediated dUTPbiotin nick end labeling (TUNEL) method was
employed to examine whether apoptotic cell death
plays a role in this process. TUNEL staining-positive
cells were frequently observed only when Smad3 was
constitutively expressed in the presence of TGF-b
(Figure 3b). Similar results were also obtained for
BEAS2B-tet-Smad3 clone 10 (data not shown).
Constitutive Smad3 expression in the presence of
TGF-b thus appeared to adversely a€ect the growth
of lung epithelial cells by inducing apoptotic cell
death.
We next wanted to know whether there is a
di€erence in the e€ect on cell proliferation and
viability between Smad2 and Smad3 proteins. For
this purpose, a tetracycline-o€ gene expression system
was employed to regulate the expression of either
Smad2 or Smad3, while a luciferase assay system was
used as an indicator of viable cells because good
correlation was observed between number of cells and
luciferase activity (correlation coecient=0.974).
BEAS2B-tet cells were transfected with 2 mg of
pT2GN-Smad2 or pT2GN-Smad3 together with
0.2 mg of pCMV-luc carrying the cytomegalovirus
(CMV) promoter-driven luciferase gene. Cells were
subsequently refed with fresh media containing
tetracycline and TGF-b. Luciferase activity was
Figure 4 The e€ect on cell proliferation and viability of the
constitutive expression of Smad2 and Smad3 in the presence or
absence of TGF-b. Constitutive Smad3 expression in the presence
of TGF-b results in a signi®cant decrease in luciferase activity
during the incubation period, indicating a continued decrease in
the number of cells. Note that luciferase activities recover
gradually under other conditions. A tetracycline-o€ gene
expression system was employed to regulate the expression of
either Smad2 or Smad3, while luciferase activity was measured as
an indicator of viability of transfected cells by using the
Luciferase Assay System (Promega, Madison, WI). Values
represent luciferase activity relative to that of each of the control
transfectants incubated under tetracycline (+) and TGF-b (7)
condition and are shown as mean+s.d. of triplicate dishes.
Similar results were obtained in three independent experiments
1745
TGF-û signaling Smads in lung epithelial cells
K Yanagisawa et al
1746
measured 24 and 48 h later. As shown in Figure 4,
induction of Smad2 and Smad3 by removal of
tetracycline in the absence of TGF-b (tet7, TGFb7) had a moderately inhibitory e€ect on cell growth
as indicated by the decrease in luciferase activity.
TGF-b treatment alone without Smad2 or Smad3
induction (tet+, TGF-b+) also showed a moderately
inhibitory e€ect on cell growth. However, in these
conditions luciferase activity increased gradually,
indicating that the number of cells were still
growing. It is noteworthy that when Smad3 was
constitutively expressed in the presence of TGF-b
(tet7, TGF-b+), luciferase activity decreased significantly from 24 to 48 h, indicating a continued
decrease in the number of cells. In contrast, the
induction of Smad2 in the presence of TGF-b
inhibited cell growth, but still allowed for a gradual
increase in the number of cells, indicating a much less
pronounced adverse e€ect than that of Smad3.
We further examined whether there might be any
di€erences between Smad2 and Smad3 in their
capacity to induce apoptosis. BEAS2B cells were cotransfected with 2 mg of either pCMV-FlagSmad2 or
pcDNA3-FlagSmad3 driven by the constitutive CMV
promoter and tagged with Flag, together with 0.2 mg
of pCMV-CD20. Similar levels of Smad2 and Smad3
expression were con®rmed by Western blot analysis
using the anti-FlagM2 antibody (IBI, Eastman
Kodak, Rochester, NY). Apoptotic cells among the
CD20-positive cells were detected by using the
TUNEL method. Transfected cells were cultured in
the presence or absence of TGF-b for 48 h. As
expected, Smad3 transfectants maintained in the
presence of TGF-b showed a marked induction of
apoptotic cell death, and 55% of the CD20-positive
cells were found to be TUNEL-staining positive
(Figure 5). Smad2 transfectants cultured in the
presence of TGF-b were also rendered to be
apoptotic but considerably less signi®cantly. These
results showed that both Smad2 and Smad3 were
capable of inducing apoptotic cell death of lung
epithelial cells, at least when constitutively overexpressed in the presence of TGF-b, while Smad3
could induce this biological response more eciently
than Smad2.
