[CANCER RESEARCH 64, 7836 –7845, November 1, 2004]
Smad3 Knockout Mice Exhibit a Resistance to Skin Chemical Carcinogenesis
Allen G. Li,1,2,3 Shi-Long Lu,1,2,3 Ming-Xiang Zhang,1,2,3 Chuxia Deng,4 and Xiao-Jing Wang1,2,3
Departments of 1Otolaryngology, 2Dermatology, and 3Cell and Developmental Biology, Oregon Health and Science University, Portland, Oregon; and 4Mammalian Genetics
Section, Genetics of Development and Disease Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland
ABSTRACT
It has been shown that Smad3 exerts both tumor-suppressive and
-promoting roles. To evaluate the role of Smad3 in skin carcinogenesis in
vivo, we applied a chemical skin carcinogenesis protocol to Smad3 knockout
mice (Smad3ⴚ/ⴚ and Smad3ⴙ/ⴚ) and wild-type littermates (Smad3ⴙ/ⴙ).
Smad3ⴚ/ⴚ mice exhibited reduced papilloma formation in comparison
with Smad3ⴙ/ⴙ mice and did not develop any squamous cell carcinomas.
Further analysis revealed that Smad3 knockout mice were resistant to
12-O-tetradecanoylphorbol-13-acetate (TPA)–induced epidermal hyperproliferation. Concurrently, increased apoptosis was observed in TPAtreated Smad3ⴚ/ⴚ skin and papillomas when compared with those of
wild-type mice. Expression levels of activator protein-1 family members
(c-jun, junB, junD, and c-fos) and transforming growth factor (TGF)-␣
were significantly lower in TPA-treated Smad3ⴚ/ⴚ skin, cultured keratinocytes, and papillomas, as compared with Smad3ⴙ/ⴙ controls. Smad3ⴚ/ⴚ
papillomas also exhibited reduced leukocyte infiltration, particularly a
reduction of tumor-associated macrophage infiltration, in comparison
with Smad3ⴙ/ⴙ papillomas. All of these molecular and cellular alterations
also occurred to a lesser extent in Smad3ⴙ/ⴚ mice as compared with
Smad3ⴙ/ⴙ mice, suggesting a Smad3 gene dosage effect. Given that
TGF-␤1 is a well-documented TPA-responsive gene and also has a potent
chemotactic effect on macrophages, our study suggests that Smad3 may be
required for TPA-mediated tumor promotion through inducing TGF-␤1–
responsive genes, which are required for tumor promotion, and through
mediating TGF-␤1–induced macrophage infiltration.
INTRODUCTION
Transforming growth factor (TGF)-␤ is a multifunctional cytokine
that regulates cell proliferation, apoptosis, tissue remodeling, and
angiogenesis, all of which are associated with cancer development
(for a recent review, see ref. 1). Smad family members are the major
intracellular mediators of TGF-␤ signaling (for a recent review, see
ref. 1). Among the Smad family members, Smad2 and Smad3 are
responsible for TGF-␤ and activin signaling (2). On binding of TGF-␤
to its receptors, Smad2 and Smad3 are phosphorylated and subsequently form heteromeric complexes with Smad4, which then translocate from the cytoplasm to the nucleus. The Smad complex functions in the nucleus as a transcription factor and regulates expression
of TGF-␤ downstream targets (2).
TGF-␤ signaling plays a complex role in cancer development. As a
potent growth inhibitor for keratinocytes, TGF-␤1 is considered to
play a tumor-suppressive role at early stages of skin carcinogenesis.
For instance, loss of TGF-␤1 expression results in early progression to
malignancy (3). Similarly, transgenic mice overexpressing the dominant negative TGF-␤ type II receptor (⌬␤RII) in the epidermis exhibited a higher susceptibility to skin chemical carcinogenesis (4, 5).
Received 4/14/04; revised 8/16/04; accepted 9/2/04.
Grant support: National Institutes of Health grants CA87849 and CA79998 (X-J.
Wang).
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.
Note: M-X. Zhang is presently in the Department of Physiology, Baylor College of
Medicine, Houston, Texas.
Requests for reprints: Xiao-Jing Wang, 3710 SW US Veterans Hospital Road, Mail
code R&D46, Portland, OR 97239. Phone: 503-220-8262, ext. 54273; Fax: 503-402-2817;
E-mail: [email protected].
©2004 American Association for Cancer Research.
Paradoxically, TGF-␤1 is overexpressed in many cancer types in
humans (6). In experimental carcinogenesis models, expression of
TGF-␤1 is also induced after application of tumor promoters, such as
12-O-tetradecanoylphorbol-13-acetate [TPA (7)]. Studies from transgenic mice show that overexpression of TGF-␤1 in the epidermis
results in inhibition of tumor formation at early stages but acceleration
of tumor progression at later stages (8, 9). These studies suggest that
TGF-␤1 has a dual role in skin carcinogenesis. It remains to be
determined, however, whether TGF-␤1 overexpression at an early
stage of skin carcinogenesis exerts any tumor promotion effect.
As the major signaling mediators of the TGF-␤ superfamily, Smads
may be actively involved in mediating the effects of TGF-␤1 in
carcinogenesis. In vitro studies have shown that TGF-␤–induced
growth inhibition in epithelial cells is preferentially mediated by
Smad3 (10, 11) and that cyclin-dependent kinase (cdk) 2 and cdk4 can
phosphorylate Smad3, thus inactivating its ability to instigate cell
cycle arrest (12). Therefore, Smad3 may mediate TGF-␤1–induced
growth inhibition during cancer development. Supporting this notion,
Smad3-null keratinocytes transduced by v-rasHa undergo accelerated
malignant conversion when they are transplanted onto nude mice (13).
Further analyses revealed that the loss of Smad3 abrogates the functions of TGF-␤1 in regulating genes critical for cell cycle control,
such as p15 and c-myc (13), suggesting that Smad3 is required for
TGF-␤–induced growth inhibition in keratinocytes. However, Smad3
knockout mice do not exhibit epithelial hyperproliferation (14), suggesting that the role of Smad3 in mediating TGF-␤1–induced growth
inhibition is dispensable in vivo. In addition, TGF-␤1 overexpression
in keratinocytes in vivo induces severe skin inflammation (15, 16),
whereas Smad3 knockout mice exhibit accelerated wound healing
with decreased inflammation in vivo (17). These studies suggest that
Smad3 may also mediate TGF-␤1–induced skin inflammation. Because inflammation plays an important role in tumor promotion (18),
affecting TGF-␤1–induced skin inflammation may significantly affect
the role of TGF-␤1 in skin carcinogenesis. Furthermore, a recent
study using breast cancer cell lines showed that Smad3 exhibits both
tumor-suppressive and -promoting roles (19). To evaluate the role of
Smad3 in TGF-␤ signaling during skin carcinogenesis in vivo, we
applied a chemical carcinogenesis protocol to Smad3 knockout mice
(designated Smad3⫹/⫺ and Smad3⫺/⫺ hereafter), in which TGF-␤1
expression is induced by TPA promotion (7). Here we demonstrate
that Smad3 knockout mice exhibited a resistance to skin chemical
carcinogenesis, correlating with decreased proliferation and increased
apoptosis in Smad3 knockout keratinocytes during TPA promotion
and reduced tumor-associated macrophage (TAM) infiltration in
Smad3 knockout tumors. Consistently, TGF-␣ and activator protein
(AP)-1 family members, which are critical factors for TPA-induced
tumor promotion, were not up-regulated in Smad3 knockout skin and
tumors. Our results suggest that Smad3 is required for TPA-mediated
tumor promotion, possibly by regulating TGF-␤–responsive genes
that are involved in tumor promotion and by mediating TGF-␤1–
induced inflammation.
MATERIALS AND METHODS
Skin Chemical Carcinogenesis Protocol. The Smad3 knockout mice were
produced in a C57BL/6 ⫻ 129SVEV background (20) and backcrossed to
7836
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 2004 American Association for Cancer
Research.
