Oncogene (2008) 27, 3527–3538
& 2008 Nature Publishing Group All rights reserved 0950-9232/08 $30.00
www.nature.com/onc
ORIGINAL ARTICLE
Reciprocal negative regulation between S100A7/psoriasin and b-catenin
signaling plays an important role in tumor progression of squamous cell
carcinoma of oral cavity
G Zhou1, T-X Xie1, M Zhao1, SA Jasser1, MN Younes1, D Sano1, J Lin1, ME Kupferman1,
AA Santillan1, V Patel2, JS Gutkind2, AK EI-Naggar3, ED Emberley4, PH Watson4,
S-I Matsuzawa5, JC Reed5 and JN Myers1,6
1
Department of Head and Neck Surgery, The University of Texas MD Anderson Cancer Center, Houston, TX, USA; 2Oral and
Pharyngeal Cancer Branch, National Institute of Dental and Craniofacial Research, National Institutes of Health, Bethesda, MD,
USA; 3Department of Pathology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA; 4Deeley Research
Center, BC Cancer Agency, Victoria, British Columbia, Canada; 5Burnham Institute for Medical Research, La Jolla, CA, USA and
6
Department of Cancer Biology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA
Overexpression of S100A7 (psoriasin), a small calciumbinding protein, has been associated with the development
of psoriasis and carcinomas in different types of epithelia,
but its precise functions are still unknown. Using human
tissue specimens, cultured cell lines, and a mouse model,
we found that S100A7 is highly expressed in preinvasive,
well-differentiated and early staged human squamous cell
carcinoma of the oral cavity (SCCOC), but little or no
expression was found in poorly differentiated, later-staged
invasive tumors. Interestingly, our results showed that
S100A7 inhibits both SCCOC cell proliferation in vitro
and tumor growth/invasion in vivo. Furthermore, we
demonstrated that S100A7 is associated with the
b-catenin complex, and inhibits b-catenin signaling
by targeting b-catenin degradation via a noncanonical
mechanism that is independent of GSK3b-mediated
phosphorylation. More importantly, our results also
indicated that b-catenin signaling negatively regulates
S100A7 expression. Thus, this reciprocal negative regulation between S100A7 and b-catenin signaling implies their
important roles in tumor development and progression.
Despite its high levels of expression in early stage
SCCOC tumorigenesis, S100A7 actually inhibits SCCOC
tumor growth/invasion as well as tumor progression.
Downregulation of S100A7 in later stages of tumorigenesis increases b-catenin signaling, leading to promotion of
tumor growth and tumor progression.
Oncogene (2008) 27, 3527–3538; doi:10.1038/sj.onc.1211015;
published online 28 January 2008
Keywords: S100A7; psoriasin; b-catenin; oral squamous
cell carcinoma; head and neck squamous cell carcinoma
Correspondence: Professor JN Myers, Department of Head and Neck
Surgery, University of Texas MD Anderson Cancer Center, 1515
Holcombe Boulevard, Unit 441, Houston, TX 77030-4009, USA.
E-mail: [email protected]
Received 17 July 2007; revised 12 November 2007; accepted 3 December
2007; published online 28 January 2008
Introduction
S100A7 (psoriasin), a low-molecular weight (11.4 kDa)
Ca2 þ -binding protein, is a member of the S100 family of
proteins and was originally identified through its brisk
overexpression in the skin of patients suffering from
psoriasis, a hyperproliferative skin disease characterized
by abnormal differentiation. Like many other members
of the S100 family, S100A7 gene is located within the
‘epidermal differentiation complex’ on human chromosome 1q21 and its protein is believed to be implicated in
Ca2 þ -dependent and -independent regulation of a
variety of intracellular and extracellular activities
(Donato, 2001, 2003; Heizmann et al., 2002). It has
also been found that S100A7 is highly overexpressed in
wound healing, and many inflammatory and hyperproliferative epidermal diseases, including psoriasis, atopic
dermatitis, mycosis fungoides, Darier’s disease and
lichen sclerosus et atrophicus as well as skin cancers
and breast cancers. However, the exact role of S100A7
in these diseases is still unclear.
S100A7 overexpression during skin and breast tumorigenesis suggests that S100A7 might have an ‘oncogenic’
role in tumor development. However, a close examination
of the expression profile of S100A7 during tumor
progression suggests a more complex picture of functional
roles for S100A7 in tumor development. Though its
expression in skin is highly induced in keratinocytes under
specific pathological circumstances, including hyperplasia
associated with inflammatory skin lesions and dysplasia
associated with neoplastic progression (Alowami et al.,
2003; Emberley et al., 2003a), S100A7 actually first
emerged in breast cancer when identified as a cDNA
downregulated in a nodal metastasis relative to a primary
invasive breast tumor (Moog-Lutz et al., 1995). Furthermore, analysis in breast cancers showed that S100A7
expression is relatively low or undetectable in normal,
benign and atypical hyperplastic proliferative ductal
lesions, but is often highly induced in the ductal epithelial
cells of premalignant, preinvasive ductal carcinoma in situ
S100A7 inhibits SCCOC progression
G Zhou et al
3528
(DCIS) (Leygue et al., 1996; Enerback et al., 2002) and
then its expression is diminished in invasive carcinomas
(Leygue et al., 1996; Emberley et al., 2003b). Such
‘biphasic expression’ suggests that its overexpression is
associated with altered differentiation in glandular epithelium of DCIS, that S100A7 may only have a role in early
breast tumor progression (Watson et al., 1998), and that
the concomitant loss of S100A7 expression in the later
stage of invasive carcinomas may be required for tumor
progression. In other words, its overexpression may
actually ‘suppress’ the tumor progression to later stages
of invasive carcinomas. Consistent with the putative
‘tumor suppressive’ role are the observations that the
expression of S100A7 is relatively low or undetectable in
most highly proliferating keratinocytes and tumor cells
in vitro, but it can be induced by agents that stimulate cell
differentiation such as retinoic acid (Tavakkol et al., 1994),
calcium (Hoffmann et al., 1994), UV light (Di Nuzzo et al.,
2000), DNA damaging agents (Kennedy et al., 2005), loss
of cell attachment to extracellular matrix, growth factor
starvation and cell confluent culture conditions (Enerback
et al., 2002; Martinsson et al., 2005). In addition, in
normal epithelium, S100A7 has been shown to be
expressed in the superficial, differentiated region of the
epithelium rather than in the highly proliferative basal
region, and its expression correlates with the degree of
keratinocyte differentiation (Broome et al., 2003). In skin
tumors, it is absent in undifferentiated basalioma and
strongly expressed in carcinoma in situ, as well as in
keratoacanthoma and differentiated squamous cell carcinoma. These findings support the hypothesis that S100A7
is involved in epithelial differentiation (Martinsson et al.,
2005) rather than in cell proliferation.
