Overholtzer M, Zhang J, Smolen GA, Muir B, Li W, Sgroi DC et al.. Transforming properties of YAP, a candidate oncogene on the chromosome 11q22 amplicon. Proc Natl Acad Sci USA 103: 12405-12410
In a screen for gene copy-number changes in mouse mammary tumors, we identified a tumor with a small 350-kb amplicon from a region that is syntenic to a much larger locus amplified in human cancers at chromosome 11q22. The mouse amplicon contains only one known gene, Yap, encoding the mammalian ortholog of Drosophila Yorkie (Yki), a downstream effector of the Hippo(Hpo)-Salvador(Sav)-Warts(Wts) signaling cascade, recently identified in flies as a critical regulator of cellular proliferation and apoptosis. In nontransformed mammary epithelial cells, overexpression of human YAP induces epithelial-to-mesenchymal transition, suppression of apoptosis, growth factor-independent proliferation, and anchorage-independent growth in soft agar. Together, these observations point to a potential oncogenic role for YAP in 11q22-amplified human cancers, and they suggest that this highly conserved signaling pathway identified in Drosophila regulates both cellular proliferation and apoptosis in mammalian epithelial cells.
Transforming properties of
, a candidate
oncogene on the chromosome 11q22 amplicon
, Jianmin Zhang
, Gromoslaw A. Smolen
, Beth Muir
, Wenmei Li
, Dennis C. Sgroi
, Joan S. Brugge*
, and Daniel A. Haber
*Department of Cell Biology, Harvard Medical School, Boston, MA 02115;
Massachusetts General Hospital Cancer Center and
Department of Pathology,
Massachusetts General Hospital Molecular Pathology Research Unit, Harvard Medical School, Charlestown, MA 02129; and
National Institute of Diabetes
and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD 20892
Contributed by Joan S. Brugge, July 4, 2006
In a screen for gene copy-number changes in mouse mammary
tumors, we identiﬁed a tumor with a small 350-kb amplicon from
a region that is syntenic to a much larger locus ampliﬁed in human
cancers at chromosome 11q22. The mouse amplicon contains only
one known gene, Yap, encoding the mammalian ortholog of
Drosophila Yorkie (Yki), a downstream effector of the Hippo(Hpo)–
Salvador(Sav)–Warts(Wts) signaling cascade, recently identiﬁed in
ﬂies as a critical regulator of cellular proliferation and apoptosis. In
nontransformed mammary epithelial cells, overexpression of hu-
man YAP induces epithelial-to-mesenchymal transition, suppres-
sion of apoptosis, growth factor-independent proliferation, and
anchorage-independent growth in soft agar. Together, these ob-
servations point to a potential oncogenic role for YAP in 11q22-
ampliﬁed human cancers, and they suggest that this highly con-
served signaling pathway identiﬁed in Drosophila regulates both
cellular proliferation and apoptosis in mammalian epithelial cells.
breast 兩 mammary 兩 transformation 兩 Yorkie
enomewide analysis of tumors for gene c opy gains and
losses by using array comparative genomic hybridization
(array CGH) enables a detailed characterization of loci impli-
cated in tumorigenesis (1). Whereas human cancers frequently
show extensive chromosomal instability, mouse tumor models
may provide a more st able baseline from which to dissect
essential tumor-related alterations. This approach may be par-
ticularly powerful when used to search for somatically acquired
genetic lesions in the background of Brca1兾Trp-53 inactivation,
a genotype associated with somatic oncogene amplification. We
have recently shown that as many as 73% of mouse Brca1兾Trp-
53-driven mammary tumors have amplification of the gene
enc oding the Met protein, pointing to gross overex pression of
this growth factor receptor as a common secondary event in
tumors with this genetic background (2). In analyzing these
mammary tumors, we also observed a tumor w ith a selective
amplification of a small region of mouse chromosome 9, syntenic
with the 11q22 amplic on commonly observed in human cancers
Amplification of 11q22 is evident in glioblastomas; oral squa-
mous-cell carcinomas; and in cancers of the pancreas, lung,
ovary, and cervix (3–11). The human amplicon tends to be large
[0.7–2.6 megabases (Mb)], including a cluster of matrix metal-
loproteinase (MMP) genes, two members of the BIRC family
(BIRC2 and BIRC3, also k nown as the cIAP family), and YAP
(3–5, 8, 10). Most analyses of this amplicon have focused on the
role of BIRC (cIAP) proteins, whose antiapoptotic functions are
well described (12). The possible contribution of the YAP gene
in driving this amplicon has not been explored.
The YAP protein was in itially isolated by virtue of its binding
to the Src family member nonreceptor t yrosine kinase YES (Yes
k inase-associated protein) (13). Additional YAP-interacting
proteins have been described more recently, including a number
of transcription factors [phosphatidylethanolamine-binding pro-
), p73, and TEA domain兾transcription en-
hancer factor (TEAD 兾TEF) family members], with which YAP
acts as a transcriptional coregulator (14–16). The Drosophila
ortholog of YAP, Yorkie (Yki), also functions as a transcrip-
tional coactivator, and it was recently described as a target of the
Hippo(Hpo)–Salvador(Sav)–Warts(Wts) pathway that nega-
tively regulates growth by simultaneously inhibiting proliferation
and promoting apoptosis (17, 18). Yki activates proliferation by
inducing the expression of cyclin E and inhibits apoptosis by
induction of the caspase-inhibitor protein DI AP1 (Drosophila
inhibitor of apoptosis) (17). The upstream Hpo–Wts kinase
cascade negatively regulates these activities (18–25).