SKIN CHEMICAL CARCINOGENESIS ON Smad3 KNOCKOUT MICE
129SVEV for ⬎12 generations during a 2-year period before their use in this
skin carcinogenesis study. Twelve hemizygous (Smad3⫹/⫺) breeding pairs
were set up to generate mice with Smad3⫹/⫹, Smad3⫹/⫺, and Smad3⫺/⫺
genotypes. In this study, 35 Smad3⫹/⫹ (15 males and 20 females), 44
Smad3⫹/⫺ (23 males and 21 females), and 16 Smad3⫺/⫺ (7 males and 9
females) mice were used. The chemical carcinogenesis protocol was applied as
described previously (21). Briefly, neonatal pups at day 3 were treated topically with 20 ␮g of 7,12-dimethylbenz(a)anthracene (DMBA; dissolved in 50
␮L of acetone; Sigma, St. Louis, MO). TPA (5 ␮g; dissolved in 50 ␮L of
acetone; Sigma) was applied to pup skin at day 4 and day 6, and application
was continued twice a week for 30 weeks.
Acute 12-O-Tetradecanoylphorbol-13-acetate Treatment on Adult
Skin. The dorsal skins of 6-week– old mice were shaved and treated with 5 ␮g
of TPA (dissolved in 50 ␮L of acetone). Forty-eight hours after TPA treatment,
mice received injection with 0.125 mg/g bromodeoxyuridine (BrdUrd; Sigma).
One hour later, the mice were euthanized, and the dorsal skins were harvested
for histologic analysis, BrdUrd labeling, immunohistochemical staining, and
RNA and protein extractions.
Tissue Histology. Dissected skin and tumor samples were fixed in 10%
neutral-buffered formalin at 4°C overnight, embedded in paraffin, sectioned to
6-␮m thickness, and stained with hematoxylin and eosin (H&E). Tumor types
were determined by H&E analysis using the criteria described previously (22)
and confirmed by Dr. C. Corless, a pathologist and the director of the Cancer
Pathology Core facility at Oregon Health and Science University. Generally,
papillomas appeared as exophytic, pedunculated proliferations consisting of
multiple finger-like hyperkeratotic epidermal projections having an intact
basement membrane creating a distinct border from the underlying dermis.
Squamous cell carcinomas (SCCs) were characterized by invasive tumor cell
proliferation into the dermis in a lobular pattern and numerous mitotic figures
within tumor lobules. SCCs were further classified into well-, moderately, and
poorly differentiated groups based on the criteria described previously (22).
Double Stain Immunofluorescence. Double stain immunofluorescence
was performed as we have described previously (23). The primary antibodies
included fluorescein isothiocyanate-conjugated BrdUrd (undiluted; BectonDickinson, Franklin Lakes, NJ), keratin 1 (K1; 1:500; ref. 24), keratin 13 (K13;
1:500; ref. 24), and nuclear factor (NF)-␬B p50 (1:50; Santa Cruz Biotechnology, Santa Cruz, CA). Briefly, paraffin-embedded sections were deparaffinized in fresh xylene and rehydrated. Antigen retrieval was performed by
microwaving slides in 10 mmol/L sodium citrate solution for 10 minutes. Each
section was incubated overnight at 4°C with a primary antibody diluted in PBS
containing 12% bovine serum albumin, together with a guinea pig antiserum
against mouse keratin 14 (K14; 1:500), the latter of which highlights the
epithelial compartment of the skin (9). The sections were then washed with
PBS and incubated with fluorescence dye-conjugated secondary antibodies, an
Alexa Fluo 488-conjugated (green) secondary antibody against the species of
the primary antibody [1:100; Molecular Probes, Eugene, OR (except for the
sections incubated with fluorescein isothiocyanate-conjugated BrdUrd antibody)] and Alexa Fluo 594-conjugated (red) anti-guinea pig secondary antibody (1:100; Molecular Probes). The fluorescence dye-conjugated secondary
antibodies were diluted in 10% bovine serum albumin-PBS and applied to the
tissue sections at room temperature for 30 minutes. After several PBS washes,
sections were visualized under a Nikon Eclipse E600W fluorescence microscope (Nikon, Melville, NY). The BrdUrd labeling index in the epidermis was
expressed as the mean number of BrdUrd-positive cells per millimeter of
basement membrane ⫾ SD.
Immunohistochemistry. To examine inflammatory cell subtypes, immunohistochemistry was performed on frozen sections, as we have described
previously (16), using primary antibodies for different leukocyte markers
including CD45 (1:20), CD4 (1:20), and Ly-6G [1:20; ref. 25 (all from BD
Biosciences, San Diego, CA)]. The BM8 antibody (1:400; BMA Biomedicals,
Augst, Switzerland), which recognizes macrophages (26), was also used.
Immunohistochemical detection of proliferating cell nuclear antigen (PCNA)
and phosphorylated Smad2 was performed on paraffin-embedded sections
using a PCNA antibody (1:100; Santa Cruz Biotechnology) and a phosphorylated Smad2 antibody (1:50; Cell Signaling, Beverly, MA). After deparaffinization, the sections were subjected to antigen retrieval as described above,
followed by incubation with 5% serum (from the species in which the secondary antibody was developed) for 1 hour at room temperature. Incubation
with primary antibodies was carried out at 4°C overnight. The sections were
then sequentially incubated with biotinylated secondary antibodies (1:250) and
an avidin-peroxidase reagent (Vector Laboratories, Burlingame, CA) at room
temperature for 10 and 5 minutes, respectively. The immune complexes in the
sections were visualized using diaminobenzidine (DAKO, Carpinteria, CA).
Quantitative measurement of positively stained cells was performed using
MetaMorph software (Universal Imaging Corp., Burnaby, British Columbia,
Canada) and expressed as the mean number of PCNA-positive cells per mm2
tumor area ⫾ SD.
Apoptosis Assays. Following the manufacturer’s instructions, apoptosis
was evaluated using terminal deoxynucleotidyltransferase-mediated uridine
nick end labeling (TUNEL) assay with the DeadEnd Fluorometric TUNEL kit
(Promega, Madison, WI). Briefly, paraffin-embedded sections were deparaffinized, rehydrated, and washed in 0.85% NaCl. The sections were then
sequentially prefixed (in 4% formaldehyde-PBS), permeabilized (in 20 ␮g/mL
proteinase K), and postfixed (in 4% formaldehyde-PBS). After several PBS
washes, the sections were incubated in a buffer that contained fluoresceindUTP and recombinant terminal deoxynucleotidyltransferase at 37°C for 1
hour in a humidified chamber in the dark. The reaction was then terminated in
2⫻ SSC for 15 minutes at room temperature, and the slides were washed in
PBS. The slides were then immersed in a 1 ␮g/mL propidium iodide solution
for 15 minutes at room temperature to counterstain the nuclei. The samples
were then analyzed under a fluorescence microscope. Quantitative measurement of apoptotic cells was performed using MetaMorph software (Universal
Imaging Corp.) and expressed as the mean number of apoptotic cells per mm2
tumor area ⫾ SD.
Keratinocyte Culture and 12-O-Tetradecanoylphorbol-13-acetate
Treatment. Primary keratinocytes were isolated from neonatal Smad3⫹/⫹,
Smad3⫹/⫺, and Smad3⫺/⫺ skin and cultured in conditioned medium supplemented with 0.05 mmol/L Ca2⫹, as we have described previously (27). On
reaching subconfluence, the cells were treated with TPA (100 nmol/L) for 12
hours before harvesting (28).
