SAGE-Hindawi Access to Research
Pathology Research International
Volume 2011, Article ID 178265, 6 pages
PARGenes:MolecularProbes to Pathological Assessmentin
Beatrice Uziely,1HagitTurm,1MyriamMaoz,1IritCohen,1Bella Maly,2
1Departments of Oncology, Hadassah-University Hospital P.O. Box 12000, Jerusalem 91120, Israel
2Departments of Pathology, Hadassah-University Hospital P.O. Box 12000, Jerusalem 91120, Israel
Correspondence should be addressed to Rachel Bar-Shavit, email@example.com
Received 15 September 2010; Accepted 4 January 2011
Academic Editor: Beiyun Chen
Copyright © 2011 Beatrice Uziely et al.ThisisanopenaccessarticledistributedundertheCreativeCommonsAttributionLicense,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Taking the issue of tumor categorization a step forward and establish molecular imprints to accompany histopathological
assessment is a challenging task. This is important since often patients with similar clinical and pathological tumors may respond
differently to a given treatment. Protease-activated receptor-1(PAR1), a G protein-coupled receptor (GPCR), is the first member of
the mammalian PAR family consisting of four genes. PAR1and PAR2play a central role in breast cancer. The release of N-terminal
peptides during activation and the exposure of a cryptic internal ligand in PARs, endow these receptors with the opportunity to
serve as a “mirror-image” index reflecting the level of cell surface PAR1&2-in body fluids. It is possible to use the levels of PAR-
released peptide in patients and accordingly determine the choice of treatment. We have both identified PAR1C-tail as a scaffold
site for the immobilization of signaling partners, and the critical minimal binding site. This binding region may be used for future
therapeuticmodalitiesinbreastcancer, sinceabrogationofthebindinginhibitsPAR1inducedbreastcancer. Altogether,bothPAR1
and PAR2may serve as molecular probes for breast cancer diagnosis and valuable targets for therapy.
The classification of a tumor differentiation level is rou-
tinely based on histopathological criteria whereby poorly
differentiated tumors generally exhibit the worst prognoses.
However, the underlying molecular pathways that regulate
the level of breast tumor development are as yet poorly
described. Until now the pathological tissue criteria that
entail tissue traits have not been defined by an appropriate
set of genes. A challenging task is to take the issue of breast
tumor categorization a step forward and establish molecular
imprints to accompany histopathological assessment. This
is important since often patients with similar clinical and
pathological tumors may have a markedly different outcome
in response to a given treatment. These differences are
encoded by and stem from the tumor genetic profile .
Individual gene signature may complement or replace the
traditional pathological assessment in evaluating tumor
behavior and risk. This is the basis for optimizing our
may refine the prediction of the course of disease and the
response to treatment . Oncotype Dx is a clinically vali-
dated and widely used multigene assay (there are also other
commercially available gene panels such as Mammaprint;
Agendia Amsterdam, Netherland, and THEROS H/I; Bio-
theranostics, San Diego, CA), that quantifies the likelihood
of breast cancer recurrence. This gene profile has been
developed specifically for women with hormone receptor-
positive (estrogen and progesterone receptor; ER, PR) and
lymph node-negative disease. The gene profile consists of 21
genes that are associated with disease recurrence. Sixteen are
cancer-related genes and 5 serve as referencegenes. This gene
that correlates with the specific likelihood of breast cancer
recurrence within 10 years from the original diagnosis.
play a central part in breast cancer biology and determine
gression . Identification of these genes will significantly
contribute to the prospect of treatment making choices.
Protease-activated receptor-1(PAR1), a G protein-coup-
led receptor (GPCR), is the first and prototype member of
2 Pathology Research International
the mammalian PAR family consisting of four genes. The
activation of PAR1 involves the release of an N-terminal
peptide and the exposure of an otherwise hindered ligand,
resulting in an exclusive mode of activation. This mode
of activation serves as a general paradigm for the entire
PAR family [4–6]. While a well-known classical observation
points to a close link between hyperactivation of the coagu-
lation system and cancer malignancies, the molecular mech-
anism that governs procoagulant tumor progression remains
poorly defined [7–10]. Thrombin is a main effector of the
coagulation cascade. In addition to cleaving fibrinogen, it
also activates cells through at least three PARs: PAR1, PAR3,
and PAR4. In contrast, PAR2is activated by multiple trypsin-
like serine proteases including the upstream coagulant pro-
teases VIIa—tissue factor (TF) and Xa, but not by thrombin.