So far very little is known about molecular mechanisms
of S100A7’s function. To study the potential biological
role of S100A7 in oral squamous epithelial tumor
progression, we first evaluated S100A7 expression in both
human and mouse specimens of human squamous cell
carcinoma of the oral cavity (SCCOC) at different stages
of tumor development. We then used both adenoviraland lentiviral-mediated gene overexpression and underexpression strategies in human SCCOC cell lines, which
identified a critical role of S100A7 in inhibiting tumor
growth and progression of SCCOC. We further established that S100A7 is an important negative modulator of
b-catenin signaling through targeting b-catenin for
degradation, and that b-catenin in turn can inhibit the
expression of S100A7. Our results suggest that although
S100A7 is overexpressed in early stages of tumorigenesis,
this protein actually plays an inhibitory role in tumor
progression of SCCOC, and loss of S100A7 expression is
associated with more advanced disease.
Results
S100A7 is overexpressed in early stages of preinvasive,
well-differentiated SCCOC but is lost in later stages of
tumor development
To investigate the role of S100A7 in the development of
human SCCOC, we first evaluated the expression
Oncogene
patterns of S100A7 in 27 archival human tongue tissue
specimens, including those of histologically normal and
dysplastic squamous mucosa and invasive SCCOC. In
normal oral epithelia, no expression of S100A7 was
observed in dividing basal layer keratinocytes, whereas
some low-level, scattered expression of S100A7 was seen
in the differentiated suprabasal layers (Figure 1a).
S100A7 expression was increased in dysplastic lesions
and in well-differentiated SCCOC, but in the later stages
of poorly differentiated invasive SCCOC, expression
was greatly reduced or absent (Figure 1a). Of the 27
SCCOC specimens we evaluated, 74% (20/27) were well
differentiated or moderately differentiated carcinomas
and they were all positive for S100A7 expression,
whereas none of the seven poorly differentiated invasive
carcinomas expressed S100A7 (Figure 1b). In addition,
S100A7 expression level was inversely correlated with
histological grade, with higher expression seen in the
well-differentiated and preinvasive carcinomas, less in
the moderately differentiated carcinomas, and none in
the poorly differentiated invasive carcinomas (Figures
1a and b). We also examined the expression of S100A7
in orthotopic SCCOC tumor models in nude mice
generated from three isogenic human SCCOC cell lines
(Tu167, JMAR and DM14) with varying tumor growth
and metastatic potentials (Supplementary Materials). As
shown in Figure 1c, the Tu167-derived tongue tumors
(less malignant) exhibited a more differentiated morphology and a higher level of S100A7 expression,
whereas JMAR-derived tongue tumors (moderately
malignant) demonstrated reduced S100A7 expression
with only scattered staining detected. In the most
malignant DM14 tumors, however, S100A7 expression
was almost entirely lost. Together, these results demonstrate an inverse relationship between levels of S100A7
expression and tumor progression in both human
SCCOC and orthotopic tumors in nude mice.
Expression of S100A7 inhibits SCCOC cell proliferation
in vitro and tumor growth and tumor progression in vivo
While in vivo expression of S100A7 is very high in
early staged, well-differentiated SCCOC, its expression
in vitro SCCOC cells (for example, Tu167, JMAR and
DM14) is very low or undetectable (Supplementary
Figure S1B), but can be induced by cell confluence or
cell detachment (Figure 6a) as previously reported in
other cell lines (Enerback et al., 2002). To elucidate the
role of S100A7 in cell growth and tumorigenesis of
SCCOC, we generated an adenoviral mediated expression system to overexpress S100A7 (Figure 2a). As
shown in Figure 2b and Supplementary Figure S1A,
when S100A7 and control green fluorescent protein
(GFP) virus were used to infect Tu167 and JMAR cells,
cell proliferation was inhibited by S100A7 overexpression in a dose-dependent manner, Similarly, when
Tu167 or immortalized human keratinocytes HOK16B
were infected with the virus, then subjected to a colony
formation and cell migration assays, the number of
colonies and cell migration were greatly suppressed by
S100A7 overexpression (Figures 2c–e). To further
extrapolate these findings in vivo, we used an orthotopic
S100A7 inhibits SCCOC progression
G Zhou et al
3529
Cornified layer
Suprabasal layer
Basal layer
Dermis
Tu167 Tumor
JMAR Tumor
DM14 Tumor
Tumor malignancy
Figure 1 S100A7 is overexpressed in the early stages of well-differentiated, preinvasive squamous cell carcinoma of the oral cavity
(SCCOC). (a) Representative results from immunohistochemical (IHC) staining of human SCCOC specimens with S100A7 antibody.
(b) Summary of S100A7 IHC staining from 27 of human SCCOC specimens. *Po0.05, Wilcoxon rank-sum test. (c) Representative
results from IHC staining with S100A7 antibody of orthotopic tongue SCCOC tumor specimens in nude mice derived from three
human SCCOC cell lines.
tongue tumor model, in which JMAR cells were infected
with the adenovirus and then orthotopically inoculated
into tongues of nude mice. Consistent with our
observation in vitro, JMAR tumor growth was suppressed by S100A7 overexpression by B5-fold as
measured by tumor volume (Figure 2f).
In addition, histological analysis revealed that tumors
derived from control cells and S100A7 cells had different
tumor morphologies (Figure 2g). In the control group
(JMAR-GFP-Ad), tumors consisted of sheets of poorly
differentiated and highly infiltrative cells. In contrast,
S100A7-infected JMAR tumor cells (JMAR-S100A7Ad) grew as well-defined, organized islands that were
well demarcated from surrounding mesenchymal tissue
and well differentiated, as shown by the presence of
keratin pearls. An immunofluorescent assay using Flag
tag antibody further demonstrated that exogenous
S100A7 was indeed expressed in JMAR-S100A7-Ad
tumors (Figure 2h). In addition, in control tumors,
immunohistochemical (IHC) staining demonstrated a
high cellular proliferative index, as determined by
extensive proliferating cell nuclear antigen (PCNA)
staining throughout the tumor (Figure 2g). However,
in JMAR-S100A7-Ad tumors, PCNA staining was
restricted to cells at the periphery of the tumor
islands that did not express S100A7, while tumor cells
expressing S100A7 exhibited no PCNA staining as
shown in consecutive sections (Figure 2g). This inverse
correlation in vivo between S100A7 and PCNA staining
further highlights that overexpression of S100A7 inhibits tumor cell proliferation, tumor cell dissemination
and invasion.