RNA Extraction, Quantitative Reverse Transcription-Polymerase
Chain Reaction, and RNase Protection Assay. Total RNA was isolated
from skin, chemically induced tumors, and cultured keratinocytes using RNAzol B (Tel-Test, Friendswood, TX), as we have described previously (11), and
further purified using a Qiagen RNeasy Mini kit (Qiagen, Valencia, CA). The
quantitative reverse transcription-polymerase chain reaction (qRT-PCR) was
achieved by combining in vitro reverse transcription with quantitative polymerase chain reaction, which was performed in a Stratagene Mx3000P thermal
cycler (Stratagene, La Jolla, CA; ref. 29). Briefly, 5 ␮g of RNA from each
sample were treated with DNase (Ambion, Austin, TX). The RNA was then
subjected to a reverse transcription reaction using avian myeloblastosis virus
reverse transcriptase (Roche, Indianapolis, MN). The resultant cDNA products
were used as templates for quantitative polymerase chain reaction to examine
the levels of transcripts of mouse Smad3, TGF-␣, and AP-1 family members
including c-jun, junB, junD, and c-fos using corresponding TaqMan Assayson-Demand probes (Applied Biosystems, Foster City, CA). Tumor RNA
samples were also assayed for gene expression levels of interleukin (IL)-1␤
and monocyte chemotactic protein (MCP-1). An 18S RNA probe was used as
an internal control, and the data (CT values) were analyzed using Stratagene
Mx3000P Comparative Quantitation software. The expression level of each
gene was normalized with 18S using a comparative CT (⌬CT) and expressed as
the difference of the CT values from a test gene (e.g., TGF-␣) and 18S
[CT(18S) minus CT(TGF-␣)]. The relative RNA expression levels were calculated using the ⌬CT method (30), and the average results from three to five
samples from three to five mice of each genotype are shown, with the
exception that only two Smad3⫺/⫺ mice developed tumors. For Smad3 gene
expression assays, the expression levels from all Smad3⫹/⫹ samples were set
as 1 arbitrary unit, which was used as a baseline to compare expression levels
of the same gene in samples with different Smad3 genotypes. In analyzing the
relative expression levels of other genes, the expression level from one
Smad3⫺/⫺ sample (unless otherwise specified) of each particular gene being
analyzed was set as 1 arbitrary unit. To examine expression levels of various
cell cycle control genes, RNase protection assay was performed using RNase
protection assay III kits (Ambion), as we have described previously (31). The
probes used included multiprobe gene sets [mCC1c, mCYC-1, and mCYC-2
(BD Biosciences)] and individual probes for c-myc, p15, and p21 (31).
7837
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 2004 American Association for Cancer
Research.
SKIN CHEMICAL CARCINOGENESIS ON Smad3 KNOCKOUT MICE
Statistics. Significant differences between the values obtained in each
assay on samples from various genotypes were determined using Student’s t
test throughout this study.
RESULTS
Smad3 Knockout Mice Exhibited Reduced Tumor Formation
and Malignant Progression during Skin Chemical Carcinogenesis. To determine the effect of loss of Smad3 on skin carcinogenesis
in vivo, we applied a chemical carcinogenesis protocol to the skin of
Smad3 knockout mice (Smad3⫹/⫺ and Smad3⫺/⫺). Because adult
Smad3⫺/⫺ mice develop a wasting syndrome (20), which could potentially affect the outcome of the skin carcinogenesis experiment, we
applied DMBA to Smad3 knockout mice 3 days after birth, followed
by applications of TPA twice a week for 30 weeks. Before 30 weeks
of age, animals of all three genotypes remained apparently healthy,
with the exception of 10% to 20% lower body weights in Smad3⫺/⫺
mice in comparison with Smad3⫹/⫺ and Smad3⫹/⫹ mice. By 30
weeks of age, the Smad3⫺/⫺ mice exhibited an approximately 30%
reduction in body weight (males, 22 ⫾ 3.3 g; females, 20 ⫾ 2.1 g;
P ⬍ 0.01), as compared with Smad3⫹/⫺ mice (males, 39 ⫾ 4.1 g;
females, 37 ⫾ 2.1 g) or Smad3⫹/⫹ mice (males, 36 ⫾ 3.7 g; females,
34 ⫾ 4.2 g). After 30 weeks of age, the wasting syndrome in some of
the Smad3⫺/⫺ mice became dominant, resulting in a rapid decline in
overall health of these mice, as described previously (20). Therefore,
tumorigenesis in mice with different Smad3 genotypes was monitored
until 30 weeks after DMBA initiation.
Smad3⫹/⫹ mice began to develop benign papillomas 6 weeks after
DMBA initiation, and by 16 weeks, approximately 60% of Smad3⫹/⫹
mice had developed papillomas (Fig. 1A). The first papillomas on
Smad3⫺/⫺ mice began to develop at 11 weeks after DMBA initiation,
and only 2 of 16 (⬃20%) Smad3⫺/⫺ mice developed papillomas by
16 weeks (Fig. 1A). Smad3⫹/⫺ mice also exhibited reduced tumorigenesis (47% by 20 weeks; Fig. 1A) as compared with Smad3⫹/⫹
mice. Additionally, Smad3⫺/⫺ mice developed significantly fewer
tumors as compared with Smad3⫹/⫹ mice. Smad3⫹/⫹ mice averaged
2.5 tumors per mouse at the end of the promotion stage (30 weeks;
Fig. 1B). In contrast, Smad3⫺/⫺ mice only averaged 0.5 tumor per
mouse by the end of TPA promotion, a 5-fold reduction in comparison
with Smad3⫹/⫹ mice (Fig. 1B; P ⬍ 0.01). Smad3⫹/⫺ mice averaged
1 tumor per mouse by 30 weeks, which was also a significant
reduction in tumor number when compared with Smad3⫹/⫹ mice
(Fig. 1B; P ⬍ 0.01). Furthermore, 40% of the tumors that developed
on Smad3⫹/⫹ mice progressed to SCCs after termination of TPA
treatment (30 weeks; Fig. 1C), with an average of 0.8 ⫾ 0.04 SCCs
per mouse (Fig. 1D). In contrast, all of the tumors dissected from
Smad3⫺/⫺ mice at the same time point were benign papillomas (Fig.
1C and D). Smad3⫹/⫺ mice also exhibited a decreased frequency of
SCC development, with only 27% of Smad3⫹/⫺ tumors progressing
to SCC (Fig. 1C) and a reduced average number of SCCs per mouse
(0.4 ⫾ 0.03; P ⬍ 0.01) in comparison with Smad3⫹/⫹ mice. Histologically, Smad3⫹/⫹ SCCs appeared more advanced and less differentiated than Smad3⫹/⫺ SCCs, which displayed features of typical
carcinomas in situ or well-differentiated SCCs (Fig. 2). Consistently,
the Smad3⫹/⫹ and Smad3⫹/⫺ tumors classified as SCCs showed a
lack of K1 expression, a marker of benign papillomas (24), whereas
all of the Smad3⫺/⫺ tumors retained uniform or at least focal K1
expression (Fig. 2). In contrast, strongly positive immunofluorescence
staining for K13, a malignancy marker in squamous epithelia (32),
was observed in tumors classified as SCCs from both Smad3⫹/⫹ and
Smad3⫹/⫺ mice, but not in tumors from Smad3⫺/⫺ mice (Fig. 2),
further confirming that the tumors from the Smad3⫺/⫺ mice were
benign papillomas.
Smad3 Deletion in Keratinocytes Resulted in a Resistance to
12-O-Tetradecanoylphorbol-13-acetate–Induced Epidermal Hyperplasia and an Increased Rate of Apoptosis. To determine
whether the resistance to skin chemical carcinogenesis in Smad3
knockout mice is due to the blockade of DMBA-induced initiation of
tumor development, we examined tumor samples from different
Smad3 genotypes for c-rasHa mutations, which are induced by DMBA
in ⬎90% of papillomas (33). Sequence analysis for a c-rasHa genomic
fragment (5) revealed that all tumor samples examined from the
different Smad3 genotypes possessed a c-rasHa mutation at codon 12,
13, or 61 (data not shown), suggesting that loss of Smad3 likely does
not affect DMBA initiation. Next, we analyzed whether Smad3 deletion contributes to a reduction of susceptibility to TPA-induced tumor
promotion by examining adult skin with different Smad3 genotypes
Fig. 1. Tumor kinetics. A, percentage of tumor-bearing
mice in each group. B, average number of tumors per
mouse. C, percentage of mice that developed carcinoma by
30 weeks after TPA treatment. None of the Smad3⫺/⫺ mice
papillomas converted to carcinomas. D, average number of
SCCs per mouse.
7838
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 2004 American Association for Cancer
Research.
SKIN CHEMICAL CARCINOGENESIS ON Smad3 KNOCKOUT MICE
Fig. 2. Histology and immunofluorescence analyses
of the most advanced tumor types from each genotype.
H&E examination revealed a poorly differentiated
SCC, an early-stage SCC, and a papilloma in
Smad3⫹/⫹, Smad3⫹/⫺, and Smad3⫺/⫺ mice, respectively. A K14 antibody (red) was used as a counterstain
for immunofluorescence. The bar in the top left panel
represents 200 ␮m for H&E sections and 60 ␮m for
immunofluorescence sections.