It is now becoming well established that human Par1, hPar
1, plays a central role in epithelial malignancies [13, 14, 16].
central assignments in breast cancer [11, 12]. High levels of
hPar1expression are directly correlated with epithelia tumor
progression in both clinically obtained biopsy specimens
and a wide spectrum of differentially metastatic cell lines
[13, 14]. PAR1also plays a role in the physiological invasion
process of placental cytotrophoblasts during implantation
into the uterus deciduas . Trophoblast invasion shares
many features with the tumor cell invasion process. It differs,
however, by the time-limited hPar1 expression, which is
confined to the trophoblast-invasive period and is shut off
immediately thereafter, when there is no need to invade
. This strongly supports the notion that the hPar1gene
is part of an invasive gene program. Surprisingly, the zinc-
that efficiently cleaves extra cellular matrix (ECM) and base-
ment membrane components, has been shown to specifically
activate PAR1 . PAR1-MMP1 axis may thus provide a
direct mechanistic link between PAR1and tumor metastasis.
The mechanism that leads to hPar1gene overexpression in
tumor is yet unclear and under current extensive investi-
gation. Although the impaired internalization of PAR1that
results with persistent signaling and invasion was previously
suggested for several breast cancer lines , an imbalanced
expression between hPar1 repressors and activators was
proposed, suggesting transcriptional regulation . We
found that the mechanism of hPar1overexpression involves
enhanced transcriptional activity, whereby enhanced RNA
chain elongation takes place in the aggressive cancer cells as
compared with the nonaggressive, low metastatic potential
cells . Indeed, we have identified the Egr-1 transcription
factor as a critical DNA-binding protein eliciting hPar1
expression in prostate cancer cells and the wt p53 tumor
suppressor as an hPar1transcription repressor [19, 20]. The
wt form of p53 thus acts as a fine-tuning regulator of hPar1
in cancer progression.
The PARs act as delicate sensors of extra cellular protease
gradient to allow the cells to respond to a proteolytically
modified environment. The fact that PAR1gene and protein
overexpression are associated with the aggressiveness of a
tumor, in vivo, reflect its potential role in cancer dissemi-
nation. Furthermore, it assigns PAR1as an attractive target
for anticancer therapy. On the other hand, the release of an
N-terminal peptide during activation and the exposure of
an otherwise cryptic internal ligand in PARs endow these
receptors with the opportunity to serve as a “mirror-image”
index reflecting in body fluids the level of PARs on the
surface of cancer cells. Hence, PAR1and PAR2peptides in
indicator for the extent of cancer progression. While the
overexpression of both PAR1 and PAR2 takes place on the
surface of cancer cells that are being constantly turned over
in the body, yet there is no current information as to the half
-life of the released peptides. It is envisioned that measuring
the level of released peptides may underline the severity of
cancer. Another aspect is that the followup levels of PAR1-
released peptides may be instrumental in demonstrating the
effectiveness of a given treatment. For example, determining
the level of the released PAR1and PAR2, through repeated
measurements in the blood stream, may serve as a base
line for a patient, and a sensitive indicator for response
to a treatment. If the released PAR peptides are becoming
gradually low and finally disappear, it may reassure that the
In contrast, if the level remains unchanged, it may indicate
that the tumor is progressing despite of a given treatment.
A critical aspect, however, that needs to be addressed is the
prospect of high released PAR1&2 peptides present during
inflammation [21, 22]. Therefore, the repeated followup
of PAR released peptides is necessary for the purpose of
demonstrating that during inflammation the high PAR-
released peptide level is transient and disappears when the
inflammatory response is over. In contrast, in the case of a
tumor, the level of PAR-released peptides remains constantly
high. The relative contribution of PAR1versus PAR2during
the process of tumor progression is as yet unknown and
is under current investigation. One approach to decisively
address this issue is by immunohistological staining (of anti-
PAR1and anti-PAR2antibodies, separately) utilizing tissue
microarray biopsy specimens on a large pool of primary
breast cancer biopsy specimens representing invasive carci-
noma. Such analysis will determine the relative percentage
of PAR-positive individuals in a given cancer patient pool.
Whether PARs join the triple negative population (ER-, PR-,
and Her-2/Neu, an indicator of disease aggressiveness)—
or perhaps stands independently as a prognostic marker—
needs to be evaluated.
