Oncogene (2013), 1–11
& 2013 Macmillan Publishers Limited All rights reserved 0950-9232/13
www.nature.com/onc
ORIGINAL ARTICLE
Infiltrating bone marrow mesenchymal stem cells increase prostate
cancer stem cell population and metastatic ability via secreting
cytokines to suppress androgen receptor signaling
J Luo1,3, S Ok Lee1,3, L Liang1, C-K Huang1, L Li1, S Wen1 and C Chang1,2
Although the contribution of the bone marrow mesenchymal stem cells (BM-MSCs) in cancer progression is emerging, their potential
roles in prostate cancer (PCa) remain unclear. Here, we showed that PCa cells could recruit BM-MSCs and consequently the metastatic
ability of PCa cells was increased. We also found that the increased metastatic ability of PCa cells could be due to the increased PCa
stem cell population. Mechanism dissection studies found that the upregulation of Chemokine ligand 5 (CCL5) expression in BM-MSCs
and PCa cells, after MSCs infiltrated into the PCa cells, subsequently downregulated androgen receptor (AR) signaling, which was due
to inhibition of AR nuclear translocation. Interruption of such signaling led to suppression of the BM-MSCs-induced PCa stem cell
population increase and thereby inhibited the metastatic ability of PCa cells. The PCa stem cell increase then led to the upregulation of
matrix metalloproteinase 9, ZEB-1, CD133 and CXCR4 molecules, and enhanced the metastatic ability of PCa cells. Therefore, we
conclude that the BM-MSCs-mediated increased metastatic ability of PCa cells can be due to the PCa stem cell increase via alteration
of the CCL5–AR signaling pathway. Together, these results uncover the important roles of BM-MSCs as key components in the prostate
tumor microenvironment to promote PCa metastasis and may provide a new potential target to suppress PCa metastasis by blocking
BM-MSCs infiltration into PCa.
Oncogene advance online publication, 24 June 2013; doi:10.1038/onc.2013.233
Keywords: bone marrow mesenchymal stem cells; CCL5; androgen receptor; prostate cancer stem cells
INTRODUCTION
Several kinds of cells in the tumor microenvironment (TME), including
macrophages, endothelial cells and fibroblasts, contribute to prostate
cancer (PCa) progression.1–3 Several studies showed that the bone
marrow-derived mesenchymal stem cell (BM-MSCs) also contributed
to various tumor progressions.4–7 However, BM-MSCs have been
shown to have both beneficial and harmful effects. Transplantation of
BM-MSCs into injured tissues exhibited promising effects,8–10 and they
also serve as therapeutic tools to effectively carry target molecules
into cancer cells.11–13 In contrast, several reports have suggested BMMSCs affect cancer progression, angiogenesis and metastasis.4–7
Importantly, no solid data have been reported to conclude whether
the BM-MSCs can affect PCa progression.
We investigated the contribution of BM-MSCs in PCa metastasis
using in vitro and in vivo approaches, and revealed that BM-MSCs
could be recruited to tumor sites and then promote PCa stem cells
population and metastatic ability. We also dissected potential
mechanisms that might be involved in the modulation of
androgen receptor (AR) signaling.
RESULTS
PCa cells have the capacity to recruit the circulating BM-MSCs
Recent reports demonstrated that circulating BM-MSCs could
migrate to various tumor sites.14,15 To investigate whether
circulating BM-MSCs could be recruited into the prostate tumor
1
site, we isolated primary BM-MSCs from mice. Flow cytometric
analysis results showed expression of the BM-MSCs markers
(Supplementary Table 1) and their ability to differentiate into
multilineage cells (Supplementary Figure 1A). We performed
Boyden chamber migration assays using these cells (Figure 1a).
After 24 h incubation, the PCa cells (LNCaP, C4-2 and CWR22Rv1),
but not the normal prostate epithelial (RWPE-1) cells, significantly
induced BM-MSCs migration (Figure 1b). We also found BM-MSCs
have higher expressions of matrix metalloproteinase 9 (MMP9),
which is a key molecule in modulation of cell motility,16 upon
coculture with PCa cells, but not with normal prostate RWPE-1 cells
(Figure 1c). These studies were further confirmed in CWR22Rv1
orthotopic xenograft mouse model. The green fluorescent protein
(GFP)-labeled BM-MSCs that had been isolated from GFP mice were
injected into the CWR22Rv1 xenografted mice tail veins after
tumors develop. As shown in Figure 1d, GFP signals were detected
in prostate tumors, but not in normal prostate tissues when
analyzed by the in vivo imaging system (IVIS). Immunohistochemical staining also showed high numbers of GFP-positively stained
cells in prostate tumors tissues (Figure 1e), but not in normal
prostate tissues. These results imply that BM-MSCs can migrate into
prostate tumor sites, but not into the normal prostate.
Proliferation of the BM-MSCs recruited by PCa cells is accelerated
Interestingly, we found that BM-MSCs growth was promoted
following PCa cells’ conditioned media (CM) treatment (Figure 1f).