48 hours after a single TPA treatment. As compared with TPA-treated
Smad3⫹/⫹ skin, which was characterized by obvious epidermal hyperplasia (8.1 ⫾ 1.6 cell layers in epidermis), Smad3⫹/⫺ (4.7 ⫾ 1.2
layers; P ⬍ 0.01; n ⫽ 3) and Smad3⫺/⫺ (2.3 ⫾ 0.4 layers; P ⬍ 0.01;
n ⫽ 3) skin showed a resistance to TPA-induced epidermal hyperplasia (Fig. 3A). Accordingly, the number of BrdUrd-labeled keratinocytes was significantly reduced in TPA-treated Smad3 knockout skin.
The number of BrdUrd-labeled keratinocytes was 189 ⫾ 23.4 per mm
of epidermis in TPA-treated Smad3⫹/⫹skin, whereas it was reduced
to 122 ⫾ 14.2 (P ⬍ 0.01; n ⫽ 3) and 35.6 ⫾ 9.0 per mm of epidermis
(P ⬍ 0.01; n ⫽ 3) in Smad3⫹/and Smad3⫺/⫺ skins, respectively (Fig.
3A). In contrast to the change in epidermal proliferation, CD45
immunostaining showed no difference in the degree of inflammatory
cell infiltration in TPA-treated skin from three groups (Fig. 3A). The
degree of infiltration by different leukocyte subsets, including CD4⫹
cells, granulocytes, and macrophages, in TPA-treated skin of three
groups was also indistinguishable (data not shown). These results
suggest that loss of Smad3 has a direct effect on keratinocyte proliferation during the promotion stage of skin chemical carcinogenesis.
Analysis of cell proliferation in Smad3 knockout papillomas revealed
that the number of PCNA-positive cells was reduced by ⬎50% in
Smad3⫹/⫺ papillomas (163 ⫾ 12.2/mm2 tumor area; P ⬍ 0.01; n ⫽ 3)
in comparison with Smad3⫹/⫹ papillomas (389 ⫾ 28.9/mm2) at the
same stage and further reduced by 90% in Smad3⫺/⫺ papillomas
(37.2 ⫾ 4.5/mm2) as compared with Smad3⫹/⫹ papillomas (Fig. 3B).
To determine whether Smad3 deletion alters the apoptotic process,
TUNEL assay was performed on TPA-treated adult skin and papillomas with different Smad3 genotypes. The numbers of apoptotic cells
in the epidermis of Smad3⫺/⫺ (24.8 ⫾ 3.53/mm2 epidermis; n ⫽ 3)
and Smad3⫹/⫺ skin (19.8 ⫾ 2.65/mm2 epidermis; n ⫽ 3) were
significantly increased as compared with the number of apoptotic cells
in Smad3⫹/⫹ skin (5.25 ⫾ 1.28/mm2 epidermis; n ⫽ 3; P ⬍ 0.01) 48
hours after TPA treatment (Fig. 3C). The apoptotic cells in these
samples were primarily located in the suprabasal layers of the epidermis (Fig. 3C). Similarly, apoptotic cells, located mainly in suprabasal
layers, were prominent in Smad3⫹/⫺ (31.3 ⫾ 2.83/mm2 tumor epithelia) and Smad3⫺/⫺ papillomas (38.7 ⫾ 3.69/mm2) but sparse in
Smad3⫹/⫹ papillomas (11.6 ⫾ 1.47/mm2; n ⫽ 3; P ⬍ 0.01; Fig. 3C).
To further confirm the correlation between the above-mentioned
effects and the inactivation of the Smad3 gene, we examined the
relative expression levels of the Smad3 transcript by qRT-PCR. As
shown in Fig. 4A, Smad3 expression levels in nontreated skin, TPAtreated skin, and papillomas, all from Smad3⫹/⫺ mice, were reduced
by about 50% as compared with the levels detected in the same tissues
from Smad3⫹/⫹ mice. Expression of Smad3 was undetectable in all
samples from Smad3⫺/⫺ mice (Fig. 4A). To elucidate whether the
effect of Smad3 deletion on skin carcinogenesis involves altered
activation of its signaling partner, Smad2, the presence of phosphorylated Smad2 (pSmad2) was examined by immunohistochemistry
using a pSmad2 antibody. Nontreated skins from different Smad3
genotypes exhibited comparable pSmad2 nuclear staining (Fig. 4B).
Noticeably, TPA-treated skin from all three Smad3 genotypes exhibited increased pSmad2 in both the epidermis and dermis (Fig. 4B),
suggesting that TPA-induced endogenous TGF-␤1 activates Smad2.
However, both TPA-treated skins and papillomas exhibited comparable pSmad2 nuclear staining among different Smad3 genotypes (Fig.
4B), suggesting that the effect of Smad3 deletion on skin chemical
carcinogenesis is independent of Smad2 activation.
Reduced Expression of Activator Protein-1 Family Members
and Transforming Growth Factor ␣ in 12-O-Tetradecanoylphorbol-13-acetate–Treated Skin and Tumors of Smad3 Knockout
Mice. To determine the molecular mechanisms of Smad3 knockout
skin resistance to TPA-induced epidermal hyperplasia and increased
apoptotic rate, we examined expression levels of proliferative/survival
factors that normally respond to TPA. In contrast to a previous report
showing decreased expression of the cdk inhibitor p15 and increased
expression of c-myc in vrasHa-transduced Smad3-null keratinocytes
(13), we did not find differences among the different Smad3 genotypes in expression of TGF-␤1 target genes involved in cell cycle
control, including the cdk inhibitors p15 and p21 and the c-myc
oncogene, in TPA-treated skin of each genotype (data not shown).
These results suggest that either TPA itself or an endogenous mechanism(s) is able to regulate expression of these molecules in the
absence of Smad3. We next examined the expression of major AP-1
family members, which are TGF-␤–responsive genes (34) that are
rapidly up-regulated in response to TPA and mediate TPA-induced
7839
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 2004 American Association for Cancer
Research.
SKIN CHEMICAL CARCINOGENESIS ON Smad3 KNOCKOUT MICE
Fig. 3. Proliferation and apoptosis in TPAtreated skin and papillomas from each group. Dotted lines denote the epidermal– dermal junction. A,
histology, BrdUrd labeling (green), and CD45
staining on TPA-treated skin. Arrows point to typical BrdUrd-labeled cells. A K14 antibody (red)
was used as a counterstain for immunofluorescence. Sections stained by immunohistochemistry
for CD45 were counterstained with hematoxylin.
The bar in the top left panel represents 100 ␮m for
H&E- and BrdUrd-stained sections and 40 ␮m for
CD45-stained sections. B, PCNA staining of representative papillomas from each genotype. The
sections are counterstained with hematoxylin. Red
arrows point to examples of positively stained
nuclei. The bar in the top left panel represents 100
␮m. C, TUNEL assay on TPA-treated skin (top
panels) and papillomas (bottom panels) from each
group. Arrows point to representative apoptotic
nuclei (green or yellowish). The bar in the top left
panel represents 40 ␮m for all sections.
tumor promotion (35). We found that expression levels of c-jun, junB,
junD, and c-fos in TPA-treated skin and papillomas were highest in
Smad3⫹/⫹ mice and lowest in Smad3⫺/⫺ mice (Fig. 5).
Because TGF-␣ overexpression induced by TPA is thought to be
critical for TPA-induced tumor promotion (36) and for protection of
tumor cells from apoptosis (37–39), we examined expression of
TGF-␣ in TPA-treated skin and tumors. Expression of TGF-␣ was
almost undetectable in untreated skin (data not shown) but was
induced in Smad3⫹/⫹ skin after TPA treatment (Fig. 5). However,
expression of TGF-␣ was 8- to 10-fold lower in TPA-treated
Smad3⫹/⫺ skin and almost undetectable in TPA-treated Smad3⫺/⫺
skin. A similar pattern of TGF-␣ expression was observed in tumors
from the three genotypes (Fig. 5).
To further determine whether changes in expression levels of these
genes represent a direct effect of Smad3 deletion on keratinocytes, we
cultured primary keratinocytes isolated from mice with different Smad3
genotypes and treated these cells with TPA for 12 hours. Expression of
the above-mentioned five genes was significantly lower in TPA-treated
Smad3⫺/⫺ keratinocytes than in Smad3⫹/⫹ cells (Fig. 5). TPA-treated
Smad3⫹/⫺ keratinocytes exhibited a reduction in the expression of these
genes by about 45% to 66% relative to Smad3⫹/⫹ samples (Fig. 5),
suggesting a dose-dependent effect of Smad3 inactivation on the transcription of these genes in keratinocytes in response to TPA.