3.PARs as Target for Therapy
Importantly, PAR1cellular trafficking and signal termination
appear to occur in a different mode than other GPCRs.
Instead of recycling back to the cell surface after ligand
stimulation, activated PAR1is sorted to the lysosomes where
it is degraded [23, 24]. While cellular trafficking of PAR1
impinges on the extent and mode of signaling, the identifica-
tion of individual PAR1signaling partners and their contri-
bution to breast cancer progression remain to be elucidated.
Pathology Research International3
Breast cancer progression
Steps in epithelia tumor progression
Normal epitheliumDysplasia Carcinoma In-Situ (high grade) Invasive carcinoma
Figure 1: Steps in breast cancer progression. Subtypes definition of breast cancer according to ER, PR, and Her2/neu status. Additional
categorization is suggested including PARs status.
We have adopted the approach of utilizing a truncated
form of hPar1gene devoid of the entire cytoplasmic tail to
demonstrate the significant role of PAR1signaling in breast
tumor progression. This was demonstrated in a xenograft
mice model of mammary gland tumor development, in vivo
. Along this line of evidence, we have identified PAR1
C-tail as a scaffold site for the immobilization of signaling
partners. In addition to identifying key partners, we have
determined the hierarchy of binding and established a region
in PAR1 C-tail critical for breast cancer signaling. This
minimal binding domain may provide a potent platform
for future therapeutic vehicles in treating breast cancer. The
above-described outcome is a brief summary of the detailed
experimental approach illustrated bellow.
The functional outcome of MCF7 cells overexpressing
various hPar1constructs in vivo was assessed by orthotopic
mammary fat pad tumor development. MCF7 cells over-
expressing either persistent hPar1 Y397Z or wt hPar1 con-
structs (e.g., MCF7/Y397Z hPar1; MCF7/wt hPar1) markedly
enhanced tumor growth in vivo following implantation into
the mammary glands, whereas MCF7 cells overexpressing
truncated hPar1, devoid of the entire cytoplasmic tail,
behaved similarly to control MCF7 cells in vector-injected
mice, which developed only very small tumors. The tumors
obtained with MCF7/wt hPar1and MCF7/Y397Z hPar1were
5 and 5.8 times larger, respectively, than tumors produced by
ination (H&E staining) showed that while both MCF7/wt
hPar1 and MCF7/Y397Z hPar1 tumors infiltrated into the
fat pad tissues of the breast, the MCF7/Y397Z hPar1tumors
produced by empty vector or truncated hPar1-transfected
cells were capsulated, with no obvious cell invasion. Tumor
growth can also be attributed to blood vessel formation [26,
27]. The hPar1-induced breast tumor vascularization was
assessed by immunostaining with antilectin and anti-CD31
antibodies, showing that both MCF7/Y397Z hPar hPar1and
MCF7/wt hPar1tumors were intensely stained. In contrast,
only few blood vessels were found in the small tumors
of empty vector or truncated hPar1. Thus, both MCF7/wt
4 Pathology Research International
Schematic presentation of PAR1
C-tail scaffold interactions with the signaling partners
Etk/Bmx and Shc: identification of the minimal binding
motif in PAR1C-tail
Figure 2: Activation of PAR1leads to the association of Etk/Bmx with PAR1C-tail. This association is mediated through Etk/Bmx PH-
domain enabling next the binding of Shc. The site of the “signal binding” domain (e.g., Etk/Bmx, as a prime signaling partner) in PAR1has
been identified. Insertion of successive replacement of A residues forming a PAR1mutant incapable of binding Etk/Bmx showed impaired
capabilities of PAR1induced invasion and migration. This site provides therefore a platform for the development of future therapeutic
medicaments in breast cancer.
hPar1and MCF7/Y397Z hPar1cells were shown to effectively
induce breast tumor growth, proliferation, and angiogenesis,
while the MCF7/truncated hPar1and MCF7/empty vector-
expressing cells had no significant effect. This experimental
results highlight the significance of PAR1signaling in PAR1-
induced breast cancer progression.
Next, in order to identify specific PAR1 signaling com-
ponents, the following approach was utilized. To detect
the putative mediator(s) linking PAR1 to potential signal-
ing pathway, we examined a custom-made antibody-array
membranes. When aggressive breast carcinoma MDA-MB-
435 cells (with high hPar1 levels) were incubated with the
antibody-array membranes before and after PAR1activation
(15 minutes), the following results were obtained. Several
activation-dependent proteins which interact with PAR1,
including ICAM, c-Yes, Shc, and Etk/Bmx, were identified.