In this study, we report that a mouse tumor-derived amplicon
defined by high-density array CGH excludes the MMP and BIRC
(cIAP) genes, pointing to YAP as a critical gene-amplification
‘‘driver.’’ Overexpression of YAP in human nontransformed
mammary epithelial cells results in phenotypic alterations that
are hallmarks of tumorigenic transformation, including epithe-
lial-to-mesenchymal transition (EMT), suppression of apoptosis,
grow th factor-independent proliferation, and anchorage-
independent g rowth in sof t agar. Collectively, these findings
suggest that YAP contributes to malignant transformation in
cancers harboring the 11q22 amplicon, and they support the
potential significance of this pathway in human cancer.
Mapping of the
-Containing Amplicon. Mammary tumors arising
in mice with a tissue-specific knockout of Brca1, engineered on
a Tr p-53-heteroz ygous backg round (Brca1
MMTV-Cre) (26), were subjected to whole-genome array CGH
analysis for gene copy-number alterations. One of 15 tumors
analyzed, CX4, harbored three distinct high-level amplifications
(Fig. 1A). The first was centered on the Met protooncogene, a
recurrent and specific genetic abnormalit y that is present in the
majorit y of Brca1兾Trp-53-driven mouse mammary tumors (2).
Whereas the amplification on chromosome 10 enc ompassed a
region of ⬎4 Mb with a large number of genes, the amplification
on chromosome 9 was centered on a single known gene, Yap (Fig.
1B). The Yap amplicon was of particular interest because
amplification of the syntenic locus on human chromosome 11q22
is found in diverse cancers, but the large size of the human
amplic ons has precluded identification of the key oncogene(s)
driving this amplification. In c ontrast, the CX4 tumor amplicon
was small (350 kb) and restricted to Yap and a neighboring
uncharacterized EST. The array CGH data were confirmed by
Conﬂict of interest statement: No conﬂicts declared.
Freely available online through the PNAS open access option.
Abbreviations: CGH, comparative genomic hybridization; EMT, epithelial-to-mesenchymal
transition; HMEC, human mammary epithelial cells; hTert, catalytic subunit of telomerase;
Mb, megabase(s); qPCR, quantitative PCR; STS, staurosporine; YAP, Yes kinase-associated
M.O., J.Z., and G.A.S. contributed equally to this work.
To whom correspondence should be addressed. E-mail: joan㛭brugge@hms.harvard.edu.
© 2006 by The National Academy of Sciences of the USA
August 15, 2006
using real-time quantitative PCR (qPCR), precisely defining the
boundaries of the amplicon (Fig. 1 B and C).
Induction of EMT in Mammary Epithelial Cells. To examine the
function of YAP in mammalian cells, we introduced this gene by
retroviral infection into the immortalized, but nontumorigenic,
human mammary epithelial cell line MCF10A. We have previ-
ously used this cell line in a three-dimensional culture model to
investigate the biological activities of known and candidate
onc ogenes within an architecture that mimics mammary acin i in
vivo (27, 28). To avoid clonal-selection effects, all ex periments
were performed with short-term cultures of drug-selected but
uncloned pools of cells, stably expressing YAP (MCF10A-YAP).
Whereas control MCF10A cells grow in epithelial-type islands
on monolayer cultures, cells overexpressing YAP displayed a loss
of cell–cell cont acts and cell scattering (Fig. 2A). YAP expression
also disrupted the morphogenesis of MCF10A cells in three-
dimensional cultures of reconstituted basement membrane (Ma-
trigel). MCF10A-YAP cells failed to for m spherical acinar-like
str uctures similar to the vector c ontrol cells (Fig. 2B). Instead,
these cells formed str uctures characterized by spike-like projec-
tions and cords of cells that invaded the basement-membrane
gel. This invasive phenotype was evident as early as day 4, and
it was detectable in ⬇50% of the structures by day 8 (Fig. 2B).
These morphological changes in monolayer and three-
dimensional cultures, i.e., a spindled morphology with cell
scattering, and invasion in Matrigel, suggested that MCF10A-
YAP cells had undergone EMT. EMT was evaluated by exam-
in ing the expression patterns of epithelial and mesenchymal
markers. The mesenchymal markers fibronectin, vimentin, and
N-cadherin were up-regulated, and the epithelial markers E-
cadherin and occludin were down-regulated in MCF10A-YAP
cells, as demonstrated by the immunoblotting analysis in Fig. 2C.
MCF10A-YAP cells also displayed disorganization of adherens
junctions, another hallmark of EMT, as shown by immunoflu-
orescence analyses of E-cadherin and actin localization (Fig.
2D). Finally, there was a 20- to 30-fold increase in the migration
of MCF10A-YAP cells c ompared with control cells in Transwell
assays (Fig. 2E). Interestingly, the increased migration was
evident only in the absence of EGF. Collectively, these morpho-
logical, biochemical, and cell-biological observations suggest
that YAP was able to induce EMT in MCF10A cells.
Overexpression Induces a Proliferative Advantage. Overexpres-
sion of the Drosophila YAP ortholog yki causes an overgrowth
phenot ype resulting from both the activation of proliferation and
the inhibition of cell death (17). Because YAP disr upted mor-
phogenesis of MCF10A cells and induced highly invasive three-
dimensional str uctures, we were unable to evaluate the effects of
YAP expression on proliferation of outer cells or cell death of the
center cells of acini in this model. To investigate further the
biological activities of YAP in mammalian cells, we examined
MCF10A-YAP cells more directly for proliferative and antiapo-
ptotic phenotypes in other assays. To assess the effects of YAP
on cell proliferation, we took advantage of the stringent require-
ment of MCF10A cells for EGF to support proliferation, and we
assayed MCF10A-YAP cells in both the presence and absence of
this grow th factor. MCF10A-YAP cells did not display an in-
creased rate of proliferation in monolayer cultures in the pres-
ence of EGF (Fig. 3A). However, these cells were able to
proliferate three-dimensionally in the absence of EGF, in con-
trast to vector control cells, which failed to proliferate under
these conditions (Fig. 3B). By 12 days in culture, MCF10A-YAP
cells had formed three-dimensional structures in the absence of
EGF that continued to grow larger until the assay was stopped
at day 30. Approximately 30% of the total input of MCF10A-
YAP cells were able to form structures af ter 30 days in culture,
whereas no control cells were able to proliferate in this assay.