Overexpression of S100A7 inhibits b-catenin signaling
To study the possible mechanisms involved in S100A7mediated tumor growth inhibition, we examined the
effects of overexpression of S100A7 on multiple signaling pathways including b-catenin signaling. We first
noticed that the tyrosine phosphorylation of b-catenin
induced by epidermal growth factor (EGF) stimulation
was concomitantly inhibited by adenovirus-mediated
expression of S100A7 in JMAR cells (Figure 3a).
Further analysis indicated that the reduced levels of
phosphorylated b-catenin could be attributed to an
overall decrease in the level of b-catenin protein
secondary to S100A7 overexpression (Figures 3b and c).
Oncogene
S100A7 inhibits SCCOC progression
G Zhou et al
3530
Figure 2 Overexpression of S100A7 inhibits cell proliferation in vitro and tumor growth in vivo. (a) Generation of recombinant
adenovirus expressing Flag-tagged S100A7. (top) Schematic of adenoviral construct encoding S100A7. (bottom left) S100A7
adenovirus-infected JMAR cells (10 multiplicity of infection) marked by green fluorescent protein (GFP) fluorescence. (bottom right)
Western blotting of JMAR cells infected with S100A7 and control GFP adenoviruses. (b) MTT assay for cell proliferation of Tu167
cells infected with S100A7 or GFP adenovirus. (c and d) Colony formation assay of Tu167 and HOK16B cells infected with different
doses of S100A7 or GFP adenovirus. (e) Cell migration assay of Tu167 cells infected with the virus. (f) Orthotopic tongue tumor
volumes derived from JMAR cells infected with or without virus. In all the above-mentioned cases, error bars represent standard
deviation. *Po0.05. (g) Consecutive tumor sections of immunohistochemical (IHC) staining of tumors derived from control GFP
(JMAR-GFP-Ad) and S100A7 virus-infected (JMAR-S100A7-Ad) tumors using antibodies as indicated (insets 200). (h)
Immunofluorescent staining of frozen sections of control and JMAR-S100A7-Ad tumors using Flag antibody as indicated. In each
group, 200 images on right hand side magnify box areas of the 50 images on left-hand side. Hematoxylin and eosin (H&E) section
is the consecutive slide of fluorescent image on the top.
The finding that S100A7 decreases b-catenin expression
led us to hypothesize that S100A7 expression could
negatively regulate b-catenin signaling through nuclear
translocation and activation of the T-cell factor (TCF)/
b-catenin transcription complex and subsequent activation
of TCF-regulated genes including c-Myc (He et al., 1998).
To test this hypothesis, a b-catenin/TCF-dependent
luciferase reporter TopFlash and its b-catenin/TCFbinding site mutant control FopFlash (Korinek et al.,
Oncogene
1997) were co-transfected into 293T, Tu167 and JMAR
cells together with b-catenin and S100A7 expression
vectors. As shown in Figure 3d, the b-catenin-activated
luciferase activity was inhibited in a dose-dependent
manner by S100A7 overexpression in all three cell lines.
Moreover, both basal and EGF-induced expressions of
c-Myc were suppressed by S100A7 overexpression in JMAR
cells (Figure 3e). Together, these findings support a role for
S100A7 expression in inhibition of b-catenin signaling.
S100A7 inhibits SCCOC progression
G Zhou et al
d
-A
00
S1
IgG
A7
Ad
- + IB:
P-Tyr
P-
+
GF
EGF -
S1
00
A7
-A
GF
PAd
d
3531
IP: β-Catenin
β-Catenin
p63
S100A7
Actin
S100A7
Actin
β-Catenin
Merge
Relative Luc Activity
Lysate
FopFlash
TopFlash
β-Catenin
S100A7
3
293T
5
Tu167
JMAR
3
GFP-Ad S100A7-Ad
0 0.5 1 2 0 0.5 1 2 EGF (h)
4
2
3
2
2
1
c-Myc
S100A7
1
1
Actin
JMAR
1
+
-
2
+
+
-
3
+
+
+
4
+
-
5
+
+
-
6
+
+
7
+
+
8
+
+
9 10 11
+ - - + +
+ - +
-
12 13
- + +
+ +
14
+
+
-
15
+
-
16
+
+
17
+
+
18
+
+
-
Figure 3 Overexpression of S100A7 inhibits b-catenin signaling. (a) Immunoprecipitation (IP) with b-catenin antibody followed by
immunoblotting (IB) with phosphotyrosine (P-Tyr) in the absence and presence of epidermal growth factor (EGF; 50 ng ml1)
stimulation (30 min) in serum-starved JMAR cells infected with control green fluorescent protein (GFP) or S100A7 adenovirus.
(b) Western blotting of b-catenin protein in JMAR cells after S100A7 and control GFP virus infection. (c) Representative confocal
microscopic images of immunofluorescent staining of JMAR cells infected with S100A7 adenovirus. (d) Luciferase assay of b-catenindependent reporter TopFlash in 293T, Tu167 and JMAR cells. (e) Western blotting of c-Myc protein in serum-starved JMAR cells
infected with virus after EGF (50 ng ml1) stimulation for different time (h).
S100A7 targets b-catenin degradation via a noncanonical
mechanism that is independent of GSK3b-mediated
phosphorylation
To further investigate the mechanism involved in the
inhibition of b-catenin signaling by S100A7, we
examined the subcellular localization of S100A7. To
visualize endogenous S100A7, we used breast cancer
MDA-MB-468 cells because this cell line has high levels
of S100A7 expression (Supplementary Figure S1B). As
shown in Figure 4a, most of the S100A7 was localized in
the cytoplasm as well as the cell cortex region.
Reciprocal immunoprecipitation (IP) assays showed
that S100A7 and b-catenin were co-precipitated from
MDA-MB-468 and confluent JMAR cell lysates
(Figure 4b), indicating that S100A7 interacts with the
b-catenin complex in these cells.
To understand whether the downregulation of bcatenin is due to transcriptional or post-transcriptional
regulation, CMV promoter-driven Akt or b-catenin
constructs were transfected with S100A7 into 293T cells.
As shown in Figure 4c, the expression of both
endogenous and transfected b-catenin was suppressed
by S100A7 expression (compare lane 3 with 1, 2 and
lane 6 with 5), whereas transfected Akt expression was
not affected (compare lanes 3, 4 with 2). Moreover,
the S100A7-mediated suppression of b-catenin was
inhibited by proteasome inhibitor MG132 (compare
lane 4 with 3, and lane 7 with 6). Taken together, all
these results demonstrated the specificity of S100A7mediated b-catenin inhibition and strongly suggested
that it might be due to the post-transcriptional regulation, which is dependent on proteasome-mediated
protein degradation.