Reduced Tumor-Associated Macrophages and Inflammation in
Smad3 Knockout Papillomas. Because Smad3 can affect infiltration
of inflammatory cells (20), we examined leukocyte infiltration in
7840
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 2004 American Association for Cancer
Research.
SKIN CHEMICAL CARCINOGENESIS ON Smad3 KNOCKOUT MICE
Fig. 4. Smad3 expression and Smad2 phosphorylation. A,
qRT-PCR results of Smad3 expression in 6-week– old mouse
skin (skin), TPA-treated skin, and papillomas from the three
different genotypes. The Smad3 expression level of all samples from Smad3⫹/⫹ mice was set as 1 arbitrary unit. B,
immunohistostaining of pSmad2. The bar in the top left panel
represents 40 ␮m for all sections. Hematoxylin was used as a
counterstain.
Fig. 5. Relative expression levels of AP-1 family
members and TGF-␣ as determined by real-time
RT-PCR. Results from two representative samples
in each group are shown. The expression level of
each molecule in one Smad3⫺/⫺ sample was set as
1 arbitrary unit. An exception was made for TGF-␣
expression levels in TPA-treated skins, in which
one Smad3⫹/⫺ skin sample was set as 1 arbitrary
unit, because TGF-␣ expression was not detectable
in TPA-treated Smad3⫺/⫺ skin.
7841
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 2004 American Association for Cancer
Research.
SKIN CHEMICAL CARCINOGENESIS ON Smad3 KNOCKOUT MICE
Fig. 6. Inflammatory response in papillomas
from different Smad3 genotypes. Immunohistochemistry was carried out to determine the presence of different leukocyte subsets. Note the reduced number of CD45⫹ cells and macrophages
(BM8) in Smad3⫹/⫺ and Smad3⫺/⫺ papillomas as
compared with Smad3⫹/⫹ papillomas. Immunofluorescence was performed to detect p50 expression
(green) in papillomas from each group. A K14
antibody (red) was used as a counterstain. The bar
in the top left panel represents 40 ␮m for immunohistochemistry sections and 100 ␮m for p50stained sections.
TPA-treated skin and papillomas from different Smad3 genotypes.
Immunohistochemistry was performed using antibodies that recognize
different surface markers of leukocytes. At the acute phase of TPAinduced inflammation (48 hours after a single TPA treatment), we did
not observe any differences in the number or type of infiltrating
inflammatory cells among different Smad3 genotypes (Fig. 3A). However, CD45 immunostaining revealed a dramatic reduction in the
number of inflammatory cells present in Smad3⫹/⫺ (216 ⫾ 19.6/mm2
tumor area; P ⬍ 0.01) and Smad3⫺/⫺ (103.3 ⫾ 9.31/mm2 tumor area;
P ⬍ 0.01) tumors as compared with Smad3⫹/⫹ tumors (723 ⫾ 67.5/
mm2 tumor area). Numerous BM8⫹ TAMs were present within
Smad3⫹/⫹ tumor epithelia, whereas only sporadic TAMs were present in tumor epithelia and stroma of Smad3⫺/⫺ papillomas (Fig. 6).
Smad3⫹/⫺ papillomas also exhibited a reduction in the number of
TAMs in comparison with Smad3⫹/⫹ tumors. In particular, the number of TAMs within tumor epithelia of Smad3⫹/⫺ (62.3 ⫾ 8.91/mm2
tumor epithelia; P ⬍ 0.01; n ⫽ 3) or Smad3⫺/⫺ papillomas
(30.4 ⫾ 2.92; P ⬍ 0.01; n ⫽ 3) was significantly reduced as compared
with that of Smad3⫹/⫹ papillomas (119 ⫾ 10.7). In contrast, the
number of CD4⫹ T cells (Smad3⫹/⫹, 76 ⫾ 4.8/mm2 tumor area;
Smad3⫹/⫺, 83 ⫾ 9.1/mm2 tumor area; Smad3⫺/⫺, 79 ⫾ 8.5/mm2
tumor area; P ⬎ 0.05; n ⫽ 3) or Ly6-G⫹ granulocytes (Smad3⫹/⫹,
67 ⫾ 5.3/mm2 tumor area; Smad3⫹/⫺, 73 ⫾ 8.0/mm2 tumor area;
Smad3⫺/⫺, 75 ⫾ 7.7/mm2 tumor area; P ⬎ 0.05; n ⫽ 3) was
statistically similar for each genotype (Fig. 6).
Furthermore, we examined nuclear translocation of the NF-␬B subunit
p50, an end point of the inflammatory cascade (40). Immunofluorescence
staining showed that ⬎90% of the Smad3⫹/⫹ papilloma cells stained
positive for p50 in the nucleus (Fig. 6). In contrast, nuclear staining of
p50 was only observed in a few Smad3⫺/⫺ papilloma cells and in only
approximately 50% of Smad3⫹/⫺ papilloma cells (Fig. 6).
Consistent with decreased inflammation in the Smad3⫺/⫺ papillomas, IL-1␤, a proinflammatory cytokine that is mainly expressed in
keratinocytes on inflammation (41), was expressed at a very low level
7842
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 2004 American Association for Cancer
Research.
SKIN CHEMICAL CARCINOGENESIS ON Smad3 KNOCKOUT MICE
Fig. 7. Relative expression levels of IL-1␤ and MCP-1 in papillomas as determined by
qRT-PCR. The expression level in one Smad3⫺/⫺ sample was set as 1 arbitrary unit.
in Smad3⫺/⫺ papillomas (Fig. 7). In contrast, IL-1␤ expression was
⬎500-fold higher in Smad3⫹/⫹ papillomas as compared with
Smad3⫺/⫺ papillomas (P ⬍ 0.01). IL-1␤ expression levels in
Smad3⫹/⫺ papillomas were approximately 400-fold less than IL-1␤
expression levels in Smad3⫹/⫹ papillomas (P ⬍ 0.01; Fig. 7). In
addition, the expression level of MCP-1, a major monocyte/macrophage-attracting chemokine (42), was expressed at approximately
8-fold higher levels in Smad3⫹/⫹ papillomas than in Smad3⫺/⫺
papillomas (P ⬍ 0.01; Fig. 7), and expression of MCP-1 in Smad3⫹/⫺
papillomas was only 2-fold higher than that in Smad3⫺/⫺ papillomas
(P ⬍ 0.01; Fig. 7).
DISCUSSION
In this report, we have shown that mice lacking Smad3 exhibited
resistance to skin chemical carcinogenesis compared with wild-type
littermates. This result contradicts a previous report that loss of Smad3
in keratinocytes accelerates malignant conversion in a tumor transplant model (13). Because nutritional status can greatly affect skin
carcinogenesis (43), we were concerned that results from a skin
chemical carcinogenesis study on adult Smad3 knockout mice, which
develop a wasting syndrome (20), may not accurately reflect the effect
of the loss Smad3 on susceptibility to skin chemical carcinogenesis.
To avoid this problem, we used neonatal mice and applied a previously described chemical carcinogenesis protocol (44) that is initiated
3 days after birth, when Smad3⫺/⫺ pups are morphologically normal
(20). By the time Smad3⫹/⫹ mice began to develop papillomas (6 to
15 weeks), Smad3⫺/⫺ mice remained free of obvious wasting syndrome symptoms yet demonstrated a resistance to tumorigenesis. In
addition, Smad3⫹/⫺ mice, which did not develop wasting syndrome
and maintained body weight similar to wild-type littermates, showed
a reduction in tumor formation and malignant conversion. Furthermore, Smad3⫺/⫺ and Smad3⫹/⫺ skin exhibited a gene dosage-dependent resistance to epidermal proliferation after acute TPA treatment.
In fact, Smad3 gene dosage effects during skin chemical carcinogenesis were seen in almost every variable we analyzed on littermates
with different Smad3 genotypes. Because activation of a Smad3
partner, Smad2, remained unaffected by Smad3 deletion (Fig. 4), our
results suggest that the resistance to skin chemical carcinogenesis in
Smad3 knockout mice is likely a direct effect of the loss of Smad3
gene expression. Supporting our observations, a recent report shows
that Smad3⫹/⫺ mice were resistant to skin chemical carcinogenesis,
whereas Smad2⫹/⫺ mice, with the same genetic background, exhibited accelerated skin carcinogenesis (45).