Interestingly, these EGF-independent three-dimensional struc-
tures did not display the invasive morphology that was observed
in the presence of EGF (Fig. 2B), suggesting that EGF is
required for the YAP-induced invasive activity.
To gain insight into the mechanism responsible for the ability
of MCF10A-YAP cells to proliferate in the absence of EGF, we
examined whether YAP expression could activate signaling
through either ERK or AKT, two of the major signaling path-
ways that can contribute to EGF-independent g rowth of
MCF10A cells (29), by immunoblotting with activation-sensitive,
phospho-specific antibodies. Whereas exogenous growth factors
were required for activation of ERK and AKT in vector control
cells, both of these proteins displayed strong activation in the
absence of growth factors in MCF10A-YAP cells (Fig. 3C). Thus,
the activation of AKT and ERK could contribute to the ability
of YAP to promote proliferation of MCF10A cells in the absence
Inhibition of Apoptosis by
Overexpression. Although yki ex-
pression inhibits apoptosis in Drosophila (17), previous reports
of YAP function in mammalian cells indicate that expression of
this gene activates apoptosis in several tumor cell lines (30, 31).
To assess the effect of YAP ex pression on apoptosis in MCF10A
cells, we exposed MCF10A-YAP and control cells to a variety of
apoptosis-inducing stresses, including the chemotherapeutic
agents Taxol (paclitaxel) and cisplatin, the pan-kinase inhibitor
Fig. 1. Yap is a candidate ‘‘driver’’ gene in the mouse chromosome 9
amplicon. (A) Whole-genome proﬁle of an individual tumor (CX4) showing
ratio of tumor DNA signal vs. normal DNA control from the same mouse; x-axis
coordinates represent oligonucleotide probes ordered by genomic map po-
sition, with the whole-genome ﬁltered median (three nearest neighbors) data
set plotted. High-level ampliﬁcations are labeled 1–3 and correspond to: 1,
2.6-Mb amplicon on chromosome 6, centering on Met (2); 2, 350-kb amplicon
on chromosome 9, centering on Yap; 3, 4.6-Mb amplicon on chromosome 10,
containing a large number of genes. (B) Mouse chromosome 9 amplicon. Filled
diamonds denote the positions of probes detecting genomic ampliﬁcation in
CX4. Open diamonds indicate the positions of closest-neighbor probes de-
tecting normal DNA copy number. Circles represent data from C. Filled circles
denote ampliﬁcation, and open circles indicate normal DNA content. (C)
Independent conﬁrmation of the amplicon boundaries by using qPCR.
www.pnas.org兾cgi兾doi兾10.1073兾pnas.0605579103 Overholtzer et al.
st aurosporine (STS), UV irradiation, and loss of matrix attach-
ment (anoik is). Surprisingly, ex pression of YAP conferred pro-
tection f rom apoptosis induced by each of these stresses as
measured by DNA fragmentation (Fig. 4A). Thus, in contrast to
reports of YAP function in tumor cell lines, overexpression of
YAP broadly inhibits cell death in MCF10A cells. To evaluate
whether the effects YAP were specific to MCF10A cells, we
examined YAP-induced effects on apoptosis in another immor-
t alized, but nontumorigenic cell line, HMECtert. Immortalized
human mammary epithelial cells (HMEC) were generated by
infection with a retrovir us enc oding the catalytic subunit of
telomerase (hTert) (32). YAP was subsequently overexpressed in
these cells by retroviral infection, and the stable cell pool
(HMECtert-YAP) was assayed for apoptosis after exposure to
STS and cisplatin. As with MCF10A cells, YAP expression in
HMEC c onferred resistance to cell death induced by both
apoptotic inducers (Fig. 4B). Thus, in contrast to the proapo-
ptotic effect of YAP reported in some cancer cell lines (30, 31),
YAP displayed antiapoptotic activity in two nontransfor med
epithelial cell lines.
Induction of Colony Formation in Soft Agar. To evaluate a more
stringent parameter of oncogen ic transformation, we examined
the effect of YAP on the ability of MCF10A cells to form colonies
in soft agar, a propert y that frequently correlates with tumori-
gen icity. As expected, MCF10A-vector control cells failed to
produce anchorage-independent colonies in soft agar. In marked
c ontrast, MCF10A-YAP cells formed large colonies after 3
weeks in soft agar (Fig. 4C), demonstrating that YAP is able to
induce a fully transformed phenotype.
In this work, we demonstrate that overexpression of YAP in
MCF10A cells induces phenotypic alterations that are commonly
associated with potent transforming oncogenes, that is, induc-
tion of anchorage-independent growth, EMT, growth factor-
independent proliferation and activation of AKT and ERK, and
inhibition of apoptosis. Notably, most other oncogenes that
display activities similar to YAP are typically c onstitutively
activated mutant variants of cellular proteins, such as the
smGTPase H-Ras (33), the tyrosine kinase Src (34, 35), and
phosphatidylinositol 3-k inase (36–39). Thus, the ability of wild-
t ype YAP to induce transfor mation of immortalized mammary
epithelial cells by mere overexpression indicates that this gene
has potent oncogenic potential. This oncogenic activity of YAP
in mammalian cells is consistent with the described functions of
the Drosophila YAP ortholog yki, whose overexpression causes an
overgrowth phenotype resulting f rom both increased prolifera-
tion and reduced cell death (17). In parallel studies, L owe and
c oworkers (40) have also demonstrated oncogenic activit y for
Yap in a mouse model of hepatocellular carcinoma where Yap
amplification contributes to the development of tumors.