Because the degradation of b-catenin is controlled by
a GSK3b-mediated phosphorylation-dependent (Hart
et al., 1999; Kitagawa et al., 1999) (Moon et al., 2004;
Willert and Jones, 2006) as well as Siah-1/Sip-mediated
GSK3b phosphorylation-independent pathways (Liu
et al., 2001; Matsuzawa and Reed, 2001; Fukushima
et al., 2006), we used both luciferase assay and western
blotting to test the possible mechanisms involved. As
shown in Figure 4d, S100A7 did not appear to influence
the activity of GSK3b, which regulates b-catenin levels
by targeting it for degradation. However, when S100A7
was transfected together with Siah-1/Sip into 293T
cells, it further decreased b-catenin protein levels and
b-catenin-dependent luciferase activity (Figure 4d).
Furthermore, when a constitutively active mutant
b-catenin that is resistant to GSK3b-mediated phosphorylation and degradation (Bienz and Clevers, 2000;
Matsuzawa and Reed, 2001) was transfected together
with S100A7 into cells, it was also inhibited by S100A7
Oncogene
S100A7 inhibits SCCOC progression
G Zhou et al
IP:
S1
00
A7
IB:
IgG
IP:
ate
Merge
-C
E-Cadherin
IgG
S100A7
nin
3532
IB:
-Catenin
S100A7
IgG
IgG
JMAR
-
+
-
+
+
-
+
+
+
-
+
+
+
5
+
+
+
-Catenin
HA (Akt)
S100A7
Actin
1
2
3
4
5
6
7
Relative Luc Activity
MG132
S100A7
HA-Akt
-Catenin
MDA-MB-468
4
3
2
1
Top Flash
-Catenin
S100A7
GSK3
Siah-1
Sip
S100A7
MT--Catenin
WT--Catenin
+
-
+
+
-
+
+
+
-
+
+
+
-
+
+
+
+
-
+
+
+
-
+
+
+
+
-
+
+
+
+
+
+
+
+
+
Flag--Catenin
Flag--Catenin
S100A7
Actin
Myc-GSK3
Flag-Siah-1
Myc-Sip
Actin
Figure 4 S100A7 targets b-catenin degradation. (a) Representative confocal microscopic images of fluorescent immunostaining with
S100A7 (green) and E-cadherin (red) antibodies in MDA-MB-468 cells. (b) Immunoprecipitation (IP) assays using S100A7 or
b-catenin antibodies as indicated. Cell lysates were from serum-starved (12 h) confluent JMAR or MDA-MB-468 cells as indicated. IB,
immunoblotting. (c) Western blotting of 293T cells 48 h after transfection with indicated DNA. In lanes 4 and 7, transfected cells were
treated with 10 mM MG132 for 3 h before harvest. (d) (top) Luciferase assay and (bottom) western blotting of 293T cells after
transfection with indicated DNA in 12-well culture plate. (e) Western blotting of 293T cell transfected with wild-type (WT) or
constitutive active mutant (MT) b-catenin in the presence or absence of increasing amounts of S100A7.
overexpression in a dose-dependent manner even though
this mutant form of b-catenin is more stable than wildtype b-catenin (Figure 4e). Together, these results
strongly indicate that S100A7 targets b-catenin for
degradation via a noncanonical mechanism that is
independent of GSK3b-mediated phosphorylation.
b-Catenin negatively regulates S100A7 expression
Since our data indicate that S100A7 inhibits b-catenin
signaling, and Wnt/b-catenin signaling functions as a
‘master switch’ for proliferation versus differentiation
(van de Wetering et al., 2002), we set out to determine
whether b-catenin signaling could in turn regulate
S100A7 expression in SCCOC cells. To assess this, we
first treated both JMAR and MDA-MB-468 cells with
LiCl, an inhibitor of GSK3b that negatively regulates
b-catenin signaling (Hart et al., 1999; Kitagawa et al.,
Oncogene
1999; Moon et al., 2004; Willert and Jones, 2006). As
shown in Figure 5a, upon treatment with LiCl, the levels
of S100A7 protein were significantly reduced in both cell
lines, and this inhibition appeared to be independent of
proteasome-mediated protein degradation since proteasome inhibitor MG132 did not affect the downregulation of S100A7 by LiCl. Consistent with this, real-time
RT-PCR analyses demonstrated that mRNA levels of
S100A7 were inhibited by LiCl (Figure 5b). Since the
inhibition of GSK3b by LiCl releases the suppression of
b-catenin signaling, we hypothesized that b-catenin
signaling may be directly implicated in the regulation
of S100A7 expression. To test this, we overexpressed a
constitutively active mutant of b-catenin in MDA-MB468 and JMAR cells, As shown in Figure 5c, retrovirusmediated overexpression of b-catenin in stable clones
reduced the S100A7 expression in these cells, suggesting
S100A7 inhibits SCCOC progression
G Zhou et al
MDA-MB-468
+
-
+
-
JMAR
+
+
-
+
+
-
+
+
LiCl
MG132
S100A7
Actin
Relative mRNA levels of S100A7
3533
- LiCl
+ LiCl
1.2
1
0.8
0.6
0.4
0.2
0
MDA 468
RN
AB
sh
RN
AA
rs
Ct
JMAR
sh
hR
NA
-C
at
e-
ab
e
pB
pB
ab
e-
-C
at
pB
ab
pB
ab
e
MDA-MB-468
-Catenin
-Catenin
-Catenin
S100A7
S100A7
S100A7
Actin
Actin
JMAR
Actin
JMAR
Figure 5 b-Catenin signaling inhibits S100A7 expression. (a) Western blotting of MDA-MB-468, and confluent JMAR cells in the
absence () or presence ( þ ) of LiCl (40 mM) and/or MG132 (10 mM) for 6 h. (b) Relative levels of S100A7 mRNA measured by realtime RT-PCR. (c) Western blotting of stable MDA-MB-468, and confluent JMAR cells generated from control pBabe retrovirus and
pBabe-b-cat retrovirus expressing a constitutive active mutant b-catenin. Stable cells are polyclonal pools established from selection
with puromycin (2 mg ml1) after retroviral infection. (d) Western blotting of stable JMAR cells infected with control (Ctr) and
b-catenin (A and B) shRNA-expressing lentivirus (see ‘Materials and methods’). Stable cells are pools of green fluorescent protein
(GFP)-positive cells sorted by fluorescence-activated cell sorting (FACS).
that activation of b-catenin signaling is sufficient to inhibit
S100A7 expression in these cell lines. Conversely, when two
independent lentivirus-mediated b-catenin shRNAs (that is,
no. A and B) were introduced into JMAR cells to
downregulate b-catenin, S100A7 expression was induced
under confluent culture condition (Figure 5d). Moreover,
levels of S100A7 expression were inversely correlated with
levels of b-catenin expression (Figure 5d). Taken together,
our results have demonstrated that S100A7 inhibits
b-catenin levels, and that b-catenin signaling in turn, inhibits
the expression of S100A7 in a negative feedback loop.
