Carcinogenesis vol.32 no.3 pp.389–398, 2011
Advance Access publication November 26, 2010
YAP is a candidate oncogene for esophageal squamous cell carcinoma
Tomoki Muramatsu1,2, Issei Imoto1,3, Takeshi Matsui4,
Ken-ichi Kozaki1,5, Shigeo Haruki1,6, Marius Sudol7,
Yutaka Shimada8, Hitoshi Tsuda5,9, Tatsuyuki Kawano6
and Johji Inazawa1,2,5,?
1Department of Molecular Cytogenetics, Medical Research Institute, Tokyo
Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-
8510, Japan,2Global Center of Excellence Program for Frontier Research on
Molecular Destruction and Reconstitution of Tooth and Bone, Tokyo Medical
and Dental University, Tokyo 113-8510, Japan,3Department of Human
Genetics and Public Health, Institute of Health Biosciences, The University of
Tokushima, Graduate school, Tokushima 770-8503, Japan,4Medical Top
Track Program, Medical Research Institute and School of Biomedical
Tissue Genome Research Center, Tokyo Medical and Dental University;
Tokyo 113-8510, Japan,6Department of Surgery, Graduate School, Tokyo
Medical and DentalUniversity; Tokyo 113-8510,Japan,7Laboratoryof Signal
Transduction and Proteomic Profiling, Weis Center for Research, Geisinger
Clinic, Danville, PA 17822, USA,8Department of Surgery and Science,
Graduate School of Medicine and Pharmaceutical Sciences for Research,
University of Toyama, Toyama 930-0194, Japan and9Clinical Laboratory
Division, National Cancer Center Hospital, Tokyo 104-0045, Japan
?To whom correspondence should be addressed. Tel: þ81 3 5803 5820;
Fax: þ81 3 5803 0244;
Yes-associated protein (YAP), the nuclear effector of the Hippo
pathway, is a key regulator of organ size and a candidate human
oncogene located at chromosome 11q22. Since we previously re-
ported amplification of 11q22 region in esophageal squamous cell
carcinoma (ESCC), in this study we focused on the clinical signif-
icance and biological functions of YAP in this tumor. Frequent
overexpression of YAP protein was observed in ESCC cells includ-
ing those with a robust amplicon at position 11q22. Overexpression
of the YAP protein was frequently detected in primary tumors
of ESCC as well. Patients with YAP-overexpressing tumors had a
worse overall rate of survival than those with non-expressing tu-
outcome in the multivariate analysis. Further analyses in cells in
which YAP was either overexpressed or depleted confirmed that
cell proliferation was promoted in a YAP isoform-independent but
YAP expression level-dependent manner. YAP depletion inhibited
cell proliferation mainly in the G0–G1phase and induced an in-
crease in CDKN1A/p21 transcription but a decrease in BIRC5/sur-
vivin transcription. Our results indicate that YAP is a putative
oncogene in ESCC and it represents a potential diagnostic and
The Hippo signaling pathway regulates the balance between cell pro-
liferation and apoptosis (1). Yes-associated protein (YAP) is the mam-
malian ortholog of Drosophila Yorkie (yki), which is a negatively
regulated effector of the Hippo pathway and functions as a transcrip-
tional coactivator or corepressor in the regulation of cell growth, pro-
liferation and apoptosis (2). Given the essential role of YAP in cellular
growth, knockout of the YAP gene in mice leads to early embryonic
death (3). YAP was described to increase the ability of p73 to induce
apoptosis as a consequence of DNA damage and be important for
c-Jun-dependent apoptosis, suggesting that, under certain conditions
of stress, YAP acts as a tumor suppressor (4–9). However, a majority
of recent studies document YAP as a bona fide oncogene: YAP ampli-
fication and overexpression were observed in various human cancers
and mouse models of cancer (10–16); overexpression of YAP in non-
transformed mammary epithelial cells induced epithelial–mesenchymal
transition, suppression of apoptosis, growth factor-independent prolif-
eration and anchorage-independent growth (11); YAP cooperated with
myc oncogene to stimulate tumor growth in nude mice (12) and trans-
genic mice with liver-targeted YAP overexpression showed a dramatic
increase in liver size that eventually resulted in hepatic tumors (17,18).
Considering those findings, YAP may have a dual role in signaling,
which depends on context, stress and/or cell lineage becauseit is placed
at the crossroad of many signaling pathways (19). In the current study,
we focused on the role of YAP gene in a cancer that is prevalent in our
Esophageal carcinoma is the sixth most frequent cause of cancer-
related deathworldwide (20), and esophageal squamous cell carcinoma
(ESCC) accounts for ?90% of esophageal carcinomas diagnosed in
Asian countries including Japan. Although accumulated evidence sug-
gests that multiple genetic alterations occurring sequentially in a cell
lineage underlie the carcinogenesis of ESCC, the repertoire of genetic
alterations identified so far in ESCC cannot fully account for the path-
ogenesis. Among the genetic alterations detected in cancer, the ampli-
fication of chromosomal DNA is one of the mechanisms capable of
activating genes whose overexpression contributes to the development
and progression of human cancer, including ESCC (21). As such, the
amplification event points to candidate tumor-promoting genes.
Before the YAP gene was mapped precisely in the human genome,
in ESCC cell lines, we detected amplification at 11q22, which con-
tains YAP gene and we also identified cellular inhibitorof apoptosis 1
(cIAP1, BIRC2) within the amplicon as a possible tumor-promoting
gene (22). Since YAP is located ?100 kb centromeric to cIAP1/BIRC2
according to the human genome database (http://www.ncbi.nlm.nih.-
gov) and seems to be amplified with cIAP1 within the 11q amplicon in
ESCC (22), YAP in concert with cIAP1 could contribute to the neo-
plastic phenotype of ESCC. In the non-neoplastic esophageal mucosa,
YAP immunoreactivity, especially in the nucleus, was shown to be
positive in the proliferating basal layer of the epithelium but negative
in the terminally differentiated mature squamous epithelium toward
the surface (10). These are novel observations as, to our knowledge,
the expression status and functional significance of YAP in the tumor-
igenesis of ESCC have not been previously characterized.