In Drosophila, the activity of Yki is negatively regulated by an
upstream kinase cascade in which the Hpo kinase, together with
its binding partner Sav, activates the Wts kinase–Mats c omplex,
which, in turn, inactivates Yki. Members of this upstream
pathway were identified before yki in genetic screens for inhib-
itors of cell growth in the Drosophila eye and wing (18–25, 41,
42). The abilit y of human YAP, like yki, to rescue pupal lethalit y
induced by overexpression of hpo and wts in Drosophila had
previously suggested that the g rowth-promoting functions of yki
Fig. 2. YAP induces an EMT. (A) YAP induces a morphology change on monolayer cultures. Representative phase-contrast images of MCF10A-YAP and vector
control cells growing in monolayer cultures are shown. (Scale bars, 100
m.) (B) YAP induces an invasive three-dimensional morphology. MCF10A-YAP and vector
control cells were cultured on Matrigel for 8 days. Representative phase-contrast images are shown, increasing in magniﬁcation from left to right. (Scale bars,
m.) (C) Expression of YAP results in loss of epithelial markers and gain of mesenchymal markers. Immunoblotting analysis reveals a decrease in E-cadherin
and occludin (epithelial markers) and concomitant increase in N-cadherin, vimentin, and ﬁbronectin (mesenchymal markers).
-Tubulin is used to show equal
loading. (D) YAP overexpression results in loss of membrane E-cadherin and cortical actin. Immunoﬂuorescence analysis shows loss of plasma membrane
E-cadherin (magniﬁcation, ⫻50) and loss of localization of actin (stained with phalloidin; magniﬁcation, ⫻100) at cortical sites adjacent to cell–cell interfaces
in MCF10A-YAP cells. (E) YAP induces Transwell migration. Control and YAP-expressing MCF10A cells were plated onto 8-
m Transwell ﬁlters and allowed to
migrate for 24 h either in the presence (⫹EGF) or in the absence (⫺EGF) of EGF. Data are the mean number of migrated cells per ⫻20 ﬁeld of four ﬁelds. Error
bars equal ⫾SD of three independent experiments.
Overholtzer et al. PNAS
August 15, 2006
are conserved in the human YAP ortholog (17). The YAP-
induced phenotypes described here in mammalian cells support
the notion that YAP, like Yki, can both activate proliferation
and inhibit apoptosis. The combination of these YAP-driven
phenot ypes is sufficient to transform the nontumorigenic human
epithelial cell line MCF10A.
Similar to the conservation between yki and YAP, the human
orthologs of wts, hpo, and mats can rescue their corresponding
Drosophila mutants, suggesting that the entire upstream Yki-
regulating pathway might be conserved in mammalian cells (18,
22, 43). In support of this hypothesis, the human Hpo ortholog,
MST2, can phosphorylate and activate the human Wts orthologs
LATS1 and LATS2 (44). Hints that this pathway might be
tumor-suppressive in mammalian cells have also been reported,
including a tumor-predisposition phenotype (soft-tissue sarco-
mas and ovarian tumors) in mice lacking one of the two wts
orthologs, Lats1 (45), and suppression of RasV12-driven trans-
for mation of NIH 3T3 cells by the second wts ortholog, Lats2
(46). Furthermore, the Hpo–Sav–Wts–Yki pathway in Drosoph-
ila was recently reported to lie downstream of signaling from
Merlin (47), which is the product of the NF2 tumor-suppressor
gene that is mut ated in humans with neurofibromatosis type II
(48, 49). Whether any of these tumor-suppressor functions in
mammalian cells c ould be mediated by the inhibition of YAP
activit y is not known. In screening for intragenic mutations in
human cancer-derived cell lines, we detected homoz ygous de-
letions in SAV in two renal cancer cell lines (24), suggesting that
this pathway c ould be targeted by the inactivation of upstream
regulators in addition to amplification of YAP.
In Drosophila, Yki functions as a transcriptional coregulator
that activates proliferation and inhibits apoptosis by increasing
the expression of the cyclin E gene and diap1 (17). We examined
the expression levels of cyclin E, cI AP1, and cIAP2 proteins in
MCF10A-YAP cells, and we found them to be similar to the
ex pression levels in control cells (data not shown). Thus, al-
though YAP appears to promote phenotypes in mammalian cells
similar to those promoted by Yki in Drosophila, the mechanism
of YAP action may be different. Previous studies of YAP in
Fig. 3. YAP overexpression promotes proliferation. (A) YAP overexpression
does not affect growth rate in the presence of EGF. MCF10A-vector cells were
grown in parallel to MCF10A-YAP cells, and their growth was assessed over an
8-day time course. (B) YAP overexpression activates EGF-independent growth.
MCF10A-YAP cells and vector control cells were grown on Matrigel in medium
without EGF for 30 days. Representative phase-contrast images are shown
from one of three independent experiments on day 12 on the Left, and both
a high-power (Center) and low-power (Right) magniﬁcation of day 30. (Scale
m.) (C) YAP overexpression results in activation of ERK1兾2 and AKT
pathways. Immunoblotting analysis with antibodies to phosphorylated
ERK1兾2 (Thr-202兾Tyr-204) and AKT (Ser-473) shows increased activation in
MCF10A-YAP cells in the absence of EGF and serum.
Fig. 4. YAP overexpression inhibits apoptosis and transforms MCF10A cells.