Expression of S100A7 and b-catenin is inversely
correlated with each other in both in vitro and in vivo
To further evaluate the reciprocal regulation of S100A7
and b-catenin, we examined the expression pattern of
S100A7 and b-catenin both in vitro and in vivo.
Although Tu167, JMAR and DM14 cells exhibit
different levels of b-catenin expression in vitro (Supplementary Figure S1B), b-catenin expression was found to
be downregulated by detachment from the extracellular
matrix, a condition which leads to elevated levels of
S100A7 (Figure 6a). Similarly, in both naturally
occurring human SCCOC and orthotopic tongue tumor
samples, b-catenin staining in S100A7-expressing areas
was found to be markedly diminished and localized to
the cell membrane when compared with the adjacent
S100A7-negative areas that exhibited higher levels of
PCNA and b-catenin staining both in cytoplasm and
nuclei (Figure 6b). The inverse correlation between
S100A7 and b-catenin staining in tissue specimens lends
further support to the hypothesis that these proteins are
reciprocally regulated.
Decreased expression of S100A7 enhances b-catenin
signaling and promotes tumor growth in an orthotopic
SCCOC xenograft mouse model
To further study the physiological role of S100A7, we
generated recombinant lentivirus expressing three independent shRNAs targeting different regions of
S100A7 gene (Supplementary Materials and methods).
Since the MDA-MB-468 cell line has a high level of
S100A7 expression (Supplementary Figure S1B), we first
used it to establish stable polyclonal cell lines to test
whether these shRNAs can suppress S100A7 expression.
As shown in Figure 7a, in stable polyclonal cell
populations generated from shRNAs nos. 2 and 3,
S100A7 expression was inhibited by more than 80%
when compared to that in cells transfected with
control shRNAs. More importantly, siRNA-mediated
Oncogene
S100A7 inhibits SCCOC progression
G Zhou et al
3534
Cell Detachment
Tu167
- +
JMAR
- +
DM14
+
S100A7
β-Catenin
Actin
β-Catenin
S100A7
β-Catenin
Human SCCOC
S100A7
X 50
X 200
Case 1
H&E
Case 2
S100A7
β-Catenin
PCNA
Tu167 Tumor
JMAR Tumor
Orthotopic SCCOC
X 50
X 200
X 50
X 200
Figure 6 Expression of S100A7 and b-catenin is inversely correlated in both in vitro and in vivo. (a) Western blotting of Tu167, JMAR
and DM14 cells treated with cell detachment. Confluent cells were treated with ( þ ) or without () cell detachment for 24 h.
(b) Representative immunohistochemical (IHC) images of consecutive tumor sections of human squamous cell carcinoma of the oral
cavity (SCCOC) and orthotopic Tu167, JMAR tongue tumors. In each group, consecutive tumor sections stained with different
antibodies are shown as indicated, in which 200 images at the bottom magnify box areas of the 50 images on the top. Note that
S100A7-expressing area has a reduced staining of b-catenin (mainly on the cell membranes) and proliferating cell nuclear antigen
(PCNA), when compared with adjacent S100A7-negative areas that has a higher level of both b-catenin (in both cytoplasm and nuclei)
and PCNA staining.
suppression of S100A7 in shRNA polyclonal cells (nos.
2 and 3) resulted in increased b-catenin as well as c-Myc
expression when compared with the control cells (no. 5)
(Figure 7b). In addition, our results indicate that
decreased expression of S100A7 also induces snail
expression and inhibits E-cadherin-mediated cell–cell
adhesion (Supplementary Figure S2). These data are
consistent with the notion that S100A7 negatively
regulates levels of b-catenin as well as its downstream
target genes in this cell line.
As shown in Figure 1c, Tu167 and JMAR tumors
exhibit different levels of differentiation, malignancy
and S100A7 expression in vivo. However the level of
S100A7 expression in these two cell lines was virtually
undetectable in vitro under subconfluent culture condition (Supplementary Figure S1B) unless induced by cell
confluence or cell detachment (Figure 6a). In spite of
this, we infected these cells with S100A7 shRNA
lentivirus and established stable shRNA clones of both
JMAR and Tu167 cells. Consistent with results in
Oncogene
MDA-MB468 cells, these cells had decreased levels of
S100A7, but elevated levels of b-catenin expression
relative to control stable cells when examined under
confluent culture conditions (Figure 7c). After orthotopic injection of these cells into the tongues of nude
mice, JMAR cells with S100A7 shRNA (nos. 2 and 3)
grew faster than control cells as measured by the tumor
volume (Figure 7d), despite that no obvious morphological differences between two groups were observed
(Supplementary Figure S3). However, since Tu167
tumor exhibits a more differentiated phenotype and a
higher level of S100A7 expression than JMAR tumor
in vivo (Figures 1c and 6b), our result demonstrated
that downregulated S100A7 by S100A7 shRNA had
a more impact on Tu167 tumor morphology, and that
S100A7 shRNA Tu167 tumors exhibit a less differentiated phenotype when compared with the control
(Figure 7e). In addition, an inverse correlation between
S100A7 and b-catenin, PCNA was also found in both
control and S100A7 shRNA tumors (Figure 7e), further
S100A7 inhibits SCCOC progression
G Zhou et al
3535
S100A7
shRNA
#1
#2
S100A7
Ctr
shRNA shRNA
#5 #3 #2
Control
shRNA
#3
#4
Control
shRNA
S100A7
#5
#4
S100A7 Control S100A7
shRNA shRNA shRNA
#2 #3
#5
#2
S100A7
β-Catenin
S100A7
β-Catenin
c-Myc
Actin
Actin
Actin
MDA-MB468
MDA-MB468
JMAR
90
80
70
60
50
40
30
20
10
0
shRNA #3
shRNA #2
ctrl shRNA #4
*
*
0
7
9
12
14
Day Post Cell Inoculation
Tu167 S100A7 shRNA #2
PCNA
β-Catenin
S100A7
H&E
Tu167 control shRNA #5
Tu167
3
Tumor Volum (mm )
JMAR
X50
X200
X50
X200
Figure 7 Decreased expression of S100A7 leads to increased levels of b-catenin, c-Myc and promotes tumor growth in an orthotopic
squamous cell carcinoma of the oral cavity (SCCOC) xenograft mouse model. (a) Western blotting of lentiviral shRNA stably
transfected MDA-MB468 cell lines. Each stable cell line was generated from pools of green fluorescent protein (GFP)-positive cells
from different S100A7 shRNAs (nos. 1, 2 and 3) and control shRNA (nos. 4 and 5) lentivirus. (b) Western blotting of lentiviral shRNA
transfected MDA-MB468 cells with different antibodies as indicated. (c) Western blotting of confluent Tu167 and JMAR shRNA
stable polyclones by indicated antibodies. (d) Tumor volumes derived from JMAR shRNA stable polyclones. *Po0.05.