Of these proteins, we chose to focus here on Etk/Bmx and
The epithelial tyrosine kinase (Etk), also known as Bmx,
is a nonreceptor tyrosine kinase that is unique by virtue of
GPCRs . This type of interaction is mainly attributed
to the pleckstrin homology (PH) which is followed by the
Src homology SH3 and SH2 domains and a tyrosine kinase
site . Etk/Bmx-PAR1 interactions were characterized
by binding of lysates exhibiting various hPar1 forms to
GST-PH-Etk/Bmx. While Y397Z hPar1and wt hPar1showed
specific association with Etk/Bmx, lysates of truncated hPar1
or JAR cells (lacking PAR1) exhibited no binding. A tight
association between the PAR1 C-tail and Etk/Bmx was
obtained, independent of whether wt or kinase-inactive
Etk/Bmx (KQ) was used [29, 30].
Next, we wished to determine the chain of events mediating
the signaling of PAR1and the binding of Shc and Etk/Bmx to
PAR1C-tail. Shc is a well-recognized cell signaling adaptor
known to associate with tyrosine-phosphorylated residues.
To this end, analysis of MCF7 cells that express little to
no hPar1were ectopically forced to overexpress hPar1gene.
When coimmunoprecipitation with anti-PAR1 antibodies
following PAR1 activation was performed, surprisingly, no
Pathology Research International5
Shc was detected in the PAR1 immunocomplex. Shc asso-
ciation with PAR1was fully rescued only when MCF7 cells
were initially cotransfected with Etk/Bmx, resulting with
abundant assembly of Shc in the immunocomplex. Thus,
Etk/Bmx is a critical component that binds first to activated
PAR1C-tail enabling the binding of Shc. Shc may bind either
to phosphorylated Etk/Bmx, via its SH2 domain, or in an
unknown manner to the PAR1C-tail, provided that Etk/Bmx
is present and is PAR1-bound complex. One cannot, how-
ever, exclude the possibility that Bmx binds first to Shc, and
only then the complex of Etk/Bmx-Shc binds to PAR1.
The functional consequences of the Etk/Bmx binding
was further evaluated by inserting mutations to the “signal-
binding” site. We prepared hPar1constructs with successive
replacement of the designated seven residues (378-384;
CQRYVYS) with A, termed as hPar1-7A. This HA tagged
mutant, HA-hPar1-7A, completely failed to immunoprecip-
itate Etk/Bmx. In contrast, in the presence of HA-wt hPar1,
potent immunoprecipitation was obtained. We thus con-
clude that the critical region for Etk/Bmx binding to PAR1
C-tail resides in the vicinity of CQRYVYS. The physiological
significance of PAR1-Etk/Bmx binding is emphasized by the
following outcome. Activated MCF7 cells that express hPar1-
7A mutant failed to invade Matrigel-coated membranes. In-
contrast, a potent invasion was obtained by activated wt
hPar1. This outcome highlights the fact that by preventing
inhibition of PAR1 pro-oncogenic functions, including the
loss of epithelial cell polarity, migration, and invasion, is
obtained (see Figures 1 and 2 for wt and mutated PAR1C-
tail and the ability to form a scaffold complexes with the
signaling partners). Elucidation of the PAR1C-tail binding
domain may therefore provide a potent platform for future
therapeutic vehicles in treating breast cancer.
The same approach may be utilized to identify a prime
signaling partner for PAR2. This will eventually lead to
characterization of a minimal PAR2 C-tail binding region.
Generation of peptides that can enter the cells via adding
Tat or penetratin, or alternatively, addition of either myris-
toylation, or another lipid moiety, will assist the peptides
to cross the cell membrane. These peptides may prove
as effective therapeutic inhibitors of PARs-induced breast
cancer growth and development. Along this line of evidence,
successful PAR1-derived peptides termed “pepducin” were
developed by the group of Kuliopulos A . This group has
demonstrated that PAR1-induced breast tumor in a mouse
model, in vivo, is blocked by the cell-penetrating lipopeptide
“pepducin,” P1pal-7, which is a potent inhibitor of cell via-
bility in breast carcinoma cells expressing PAR1. It has been
shown that P1pal-7 is capable of promoting apoptosis in
breast tumor xenografts and significantly inhibits metastasis
to the lung.