(A) YAP overexpression inhibits apoptosis in MCF10A cells. Control and
MCF10A-YAP monolayer cells were treated with STS, Taxol, UV light, or
cisplatin (Cisplat), or they were detached from matrix (anoikis) to induce
apoptosis. (A Left) Data are the mean percentages of sub-G
DNA content cells
determined by ﬂow-cytometric analysis of propidium iodide-stained samples
collected after the indicated treatments. (A Right) Cell-death data were
quantiﬁed by using the cell-death detection ELISA kit, which measures DNA
fragmentation; the data are the mean absorbance readings at 405– 490 nm
relative to vector control cells after 48 h of anoikis. att, attached control
monolayer cells. All error bars equal ⫾SD of three independent experiments.
(B) YAP overexpression inhibits apoptosis in HMEC. HMECtert-vector cells and
HMECtert-YAP cells were treated with STS and cisplatin to induce apoptosis.
Data are the mean percentages of the sub-G
content cells determined by ﬂow
cytometry as in A. Error bars equal ⫾SD of three independent experiments. (C)
YAP overexpression induces anchorage-independent growth in soft agar.
MCF10A-YAP cells and vector control cells were plated in soft-agar assays and
allowed to grow for 21 days. (Left) Data are the mean number of colonies per
six-well culture of either 10,000 cells (10k) or 50,000 cells (50k). Error bars equal
⫾SD of three independent experiments. (Right) Representative wells stained
with 0.02% iodonitrotetrazolium chloride are shown.
www.pnas.org兾cgi兾doi兾10.1073兾pnas.0605579103 Overholtzer et al.
mammalian cells have unc overed a variety of seemingly non-
overlapping functions. Since its discovery as a YES-binding
protein (13), other cy toplasmic functions for YAP have been
revealed, including the recruitment of Smad7 to TGF-
I (50), as well as nuclear functions, where YAP is a binding
partner and cotranscriptional regulator of a variety of transcrip-
tion factors, including PEBP2
, TEAD兾TEF, and p73 family
proteins (14–16). Whether any of these reported YAP functions
or binding partners might c ontribute to the phenotypes reported
here remains to be determined. The description of YAP acting
as an inhibitor of apoptosis is in direct contrast to previous
reports in which YAP was an activator of cell death in mam-
malian cells. YAP was previously shown to activate apoptosis in
response to DNA damage by interacting with p73 in several
cancer cell lines (30, 31). Suppressive interactions between p73
and endogenously high levels of p63 isoforms expressed in the
nontransfor med cells used here may have modulated the ob-
served effects (51).
Finally, although these data predict a role for YAP in human
cancer, this role remains to be clearly defined. As discussed
above, amplification of the 11q22 chromosomal locus including
YAP is observed in multiple cancer types. The Brca1兾Tr p-53-
driven mouse mammary tumor with selective amplification of
YAP led us to focus on this proposed oncogene by using
mammary epithelial transformation assays. However, using
qPCR analysis of microscopically dissected specimens, we did
not detect YAP amplification in ⬎100 sporadic human breast
cancers (data not shown). Thus, it is possible that the physio-
logical significance of YAP amplification may be more relevant
for other cancers that are more commonly known to have
amplification of the 11q22 locus, such as oral squamous-cell
carcinomas, where it is present in 5–15% of primary tumors (3,
10). In addition, although the specific YAP amplification in the
mouse model emphasized the unique contribution of this gene
to tumorigenesis, the larger size of the common human amplicon
points to additional genes that may jointly contribute to malig-
nanc y. For example, in some contexts the BIRC2 and BIRC3
genes encoding the apoptotic inhibitors cIAP1 and cIAP2 might
also be relevant targets of this amplicon. This indeed appears to
be the case in parallel work from a hepatocellular carcinoma
model, where the Yap and cIAP1 genes are coamplified and
jointly contribute to the development of tumors in mice (40).
Together, our studies point to potent oncogen ic effects of YAP,
a component of a highly c onserved pathway regulating prolif-
eration and apoptosis in Drosophila. The precise mechanisms
underlying the onc ogenic effects of YAP itself and the cellular
c ontexts in which it can contribute to malignanc y remain to be
Materials and Methods
Mammary Tumor Analysis. Experiment al mice (26) and array CGH
screen ing of tumors (2) have been previously described.
qPCR. The sequences of the PCR primer pairs and fluorogenic
MGB probes (Applied Biosystems, Foster City, CA) (all listed
f rom 5⬘ to 3⬘) used for DNA copy number analyses were:
Mm.Yap㛭F, CCTATGACCTCGCAGCATTCT; Mm.Yap㛭R,
GGA AACCTCCTCCCGTGTCT; Mm.Yap㛭 probe, VIC-
CCCCAGGGTCCACTC-MGBNFQ; Mm.Birc3㛭 F, CCC-
CTGAGCCTTCCAACA; Mm.Birc3㛭R, ATTGCACAAAAT-
TGAGGGCTTT; Mm.Birc3㛭 probe, VIC-ACAGCAGATTT-
TAAACACTT-MGBNFQ; Mm.EST1㛭 F, GGCAAGGAAT-
GACGGTCACT; Mm.EST1㛭 R, AATGT TGCCTCCTAC-
CCA ACA; Mm.EST1㛭 probe, VIC-CACA A A ACTGA A-
CACTTTACCTA-MGBNFQ; Mm.EST2㛭F, TCTGTTGTAT-
TCTGT TGCTGATGCT; Mm.EST2㛭 R, A ACACCGGAG-
ATAGAGACCCTAGA; Mm.EST2㛭probe, VIC-TGTATGGT-
TCTAGAT T TC-MGBNFQ; Mm.Trpc6㛭 F, AAGTACCT-
GA ACGCCCAT TT TC; Mm.Trpc6㛭 R, GA ATGATGGC-
GTCTTTCA AGTG; Mm.Trpc6㛭 probe, VIC-TCCTGAGT-
CTAATGCCT-MGBNFQ; Mm.Edem㛭 F, GTTTCCACAC-
CACCTTTGATTCT; Mm.Edem㛭R, GTCAGGAGGAACAC-
CTGTCTTCA; and Mm.Edem㛭 probe, VIC-CCCACTGCAG-
A ll samples were analyzed in triplicate, and the relative copy
number was derived by standardizing the input DNA to the
c ontrol signal (Edem1, a gene on chromosome 6 that is genomi-
cally stable based on the array CGH analysis).