In the present study, we demonstrated that
expression of a TGF-b signaling mediator, Smad3,
but not its close relative, Smad2, was down-regulated
by the TGF-b mediated signals themselves. Reduction
of Smad3 mRNA levels did not appear to result from
shortening of its half-life. We could also show that
forced constitutive expression of Smad3 in the
presence of TGF-b induced marked apoptotic cell
death in human normal lung epithelial cells. One
could speculate that down-regulation of Smad3
expression in response to TGF-b might be a negative
feedback mechanism by which lung cells can avoid
excess TGF-b signals, although the underlying
assumption that protein expression of Smad3 is
indeed diminished in response to TGF-b needs to be
con®rmed. The present study thus shows clear
distinctions between Smad2 and Smad3, in that a
qualitative di€erence was observed with regard to the
regulation of their mRNA expression in response to
TGF-b, while Smad2 and Smad3 appeared to be
quantitatively di€erent in their adverse e€ect on cell
growth of lung epithelial cells. We note that Zhang et
al. (1996) and Yingling et al. (1997) previously showed
that co-overexpression of Smad3/Smad4, but not
Smad2/Smad4, induced transcriptional activation of
plasminogen activator inhibitor-1 promoter, while
Imamura et al. (1997) reported selective binding of
Smad6 to Smad2. Taken together, these ®ndings
suggest that these two highly related Smads may
have distinct functional roles. Since TGF-b-mediated
signals are known to play important roles in lung
development, future studies are warranted to investigate distinctions between in situ expression of Smad2
and Smad3 during the developmental process. In
addition, although our extensive search did not reveal
any Smad3 mutations in lung cancers (unpublished
observation), in contrast to the occurrence of Smad2
mutations (Uchida et al., 1996), it would also be
interesting to examine whether there are any
alterations in the regulation of Smad3 expression in
lung cancer cells.
Figure 5 Apoptotic cell death is induced by the forced
expression of both Smad2 and Smad3 in the presence of TGFb, while Smad3 appears more e€ective for induction of this
biological response. BEAS2B cells were co-transfected with 2 mg
of pCMV-FlagSmad2 and pcDNA3-FlagSmad3 driven by the
constitutive CMV promoter and tagged with Flag, together with
0.2 mg of pCMV-CD20. The construct of Smad2 was generated by
the following oligonucleotide primers: Smad2 (sense), 5'AGCTCTAGATGGCTTGCTGCCTTTGGTAAG;
Smad2AS
(antisense), 5'-AGCGAATTCAAGTCTTTTCCATGGGACTTG.
Five hours after transfection, cells were placed in the conditions
indicated with the use of 1 mg/ml of tetracycline and 1 ng/ml of
TGF-b. Forty-eight hours later, transfected cells were examined
by Western blot analysis to con®rm equal expression of Smad2
and Smad3 using 50 mg of the solubilized proteins and antiFlagM2 antibody (Eastman, Kodak). Transfected cells were also
examined by the TUNEL method. In brief, cells were ®xed in PBS
containing 4% paraformaldehyde for 30 min at room temperature
and detected by using the anti-CD20 antibody (DAKO A/S,
Glostrup, Denmark) and subsequent application of biotinylated
anti-mouse IgG (Vector, Burlingame, CA) and tetramethylrodamine-isothiocyanate-isomer R conjugated-streptavidin (Cosmo
Bio Co. Ltd., Tokyo, Japan). Apoptotic cell deaths were
examined as described in the legend for Figure 3. The number
of TUNEL-stained CD20-positive cells was determined by
counting at least 1000 CD20-positive cells. Values represent
apoptotic cells as a percentage of CD20-positive cells and are
shown as means+s.d. of triplicate dishes. Three independent
experiments gave similar results
TGF-û signaling Smads in lung epithelial cells
K Yanagisawa et al
Acknowledgements
We would like to thank Dr T Hayakawa (Nagoya
University School of Medicine) for his encouragement
throughout this study, Dr J Yokota (National Cancer
Center Research Institute) for his kind gifts of pTA-hyg
and pT2GN, and Dr R Derynck (University of California,
San Francisco) for the wild-type Smad3 cDNA. This work
was supported in part by a Grant-in-Aid for Scienti®c
Research on Priority Areas from the Ministry of
Education, Science, Sports and Culture, Japan; a Grantin-Aid for the Second Team Comprehensive Ten-Year
Strategy for Cancer Control and a Grant-in-Aid for
Cancer Research from the Ministry of Health and
Welfare, Japan.
References
Baker JC, Harland RM, Norgaard P, Spang TM, Poulsen
HS, Hirakata Y and Kitamura S. (1996). Genes Dev., 10,
1880 ± 1889.
