Kohashi, K., Oda, Y., Yamamoto, H., Tamiya, S.,
Takahira, T., Takahashi, Y., Tajiri, T., Taguchi, T.,
Suita, S., and Tsuneyoshi, M. (2008). J. Cancer
Res. Clin. Oncol. 134, 1097–1103.
Kuang, S., Kuroda, K., Le Grand, F., and Rudnicki,
M.A. (2007). Cell 129, 999–1010.
Rubin, B.P., Nishijo, K., Chen, H.-I.H., Yi, X.,
Scheutze, D.P., Pal, R., Prajapati, S.I., Abraham,
J., Arenkiel, B.R., Chen, Q.-R.,et al. (2011). Cancer
Cell 19, this issue, 177–191.
Stratton, M.R., Fisher, C., Gusterson, B.A., and
Cooper, C.S. (1989). Cancer Res. 49, 6324–6327.
Tiffin, N., Williams, R.D., Shipley, J., and Pritchard-
Jones, K. (2003). Br. J. Cancer 89, 327–332.
Wang, H., Garzon, R., Sun, H., Ladner, K.J., Singh,
R., Dahlman, J., Cheng, A., Hall, B.M., Qualman,
S.J., Chandler, D.S., et al. (2008). Cancer Cell 14,
One NOTCH Further: Jagged 1 in Bone Metastasis
Jianning Tao,1Ayelet Erez,1and Brendan Lee1,2,*
1Department of Molecular and Human Genetics
2Howard Hughes Medical Institute
Baylor College of Medicine, Houston, Texas 77030, USA
The outgrowth of metastatic cells to bone depends on the interaction between multiple intrinsic and host
factors. In this issue of Cancer Cell, Sethi and colleagues report Notch signaling in bone cells as responsible
for promoting this outgrowth and provide evidence for a beneficial treatment effect of NOTCH inhibitors.
Metastasis, the last and most devastating
stage of tumor progression, remains the
cause of 90% death in cancer patients.
The development of metastases requires
a series of sequential rate-limiting steps
through which malignant tumor cells
from the primary site invade into blood
and lymphatic vasculature, survive in the
circulation, lodge at distant organs, and
outgrow. The revised ‘‘seed and soil’’
theory, originally proposed by Steven
Paget a century ago, hypothesizes that
the outcome of metastasis depends on
crosstalk between predetermined cancer
cells (the ‘‘seeds’’) and specific organ
microenvironments (the ‘‘soil’’), which
release homeostatic factors (Fidler and
Poste, 2008). In the soil organ, the seed
cancer cells can enter either latent phase
or outgrowth phase.
habitat of the hematopoietic stem cell as
well as the most common metastatic site
for breast cancer. Recently, a positive
feedback loop causing a ‘‘vicious cycle’’
has been identified, in which the out-
growth phase of the bone metastases is
determined by a bidirectional interaction
between the cancer cells and the bone
et al., 2003; Zhang et al., 2009).
patients are affected by bone metastasis,
manifested by skeletal-related events
such as severe bone pain and patholog-
ical fractures (Mundy, 2002). There is
much evidence that metastatic cancer
modeling through stimulation of both
osteoclasts that resorb bone and osteo-
blasts that deposit bone, and the net
outcome of lesions depends on the
relative contribution of each cell type. In
multiple growth factors, including trans-
forming growth factor b (TGFb) and
insulin-like growth factor 1 (IGF1), are
released from the degraded bone matrix.
TGFb and IGF1 both enhance the growth
of cancer cells and stimulate them to
(PTHrP), connective tissue growth factor
(CTGF), and interleukin 11 (IL11). PTHrP
clastogenic factors receptor activator of
nuclear factor-kB ligand (RANKL) and os-
whereas CTGF mediates both angiogen-
esis and invasion (Massague, 2008).
Additional cells, such as bone borrow-
derived stromal, endothelial, and hemato-
poietic cells, have all been shown to
rometastases and the production of
prometastatic factors (Joyce and Pollard,
2009). On the other hand, the processes
by which metastatic cancer cells directly
communicate with various types of cells
in the bone and bone marrow remains
an enigma in the field of bone metastasis.
Undoubtedly, fully answering such ques-
tions will provide the insight necessary
for the development of effective therapies
against bone metastasis.
Sethi et al. (2011) now provide both
experimental and preclinical evidence
that the Notch ligand Jagged1 plays
a critical role in the promotion of bone
metastatic outgrowth of breast cancer.