Plasmid Construction. The human YAP ORF was cloned into the
pBABEpuro vector as an EcoRI–BamHI f ragment. A single
FLAG t ag was added to the N ter minus during PCR with the
following primers: Hs.YAP.F, CCGGGATCCACCATGGAT-
AGCAGCCCGCCGC; and Hs.YAP.R, CCGGAATTCCTA-
Immunoblotting Analysis. Cells were lysed in NETN lysis buffer
[150 mM NaCl兾1 mM EDTA兾20 mM Tris, pH 8兾0.5% Nonidet
P-40 兾1⫻ protease inhibitor mixture (Roche, Indianapolis, IN)].
Lysates were run on an SDS兾10% polyacrylamide gel and
transferred onto PVDF membranes (Millipore, Bedford, MA),
and immunoblots were visualized w ith a Western Lightning Plus
chemiluminescence kit (PerkinElmer, Boston, MA).
Immunofluorescence Analysis. Immunofluorescence was analyzed
as described in ref. 52.
Antibodies. Phospho-AKT (Ser-473), AKT, phospho-ERK1兾2
(Thr-202兾Tyr-204), and ERK1兾2 antibodies were purchased from
Cell Signaling Technology (Beverly, MA). Anti-occludin (His-279)
was from Santa Cruz Biotechnology (Santa Cruz, CA). The
bulin monoclonal antibody was from Upstate Biotechnology (Lake
Placid, NY). The fibronectin and FLAG (M2) antibodies were from
Sigma (St. Louis, MO). The E-cadherin, N-cadherin, and vimentin
antibodies were from BD Biosciences (San Jose, CA).
Cell Culture. MCF10A cells were cultured as described in ref. 53.
HMEC were obtained f rom Cambrex (East Rutherford, NJ) and
cultured in MEGM (Cambrex). HMEC were immortalized by
retroviral introduction of hTert (HMECtert). A retroviral con-
str uct encoding hTert (pBABEhygro-hTert) was a k ind gift from
W. C. Hahn (Dana–Farber Cancer Institute, Boston, MA).
Vesicular stomatitis virus glycoprotein-pseudot yped retrovi-
r uses were generated as described in ref. 53 and used to infect
MCF10A and HMEC. Cell lines were selected with 2
puromycin for MCF10A and 1
g兾ml puromycin or 50
hygromycin for HMEC.
Three-Dimensional Morphogenesis Assays. Cells were cultured in
growth factor-reduced reconstituted basement membrane (Ma-
trigel; BD Biosciences) as described in ref. 53. Cell lines were
assayed in three independent experiments.
Transwell Migration Assays. Transwell migration assays were per-
for med essentially as described in ref. 54. Cells (2 ⫻ 10
plated w ithout EGF, or 2.5 ⫻ 10
cells were plated with 5 ng兾ml
EGF on Transwell filters (8-
m pore size; Corning, Corning,
NY) in three-dimensional medium (described above). Assays
were stained and quantified after cells mig rated for 24 h.
Two-Dimensional Cell Proliferation. Two-dimensional cell prolifer-
ation was measured by using the fluorescent nuclear stain Syto
60 (Molecular Probes, Eugene, OR) as described in ref. 52.
Overholtzer et al. PNAS
August 15, 2006
EGF-Independent Proliferation. Cells were plated on Matrigel as for
three-dimensional morphogenesis assays but without EGF. Cells
were fed every 4 days with medium lacking EGF for the duration
of the experiment.
Cell Death Assays. Monolayer MCF10A cultures were assayed for
DNA fragmentation after treatment with 0.5
M STS (Sigma) for
18 h, 100 nM Taxol (Sigma) for 48 h, 100
UV light [45-s exposure of UV-C (254 nm) from a 30-W G30T8
bulb (VWR, Bridgeport, NJ] for 24 h. HMECtert were assayed
after treatment with 0.5
M STS or 50
M cisplatin for 24 h.
Floating cells were collected and combined with trypsinized cells,
fixed in 75% ethanol, treated with RNase A (0.25 mg兾ml), stained
with propidium iodide (10
g兾ml), and analyzed on a FACSCalibur
flow cytometer (BD Biosciences) for percentage of cells with
DNA content. Data were analyzed by using CellQuest (BD
Biosciences). For anoikis, MCF10A cells were plated in growth
medium on tissue-culture plates pretreated w ith poly(2-
hydroxyethyl methacrylate) [poly-HEMA from Sigma–Aldrich (St.
Louis, MO) (6 mg兾ml in 95% ethanol at 37°C until dry)] to prevent
adherence. After 48 h, cells were collected and analyzed for DNA
fragmentation by using the cell-death detection ELISA kit (Roche
Diagnostics, Mannheim, Germany). All cell-death assays were
performed in three independent experiments.