(e) Representative immunohistochemical (IHC) images of consecutive sections of Tu167 tumors derived from S100A7 shRNA (no.
2) and control (no. 5) stable polyclones. In each panel, consecutive sections stained with different antibodies are shown as indicated, in
which 200 images on right hand side magnify box areas of the 50 images on left hand side. Note that S100A7-expressing area has a
reduced staining of b-catenin (mainly on the cell membranes) and proliferating cell nuclear antigen (PCNA), when compared with
adjacent S100A7-negative areas that has higher levels of both b-catenin (in both cytoplasm and nuclei) and PCNA staining.
supporting that S100A7 expression inhibits b-catenin
signaling and tumor cell proliferation in vivo. Therefore,
from our vitro and in vivo analyses of overexpression
and downregulation of S100A7, we conclude that
S100A7 negatively regulates tumor growth and progression by targeting b-catenin signaling.
Discussion
In this study, we have shown that the reciprocal negative
regulation of S100A7 and b-catenin signaling plays an
important role in tumor progression of SCCOC. On the
basis of our results, we have proposed a working model
for S100A7 function, which is summarized in Figure 8.
In this model, S100A7 functions as a putative
‘tumor suppressor’ inhibiting tumor growth and tumor
progression, despite its overexpression in the early
stage of tumor development. In support of this model
are our results showing that S100A7 inhibits SCCOC
cell proliferation in vitro, promotes tumor differentiation and suppresses tumor growth, tumor cell
invasion in vivo in an orthotopic xenograft SCCOC
tumor model. Moreover, we have demonstrated that
S100A7 is associated with the b-catenin complex, and it
negatively regulates b-catenin signaling by targeting its
Oncogene
S100A7 inhibits SCCOC progression
G Zhou et al
3536
Pre-malignancy
carcinoma in situ
Malignancy
Invasive carcinoma
S100A7
S100A7
β-Catenin
Tumor Progression
β-Catenin
EMT
Degradation
X
TCF
Differentiation
β-Catenin
TCF
c-myc
cyclin D1
etc.
Proliferation
Invasion
Metastasis
Figure 8 Working model of S100A7 in oral cancer tumor
progression. Despite its overexpression in premalignant tumors,
S100A7 actually suppresses tumor progression by targeting bcatenin signaling. Conversely, its downregulation by b-catenin
signaling and/or other unknown mechanisms promotes tumor
progression.
degradation via a noncanonical mechanism that is
independent of GSK3b-mediated phosphorylation.
Conversely, downregulation of S100A7 expression by
siRNA was shown to enhance b-catenin signaling,
including increased b-catenin and c-Myc, promoting
tumor growth and tumor progression in vivo. Finally,
we demonstrated that not only does S100A7 inhibit
b-catenin, but b-catenin signaling can negatively regulate
S100A7 expression to form a negative feedback loop.
Therefore, the reciprocal negative regulation between
S100A7 and b-catenin signaling highlights their important roles in tumor development and tumor progression.
In contrast to our findings are previous studies of
breast cancer MDA-MB-231 and MDA-MB-468 cells
that suggested that S100A7 expression promotes tumor
growth in nude mice (Emberley et al., 2003a, 2005; Krop
et al., 2005). This discrepancy may be attributed to the
biological differences between squamous and secretory
epithelial cells, which are derived from distinct stem cell
lineages. It should be noted that in agreement with our
current results, a prior study of MDA-MB-468 cells
indicated that S100A7 expression inhibited cell anchorage-independent growth, motility and invasion in
in vitro assays (Krop et al., 2005). However, in the same
study, downregulation of S100A7 expression by
shRNAs did not increase MDA-MB-468 cell tumorigenicity in nude mice. The different in vitro and in vivo
phenotypic behaviors of MDA-MB-468 cells remain
unexplained. Unlike most cultured tumor cell lines that
do not normally demonstrate constitutive S100A7
expression unless induced by differentiation stimuli
(Hoffmann et al., 1994; Tavakkol et al., 1994; Di Nuzzo
et al., 2000; Enerback et al., 2002; Kennedy et al., 2005),
MDA-MB-468 is a relatively differentiated cell line that
Oncogene
constitutively expresses extremely high levels of S100A7
protein (Supplementary Figure S1B) (Enerback et al.,
2002; Emberley et al., 2003a), and exhibits a poorly
tumorigenic phenotype in subcutaneous mouse model
(Krop et al., 2005) when compared with our orthotopic
SCCOC model (10 versus 2 weeks). Although our results
indicate that decreased expression of S100A7 in MDAMB-468 cells increased b-catenin signaling in vitro
(Figures 7a and b), how this will impact tumor growth
in vivo in this particular cell line needs to be further
investigated. However, given its unusually high levels of
S100A7 expression when compared with most in vitro
tumor cell lines, it is quite possible that MDA-MB-468
cells have developed certain genetic or epigenetic
modifications that render them resistant to growth
inhibition by S100A7 overexpression.
It has been shown that activation of Wnt/b-catenin
signaling functions as a master switch for proliferation
versus differentiation (van de Wetering et al., 2002), and
this plays a pivotal role in the initiation and progression of
cancers (Morin and Weeraratna, 2003; Taketo, 2004). In
normal epithelium, Wnt/b-catenin signaling pathways
have been shown to control the specification, maintenance, and activation of stem cells, and dysregulation of
the pathway often results in the development of familial
and/or sporadic epithelial cancers (Blanpain et al., 2007).
Our finding that S100A7 negatively regulates b-cateninsignaling pathways suggests that S100A7 functions
through these pathways to inhibit cell proliferation and
promote cell differentiation. In further support of this
hypothesis, we have shown that S100A7 overexpression
inhibited expression of the b-catenin target gene, c-Myc
(Figure 3e). c-Myc, an oncogenic protein, has the ability
to drive unrestricted cell proliferation, inhibit cell
differentiation, stimulate cell growth and vasculogenesis,
reduce cell adhesion and promote metastasis and genomic
instability (Patel et al., 2004; Adhikary and Eilers, 2005).