Using a bioinformatic approach that
correlates the gene expression pattern
of Notch signaling pathway components
targets) to bone metastasis, the authors
Jagged1, which was highly correlated
with human breast cancer metastases
to bone. To investigate the functional
role of Jagged1 in the development of
bone metastasis, the authors applied
two different types of Jagged1-express-
ing human breast cancer cell lines in
a xenograft mouse model. In the strongly
bone tropic cell lines with high levels of
Jagged1 expression, stable knockdown
of Jagged1 resulted in a reduction of
Cancer Cell 19, February 15, 2011 ª2011 Elsevier Inc.
tumor outgrowth of bone lesions, while in
the weakly bone tropic cell lines with low
expression of Jagged1, overexpression
of Jagged1 resulted in a significant
lesions. In both cases, there was no
change in cells’ proliferative and invasive
abilities, either in culture or as primary
necessary and sufficient to promote os-
teolytic bone metastasis.
TGFb is a central component of the
found that Jagged1 is one of ten core
genes that respond to TGFb, among
them IL11 and CTGF, which have been
previously identified as targets of TGFb
in metastasis (Kang et al., 2003). Indeed,
Jagged1 mRNA and protein levels in
cultured breast cancer cells significantly
increased in the presence of excess
active TGFb, mimicking the in vivo patho-
logical environment of osteolytic bone
metastases. To confirm this finding, Sethi
et al. (2011) applied pharmacological
inhibitors and an inducible Smad4 knock-
down system to block TGFb signaling
in vitro and in vivo, verifying that Jagged1
is an essential target of the TGF-b
pathway. Notably, Jagged1 overexpres-
sion restored the ability of Smad4 knock-
down cells to generate osteolytic bone
Jagged1 as a critical component in this
The authors observed severe osteolytic
bone metastases, which is attributable
to elevated osteoclastic activity. The
classic view suggests that metastasized
cancer cells indirectly regulate osteo-
sion of osteoblast-derived RANKL and
OPG (Mundy, 2002). Augmented RANKL
to OPG ratio is responsible for one of the
tion. Using a co-culture system that
contains both tumor cells and preosteo-
clastic cells, Sethi et al. (2011) demon-
strated that Jagged1-expressing cells
directly interact with preosteoclast to
increase thisactivitybyaccelerating oste-
oclast differentiation and maturation. In
contrast, the authors show that RANKL
and OPG levels are not changed when
osteoblasts are co-cultured with tumor
cells. Furthermore, the pharmacological
inhibitor of Notch signaling suppresses
Although this finding still needs to be
confirmed in vivo, this new mechanism
processes that underlie osteolytic bone
A persistent outgrowth of bone metas-
tases likely depends on many intrinsic
factors in the bone microenvironment
that promote outgrowth of metastatic
breast cancer, the authors performed
a series of elegant experiments using
a co-culture system that contains both
tumor cells and osteoblasts, the major
secretory cell of the skeletal system.
They first found that cultured tumor cells
acquired a growth advantage via acti-
vating Notch pathway in osteoblasts.
Figure 1. TheRoleofJagged1inthe‘‘ViciousCycle’’ofBreastCancerOsteolyticMetastases
Theupper panelschematicallydescribes thestepsrequiredfortumorprogression from cancerinitiationto
osteolytic metastases. After cancer initiation, the cells invade the blood vessel, home in the bone marrow,
‘‘vicious cycle,’’ which is the focus of the article by Sethi et al (2011). Here, the breast cancer cells
expressing Jagged1 directly interact with osteoblasts and preosteoclasts, resulting inthe release of active
IL6and TGFb,respectively, promoting thetumoroutgrowth.Thispositivefeedback/crosstalk between the
bone macrometastases. The inhibitors of Notch signaling, IL6, and TGFbeta, which have been extensively
used in this study, are demonstrated in this figure, as they might be beneficial clinically in the future.
Cancer Cell 19, February 15, 2011 ª2011 Elsevier Inc.
Next, they performed microarray profiling Download full-text
of the co-cultured osteoblasts and found
IL6, a proproliferative cytokine involved in
bone metastasis (Ara et al., 2009), among
the upregulated genes. Using a combina-
tion of strategies, the authors nicely
demonstrate that IL6 is a downstream
target of the Jagged1-Notch-Rbpj-Hey1
to promote tumor proliferation.