Soft-Agar Assays. Cells (1 ⫻ 10
or 5 ⫻ 10
) were added to 1.5 ml
of growth medium with 0.4% agar and layered onto 2 ml of 0.5%
agar beds in six-well plates. Cells were fed with 1 ml of medium
with 0.4% agar every 7 days for 3 weeks, after which colonies
were stained with 0.02% iodon itrotetrazolium chloride (Sigma–
A ldrich) and photographed. Colonies larger than 50
diameter were counted as positive for g rowth. Assays were
c onducted in duplicate in three independent experiments.
We thank Scott Lowe for sharing unpublished results and Lynda Chin for
array CGH analysis. This work was supported by National Cancer
Institute兾National Institutes of Health Grants CA080111 and CA089393
and the Breast Cancer Research Foundation (to J.S.B.); National
Institutes of Health Grant P01 95281, the Doris Duke Foundation
Distinguished Clinical Investigator Award, and a National Foundation
for Cancer Research grant (to D.A.H.); National Institutes of Health
Grant F32 CA117737 (to G.A.S.); and National Cancer Institute兾
National Institutes of Health Institutional Training Grant T32CA09361
1. Pinkel, D., Segraves, R., Sudar, D., Clark, S., Poole, I., Kowbel, D., Collins, C.,
Kuo, W. L., Chen, C., Zhai, Y., et al. (1998) Nat . Genet. 20, 207–211.
2. Smolen, G. A., Muir, B., Mohapatra, G., Barmettler, A., K im, W. J., Rivera,
M. N., Haserlat, S. M., Okimoto, R. A., Kwak, E., Dahiya, S., et al. (2006)
Cancer Res. 66, 3452–3455.
3. Baldw in, C., Garn is, C., Zhang, L., Rosin, M. P. & Lam, W. L. (2005) Cancer
Res. 65, 7561–7567.
4. Bashyam, M. D., Bair, R., Kim, Y. H., Wang, P., Hernandez-Boussard, T., Karikari,
C. A., Tibshirani, R., Maitra, A. & Pollack, J. R. (2005) Neoplasia 7, 556–562.
5. Dai, Z., Zhu, W. G., Morrison, C. D., Brena, R. M., Smiraglia, D. J., Raval, A., Wu,
Y. Z., Rush, L. J., Ross, P., Molina, J. R., et al. (2003) Hum. Mol. Genet. 12, 791–801.
6. Hermsen, M., A lonso Guervos, M., Meijer, G., van Diest, P., Suarez Nieto, C.,
Marcos, C. A. & Sampedro, A. (2005) Cell. Oncol. 27, 191–198.
7. Imoto, I., Tsuda, H., Hirasawa, A., Miura, M., Sakamoto, M., Hirohashi, S. &
Inazawa, J. (2002) Cancer Res. 62, 4860–4866.
8. Imoto, I., Yang, Z. Q., Pimkhaokham, A., Tsuda, H., Shimada, Y., Imamura,
M., Ohk i, M. & Inazawa, J. (2001) Cancer Res. 61, 6629–6634.
9. Lambros, M. B., Fiegler, H., Jones, A., Gorman, P., Roylance, R. R., Carter,
N. P. & Tomlinson, I. P. (2005) J. Pathol. 205, 29–40.
10. Snijders, A. M., Schmidt, B. L., Fridlyand, J., Dekker, N., Pinkel, D., Jordan,
R. C. & Albertson, D. G. (2005) Oncogene 24, 4232–4242.
11. Weber, R. G., Sommer, C., Albert, F. K., Kiessling, M. & Cremer, T. (1996)
Lab. Invest . 74, 108–119.
12. Deveraux, Q. L. & Reed, J. C. (1999) Genes Dev. 13, 239–252.
13. Sudol, M. (1994) Oncogene 9, 2145–2152.
14. Yagi, R., Chen, L. F., Shigesada, K., Murakami, Y. & Ito, Y. (1999) EMBO J.
15. Strano, S., Munarriz, E., Rossi, M., Castagnoli, L., Shaul, Y., Sacchi, A., Oren,
M., Sudol, M., Cesareni, G. & Blandino, G. (2001) J. Biol. Chem. 276,
16. Vassilev, A., Kaneko, K. J., Shu, H., Zhao, Y. & DePamphilis, M. L. (2001)
Genes Dev. 15, 1229–1241.
17. Huang, J., Wu, S., Barrera, J., Matthews, K. & Pan, D. (2005) Cell 122, 421–434.
18. Wu, S., Huang, J., Dong, J. & Pan, D. (2003) Cell 114, 445–456.
19. Harvey, K. F., Pfleger, C. M. & Hariharan, I. K. (2003) Cell 114, 457–467.
20. Jia, J., Zhang, W., Wang, B., Trinko, R. & Jiang, J. (2003) Genes Dev. 17, 2514–2519.
21. Kango-Singh, M., Nolo, R., Tao, C., Verstreken, P., Hiesinger, P. R., Bellen,
H. J. & Halder, G. (2002) Development (Cambridge, U.K.) 129, 5719–5730.
22. Lai, Z. C., Wei, X., Shimizu, T., Ramos, E., Rohrbaugh, M., Nikolaidis, N., Ho,
L. L. & Li, Y. (2005) Cell 120, 675–685.
23. Pantalacci, S., Tapon, N. & Leopold, P. (2003) Nat. Cell Biol . 5, 921–927.
24. Tapon, N., Harvey, K. F., Bell, D. W., Wahrer, D. C., Schiripo, T. A., Haber,
D. A. & Hariharan, I. K. (2002) Cell 110, 467–478.
25. Udan, R. S., Kango-Singh, M., Nolo, R., Tao, C. & Halder, G. (2003) Nat. Cell
Biol. 5, 914–920.
26. Brodie, S. G., Xu, X., Qiao, W., Li, W. M., Cao, L. & Deng, C. X. (2001)
Oncogene 20, 7514–7523.