Here, we demonstrate that four isoforms of YAP, including one
newly identified isoform, were frequently overexpressed in ESCC.
We also showed that patients with YAP-overexpressing tumors had
a worse overall rate of survival than those with non-expressing tu-
mors. We also report proliferation-promoting activity of YAP, at least
in part, through the inhibition of CDKN1A/p21 transcription and in-
duction of BIRC5/survivin transcription, either directly or indirectly.
Interestingly, the YAP activity was isoform independent. These find-
ing suggest YAP to be a potential target in the treatment of ESCC.
Materials and methods
Cell culture and primary tissue samples
A total of 43 ESCC cell lines were used, of which 31 belonged to the KYSE
series established from surgically resected tumors and 12 were TE series lines
provided by the Cell Resource Center for Biomedical Research, Institute of
Development, Aging and Cancer, Tohoku University (23).
All 120 primary tumor samples of ESCC had been obtained from ESCC
patients treated at Tokyo Medical and Dental University (Tokyo, Japan) be-
tween 2000 and 2005 and embedded in paraffin after 24 h of formalin fixation.
Relevant clinical and survival data were available for all patients (supplemen-
tary Tables S1 and S2 are available at Carcinogenesis Online). After approval
by the local ethics committee, a formal written consent was always obtained
Abbreviations: cIAP1, cellular inhibitor of apoptosis 1; ESCC, esophageal
squamous cell carcinoma; HCC, hepatocellular carcinoma; mRNA, messenger
RNA; RT–PCR, reverse transcription–polymerase chain reaction; siRNA,
small interfering RNA; siRNA-NC, negative control siRNA;YAP, Yes-associated
protein; Yki, yorkie.
? The Author 2010. Published by Oxford University Press. All rights reserved. For Permissions, please email: email@example.com
by guest on February 21, 2013
from patients. None of these patients underwent endoscopic mucosal resection,
palliative resection, preoperative chemotherapy or radiotherapy, and none had
synchronous or metachronous multiple cancers in other organs. Disease stage
was defined in accordance with the International Union against Cancer and
following the tumor-lymph node-metastases classification. In this series, all the
M1 tumors had distant lymph node metastases and there was no organ metas-
tasis. The median follow-up period for the surviving patients was 19 (ranging
from 1 to 103) months.
The anti-YAP (H-125),anti-cIAP1 (H-83), anti-p21 (C-19), anti-survivin (D-8)
and Lamin B (C-20) antibodies were purchased from Santa Cruz Biotechnol-
ogy (Santa Cruz, CA); anti-b-actin antibody from Sigma (St Louis, MO); anti-
Ki67 antibody from DAKO (Carpinteria, CA) and anti-phospho-YAP (Ser127)
and anti-b-tubulin antibodies from Cell Signaling Technology (Danvers, MA).
Cells were lysed in Tris buffer (50 mM, pH 7.5) containing 150 mM NaCl,
1 mM ethylenediaminetetraacetic acid, 0.5% NP-40, 10% glycerol, 100 mM
NaF, 10 mM sodium pyrophosphate, 2 mM Na2VO3and a protease inhibitor
cocktail (Roche, Tokyo, Japan), and lysates were analyzed as described else-
where (23). Nuclear and cytoplasmic proteins were extracted from cells sep-
arately using NE-PER Nuclear and Cytoplasmic Extraction Reagents (Pierce,
Cells were fixed in 10% trichloroacetic acid, permeabilized with 0.2% Triton
X-100 and treated with blocking solution (1% bovine serum albumin, in phos-
phate-buffered saline) and then incubated with the primary antibodies (YAP,
1:100 and/or Ki67, 1:200) for 1 h. The bound antibody was visualized using
a fluorescein isothiocyanate-conjugated or Cy3-conjugated secondary anti-
body (1:1000). After being mounted with 4#,6#-diamidino-2-phenylindole to
stain nuclei, the cells were observed under a fluorescence microscope (BZ-
8100; Keyence, Osaka, Japan).
embedded in paraffin, sectioned into 4 lm thick slices and subjected to immu-
nohistochemical staining of YAP protein with the avidin–biotin–peroxidase
method (24). In brief, endogenous peroxidases were quenched by incubating
the sections for 20 min in 3% H2O2. Antigen retrieval was performed by heating
the samples in 10 mM citrate buffer (pH 6.0) at 120?C for 15 min using an
autoclave. After treatment with Block Ace (Dainippon Sumitomo Pharmaceu-
tical, Osaka, Japan) for 30 min at room temperature, the sections were incubated
complex system (Vectastain Elite ABC kit; Vector Laboratories, Burlingame, CA)
was used for color development with diaminobenzidine tetrahydrochloride. The
line overexpressing YAP (KYSE170) was used as a positive control, whereas
a formalin-fixed ESCC cell line with low expression of YAP (KYSE1240) was
included as a negative control. The percentage of the total cell population that
expressed YAP was evaluated for each case at ?200 magnification. Expression of
YAP protein was graded as either positive (?30% of the nucleus or cytoplasm
showing immunopositivity; cytoplasmic or nuclear positive, respectively) or neg-
ative (,30% of the nucleus or cytoplasm showing immunopositivity or no stain-
ing; cytoplasmic or nuclear negative, respectively).
Reverse transcription–polymerase chain reaction
Single-stranded complementary DNA was generated from total RNA. The
messenger RNA (mRNA) expression pattern was analyzed by reverse tran-
scription–polymerasechain reaction(RT–PCR)usinggene-specific primersets
(supplementary Table S3 is available at Carcinogenesis Online). PCR products
were electrophoresed in 3% agarose gels, and bands were quantified using
LAS-3000 (Fujifilm, Tokyo, Japan) and Multi Gauge software (Fujifilm). Lev-
els of mRNA expression were measured with a quantitative real-time fluores-
cence detection method (ABI PRISM 7500 sequence detection System;
Applied Biosystems, Foster City, CA) using TaqMan Gene Expression Assays
(Hs00193201_m1 and Hs00231069_m1 for CDKN1A/p21 and BIRC5/survi-
vin, respectively; Applied Biosystems) according to the manufacturer’s in-
structions. Gene expression values are expressed as ratios between the genes
of interest and an internal reference gene (Hs99999903_m1 for beta-actin,
ACTB; Applied Biosystems) that provides a normalization factor for the
amount of RNA isolated from a specimen and subsequently normalized with
the value in the controls (relative expression level). Each assay was performed
in duplicate for each sample.