In summary, PARs may provide a timely effective chal-
lenge for developing valuable prognostic vehicles and also
critical targets for therapy in breast cancer. While the PAR
prognostic vehicles stem from the extracelluar portion of the
receptors, we offer the intracellular C-tail site as potential
targets for therapy in breast cancer. What is the relative
contribution of PAR1 versus PAR2 in breast cancer tumor
growth and development is yet an open question and a
subject of current evaluation.
Conflict of Interests
The authors have declared that no conflict of interests exists.
This work was supported by grants from the Israel Science
Foundation (Grant no. 1313/07), Fritz-Thyssen Foundation,
and Israel Cancer Research Fund (granted to R. B.). The
funds had no role in the study design, data collection,
analysis, decision to publish, or preparation of the paper.
 S. Hanash, “Integrated global profiling of cancer,” Nature
Reviews Cancer, vol. 4, no. 8, pp. 638–644, 2004.
 C. Oakman, S. Bessi, E. Zafarana, F. Galardi, L. Biganzoli,
and A. Di Leo, “Recent advances in systemic therapy: new
diagnostics and biological predictors of outcome in early
breast cancer,” Breast Cancer Research, vol. 11, no. 2, p. 205,
 S. Paik, S. Shak, G. Tang et al., “A multigene assay to predict
recurrence oftamoxifen-treated,node-negative breastcancer,”
New England Journal of Medicine, vol. 351, no. 27, pp. 2817–
 S. R. Coughlin, “Thrombin signalling and protease-activated
receptors,” Nature, vol. 407, no. 6801, pp. 258–264, 2000.
 P. Arora, B. D. Cuevas, A. Russo, G. L. Johnson, and J.
Trejo, “Persistent transactivation of EGFR and ErbB2/HER2
by protease-activated receptor-1 promotes breast carcinoma
cell invasion,” Oncogene, vol. 27, no. 32, pp. 4434–4445, 2008.
 M. L. Nierodzik and S. Karpatkin, “Thrombin induces
tumor growth, metastasis, and angiogenesis: evidence for a
thrombin-regulated dormant tumor phenotype,” Cancer Cell,
vol. 10, no. 5, pp. 355–362, 2006.
 F. R. Rickles and R. L. Edwards, “Activation of blood
coagulation in cancer: trousseau’s syndrome revisited,” Blood,
vol. 62, no. 1, pp. 14–31, 1983.
 E. Camerer, D. N. Duong, J. R. Hamilton, and S. R. Coughlin,
“Combined deficiency of protease-activated receptor-4 and
fibrinogen recapitulates the hemostatic defect but not the
embryonic lethality of prothrombin deficiency,” Blood, vol.
103, no. 1, pp. 152–154, 2004.
 J. S. Palumbo, K. W. Kombrinck, A. F. Drew et al., “Fibrinogen
is an important determinant of the metastatic potential of
circulating tumor cells,” Blood, vol. 96, no. 10, pp. 3302–3309,
 M. Riewald, V. V. Kravchenko, R. J. Petrovan et al., “Gene
induction by coagulation factor Xa is mediated by activation
of protease-activated receptor 1,” Blood, vol. 97, no. 10, pp.
 S. Su, Y. Li, Y. Luo et al., “Proteinase-activated receptor 2
expression in breast cancer and its role in breast cancer cell
migration,” Oncogene, vol. 28, no. 34, pp. 3047–3057, 2009.
 M. E.W. Collier, C. Li, and C. Ettelaie, “Influence of exoge-
nous tissue factor on estrogen receptor alpha expression
in breast cancer cells: involvement of beta1-integrin, PAR2,
and mitogen-activated protein kinase activation,” Molecular
Cancer Research, vol. 6, no. 12, pp. 1807–1818, 2008.
6 Pathology Research International Download full-text
 S. Even-Ram, B. Uziely, P. Cohen et al., “Thrombin receptor
overexpression in malignant and physiological invasion pro-
cesses,” Nature Medicine, vol. 4, no. 8, pp. 909–914, 1998.
 S. Grisaru-Granovsky, Z. Salah, M. Maoz, D. Pruss, U.
Beller, and R. Bar-Shavit, “Differential expression of Protease
activated receptor 1 (Par1) and pY397FAK in benign and
malignant human ovarian tissue samples,” International Jour-
nal of Cancer, vol. 113, no. 3, pp. 372–378, 2005.
 S. C. Even-Ram, S. Grisaru-Granovsky, D. Pruss et al., “The
pattern of expression of protease-activated receptors (PARs)
during early trophoblast development,” Journal of Pathology,
vol. 200, no. 1, pp. 47–52, 2003.