Conversely, the loss of Myc proteins not only inhibits cell
proliferation and growth but also accelerates differentiation and increases cell adhesion (Patel et al., 2004;
Adhikary and Eilers, 2005). Therefore, the tumor
suppressive phenotype of S100A7 overexpression shown
in our study is consistent with the phenotype derived from
the loss of Myc function, supporting the involvement of
S100A7 in the negative regulation of the Wnt/b-catenin/
Myc pathways. It should be noted that despite S100A7
overexpression in skin cells of patients suffering from
psoriasis, a hyperproliferative skin disease characterized
by abnormal differentiation, its role in psoriasis is largely
unknown. On the basis of our observation, however, we
believed that it plays a role in ‘abnormal differentiation’
rather than ‘proliferation’ in this disease. Consistent with
this, both decreased b-catenin and lack of activation of
Wnt/b-catenin signaling were observed in psoriatic skin
lesions (Reischl et al., 2007).
While our preliminary results indicate that S100A7
might interact with the b-catenin complex and target bcatenin degradation in a phosphorylation-independent
manner (Figure 4), the detailed mechanisms involved are
still largely unknown. Previously, other members of
S100A7 proteins such as S100A1, S100A6, S100A12,
S100A7 inhibits SCCOC progression
G Zhou et al
3537
S100B and S100P, were shown to interact with SIP in a
calcium-dependent manner (Filipek et al., 2002). It
remains to be determined whether S100A7 can also
interact with SIP and participate in Siah-SIP-mediated
b-catenin degradation as previously reported (Liu et al.,
2001; Matsuzawa and Reed, 2001; Fukushima et al.,
2006). Alternatively, S100A7 may target additional
components of b-catenin-signaling or other important
cell signaling pathways, which remain to be identified.
Furthermore, although our results unambiguously
demonstrate that b-catenin signaling provides negative
feedback to inhibit S100A7 expression (Figure 5), the
mechanisms involved are still unknown. Interestingly,
c-Myc, one of downstream targets of b-catenin signaling, was recently shown to directly inhibit S100A7
expression (Kennedy et al., 2005). However, whether
c-Myc or/and other downstream targets are also
implicated in the inhibition of S100A7 expression by
b-catenin signaling still remain to be further investigated.
The exact mechanisms by which the biphasic expression (up- and downregulation) of S100A7 during the
SCCOC tumorigenesis is achieved and their biological
significance also remain to be determined. However,
through this study, we have identified a critical role of
S100A7 in a negative regulation of b-catenin signaling,
and vice versa, in head and neck cancer cells. On the
basis of our results, we believe that the initial increase in
S100A7 expression in premalignant/highly differentiated
tumors may represent a natural defense mechanism by
which the cells may attempt to promote differentiation
and restrict cell migration. When increased levels of
S100A7 are lost by activated b-catenin signaling, the
cells are then relieved from S100A7’s pro-differentiating/growth inhibitory effect, thereby unleashing their
growth and invasive capacity. Therefore, despite its high
levels of expression in early stage SCCOC tumorigenesis, S100A7 appears to function as a tumor suppressor
by inhibiting tumor growth and tumor progression.
Together, our findings contribute to elucidating the
mechanistic roles of S100A7 and b-catenin signaling
during tumor development and progression of SCCOC,
and may help us to eventually develop diagnostic and
therapeutic strategies for SCCOC.
Materials and methods
Materials and Plasmids
Materials and plasmids are described in Supplementary
Materials.
Cell culture, transient DNA and siRNA transfections
Human SCCOC cell lines Tu167 and JMAR were cultured in
Dulbecco’s modified Eagle’s medium, and human mammary
adenocarcinoma cell line MDA-MB468 was cultured in RPMI
1640 medium. The spontaneously immortalized human oral
keratinocyte cell line HOK16B was maintained in keratinocyte
serum-free medium (Life Technologies, Bethesda, MD, USA)
supplemented with EGF and bovine pituitary extract. The
background information of these cells and transfection
methods are described in Supplementary Materials.
Generation of recombinant adenovirus, retrovirus and lentivirus
The detail for generation of adenovirus, retrovirus and
lentivirus for overexpression or suppression of S100A7 and
b-catenin is described in Supplementary Materials.
Cell proliferation, cell migration and colony formation assays
Cell proliferation was determined using the tetrazolium-based
(that is, MTT) assay as described in Yigitbasi et al. (2004) or
cell counting. Cell migration and colony formation assays are
described in Supplementary Materials
Luciferase reporter assay
Luciferase reporter assay is described in Supplementary
Materials.
Western blotting, IP and quantitative real-time RT-PCR
The details of western blotting, IP and quantitative real-time
RT-PCR are described in Supplementary Materials.
Orthotopic injections of oral tumors
The procedures for cell injection, tumor measurements and
subsequent analyses were conducted as described in Yigitbasi
et al. (2004) and Supplementary Materials.
Tumor specimens, IHC and immunofluorescence microscopy
The detail of tumor specimens, IHC and immunofluorescence
microscopy is described in Supplementary Materials.
Acknowledgements
This work was supported by the National Institute of Dental
and Craniofacial Research—NIH grant RO1DE01461301
and the NIH Cancer Center support grant CA16672
and CA69381. We thank Mrs Carol Johnston for the excellent
technical assistance. We thank Dr PD McCrea, Dr M-C Hung,
Dr X-W Wu, Dr E Fearon and Dr GP Nolan for providing
plasmids.
References
Adhikary S, Eilers M. (2005). Transcriptional regulation and
transformation by Myc proteins. Nat Rev Mol Cell Biol 6:
635–645.
Alowami S, Qing G, Emberley E, Snell L, Watson PH. (2003).
Psoriasin (S100A7) expression is altered during skin tumorigenesis.
BMC Dermatol 3: 1.
Bienz M, Clevers H. (2000). Linking colorectal cancer to Wnt
signaling. Cell 103: 311–320.
Blanpain C, Horsley V, Fuchs E. (2007). Epithelial stem cells: turning
over new leaves. Cell 128: 445–458.
Broome AM, Ryan D, Eckert RL. (2003). S100 protein subcellular
localization during epidermal differentiation and psoriasis.
J Histochem Cytochem 51: 675–685.
Di Nuzzo S, Sylva-Steenland RM, Koomen CW, De Rie MA, Das PK,
Bos JD et al. (2000). Exposure to UVB induces accumulation
of LFA-1+ T cells and enhanced expression of the chemokine
psoriasin in normal human skin. Photochem Photobiol 72: 374–382.
Donato R. (2001). S100: a multigenic family of calcium-modulated
proteins of the EF-hand type with intracellular and extracellular
functional roles. Int J Biochem Cell Biol 33: 637–668.
Oncogene
S100A7 inhibits SCCOC progression
G Zhou et al
3538
Donato R. (2003). Intracellular and extracellular roles of S100
proteins. Microsc Res Tech 60: 540–551.