Expression constructs and colony formation assay
Plasmids expressing each of the YAP isoforms (pcDNA3.1-YAP-a, -b, -c and
-d) were obtained by cloning the full coding sequences for wild-type YAP-a,
-b, -c and -d, respectively, into the vector pcDNA3.1(?) (Invitrogen, St Louis,
MO). pcDNA3.1-YAP-a, -b, -c or -d or the empty vector (pcDNA3.1-mock),
as a control, was introduced into ESCC cells as described previously (23). The
expression of YAP protein in transfected cells was confirmed by western
blotting. After 3 weeks of incubation with appropriate concentrations of
G418, cells were fixed and stained with crystal violet.
Loss-of-function by small interfering RNA and cell growth analysis
Loss-of-function screening was done using small interfering RNAs (siRNAs)
purchased from Invitrogen targeting the YAP gene (HSS115942, HSS115943
or HSS115944) or the cIAP1 gene (HSS100559) and a universal negative
control (46-2001). Each siRNA (total 20 nM) was transfected into ESCC cells
using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s in-
structions. The knockdown of a target gene was confirmed by quantitative
real-time RT–PCR and western blotting. The numbers of viable cells at various
time points after transfection were assessed by a colorimetric water-soluble
tetrazolium salt assay as described elsewhere (23). The cell cycle was evalu-
ated 72 h after transfection by fluorescence-activated cell sorting as described
The clinicopathologicalvariablespertainingto the correspondingpatientswere
analyzed for statistical significance by the v2or Fisher’s exact test. For the
analysis ofsurvival, Kaplan–Meier survival curves were constructedforgroups
based on univariate predictors and differences between the groups were tested
with the log-rank test. Univariate and multivariate survival analyses were
performed using the likelihood ratio test of the stratified Cox proportional
hazards model. Differences between subgroups were tested with the Student’s
t-test. For multiple group comparisons, a one-way analysis of variance, fol-
lowed by Scheffe ´’s post-hoc test, was used. Differences were assessed with
a two-sided test and considered significant at the P , 0.05 level.
YAP protein is frequently overexpressed in human ESCC
To determine whether YAP may be involved in human ESCC, we first
performed protein expression analyses ina panel ofcell linesof ESCC
using an antibody raised against 125 amino acid long, C-terminal
fragment of YAP. We observed frequent high expression of the YAP
protein in ESCC cell lines (Figure 1A) including thosewith high-copy
amplification at 11q22 (KYSE170 and KYSE2270) (22). At least two
bands (arrowheads in Figure 1A) at ?75 kD (major) and 65 kD
(minor) were detected. Since YAP protein were overexpressed in cells
with remarkable amplification of this gene (22) as well as in cells
without amplification (Figure 1A), YAP amplification seems to be
only part of the mechanism to induce its overexpression/activation.
The YAP protein was predominantly located in the cytoplasm of
ESCC cells with the 11q22 amplification (Figure 1B and C).
Immunohistochemical analysis (Figure 1D–J) revealed that YAP
protein is frequently overexpressed in primary tumors of ESCC com-
pared with nontumorous esophageal epithelia in either the cytoplasm
or nucleus. In 120 primary ESCC tumors, positive nuclear and cyto-
plasmic YAP immunoreactivity was detected in 69 (57.5%) and 74
(61.7%) cases, respectively (supplementary Tables S1 and S2 are
available at Carcinogenesis Online). Notably, basal cell layer cells
of nontumorous epithelia also clearly expressed YAP protein, pre-
dominantly in the nucleus (Figure 1F).
Nuclear YAP protein overexpression associated with short overall
We examined the clinicopathological significance of the nuclear or
cytoplasmic overexpression of YAP in 120 primary ESCC tumors
based on the immunohistochemical staining pattern of this protein
(supplementary Tables S1 and S2 are available at Carcinogenesis
Online). No significant association between any clinicopathological
characteristics and nuclear or cytoplasmic YAP immunoreactivity was
observed in our set of primary tumors. However, Kaplan–Meier sur-
vival estimates (Figure 1K) showed that nuclear immunoreactivity of
YAP in tumor cells was significantly associated with a worse overall
survival in all cases (P 5 0.0213, log-rank test). In the Cox propor-
tional hazard regression model (Table I), univariate analyses demon-
strated that YAP protein expression, gender and pN category and stage
T.Muramatsu et al.
by guest on February 21, 2013
in the tumor-lymph node-metastases classification were significantly
associated with overall survival. In the multivariate analysis using
a stepwise Cox regression procedure, nuclear YAP expression status,
gender and tumor stage according to the tumor-lymph node-metasta-
ses classification were identified as independently selected predictive
factors for overall survival in both forward and backward procedures
Fig. 1. Expression of YAP in cell lines and primary tumors of ESCC. (A) Expression of YAP protein detected by western blotting in a panel of ESCC cell lines
with an immortalized esophageal epithelial cell line Het-1A [American Type Culture Collection (Manassas, VA)]. Asterisks indicate cell lines with remarkable
11q22 amplification (22). Note that at least two bands (arrowheads) at ?75 kD (major) and 65 kD (minor) were detected. The bands were quantified by
densitometry using an image analyzer (LAS3000; Fujifilm), and results of YAP protein expression level normalized with b-actin are shown with values relative to
that for the control Het-1A cell line. Red, ?2-fold increase in YAP protein expression compared with the Het-1A cell line. (B) Immunoblotting using separately
extracted nuclear and cytoplasmic proteins and whole cell lysate of KYSE170, both YAP and phospho-YAP (Ser127). (C) Immunofluorescent cytochemical
staining of endogenous YAP using anti-YAP antibody (Red). (D–J) Representative results of immunohistochemical staining of YAP in KYSE1240 cells [negative
control, (D)], KYSE170 cells [positive control, (E)], normal esophageal mucosa (F), ESCC with negative staining (G), ESCC with positive staining in the
cytoplasm (H), ESCC with positive staining in the nucleus (I) and ESCC with positive staining in both the nucleus and cytoplasm (J). Bars, 100 or 200 lm.