 A. Boire, L. Covic, A. Agarwal, S. Jacques, S. Sherifi, and A.
Kuliopulos, “PAR1 is a matrix metalloprotease-1 receptor that
promotes invasion and tumorigenesis of breast cancer cells,”
Cell, vol. 120, no. 3, pp. 303–313, 2005.
 M. A. Booden, L. B. Eckert, C. J. Der, and J. Trejo, “Persistent
signaling by dysregulated thrombin receptor trafficking pro-
motes breast carcinoma cell invasion,” Molecular and Cellular
Biology, vol. 24, no. 5, pp. 1990–1999, 2004.
 C. Tellez, M. McCarty, M. Ruiz, and M. Bar-Eli, “Loss of
activator protein-2alpha results in overexpression of protease-
activated receptor-1 and correlates with the malignant phe-
notype of human melanoma,” Journal of Biological Chemistry,
vol. 278, no. 47, pp. 46632–46642, 2003.
 Z. Salah, M. Maoz, G. Pizov, and R. Bar-Shavit, “Transcrip-
tional regulation of human protease-activated receptor 1:
a role for the early growth response-1 protein in prostate
cancer,” Cancer Research, vol. 67, no. 20, pp. 9835–9843, 2007.
 Z. Salah, S. Haupt, M. Maoz et al., “p53 controls hPar1
function and expression,” Oncogene, vol. 27, no. 54, pp. 6866–
 R. Saban, M. R. D’Andrea, P. Andrade-Gordon et al., “Manda-
tory role of proteinase-activated receptor 1 in experimental
bladder inflammation,” BMC Physiology, vol. 7, p. 4, 2007.
 X. Su, E. Camerer, J. R. Hamilton, S. R. Coughlin, and M.
A. Matthay, “Protease-activated receptor-2 activation induces
acute lung inflammation by neuropeptide-dependent mecha-
nisms,” Journal of Immunology, vol. 175, no. 4, pp. 2598–2605,
 J. Trejo and S. R. Coughlin, “The cytoplasmic tails of protease-
activated receptor-1 and substance P receptor specify sorting
to lysosomes versus recycling,” Journal of Biological Chemistry,
vol. 274, no. 4, pp. 2216–2224, 1999.
 L. Hein, K. Ishii, S. R. Coughlin, and B. K. Kobilka, “Intra-
cellular targeting and trafficking of thrombin receptors. A
novel mechanism for resensitization of a G protein-coupled
receptor,” Journal of Biological Chemistry, vol. 269, no. 44, pp.
 I. Cohen, M. Maoz, H. Turm et al., “Etk/Bmx regulates
proteinase-activated-receptor1 (PAR1) in breast cancer inva-
sion: signaling partners, hierarchy and physiological signifi-
cance,” Plos One, vol. 5, no. 6, Article ID e11135, 2010.
 C. T. Griffin, Y. Srinivasan, Y. W. Zheng, W. Huang, and
S. R. Coughlin, “A role for thrombin receptor signaling in
endothelial cells during embryonic development,” Science, vol.
293, no. 5535, pp. 1666–1670, 2001.
 A. J. Connolly, H. Lshihara, M. L. Kahn, R. V. Farese, and S.
R. Coughlin, “Role of the thrombin receptor an development
pp. 516–519, 1996.
 Y. Qiu and H. J. Kung, “Signaling network of the Btk family
kinases,” Oncogene, vol. 19, no. 49, pp. 5651–5661, 2000.
 Y. Qiu, D. Robinson, T. G. Pretlow, and H. J. Kung, “Etk/Bmx,
a tyrosine kinase with a pleckstrin-homology domain, is
an effector of phosphatidylinositol 3?-kinase and is involved
in interleukin 6-induced neuroendocrine differentiation of
prostate cancer cells,” Proceedings of the National Academy of
 Y. T. Tsai, YI. H. Su, S. S. Fang et al., “Etk, a Btk family
tyrosine kinase, mediates cellular transformation by linking
src to STAT3 activation,” Molecular and Cellular Biology, vol.
20, no. 6, pp. 2043–2054, 2000.
 E. Yang, A. Boire, A. Agarwal et al., “Blockade of PAR1
pathways in breast cancer cells and suppresses tumor survival
and metastasis,” Cancer Research, vol. 69, no. 15, pp. 6223–