Lack of AP-1 Activation and TGF-␣ Overexpression in Smad3
Knockout Keratinocytes Contributes to Resistance to TPA-Induced Tumor Promotion. Unlike v-rasHa-transduced keratinocytes,
which can progress to benign papillomas on grafting onto the athymic
mouse skin (13), normal keratinocytes harboring a point mutation in
the c-rasHa gene induced in vivo by a subcarcinogenic dose of DMBA
do not typically form papillomas in normal mice (5), and further
tumor promotion is required for these cells to undergo tumorigenesis.
The tumor promotion effect of TPA depends mainly on the ability of
TPA to induce keratinocyte hyperproliferation, which expands the
population of initiated stem cells [e.g., stem cells carrying a c-rasHa
mutation (46, 47)]. On TPA application, TGF-␤1 expression is elevated (7). Although TGF-␤1 directly inhibits keratinocyte proliferation, it also activates expression of genes closely related to TPAinduced tumor promotion, e.g., the AP-1 family members (34). Smad3
has been found to mediate the effect of TGF-␤1 on transactivation of
c-fos, c-jun, junB, and junD by binding to their promoters (48 –50). In
our current study, we show that induction of these AP-1 family
members did not occur in TPA-treated Smad3 knockout skin and
tumors or TPA-treated Smad3 knockout keratinocytes in vitro. Thus,
Smad3 appears to be required for transcriptional activation of these
genes during TPA promotion. In addition to direct transcriptional
regulation by Smad3, the AP-1 family members are also transcriptionally regulated on activation of epidermal growth factor receptor
(51). Therefore, the lack of overexpression of TGF-␣, the epidermal
growth factor receptor ligand, in Smad3 knockout skin and tumors
during TPA promotion may also contribute to the low levels of AP-1
family members in Smad3 knockout skin and tumors. Each AP-1
family member has been suggested to play an essential role in skin
carcinogenesis through regulating cellular proliferation and apoptosis
(52). For instance, c-fos, c-jun, and junD proteins positively regulate
cell proliferation, whereas junB may negatively regulate cell proliferation in the presence of c-jun (52). Thus, it could be predicted that
decreased expression of these molecules in TPA-treated skin and
tumors of Smad3 knockout mice can cause a decrease in keratinocyte
proliferation. With respect to apoptosis, the effects of AP-1 family
members are somewhat contradictory and highly tissue specific. c-jun
knockout embryos exhibit massive apoptosis in hepatocytes, indicating an antiapoptotic activity of c-jun protein. A previous study has
shown an antiapoptotic effect of jun/fos proteins on skin carcinogenesis (53). Overall, the role of AP-1 family members in skin carcinogenesis has been at least partially attributed to their ability to promote
keratinocyte proliferation and inhibit apoptosis (52, 54). In vivo experiments showed that c-jun is required for the development of
papillomas (55), whereas c-fos is required for malignant transformation (56). Thus, lack of AP-1 expression/activation during TPA promotion in Smad3 knockout skin appears to greatly contribute to
resistance to papilloma formation as well as malignant progression
during skin chemical carcinogenesis. In the case of constant overexpression of v-rasHa in transplanted keratinocytes (13), the tumor
promotion event can bypass TPA-induced AP-1 activation; thus, the
effect of Smad3 on AP-1 gene expression is dispensable. Under this
circumstance, the effect of the loss of Smad3 appears to predominantly abrogate TGF-␤1–induced growth inhibition, resulting in accelerated malignant conversion.
Another growth factor that is typically rapidly induced by TPA
treatment of normal keratinocytes is TGF-␣. TPA up-regulates TGF-␣
expression through signaling via protein kinase C (PKC)-dependent
and -independent pathways (57). In the latter case, it is well-documented that TGF-␣ is an autoinductive cytokine and that the TGF-␣
7843
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 2004 American Association for Cancer
Research.
SKIN CHEMICAL CARCINOGENESIS ON Smad3 KNOCKOUT MICE
gene promoter contains a TGF-␣/epidermal growth factor-responsive
element (58). It has also been shown that TGF-␤1 transcriptionally
regulates TGF-␣ expression (59). Because PKC is rapidly downregulated after TPA application (60), it is reasonable to believe that
the PKC-independent pathways are primarily responsible for the
persistent TGF-␣ overexpression. Although the exact mechanism by
which Smad3 deletion blocks TGF-␣ overexpression remains to be
determined, it is likely that loss of Smad3 abrogates TGF-␤1–induced
TGF-␣ overexpression, consequently perturbing the autoinduction of
TGF-␣ expression. TGF-␣ is overexpressed in many cancer types and
functions as a potent mitogen for cancer cell proliferation (61, 62). In
addition, TGF-␣ is a critical survival factor against apoptosis for
cancer cells (37, 38). Thus, persistent TGF-␣ overexpression is
thought to be critical for the TPA-induced tumor promotion effect
(36). Supporting this, TGF-␣– deficient mice (Wa-1 mutant) exhibit a
resistance to TPA promotion (63). Consistent with the above-mentioned studies, lack of TPA- induced TGF-␣ overexpression in Smad3
knockout skin and papillomas correlated with decreased proliferation
and increased apoptosis in keratinocytes, which apparently contributed to the resistance of Smad3 knockout mice to skin chemical
carcinogenesis. Because TGF-␣ overexpression is critical for activation of several signal transduction pathways, such as the ras-mitogenactivated protein kinase pathway (64), the lack of TGF-␣ overexpression in Smad3 knockout skin/tumors may also explain the
contradictory data between our present study and a previous report
showing accelerated malignant conversion in skin tumors that resulted
from transplantation of Smad3-null keratinocytes transduced with
v-rasHa (13). In that study, consistently high levels of v-rasHa expression in keratinocytes should be able to bypass the effect of TGF-␣
overexpression as seen during TPA-induced tumor promotion.
Reduced Inflammation in Smad3 Knockout Skin May Also
Contribute to Resistance to Skin Carcinogenesis In vivo. Given the
importance of inflammation in cancer development (18), TPAinduced skin inflammation is expected to contribute to the tumor
promotion effect of TPA. In keratinocytes, TPA application induces
expression of inflammatory cytokines, including IL-1 (65), which not
only induces inflammation but also stimulates keratinocyte proliferation (66). In TPA-treated preneoplastic skin, we did not find differences in inflammation among the different Smad3 genotypes. This
suggests that TPA-induced inflammation may not be completely
mediated by TGF-␤1 overexpression, and therefore the loss of Smad3
cannot attenuate TPA-induced inflammation, at least at the acute
phase. At this stage, the resistance to skin carcinogenesis in Smad3
knockout mice may be mainly a result of altered keratinocyte properties (i.e., resistance to proliferation and increased apoptosis). However, because TGF-␤1 overexpression has been shown to induce skin
inflammation (15, 16), persistent TGF-␤1 overexpression at later
stages of skin carcinogenesis may maintain a certain level of inflammation that is required for tumor development (18). In particular,
TAMs play an important role in tumor promotion and progression
(18). Thus, a reduction in TAM infiltration in Smad3 knockout tumors
may also explain, at least in part, the resistance of these tumors to
malignant conversion. The impaired chemotactic effect of TGF-␤1 on
Smad3⫺/⫺ monocytes (20) is possibly responsible for reduced TAM
infiltration in Smad3 knockout tumors. Supporting this notion, reduced local inflammation associated with decreased monocyte infiltration has also been reported in Smad3 knockout wounds as compared with Smad3⫹/⫹ wounds (17). TGF-␤1 has a potent chemotactic
effect on monocytes (67) and may induce expression of multiple
chemokines in the skin (16, 68). MCP-1, which was down-regulated
in Smad3 knockout tumors, is a chemokine that attracts monocytemacrophage infiltration to sites of inflammation and can be upregulated on TGF-␤1 overexpression in the skin (16). It is thus likely
that Smad3⫺/⫺ monocytes do not respond to the chemotactic effect of
TGF-␤1. Alternatively, Smad3⫺/⫺ stromal cells and endothelial cells
may affect monocyte infiltration. In either case, reduced TAM infiltration would not occur in tumors generated in mice in which Smad3
is deleted only in keratinocytes, as indicated in a previous study that
reported accelerated skin carcinogenesis of transplanted Smad3-null
keratinocytes (13). However, because inflammation was reduced in
Smad3 knockout papillomas, as evidenced by reduced IL-1␤ and the
lack of NF-␬B activation, reduced TAM infiltration in Smad3 knockout papillomas may also be a consequence of reduced expression
levels of inflammatory cytokines and chemokines in Smad3⫺/⫺ tumor
epithelia. All of the above-mentioned possibilities can be tested in the
future by generating mice with epidermal-specific deletion of Smad3
and performing carcinogenesis experiments on these mice.