(K) Kaplan–Meier curves for overall survival rates of patients with primary ESCC at all stages according to the nuclear expression of YAP. Nuclear YAP
immunoreactivity of tumor cells was significantly associated with a worse overall survival at all stages (P 5 0.0213, log-rank test).
YAP oncogene in esophageal cancer
by guest on February 21, 2013
(P 5 0.0420, 0.0452 and 0.002, respectively). In contrast to the nu-
clear YAP immunoreactivity association with survival, no significant
association was observed between cytoplasmic immunoreactivity of
YAP in tumor cells and overall survival (supplementary Figure S1 is
available at Carcinogenesis Online).
These observations are consistent with the prevailing notion that
proliferative or oncogenic signaling of the Hippo pathway requires
nuclear localization of YAP (18).
Four major isoforms of YAP include two novel types
Since several forms of YAP protein were observed in ESCC cells
(Figure 1A and data not shown) and some isoforms of human YAP
were previously reported (24,25), we decided to clone the full-length
coding sequence for each of the YAP isoforms using RT–PCR. Con-
sequently, we identified four variants (Figure 2A and B), designated
YAP-a, -b, -c and -d in order of increasing size. YAP-a and -d have
been reported as isoforms 1 and 2, respectively, in the human genome
database (http://www.ncbi.nlm.nih.gov/), and YAP-d is the same as
YAP2L reported previously (24,25). YAP-b and YAP-c (GenBank
accession number, AB567720 and AB567721, respectively) are novel
isoforms. YAP-c and -d contain two WW domains, whereas YAP-a
and -b contain one. YAP-b and -d contain a short insert sequence (16
amino acids) compared with YAP-a and -c (Figure 2B). To determine
the relative transcription levels of the four YAP isoforms, we designed
primer sets (Figure 2B; supplementary Table S3 is available at Car-
cinogenesis Online) for RT–PCR. All four isoforms were detectable
in normal esophagus and the ESCC cell line with the 11q22 amplifi-
cation (KYSE170), although YAP-c was predominantly expressed in
the normal esophagus, whereas YAP-d was predominantly expressed
in KYSE170 cells (Figure 2C). The mRNA expression pattern of the
four isoforms of YAP was similar among all ESCC cell lines exam-
ined regardless of the 11q22 amplification (Figure 2D). Since we did
not detect smaller protein products of previously identified two puta-
tive, short isoforms of human YAP (GenBank accession number,
BAG62143 and BAG65295), which miss in part, or completely, the
WW domain region, in ESCC cell lines and tissues (data not shown),
we did not follow the analysis of these two isoforms extensively.
Proliferation-promoting effect of YAP overexpression in ESCC cells
Increased expression was observed for all four major isoforms in
ESCC cell lines, and the pattern of distribution was similar among
cell lines examined, making it unclear whether all the YAP isoforms
or some specific isoforms play a role in the tumorigenesis of ESCC
through overexpression. To assess the effects of YAP on cell prolif-
eration especially in ESCC cells, we conducted colony formation
assays using expression construct encoding full-length sequence of
each YAP isoform in an ESCC cell line with relatively low endogenous
expression of this gene (KYSE1240; Figures 1A and 2D). Transient
transfection of the expression construct for each isoform produced
the corresponding protein, which localized predominantly in the cyto-
(Figure 3A and B). YAP-expressing constructs produced consistently
more colonies than did the empty plasmid, regardless of the isoform
(Figure 3C). In addition, no difference was observed in the occupancy
of the stained area among the YAP isoforms (Figure 3D), suggesting
that each isoform promotes cell proliferation in ESCC cells indepen-
dently of the number of WW domains and the existence of short insert
Suppression of cell growth by downregulation of YAP expression
in ESCC cells
We next performed a cell growth assay using YAP-specific siRNA to
investigate whether the knockdown of YAP expression would sup-
press the proliferation of KYSE170 cells showing amplification/over-
expression of YAP. Since each YAP isoform seems to show similar
growth-promoting activity in ESCC cells, we used YAP-specific
siRNA, which can knockdown all isoforms of YAP. In KYSE170
cells, the expression of all the isoforms was efficiently knocked
down 24–72 h after the transient introduction of YAP-specific siRNAs
(siRNA-YAP#1, #2 and #3) compared with a negative control siRNA
(siRNA-NC, Figure 4A). The proliferation of KYSE170 cells was
decreased after the knockdown of endogenous YAP expression com-
pared with the control (Figure 4B). On the other hand, knockdown
of YAP had a minimal effect on proliferation in KYSE1240 cells
expressing relatively low levels of this gene (Figure 4Aand B),
Table I. Coxproportional hazard regression analysis for overall survival
Hazard ratio (95% confidence interval)P-value
Male versus female
?60 versus ,60
Poor versus well-moderate
1–3 versus 0
1–3 versus 0
pT2–4 versus pT1
pN1 versus pN0
pM1 versus pM0
III þ IV versus I þ II
Nuclear YAP expressionb
Positive versus negative
1.385 (0.819–2.342) 0.2238—
Note. Statistically significant values are in boldface type.
aForward- and backward-stepwise analyses were used for mutivariate analysis.
bYAP expression was evaluated by immunohistochemical analysis as described in Materials and methods.
?P-values are from two-sided tests and were statistically significant at ,0.05.
T.Muramatsu et al.
by guest on February 21, 2013
suggesting that the growth-promoting effect of YAP in ESCC cells
depends on its expression level. In order to unequivocally confirm the
specificity of siRNA for YAP and assess the difference in the growth-
promoting effect of each YAP isoform, we performed YAP-siRNA
rescue experiments using cells with siRNA-induced YAP-depletion
(supplementary Figure S2 is available at Carcinogenesis Online). We
stably introduced individual expression constructs producing each of
the four isoforms of YAP, which are refractory to siRNA-YAP#2, into
KYSE170 cells and then depleted the endogenous YAP by transiently
introducing siRNA-YAP#2. The expression of each YAP isoform was
detected, and each isoform partially restored YAP depletion inhibited
cell proliferation possibly due to lower expression level of each iso-
form in stable transfectants compared with their parental cells. How-
ever, no significant differences were observed among YAP isoforms.