27. Debnath, J. & Brugge, J. S. (2005) Nat. Rev. Cancer 5, 675–688.
28. Witt, A. E., Hines, L. M., Collins, N. L., Hu, Y., Gunawardane, R. N., Moreira,
D., Raphael, J., Jepson, D., Koundinya, M., Rolfs, A., et al. (2006) J. Proteome
Res. 5, 599–610.
29. Debnath, J., Walker, S. J. & Br ugge, J. S. (2003) J. Cell Biol. 163, 315–326.
30. Basu, S., Totty, N. F., Irwin, M. S., Sudol, M. & Downward, J. (2003) Mol. Cell 11, 11–23.
31. Strano, S., Monti, O., Pediconi, N., Baccarini, A., Fontemaggi, G., Lapi, E.,
Mantovani, F., Damalas, A., Citro, G., Sacchi, A., et al. (2005) Mol. Cell 18,
32. Elenbaas, B., Spirio, L., Koerner, F., Fleming, M. D., Zimonjic, D. B., Donaher, J. L.,
Popescu, N. C., Hahn, W. C. & Weinberg, R. A. (2001) Genes Dev. 15, 50–65.
33. Taparowsky, E., Suard, Y., Fasano, O., Shimizu, K., Goldfarb, M. & Wigler, M.
(1982) Nature 300, 762–765.
34. Iba, H., Takeya, T., Cross, F. R., Hanafusa, T. & Hanafusa, H. (1984) P roc.
Natl. Acad. Sci. USA 81, 4424–4428.
35. Levy, J. B., Iba, H. & Hanafusa, H. (1986) Proc. Natl. Acad. Sci . USA 83,
36. Samuels, Y., Wang, Z., Bardelli, A., Silliman, N., Ptak, J., Szabo, S., Yan, H.,
Gazdar, A., Powell, S. M., Riggins, G. J., et al. (2004) Science 304, 554.
37. Kang, S., Bader, A. G. & Vogt, P. K. (2005) Proc. Natl. Acad. Sci. USA 102, 802–807.
38. Ikenoue, T., Kanai, F., Hikiba, Y., Obata, T., Tanaka, Y., Imamura, J., Ohta,
M., Jazag, A., Guleng, B., Tateishi, K., et al. (2005) Cancer Res. 65, 4562–4567.
39. Isakoff, S. J., Engelman, J. A., Irie, H. Y., Luo, J., Brachmann, S. M., Pearline,
R. V., Cantley, L. C. & Brugge, J. S. (2005) Cancer Res. 65, 10992–11000.
40. Zender, L., Spector, M., Xue, W., Flemming, P., Cordon-Cardo, C., Silke, J.,
Fan, S., Luk, J., Wigler, M., Hannon, G., et al. (2006) Cell 125, 1253–1267.
41. Justice, R. W., Zilian, O., Woods, D. F., Noll, M. & Bryant, P. J. (1995) Genes
Dev. 9, 534–546.
42. Xu, T., Wang, W., Zhang, S., Stewart, R. A. & Yu, W. (1995) Development
(Cambridge, U.K.) 121, 1053–1063.
43. Tao, W., Zhang, S., Turenchalk, G. S., Stewart, R. A., St. John, M. A. R., Chen,
W. & Xu, T. (1999) Nat . Genet. 21, 177–181.
44. Chan, E. H., Nousiainen, M., Chalamalasetty, R. B., Schafer, A., Nigg, E. A.
& Sillje, H. H. (2005) Oncogene 24, 2076–2086.
45. St. John, M. A. R., Tao, W., Fei, X., Fukumoto, R., Carcangiu, M. L.,
Brownstein, D. G., Parlow, A. F., McGrath, J. & Xu, T. (1999) Nat. Genet. 21,
46. Li, Y., Pei, J., Xia, H., Ke, H., Wang, H. & Tao, W. (2003) Oncogene 22,
47. Hamaratoglu, F., Willecke, M., Kango-Singh, M., Nolo, R., Hyun, E., Tao, C.,
Jafar-Nejad, H. & Halder, G. (2006) Nat. Cell. Biol. 8, 27–36.
48. Rouleau, G. A., Merel, P., Lutchman, M., Sanson, M., Zucman, J., Marineau,
C., Hoang-Xuan, K., Demczuk, S., Desmaze, C., Plougastel, B., et al. (1993)
Nature 363, 515–521.
49. Trofatter, J. A., MacCollin, M. M., Rutter, J. L., Murrell, J. R., Duyao, M. P.,
Parry, D. M., Eldridge, R., Kley, N., Menon, A. G., Pulaski, K., et al. (1993) Cell
50. Ferrigno, O., Lallemand, F., Verrecchia, F., L’Hoste, S., Camonis, J., Atfi, A.
& Mauviel, A. (2002) Oncogene 21, 4879–4884.
51. Rocco, J. W., Leong, C. O., Kuperwasser, N., DeYoung, M. P. & Ellisen, L. W.
(2006) Cancer Cell 9, 45–56.
52. Smolen, G. A., Sordella, R., Muir, B., Mohapatra, G., Barmettler, A.,
Archibald, H., Kim, W. J., Ok imoto, R. A., Bell, D. W., Sgroi, D. C., et al. (2006)
Proc. Natl. Acad. Sci. USA 103, 2316–2321.
53. Debnath, J., Muthuswamy, S. K. & Brugge, J. S. (2003) Methods 30, 256 –268.
54. Gunawardane, R. N., Sgroi, D. C., Wrobel, C. N., Koh, E., Daley, G. Q. &
Brugge, J. S. (2005) Cancer Res. 65, 11572–11580.
www.pnas.org兾cgi兾doi兾10.1073兾pnas.0605579103 Overholtzer et al.