Emberley ED, Niu Y, Curtis L, Troup S, Mandal SK, Myers JN et al.
(2005). The S100A7-c-Jun activation domain binding protein 1
pathway enhances prosurvival pathways in breast cancer. Cancer
Res 65: 5696–5702.
Emberley ED, Niu Y, Leygue E, Tomes L, Gietz RD, Murphy LC
et al. (2003a). Psoriasin interacts with Jab1 and influences breast
cancer progression. Cancer Res 63: 1954–1961.
Emberley ED, Niu Y, Njue C, Kliewer EV, Murphy LC, Watson PH.
(2003b). Psoriasin (S100A7) expression is associated with poor
outcome in estrogen receptor-negative invasive breast cancer. Clin
Cancer Res 9: 2627–2631.
Enerback C, Porter DA, Seth P, Sgroi D, Gaudet J, Weremowicz S
et al. (2002). Psoriasin expression in mammary epithelial cells in
vitro and in vivo. Cancer Res 62: 43–47.
Filipek A, Jastrzebska B, Nowotny M, Kuznicki J. (2002).
CacyBP/SIP, a calcyclin and Siah-1-interacting protein, binds
EF-hand proteins of the S100 family. J Biol Chem 277:
28848–28852.
Fukushima T, Zapata JM, Singha NC, Thomas M, Kress CL,
Krajewska M et al. (2006). Critical function for SIP, a ubiquitin E3
ligase component of the beta-catenin degradation pathway, for
thymocyte development and G1 checkpoint. Immunity 24: 29–39.
Hart M, Concordet JP, Lassot I, Albert I, del los Santos R, Durand H
et al. (1999). The F-box protein beta-TrCP associates with
phosphorylated beta-catenin and regulates its activity in the cell.
Curr Biol 9: 207–210.
He TC, Sparks AB, Rago C, Hermeking H, Zawel L, Da Costa LT
et al. (1998). Identification of c-MYC as a target of the APC
pathway. Science 281: 1509–1512.
Heizmann CW, Fritz G, Schafer BW. (2002). S100 proteins: structure,
functions and pathology. Front Biosci 7: d1356–d1368.
Hoffmann HJ, Olsen E, Etzerodt M, Madsen P, Thogersen HC,
Kruse T et al. (1994). Psoriasin binds calcium and is upregulated
by calcium to levels that resemble those observed in normal skin.
J Invest Dermatol 103: 370–375.
Kennedy RD, Gorski JJ, Quinn JE, Stewart GE, James CR, Moore S
et al. (2005). BRCA1 and c-Myc associate to transcriptionally
repress psoriasin, a DNA damage-inducible gene. Cancer Res 65:
10265–10272.
Kitagawa M, Hatakeyama S, Shirane M, Matsumoto M, Ishida N,
Hattori K et al. (1999). An F-box protein, FWD1, mediates
ubiquitin-dependent proteolysis of beta-catenin. EMBO J 18:
2401–2410.
Korinek V, Barker N, Morin PJ, van Wichen D, De Weger R,
Kinzler KW et al. (1997). Constitutive transcriptional activation by
a beta-catenin-Tcf complex in APC/ colon carcinoma. Science
275: 1784–1787.
Krop I, Marz A, Carlsson H, Li X, Bloushtain-Qimron N, Hu M et al.
(2005). A putative role for psoriasin in breast tumor progression.
Cancer Res 65: 11326–11334.
Leygue E, Snell L, Hiller T, Dotzlaw H, Hole K, Murphy LC et al.
(1996). Differential expression of psoriasin messenger RNA
between in situ and invasive human breast carcinoma. Cancer Res
56: 4606–4609.
Liu J, Stevens J, Rote CA, Yost HJ, Hu Y, Neufeld KL et al. (2001).
Siah-1 mediates a novel beta-catenin degradation pathway linking
p53 to the adenomatous polyposis coli protein. Mol Cell 7: 927–936.
Martinsson H, Yhr M, Enerback C. (2005). Expression patterns of
S100A7 (psoriasin) and S100A9 (calgranulin-B) in keratinocyte
differentiation. Exp Dermatol 14: 161–168.
Matsuzawa SI, Reed JC. (2001). Siah-1, SIP, and Ebi collaborate in a
novel pathway for beta-catenin degradation linked to p53 responses.
Mol Cell 7: 915–926.
Moog-Lutz C, Bouillet P, Regnier CH, Tomasetto C, Mattei MG,
Chenard MP et al. (1995). Comparative expression of the psoriasin
(S100A7) and S100C genes in breast carcinoma and co-localization
to human chromosome 1q21-q22. Int J Cancer 63: 297–303.
Moon RT, Kohn AD, De Ferrari GV, Kaykas A. (2004). WNT and
beta-catenin signalling: diseases and therapies. Nat Rev Genet 5:
691–701.
Morin PJ, Weeraratna AT. (2003). Wnt signaling in human cancer.
Cancer Treat Res 115: 169–187.
Patel JH, Loboda AP, Showe MK, Showe LC, McMahon SB. (2004).
Analysis of genomic targets reveals complex functions of MYC. Nat
Rev Cancer 4: 562–568.
Reischl J, Schwenke S, Beekman JM, Mrowietz U, Sturzebecher S,
Heubach JF. (2007). Increased expression of Wnt5a in psoriatic
plaques. J Invest Dermatol 127: 163–169.
Taketo MM. (2004). Shutting down Wnt signal-activated cancer. Nat
Genet 36: 320–322.
Tavakkol A, Zouboulis CC, Duell EA, Voorhees JJ. (1994). A retinoic
acid-inducible skin-specific gene (RIS-1/psoriasin): molecular cloning and analysis of gene expression in human skin in vivo and
cultured skin cells in vitro. Mol Biol Rep 20: 75–83.
van de Wetering M, Sancho E, Verweij C, De Lau W, Oving I,
Hurlstone A et al. (2002). The beta-catenin/TCF-4 complex imposes
a crypt progenitor phenotype on colorectal cancer cells. Cell 111:
241–250.
Watson PH, Leygue ER, Murphy LC. (1998). Psoriasin (S100A7).
Int J Biochem Cell Biol 30: 567–571.
Willert K, Jones KA. (2006). Wnt signaling: is the party in the nucleus?
Genes Dev 20: 1394–1404.
Yigitbasi OG, Younes MN, Doan D, Jasser SA, Schiff BA,
Bucana CD et al. (2004). Tumor cell and endothelial cell therapy
of oral cancer by dual tyrosine kinase receptor blockade. Cancer Res
64: 7977–7984.
Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc).
Oncogene