This result is consistent with that observed in colony formation assays
using transiently introduced constructs producing each YAP isoform.
Since our previous studies and the current study demonstrated that
YAP and cIAP1, two possible oncogenes located at 11q22, were co-
overexpressed in some of the ESCC cell lines, including KYSE170
and KYSE2270havingthe 11q22amplification, we tested whether the
two genes could cooperate to promote the proliferation of ESCC cells.
We accomplished that test by using a double-knockdown with siRNA
specific for each gene (supplementary Figure S3 is available at Car-
cinogenesis Online). In KYSE170 cells, the transient introduction of
10 nM of YAP- or cIAP1-specific siRNA alone with 10 nM of an
siRNA-NC inhibited cell proliferation compared withthat of20 nM of
siRNA-NC. In addition, the introduction of both YAP- and cIAP1-
specific siRNAs (10 nM each) induced a larger decrease in cell pro-
liferation compared with that of each siRNA (10 nM with 10 nM of
a siRNA-NC), suggesting YAP and cIAP1 cooperatively promote
tumorigenesis in ESCC.
Mechanisms of YAP-induced growth-promoting activity in ESCC cells
To gain further insight into the potential mechanism of YAP as an
oncogene in esophageal carcinogenesis, we performed fluorescence-ac-
tivated cell sorting, fluorescence immunocytochemical and expression
analyses in ESCC cells treated with YAP-specific siRNA. Consistent
with the results of the cell proliferation experiment, siRNA-YAP treat-
ment resulted in an accumulation of cells in the G0–G1phase com-
pared with siRNA-NC-treated counterparts in the KYSE170 cell line
in a fluorescence-activated cell sorting analysis, whereas a minimal
effect was observed in siRNA-YAP-treated KYSE1240 cells (Figure
4C). In addition, we stained siRNA-YAP-treated KYSE170 cells
with anti-YAP antibody and a cell proliferation marker, anti-Ki67
antibody, and found that Ki67 expression status was clearly correlated
with YAP expression status: cells which retained the expression of
YAP were positive for Ki67, whereas cells in which YAP was effec-
tively knocked down tested negative (Figure 4D). Since CDKN1A/p21
Fig. 2. YAP has four isoforms. (A and B) Schematic structure of the YAP-a and YAP-d genes (A) and proteins (B), together with the novel isoforms YAP-b and
YAP-c. P-rich, proline-rich domain; TEAD-B, TEAD-binding region; WW, WW domain; SB, SH3-binding motif; IS, uncharacterized insertion sequence; CC,
a coiled-coildomain;TAD,transcriptionactivationdomainand PDZ-B,PDZ-bindingmotif.(C) mRNA expressionofYAP isoforms(arrows) detectedby RT–PCR
using the primer set producing four products, which can be recognized by size (see Figure 2B). (D) Expression pattern of YAP isoforms detected by RT–PCR
(arrows) using the same primer set as in Figure 2D in a panel of ESCC cell lines. Asterisks indicate cell lines with remarkable 11q22 amplification (22).
YAP oncogene in esophageal cancer
by guest on February 21, 2013
and BIRC5/survivin were reported as possible cell cycle-associated
targets for transcriptional suppression and activation by YAP, respec-
tively (13,26–28), we investigated CDKN1A/p21 and BIRC5/survivin
expression after treatment with siRNA for YAP. After the downregu-
lation of YAP expression, CDKN1A/p21 and BIRC5/survivin expres-
sion was remarkably induced and suppressed, respectively, at both the
mRNA and protein levels in KYSE170 cells (Figure 5A and B) but
only slightly affected in KYSE1240 cells (supplementary Figure S4 is
available at Carcinogenesis Online). The results confirm YAP to in-
duce and reduce transcription of CDKN1A/p21 and BIRC5/survivin in
a YAP expression level-dependent manner, resulting in G0–G1arrest,
although it remains unclear whether these are direct or indirect effects
The Hippo signaling pathway is gaining recognition as an important
player in both organ size control and tumorigenesis, which are phys-
iological and pathological processes that share common cellular sig-
naling mechanisms (29). Since the disruption of any factor in this
pathway can lead to tumorigenesis, it is not surprising that YAP func-
tions as an oncogene and the major downstream effector of the Hippo
pathway. As it is the case for ESCC, the YAP locus is also amplified in
other human cancers, including intracranial ependymomas, oral squa-
mous cell carcinomas and medulloblastomas (16,30–32). Two reports
identified YAP as a driving oncogene of the 11q22 amplicon in human
hepatocellular carcinoma (HCC) and breast cancer (11,12). Elevated
Fig. 3. Colony formation assays using the KYSE1240 cell line showing relatively low YAP expression. (A) KYSE1240 cells were transiently transfected with
expression constructs containing empty vector [pcDNA3.1(?)-mock] or each isoform of YAP [pcDNA3.1(?)-YAP-a, -b, -c or -d]. YAP expression was detected
by immunoblotting using 10 lg of protein extract and anti-YAP antibody. (B) Immunofluorescence staining of each exogenously expressed YAP isoform in
KYSE1240 cells using anti-YAP antibody (Red). All isoforms were predominantly detected in the cytoplasm. (C) Representative image of a colony formation
assay. Cells were transiently transfected with each construct and selected with appropriate concentrations of G418 for 3 weeks. The drug-resistant colonies formed
by the YAP-transfected cells were more numerous than those formed by empty vector-transfected cells. (D) Quantitative analysis of occupancy of the stained area.
The stained area was calculated by densitometry using an image analyzer (LAS3000; Fujifilm) and Black area calculation STD software (Gougasha, Kyoto,
Japan). Columns, means of three separate experiments, each performed in triplicate; bars, standard deviation (histogram). Differences among multiple
comparisons were analyzed by one-way analysis of variance with subsequent Scheffe ´’s tests: asterisks represents P , 0.05 versus empty vector (Mock).