Derynck R, Gelbart WM, Harland RM, Heldin CH, Kern
SE, Massague J, Melton DA, Mlodzik M, Padgett RW,
Roberts AB, Smith J, Thomsen GH, Vogelstein B and
Wang XF. (1996). Cell, 87, 173.
Eppert K, Scherer SW, Ozcelik H, Pirone R, Hoodless P,
Kim H, Tsui LC, Bapat B, Gallinger S, Andrulis IL,
Thomsen GH, Wrana JL and Attisano L. (1996). Cell, 86,
543 ± 552.
Gra€ JM, Bansal A and Melton DA. (1996). Cell, 85, 479 ±
487.
Hahn SA, Schutte M, Hoque AT, Moskaluk CA, da Costa
LT, Rozenblum E, Weinstein CL, Fischer A, Yeo CJ,
Hruban RH and Kern SE. (1996). Science, 271, 350 ± 353.
Hayashi H, Abdollah S, Qiu Y, Cai J, Xu YY, Grinnell BW,
Richardson MA, Topper JN, Gimbrone Jr, MA, Wrana
JL, Anderson KR, Deeds JD, Feeley R, Gimeno CJ,
Woolf EA, Tayber O, Mays GG, Sampson BA, Schoen FJ
and Falb D. (1997). Cell, 89, 1165 ± 1173.
Hoodless PA, Haerry T, Abdollah S, Stapleton M,
O'Connor MB, Attisano L and Wrana JL. (1996). Cell,
85, 489 ± 500.
Imamura T, Takase M, Nishihara A, Hanai J, Kawabata M
and Miyazono K. (1997). Nature, 389, 622 ± 626.
Kingsley DM. (1994). Trends Genet., 10, 16 ± 21.
Lechleider RJ, de Caestecker MP, Dehejia A, Polymeropoulos MH and Roberts AB. (1996). J. Biol. Chem., 271,
17617 ± 17620.
Liu F, Hata A, Baker JC, Doody J, Carcamo J, Harland RM
and Massague J. (1996). Nature, 381, 620 ± 623.
Massague J. (1996). Cell, 85, 947 ± 950.
Masuda A, Kondo M, Saito T, Yatabe Y, Kobayashi T,
Okamoto M, Suyama M, Takahashi T and Takahashi T.
(1997). Cancer Res., 57, 4898 ± 4904.
Masui T, Wake®eld LM, Lechner JF, LaVeck MA, Sporn
MB and Harris CC. (1986). Proc. Natl. Acad. Sci. USA,
83, 2438 ± 2442.
Nagatake M, Takagi Y, Osada H, Uchida K, Mitsudomi T,
Saji S, Shimokata K, Takahashi T and Takahashi T.
(1996). Cancer Res., 56, 2718 ± 2720.
Ookawa K, Tsuchida S, Adachi J and Yokota J. (1997).
Oncogene, 14, 1389 ± 1396.
Reddel RR, Ke Y, Gerwin BI, McMenamin MG, Lechner
JF, Su RT, Brash DE, Park JB, Rhim JS and Harris CC.
(1988). Cancer Res., 48, 1904 ± 1909.
Riggins GJ, Thiagalingam S, Rozenblum E, Weinstein CL,
Kern SE, Hamilton SR, Willson JK, Markowitz SD,
Kinzler KW and Vogelstein B. (1996). Nature Genet., 13,
347 ± 349.
Roberts AB and Sporn MB. (1993). Growth Factors, 8, 1 ± 9.
Thomsen GH. (1996). Development, 122, 2359 ± 2366.
Topper JN, Cai J, Qui Y, Anderson KR, Xu YY, Deeds JD,
Feeley R, Gimeno CJ, Woolf EA, Tayber O, Mays GG,
Sampson BA, Schoen FJ, Gimbrone Jr, MA and Falb D.
(1997). Proc. Natl. Acad. Sci. USA, 94, 9314 ± 9319.
Uchida K, Nagatake M, Osada H, Yatabe Y, Kondo M,
Mitsudomi T, Masuda A, Takahashi T and Takahashi T.
(1996). Cancer Res., 56, 5583 ± 5585.
Wrana J and Pawson T. (1997). Nature, 388, 28 ± 29.
Yingling JM, Das P, Savage C, Zhang M, Padgett RW and
Wang XF. (1996). Proc. Natl. Acad. Sci. USA, 93, 8940 ±
8944.
Yingling JM, Datto MB, Wong C, Frederick JP, Liberati NT
and Wang XF. (1997). Mol. Cell. Biol., 17, 7019 ± 7028.
Zhang Y, Feng X, We R and Derynck R. (1996). Nature, 383,
168 ± 172.
1747