Notch signaling is a central pathway for
embryonic development and is also
involved in the pathogenesis of skeletal
diseases (Tao et al., 2010). Hence, it is
not unexpected that pharmacological
signaling will have a beneficial effect in
tumor cells and bone cells, as well as in
the mouse model. However, a remaining
clinically relevant question is whether
these inhibitors can inhibit the outgrowth
of bone metastasis in patients. Another
important area that is critical for clinical
translation would be the effect of Notch
inhibition in other components of the skel-
etal system (Engin et. al., 2008), as well as
elsewhere in the body.
In summary, the work presented by
Sethi et al. (2011) identified a novel ‘‘seed
and soil’’ crosstalk mediated by the
network that promotes the outgrowth of
bone metastasis. The knowledge gained
in this study contributes to our under-
standing of the pathogenesis of bone
metastases and aids in finding therapy
against it. In addition, it opens doors for
many remaining unanswered questions.
Does Jagged1 activate Notch signaling
in other bone marrow-residing cells?
How does Notch pathway interplay with
Will the use of Notch inhibitors together
with inhibitors of IL6 or/and TGFb be
synergistic in halting bone metastasis?
Ara, T., Song, L., Shimada, H., Keshelava, N., Rus-
sell, H.V., Metelitsa, L.S., Groshen, S.G., Seeger,
R.C., and DeClerck, Y.A. (2009). Cancer Res. 69,
Engin, F., Yao, Z., Yang, T., Zhou, G., Bertin, T.,
R.E., et al. (2008). Nat. Med. 14, 299–305.
Fidler, I.J., and Poste, G. (2008). Lancet Oncol. 9,
Joyce, J.A., and Pollard, J.W. (2009). Nat. Rev.
Cancer 9, 239–252.
Kang, Y., Siegel, P.M., Shu, W., Drobnjak, M., Ka-
konen, S.M., Cordon-Cardo, C., Guise, T.A., and
Massague, J. (2003). Cancer Cell 3, 537–549.
Massague, J. (2008). Cell 134, 215–230.
G.R.(2002).Nat. Rev. Cancer2,
Sethi, N., Dai,X., Winter, C.G.,and Kang, Y. (2011).
Cancer Cell 19, this issue, 192–205.
Tao, J., Chen, S., and Lee, B. (2010). Ann. N Y
Acad. Sci. 1192, 257–268.
Zhang, X.H., Wang, Q., Gerald, W., Hudis, C.A.,
Norton, L., Smid, M., Foekens, J.A., and Mas-
sague, J. (2009). Cancer Cell 16, 67–78.
HiJAKing the Methylosome
in Myeloproliferative Disorders
Radek C. Skoda1,* and Ju ¨rg Schwaller1
1Department of Biomedicine, University Hospital Basel, Hebelstrasse 20, 4031 Basel, Switzerland
JAK2 gain-of-function mutations have been shown to cause myeloproliferative neoplasms. In this issue of
Cancer Cell, Liu et al. (2011) demonstrate that these JAK2 mutants, but not wild-type JAK2, directly phos-
phorylate PRMT5 and inhibit its arginine methyltransferase activity, establishing a link between mutant
JAK2 and histone arginine methylation.
pressed intracellular tyrosine kinase that
associates with the cytoplasmic domains
of hematopoietic cytokine receptors and
becomes activated upon these receptors
binding to their cognate ligands. Acti-
in the majority of patients with myelo-
proliferative neoplasms (MPN), which
cell diseases characterized by increased
proliferation of the erythroid, megakar-
yocytic, or myeloid lineages. The vast
majority of patients with MPN (about
80%) carry mutations in JAK2 codon
This mutated JAK2-V617F is an activated
tyrosine kinase that renders hemato-
poietic cells hypersensitive for signals
from upstream cytokine receptors (Epo,
Tpo, and G-CSF) and phosphorylates
(STAT3 and STAT5), as well as other
a minority of patients (less than 3% of
MPNs), JAK2 mutations have been found
in codon 539 (e.g., JAK2-K539L) or in
neighboring codons, leading to a variant
numbers of erythroid cells. The mutant
JAK2 proteins were shown to activate
proliferation, inhibit apoptosis, and inter-
fere with genome stability. However, all
Cancer Cell 19, February 15, 2011 ª2011 Elsevier Inc.