T.Muramatsu et al.
by guest on February 21, 2013
Fig. 4. YAP overexpression-dependent proliferation of KYSE170 cells. (A) knockdown of YAP protein expression by YAP-specific siRNA as confirmed by
western blotting in KYSE170 (left) and KYSE1240 cells (right). The protein expression of all endogenous YAP isoforms was downregulated by either siRNA-
YAP#1, #2 or #3 in KYSE170 cells. Since siRNA-YAP#2 had the greatest effect, it was in further knockdown experiments. (B) The number of viable cells 24–72 h
after transfection of YAP-specific siRNAs or an siRNA-NC was determined by the water-soluble tetrazolium salt assay at the indicated times. Results are shown
with means ± standard deviations (bars) for quadruplicate experiments. Differences among multiple comparisons were analyzed by one-way analysis of variance
with subsequent Scheffe ´’s tests: asterisks represent P , 0.05 versus siRNA-NC transfectants. (C) Representative results of the population in each phase of the cell
cells. Note that the transfection of YAP-specific siRNA mainly induced the accumulation of G0/G1phase cells and a decrease in S and G2/M phase cells in the
KYSE170 cell line. In KYSE1240 cells, a much smaller alteration was observed in the population in each phase compared with KYSE170 cells. Results of
quantitative analysis are shown with means ± standard deviations (bars) for quadruplicate experiments. Differences among multiple comparisons were analyzed by
one-way analysis of variance with subsequent Scheffe ´’s tests (KYSE170 cells) or Student’s t-test (KYSE1240 cells): asterisks represents P , 0.05 versus siRNA-NC
transfectants. (D) Expression status of YAP and Ki67 assessed with immunofluorescence staining in KYSE170 cells treated with siRNA-NC or siRNA-YAP#2 for 72
h. Ki67 immunopositivity was clearly correlated with YAP expression: cells which retained the expression of treatment with YAP even after YAP-specific siRNA
showed positive Ki67 immunoreactivity (arrows), whereas cells that effectively lost the YAP protein showed negative Ki67 immunoreactivity (arrowheads).
YAP oncogene in esophageal cancer
by guest on February 21, 2013
YAP expression and nuclear localization have been observed in mul-
tiple types of cancer, including HCC, colon cancer, ovarian cancer,
lung cancer and prostate cancer (12,13,14,18,33,34). Only for HCC
patients, however, YAP was determined to be an independent prog-
nostic marker for overall survival and disease-free survival (33). In
this study, we demonstrated that overexpression and nuclear localiza-
tion of YAP in ESCC tissues was associated with a poor prognosis for
patients with this disease. In addition, we revealed that YAP’s over-
expression in ESCC cells promoted cell proliferation, whereas its
depletion impaired cell viability through the transcriptional regulation
Fig. 5. YAP regulates transcription of CDKN1A/p21 and BIRC5/survivin. (A) Relative mRNA levels of CDKN1A/p21 or BIRC5/survivin 24–72 h after
transfection of YAP-specific siRNA (siRNA-YAP#1, #2 or #3) or an siRNA-NC in KYSE170 cells, determined by real-time RT–PCR. Results are shown with
means ± standard deviations (bars) for triplicate experiments relative to thevalue for KYSE170 cells treated with siRNA-NC for 24 h. Although a different pattern
was observed more or less in both genes among siRNAs specific for YAP, knockdown of YAP expression predominantly induced CDKN1A/p21 mRNA expression
and inhibited BIRC5/survivin mRNA expression in KYSE170 cells. (B) Representative results of CDKN1A/p21 and BIRC5/survivin protein expression in
KYSE170 cells 24–72 h after treatment with YAP-specific siRNA (siRNA-YAP#1, #2 and #3) or an siRNA-NC by western blotting. Efficient knockdown of YAP
was observed in cells 24–72 h after the transfection of each YAP-specific siRNA compared with the siRNA-NC-transfected counterparts. Note that an increase in
CDKN1A/p21 protein and a decrease in BIRC5/survivin protein were observed in cells transfected with YAP-specific siRNA (siRNA-YAP#1, #2 and #3), whereas
almost no change was observed in the expression of either protein in the siRNA-NC-transfected counterparts. (C) Hypothetical model of the overexpression/
activation of YAP in ESCC cells. Overexpressed/activated YAP possibly binds to transcription factors (TFs, X or Y), cooperatively represses or enhances
transcription of target genes, such as CDKN1A/p21 or BIRC5/survivin, respectively, and promote cell proliferation although it remains unknown which
transcription factors work cooperatively and whether those candidate targets are regulated directly or indirectly by YAP-mediated transcriptional regulation. The
dotted line indicates the repression of BIRC5/survivin expression by CDKN1A/p21 because CDKN1A/21 was reported to mediate negative regulation of gene
expression including BIRC5/survivin expression by p53 (28).
T.Muramatsu et al.
by guest on February 21, 2013
of putative target genes. At present, the available data are still frag-
mentary and insufficient to delineate the tissue specificity and fre-
quency of mutations/alterations affecting the components of the
Hippo pathway, including YAP, in human cancers. A previous report
(33) and the present study indicate that dysregulation of YAP plays
a role in tumorigenesis and may be useful as an independent prog-
nostic marker, at least in HCC and ESCC. Although it remains unclear
whether this is due to a tissue-specific function of the Hippo pathway
or simply due to a lack of studies of this pathway in other tissues.
In this study, we identified two novel isoforms of YAP, which have
not been described in previous reports, although no functional differ-
ences was found among the YAP isoforms in cell growth-promoting
activity assay. YAP has an N-terminal proline-rich domain, a TEAD-
binding region, one (YAP-a and -b) or two (YAP-c and -d) WW do-
mains, a Src homology domain 3- (SH3-) binding motif, an insertion
sequence of unknown function (YAP-b and -d), a coiled-coil domain,
a transcription activation domain and a C-terminal PDZ-binding motif.
Our results suggest that neither the number of WW domains nor the
uncharacterized insertion sequence affects the growth-promoting activ-
ity of the YAP protein. Although it is still not clear how cells use the
YAP isoforms differently (35), normal esophagus and ESCC cells pre-
dominantly express YAP-c and -d unrelated to the YAP level. Among
these two isoforms, interestingly, the normal esophagus mainly ex-
presses YAP-c, whereas ESCC cells mostly express YAP-d. Therefore,
it is possible that unknown functions of YAP other than in cell prolifer-
and/or context-specific manner through transcriptional regulation.