In summary, the present study shows that loss of Smad3 in vivo
significantly reduced tumor formation and malignant conversion during skin chemical carcinogenesis. Our study suggests that, at early
stages of skin carcinogenesis, TGF-␤1 overexpression induced by
TPA may also have a tumor promotion effect via activation of AP-1
family members in keratinocytes and inflammation in the stroma, both
of which may require wild-type Smad3. Our study instigates future
studies to investigate the complex nature of the role of Smad3 in
cancer development and its underlying molecular mechanisms under
different pathological conditions.
ADDENDUM
While this article was under review, a report was published (45) showing that
Smad3⫹/⫺ mice were resistant to skin chemical carcinogenesis.
REFERENCES
1. Siegel PM, Massague J. Cytostatic and apoptotic actions of TGF-beta in homeostasis
and cancer. Nat Rev Cancer 2003;3:807–21.
2. Derynck R, Zhang YE. Smad-dependent and Smad-independent pathways in TGFbeta family signalling. Nature (Lond) 2003;425:577– 84.
3. Glick AB, Lee MM, Darwiche N, et al. Targeted deletion of the TGF-beta 1 gene
causes rapid progression to squamous cell carcinoma. Genes Dev 1994;8:2429 – 40.
4. Amendt C, Schirmacher P, Weber H, Blessing M. Expression of a dominant negative
type II TGF-beta receptor in mouse skin results in an increase in carcinoma incidence
and an acceleration of carcinoma development. Oncogene 1998;17:25–34.
5. Go C, Li P, Wang XJ. Blocking transforming growth factor beta signaling in
transgenic epidermis accelerates chemical carcinogenesis: a mechanism associated
with increased angiogenesis. Cancer Res 1999;59:2861– 8.
6. Cohen MM Jr. TGF beta/Smad signaling system and its pathologic correlates. Am J
Med Genet 2003;116A:1–10.
7. Patamalai B, Burow DL, Gimenez-Conti I, et al. Altered expression of transforming
growth factor-beta 1 mRNA and protein in mouse skin carcinogenesis. Mol Carcinog
1994;9:220 –9.
8. Cui W, Fowlis DJ, Bryson S, et al TGFbeta1 inhibits the formation of benign skin
tumors, but enhances progression to invasive spindle carcinomas in transgenic mice.
Cell 1996;86:531– 42.
9. Weeks BH, He W, Olson KL, Wang XJ. Inducible expression of transforming growth
factor beta1 in papillomas causes rapid metastasis. Cancer Res 2001;61:7435– 43.
10. Kretschmer A, Moepert K, Dames S, et al. Differential regulation of TGF-beta
signaling through Smad2, Smad3 and Smad4. Oncogene 2003;22:6748 – 63.
11. He W, Cao T, Smith DA, Myers TE, Wang XJ. Smads mediate signaling of the
TGFbeta superfamily in normal keratinocytes but are lost during skin chemical
carcinogenesis. Oncogene 2001;20:471– 83.
12. Matsuura I, Denissova NG, Wang G, et al. Cyclin-dependent kinases regulate the
antiproliferative function of Smads. Nature (Lond) 2004;430:226 –31.
13. Vijayachandra K, Lee J, Glick AB. Smad3 regulates senescence and malignant
conversion in a mouse multistage skin carcinogenesis model. Cancer Res 2003;63:
3447–52.
14. Sporn MB, Roberts AB. Transforming growth factor-beta: recent progress and new
challenges. J Cell Biol 1992;119:1017–21.
15. Liu X, Alexander V, Vijayachandra K, et al. Conditional epidermal expression of
TGFbeta 1 blocks neonatal lethality but causes a reversible hyperplasia and alopecia.
Proc Natl Acad Sci USA 2001;98:9139 – 44.
16. Li AG, Wang D, Feng XH, Wang XJ. Latent TGFbeta1 overexpression in keratinocytes results in a severe psoriasis-like skin disorder. EMBO J 2004;23:1770 – 81.
17. Ashcroft GS, Yang X, Glick AB, et al. Mice lacking Smad3 show accelerated wound
healing and an impaired local inflammatory response. Nat Cell Biol 1999;1:260 – 6.
18. Coussens LM, Werb Z. Inflammation and cancer. Nature (Lond) 2002;420:860 –7.
7844
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 2004 American Association for Cancer
Research.
SKIN CHEMICAL CARCINOGENESIS ON Smad3 KNOCKOUT MICE
19. Tian F, DaCosta BS, Parks WT, et al. Reduction in Smad2/3 signaling enhances
tumorigenesis but suppresses metastasis of breast cancer cell lines. Cancer Res
2003;63:8284 –92.
20. Yang X, Letterio JJ, Lechleider RJ, et al. Targeted disruption of SMAD3 results in
impaired mucosal immunity and diminished T cell responsiveness to TGF-beta.
EMBO J 1999;18:1280 –91.
21. Bottinger EP, Letterio JJ, Roberts AB. Biology of TGF-beta in knockout and transgenic mouse models. Kidney Int 1997;51:1355– 60.
22. Aldaz CM, Conti CJ, Klein-Szanto AJ, Slaga TJ. Progressive dysplasia and aneuploidy are hallmarks of mouse skin papillomas: relevance to malignancy. Proc Natl
Acad Sci USA 1987;84:2029 –32.
23. Wang XJ, Liefer KM, Tsai S, O’Malley BW, Roop DR. Development of gene-switch
transgenic mice that inducibly express transforming growth factor beta1 in the
epidermis. Proc Natl Acad Sci USA 1999;96:8483– 8.
24. Greenhalgh DA, Wang XJ, Rothnagel JA, et al. Transgenic mice expressing targeted
HPV-18 E6 and E7 oncogenes in the epidermis develop verrucous lesions and
spontaneous, rasHa-activated papillomas. Cell Growth Differ 1994;5:667–75.
25. Ledbetter JA, Herzenberg LA. Xenogeneic monoclonal antibodies to mouse lymphoid
differentiation antigens. Immunol Rev 1979;47:63–90.
26. Malorny U, Michels E, Sorg C. A monoclonal antibody against an antigen present on
mouse macrophages and absent from monocytes. Cell Tissue Res 1986;243:421– 8.
27. Wang XJ, Greenhalgh DA, Bickenbach JR, et al. Expression of a dominant-negative
type II transforming growth factor beta (TGF-beta) receptor in the epidermis of
transgenic mice blocks TGF-beta-mediated growth inhibition. Proc Natl Acad Sci
USA 1997;94:2386 –91.
28. Rennecke J, Rehberger PA, Furstenberger G, et al. Protein-kinase-Cmu expression
correlates with enhanced keratinocyte proliferation in normal and neoplastic mouse
epidermis and in cell culture. Int J Cancer 1999;80:98 –103.
29. Lu SL, Reh D, Li AG, et al. Overexpression of transforming growth factor beta1 in
head and neck epithelia results in inflammation, angiogenesis, and epithelial hyperproliferation. Cancer Res 2004;64:4405–10.
30. Wong MH, Huelsken J, Birchmeier W, Gordon JI. Selection of multipotent stem cells
during morphogenesis of small intestinal crypts of Lieberkuhn is perturbed by
stimulation of Lef-1/beta-catenin signaling. J Biol Chem 2002;277:15843–50.
31. Go C, He W, Zhong L, et al. Aberrant cell cycle progression contributes to the
early-stage accelerated carcinogenesis in transgenic epidermis expressing the dominant negative TGFbetaRII. Oncogene 2000;19:3623–31.
32. Nischt R, Roop DR, Mehrel T, et al. Aberrant expression during two-stage mouse skin
carcinogenesis of a type I 47-kDa keratin, K13, normally associated with terminal
differentiation of internal stratified epithelia. Mol Carcinog 1988;1:96 –108.
33. Quintanilla M, Brown K, Ramsden M, Balmain A. Carcinogen-specific mutation and
amplification of Ha-ras during mouse skin carcinogenesis. Nature (Lond) 1986;322:
78 – 80.