Knockdown of the overexpressed YAP inhibits the proliferation of
ESCC cells mainly by inducing G0–G1arrest, not by inducing apo-
ptosis, in an expression-dependent manner, suggesting that ESCC
cells overexpressing YAP follow a highly activated YAP-mediated
pathway of proliferation. Since CDKN1A/p21 transcription was in-
duced by YAP’s knockdown and YAP is a transcription cofactor, which
itselfhas no DNA-binding activity, YAPmay directlyinhibitCDKN1A/
p21 transcription as a transcription corepressor regulating specific tran-
scription factors and resulting in the promotion of cell proliferation
(Figure 5C). One candidate for such transcription factors is RUNX2,
which was reported recently (27). In a preliminary experiment using
a nuclear extract from KYSE170 cells, indeed, endogenous RUNX2
was co-immunoprecipitated with YAP (supplementary Figure S5 is
available at Carcinogenesis Online). However, it is unclear whether
RUNX2 is a critical transcription factor for YAP-induced CDKN1A/
p21 suppression in the esophagus because the expression level of
RUNX2isrelatively low inthe esophagusand Vitoloet al.(27) showed
that YAP overexpression promoted cell proliferation but prevented
RUNX2-mediated repression of CDKN1A/p21 transcription. TEAD
family (TEAD1–4), four highly homologous proteins sharing a con-
served DNA-binding TEA domain in humans, has been demonstrated
to be important for the growth-promoting function of YAP (2,36).
Knockdown of TEAD aborts expression of the majority of YAP-in-
ducible genes and largely attenuates YAP-induced overgrowth, epithe-
lial–mesenchymal transition and oncogenic transformation (2).
Furthermore, the phenotype of TEAD1/TEAD2 double-knockout mice
resembles that of YAP-knockout mice (37), suggesting the TEAD pro-
teins to be a key downstream transcription factors mediating YAP’s
cellular function. However, in Drosophila, yki mutant cells have more
severe growth defects than scalloped, a Drosophila TEAD homolog,
Yki-S97A elicits a reduced but still obvious overgrowth in Drosophila
eyes and wings (2). Consistently, the TEAD-binding-defective YAP-
S94A mutant can still induce an expression of a fraction of YAP-reg-
ulated genes (2), indicating that besides TEAD, additional transcription
factors may be used by YAP/yki to stimulate cell and tissue growth.
Since all four functionally redundant members of the TEAD family are
ubiquitously expressed in ESCC cells (data not shown), it is difficult to
knockdown all TEAD proteins to assess whether the inactivation of
TEAD can abort expression of the majority of YAP-inducible genes
and largely attenuate YAP-induced growth promotion in ESCC cells.
Further analysis will be needed to screen all possible transcription
factors, which activate or repress transcription of target genes, working
with YAP in ESCC cells.
Dong et al. (13) used a microarray to identify YAP-induced gene
expression in murine livers and demonstrated that YAP induced the
transcription of many genes, which are normally associated with he-
patocyte proliferation, such as Ki67, c-myc, SOX4, H19 and AFP,
and several negative regulators of apoptosis, such as the IAP family
members BIRC5/survivin and BIRC2/cIAP1. Our expression analy-
ses using real-time RT–PCR in ESCC cells with the amplification/
overexpression of YAP (KYSE170) demonstrated that knockdown of
YAP reduced survivin transcription but did not affect cIAP1 expres-
sion (supplementary Figure S4A is available at Carcinogenesis On-
line), suggesting that survivin but not cIAP1 is a possible direct target
mediating YAP function at least in ESCC cells, possibly due to cel-
lular context. Our double-knockdown experiments in ESCC cells
(supplementary Figure S4 is available at Carcinogenesis Online)
and a previous study with in mouse models (12) demonstrate that
YAP and cIAP1 cooperatively contribute to tumorigenesis and pro-
gression possibly by mediating different pathways. Although diap1,
a Drosophila homologue of cIAP1, is known to be a direct target for
yki and contributes to tissue overgrowth, our results in ESCC cells
indicate that cIAP1 is co-overexpressed with YAP through co-ampli-
fication within the same amplicon and not through direct targeting of
YAP at least in esophageal carcinogenesis.
Supplementary Figures S1–S5 and Tables S1–S3 can be found at
Grants-in-aid for Scientific Research on Innovative Areas (4201),
Global Center of Excellence Program; International Research Center
for Molecular Science in Tooth and Bone Diseases and Integrative
project of advanced genome research and nanoscience for hard tissue
diseases from the Ministry of 520 Education, Culture, Sports, Science,
and Technology, Japan; a Health and Labor Sciences Research Grant
by the Ministry of Health, Labor and Welfare, Japan; a grant from
Core Research for Evolutional Science and Technology of Japan Sci-
ence and Technology Corporation; a grant from the New Energy and
Industrial Technology Development Organization. PA Breast Cancer
Coalition and Geisinger Clinic to M.S.
We thank Ayako Takahashi, Rumi Mori and Kathy Masker for technical as-
Conflict of Interest Statement: None declared.
1.Harvey,K. et al. (2007) The Salvador-Warts-Hippo pathway: an emerging
tumor-suppressor network. Nat. Rev. Cancer, 7, 182–191.
2.Zhao,B. et al. (2008) TEAD mediates YAP-dependent gene induction and
growth control. Genes Dev., 22, 1962–1971.
3.Morin-Kensicki,E.M. et al. (2006) Defects in yolk sac vasculogenesis,
chorioallantoic fusion, and embryonicaxis elongation in micewith targeted
disruption of Yap65. Mol. Cell. Biol., 26, 77–87.
4.Strano,S.et al. (2001) Physical interaction with Yes-associated protein en-
hances p73 transcriptional activity. J. Biol. Chem., 276, 15164–15173.
5.Matallanas,D. et al. (2007) RASSF1A elicits apoptosis through an MST2
pathway directing proapoptotic transcription by the p73 tumor suppressor
protein. Mol. Cell, 27, 962–975.
6.Danovi,S.A. et al. (2008) Yes-associated protein (YAP) is a critical medi-
ator of c-Jun-dependent apoptosis. Cell Death Differ., 15, 217–219.
7.Lapi,E. et al. (2008) PML, YAP, and p73 are components of a proapoptotic
autoregulatory feedback loop. Mol. Cell, 32, 803–814.
8.Oka,T. et al. (2008) Mst2 and Lats kinases regulate apoptotic function of
Yes kinase-associated protein (YAP). J. Biol. Chem., 283, 27534–27546.
YAP oncogene in esophageal cancer
by guest on February 21, 2013
9.Yuan,M. et al. (2008) Yes-associated protein (YAP) functions as a tumor Download full-text
suppressor in breast. Cell Death Differ., 15, 1752–1759.
10.Lam-Himlin,D. M. et al. (2006) The hippo pathway in human upper gas-
trointestinal dysplasia and carcinoma: a novel oncogenic pathway. Int. J.
Gastrointest. Cancer, 37, 103–109.
11.Overholtzer,M. et al. (2006) Transforming properties of YAP, a candidate
oncogene on the chromosome 11q22 amplicon. Proc. Natl Acad. Sci. USA,
12.Zender,L. et al. (2006) Identification and validation of oncogenes in liver
cancer using an integrative oncogenomic approach. Cell, 125, 1253–1267.
13.Dong,J. et al. (2007) Elucidation of a universal size-control mechanism in
Drosophila and mammals. Cell, 130, 1120–1133.
14.Steinhardt,A.A. et al. (2008) Expression of Yes-associated protein in com-
mon solid tumors. Hum. Pathol., 39, 1582–1589.
15.Da,C.L. et al. (2009) Significance and relationship between Yes-associated
protein and survivin expression in gastric carcinoma and precancerous
lesions. World J. Gastroenterol., 15, 4055–4061.
16.Fernandez-L,A. et al. (2009) YAP1 is amplified and up-regulated in hedge-
hog-associated medulloblastomas and mediates Sonic hedgehog-driven
neural precursor proliferation. Genes Dev., 23, 2729–2741.
17.Camargo,F.D. et al. (2007) YAP1 increases organ size and expands un-
differentiated progenitor cells. Curr. Biol., 17, 2054–2060.
18.Zhao,B. et al. (2007) Inactivation of YAP oncoprotein by the Hippo path-
way is involved in cell contact inhibition and tissue growth control. Genes
Dev., 21, 2747–2761.
19.Bertini,E. et al. (2009) YAP at the crossroad between transformation and
tumor suppression. Cell Cycle, 8, 49–57.
20.Parkin,D.M. et al. (2005) Global cancer Statistics, 2002. CA Cancer J.
Clin., 55, 74–108.
21.Lam,A.K. (2000) Molecular biology of esophageal squamous cell carci-
noma. Crit. Rev. Oncol. Hematol., 33, 71–90.
22.Imoto,I. et al. (2001) Identification of cIAP1 as a candidate target gene
within an amplicon at 11q22 in esophageal squamous cell carcinomas.
Cancer Res., 61, 6629–6634.
23.Komatsu,S. et al. (2009) Overexpression of SMYD2 relates to tumor cell
proliferation and malignant outcome of esophageal squamous cell carci-
noma. Carcinogenesis, 30, 1139–1146.
24.Naoi,Y. et al. (2008) Connexin26 expression is associated with aggressive
phenotype in human papillary and follicular thyroid cancers. Cancer Lett.,
25.Sudol,M. et al. (1995) Characterization of the mammalian YAP (Yes-
associated Protein) gene and Its role in defining a novel protein module,
the WW Domain. J. Biol. Chem., 270, 14733–14741.
26.Komuro,A. et al. (2003) WW domain-containing protein YAP associates
with ErbB-4 and acts as a co-transcriptional activator for the carboxy-
terminal fragment of ErbB4 that translocates to the nucleus. J. Biol. Chem.,
27.Vitolo,M.I. et al. (2007) The RUNX2 transcription factor cooperates with
the YES-associated protein, YAP65, to promote cell transformation. Can-
cer Biol. Ther., 6, 856–863.
28.Lo ¨hr,K. et al. (2003) p21/CDKN1A mediates negative regulation of tran-
scription by p53. J. Biol. Chem., 278, 32507–32516.
29.Zhao,B. et al. (2010) The Hippo-YAP pathway in organ size control and
tumorigenesis: an updated version. Genes Dev., 24, 862–874.
30.Baldwin,C. et al. (2005) Multiple microalterations detected at high fre-
quency in oral cancer. Cancer Res., 65, 7561–7567.
31.Snijders,A.M. et al. (2005) Rare amplicons implicate frequent deregulation
of cell fate specification pathways in oral squamous cell carcinoma. Onco-
gene, 24, 4232–4242.
32.Modena,P. et al. (2006) Identification of tumor-specific molecular signa-
tures in intracranial ependymoma and association with clinical character-
istics. J. Clin. Oncol., 24, 5223–5233.
33.Xu,M.Z. et al. (2009) Yes-associated protein is an independent prognostic
marker in hepatocellular carcinoma. Cancer, 115, 4576–4585.
34.Wang,Y. et al. (2010) Overexpression of yes-associated protein contributes
to progression and poor prognosis of non-small-cell lung cancer. Cancer
Sci., 101, 1279–1285.
35.Sudol,M. et al. (2010) Modularity in the Hippo signaling pathway. Trends
Biochem. Sci., 35, 627–633.
36.Ota,M. et al. (2008) Mammalian Tead proteins regulate cell proliferation
and contact inhibition as transcriptional mediators of Hippo signaling.
Development, 135, 4059–4069.
37.Sawada,A. et al. (2008) Redundant roles of Tead1 and Tead2 in notochord
development and the regulation of cell proliferation and survival. Mol. Cell.
Biol., 28, 3177–3189.
38.Huang,J. et al. (2005) The Hippo signaling pathway coordinately regulates
cell proliferation and apoptosis by inactivating Yorkie, the Drosophila ho-
molog of YAP. Cell, 122, 421–434.
Received August 3, 2010; revised October 4, 2010; accepted October 19, 2010
T.Muramatsu et al.
by guest on February 21, 2013