34. Hocevar, AH, Howe PH. Regulation of AP-1 activity by TGF-beta. In: Howe PH,
editor. Transforming growth factor-beta protocols. Totowa, NJ: Human Press; 2001.
p. 97–108.
35. Hsu TC, Young MR, Cmarik J, Colburn NH. Activator protein 1 (AP-1)- and nuclear
factor kappaB (NF-kappaB)-dependent transcriptional events in carcinogenesis. Free
Radic Biol Med 2000;28:1338 – 48.
36. DiGiovanni J, Rho O, Xian W, Beltran L. Role of the epidermal growth factor
receptor and transforming growth factor alpha in mouse skin carcinogenesis. Prog
Clin Biol Res 1994;387:113–38.
37. Reinartz J, Bechtel MJ, Kramer MD. Tumor necrosis factor-alpha-induced apoptosis
in a human keratinocyte cell line (HaCaT) is counteracted by transforming growth
factor-alpha. Exp Cell Res 1996;228:334 – 40.
38. Kanda D, Takagi H, Toyoda M, et al. Transforming growth factor alpha protects
against Fas-mediated liver apoptosis in mice. FEBS Lett 2002;519:11–5.
39. Amundadottir LT, Nass SJ, Berchem GJ, Johnson MD, Dickson RB. Cooperation of
TGF alpha and c-Myc in mouse mammary tumorigenesis: coordinated stimulation of
growth and suppression of apoptosis. Oncogene 1996;13:757– 65.
40. Tak PP, Firestein GS. NF-kappaB: a key role in inflammatory diseases. J Clin
Investig 2001;107:7–11.
41. Murphy JE, Robert C, Kupper TS. Interleukin-1 and cutaneous inflammation: a
crucial link between innate and acquired immunity. J Investig Dermatol 2000;114:
602– 8.
42. Bonifati C, Ameglio F. Cytokines in psoriasis. Int J Dermatol 1999;38:241–51.
43. Liu Y, Duysen E, Yaktine AL, et al. Dietary energy restriction inhibits ERK but not
JNK or p38 activity in the epidermis of SENCAR mice. Carcinogenesis (Lond)
2001;22:607–12.
44. Rodriguez-Villanueva J, Greenhalgh D, Wang XJ, et al. Human keratin-1.bcl-2
transgenic mice aberrantly express keratin 6, exhibit reduced sensitivity to keratino-
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
cyte cell death induction, and are susceptible to skin tumor formation. Oncogene
1998;16:853– 63.
Tannehill-Gregg SH, Kusewitt DF, Rosol TJ, Weinstein M. The roles of Smad2 and
Smad3 in the development of chemically induced skin tumors in mice. Vet Pathol
2004;41:278 – 82.
Yuspa SH. Alterations in epidermal functions resulting from exposure to initiators
and promoters of carcinogenesis. Curr Probl Dermatol 1983;11:227– 41.
Kiguchi K, Beltran LM, You J, Rho O, DiGiovanni J. Elevation of transforming
growth factor-alpha mRNA and protein expression by diverse tumor promoters in
SENCAR mouse epidermis. Mol Carcinog 1995;12:225–35.
Verrecchia F, Tacheau C, Schorpp-Kistner M, Angel P, Mauviel A. Induction of the
AP-1 members c-Jun and JunB by TGF-beta/Smad suppresses early Smad-driven
gene activation. Oncogene 2001;20:2205–11.
Liberati NT, Datto MB, Frederick JP, et al. Smads bind directly to the Jun family of
AP-1 transcription factors. Proc Natl Acad Sci USA 1999;96:4844 –9.
Wong C, Rougier-Chapman EM, Frederick JP, et al. Smad3-Smad4 and AP-1
complexes synergize in transcriptional activation of the c-Jun promoter by transforming growth factor beta. Mol Cell Biol 1999;19:1821–30.
Jorissen RN, Walker F, Pouliot N, et al. Epidermal growth factor receptor: mechanisms of activation and signalling. Exp Cell Res 2003;284:31–53.
Shaulian E, Karin M. AP-1 as a regulator of cell life and death. Nat Cell Biol
2002;4:E131– 6.
Wach S, Schirmacher P, Protschka M, Blessing M. Overexpression of bone morphogenetic protein-6 (BMP-6) in murine epidermis suppresses skin tumor formation by
induction of apoptosis and downregulation of fos/jun family members. Oncogene
2001;20:7761–9.
Eferl R, Wagner EF. AP-1: a double-edged sword in tumorigenesis. Nat Rev Cancer
2003;3:859 – 68.
Young MR, Li JJ, Rincon M, et al. Transgenic mice demonstrate AP-1 (activator
protein-1) transactivation is required for tumor promotion. Proc Natl Acad Sci USA
1999;96:9827–32.
Saez E, Rutberg SE, Mueller E, et al. c-fos is required for malignant progression of
skin tumors. Cell 1995;82:721–32.
Klein SB, Fisher GJ, Jensen TC, et al. Regulation of TGF-alpha expression in human
keratinocytes: PKC-dependent and -independent pathways. J Cell Physiol 1992;151:
326 –36.
Awwad R, Humphrey LE, Periyasamy B, et al. The EGF/TGFalpha response element
within the TGFalpha promoter consists of a multi-complex regulatory element.
Oncogene 1999;18:5923–35.
Lynch MJ, Pelosi L, Carboni JM, et al. Transforming growth factor-beta 1 induces
transforming growth factor-alpha promoter activity and transforming growth factoralpha secretion in the human colon adenocarcinoma cell line FET. Cancer Res
1993;53:4041–7.
Wang XJ, Warren BS, Rupp T, Beltran LM, DiGiovanni J. Loss of mouse epidermal
protein kinase C isozyme activities following treatment with phorbol ester and
non-phorbol ester tumor promoters. Carcinogenesis (Lond) 1994;15:2795– 803.
Sanders S, Thorgeirsson SS. Promotion of hepatocarcinogenesis by phenobarbital in
cmyc/TGF-alpha transgenic mice. Mol Carcinog 2000;28:168 –73.
Wang XJ, Greenhalgh DA, Lu XR, Bickenbach JR, Roop DR. TGF alpha and v-fos
cooperation in transgenic mouse epidermis induces aberrant keratinocyte differentiation and stable, autonomous papillomas. Oncogene 1995;10:279 – 89.
Kiguchi K, Beltran L, Dubowski A, DiGiovanni J. Analysis of the ability of 12-Otetradecanoylphorbol-13-acetate to induce epidermal hyperplasia, transforming
growth factor-alpha, and skin tumor promotion in wa-1 mice. J Investig Dermatol
1997;108:784 –91.
Keshamouni VG, Mattingly RR, Reddy KB. Mechanism of 17-beta-estradiol-induced
Erk1/2 activation in breast cancer cells. A role for HER2 AND PKC-delta. J Biol
Chem 2002;277:22558 – 65.
Updyke LW, Yoon HL, Chuthaputti A, Pfeifer RW, Yim GK. Induction of interleukin-1 and tumor necrosis factor by 12-O-tetradecanoylphorbol-13-acetate in phorbol
ester- sensitive (SENCAR) and resistant (B6C3F1) mice. Carcinogenesis (Lond)
1989;10:1107–11.
Cooper KD, Hammerberg C, Baadsgaard O, et al. Interleukin-1 in human skin:
dysregulation in psoriasis. J Investig Dermatol 1990;95:24S– 6S.
Wahl SM. Transforming growth factor beta (TGF-beta) in inflammation: a cause and
a cure. J Clin Immunol 1992;12:61–74.
Wahl SM, Allen JB, Weeks BS, Wong HL, Klotman PE. Transforming growth factor
beta enhances integrin expression and type IV collagenase secretion in human
monocytes. Proc Natl Acad Sci USA 1993;90:4577– 81.
7845
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 2004 American Association for Cancer
Research.
Smad3 Knockout Mice Exhibit a Resistance to Skin Chemical
Carcinogenesis
Allen G. Li, Shi-Long Lu, Ming-Xiang Zhang, et al.
Cancer Res 2004;64:7836-7845.
Updated version
Cited articles
Citing articles
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/64/21/7836
This article cites 66 articles, 23 of which you can access for free at:
http://cancerres.aacrjournals.org/content/64/21/7836.full#ref-list-1
This article has been cited by 10 HighWire-hosted articles. Access the articles at:
http://cancerres.aacrjournals.org/content/64/21/7836.full#related-urls
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 2004 American Association for Cancer
Research.