George Whipple Lab for Cancer Research, Departments of Pathology, Urology, Radiation Oncology, and The Wilmot Cancer Center, University of Rochester Medical Center, Rochester,
NY, USA and 2Sex Hormone Research Center, China Medical University and Hospital, Taichung, Taiwan. Correspondence: Dr C Chang, George Whipple Lab for Cancer Research,
Departments of Pathology, Urology, Radiation Oncology, and The Wilmot Cancer Center, University of Rochester Medical Center, 601 Elmwood Avenue, Rochester, NY 14642, USA.
E-mail: chang@urmc.rochester.edu
3
These authors contributed equally to this work.
Received 24 October 2012; revised 22 March 2013; accepted 25 March 2013
BM-MSCs increase stem cells PCa population and metastatic ability
J Luo et al
2
Figure 1. BM-MSCs migration to prostate tumors. (a) The cartoon demonstrating BM-MSCs migration assay. The mouse primary BM-MSCs
(1 106) were placed in the upper chamber of the transwell plates (8 mm membrane) whereas PCa cells (normal prostate epithelial cells, RWPE, as
control) were placed in the lower chamber. (b) The BM-MSCs migration assay result. The LNCaP, C4-2 and CWR22Rv1 cells (RWPE cells as control)
were assayed. After 24 h incubation, the cells migrated to the bottom chamber were visualized by staining. Quantification is shown at right.
(c) Quantitative real-time PCR analysis of MMP9 mRNA expression. The primary BM-MSCs were treated with conditioned media (CM) of PCa cells
(LNCaP, C4-2 and CWR22Rv1) or normal RWPE cells for 48 h, total RNAs were extracted from the BM-MSCs and used for analysis. (d) GFP-labeled
BM-MSCs recruitment in normal prostate and prostate tumors in CWR22Rv1-xenografted mice. After tumors develop, the GFP-BM-MSCs (isolated
from the GFP mice) were injected through tail vein, and 3 days after GFP-BM-MSCs injection, green fluorescence was monitored using the IVIS.
(e) Immunohistochemical (IHC) staining of GFP signal in tumor tissues obtained from the CWR22Rv1-xenografted mice using GFP antibody as
arrows point out. (f) MTT assay result of the primary BM-MSCs upon coculture with CM from PCa cells. The primary BM-MSCs (1 105) were
placed in 24-well culture plates and treated with either PCa CM or the control medium and cell growth was analyzed by MTT assay. (g) 5-bromo2’-deoxyuridine (BrdU) labeling assay. The primary BM-MSCs were treated with either PCa CM or control medium for 48 h, and the cells were
labeled with BrdU for detection by IHC staining, according to the manufacturer’s protocol. Quantification is shown at right. Result values are
presented as means±s.d. Statistical analyses in b, c, f and g were done by two-tailed Student’s t-test. *Po0.05; **Po0.01.
Consistently, we detected higher numbers of the 5-bromo-2’deoxyuridine-stained BM-MSCs cells upon PCa cells CM treatment
(Figure 1g), suggesting that PCa cells could promote BM-MSCs
proliferation when recruited to tumor sites. This is important as there
are limited numbers of circulating BM-MSCs; accelerated proliferation by PCa cells might be favorable for exerting their action.
The recruited BM-MSCs influence PCa cell invasion
We next studied the influence of the infiltrated BM-MSCs on
in vitro invasion ability of PCa cells and in vivo metastasis. We
found the in vitro invasion abilities of PCa cells were significantly
increased upon mouse primary BM-MSCs (Figures 2b and c) and
human BM-MSCs (Figure 2d) coculture.
To confirm the above in vitro cell lines results in in vivo mice
studies, we developed a CWR22Rv1 orthotopic xenograft mouse
model. For monitoring metastasis, luciferase-incorporated
CWR22Rv1 (luc-CWR22Rv1) cells were developed and implanted,
either alone or as the mixture of luc-CWR22Rv1 cells and primary
BM-MSCs (CWR22Rv1: BM-MSCs, 10:1) for xenografts. After 3 weeks,
we started to use the IVIS system to monitor the luminescence
signal of the tumors once a week (3, 4, 5 and 6 weeks). At week 5,
Oncogene (2013), 1 – 11
we found that the two groups of mice developed similar-sized
tumors, but the CWR22Rv1-/BM-MSCs-co-implanted mice showed
higher luminescence signals at proximal lymph nodes and distant
organs compared with control mice (Figure 2e and Supplementary
Figure 2A). We confirmed higher metastasis in the co-implantation
group after killing (Figures 2f–g). The luciferase staining results of
tissues obtained from diaphragms of the co-implanted group
confirmed the origin of the metastasized tumors (Figure 2h). We also
found higher MMP9-positively stained cells in prostate tumor tissues
of the co-implanted group compared with control mice (Figure 2i).
Detection of tumor cells in the ascitic fluids have been reported to
be an important indication of tumor metastasis.17 We found ascitic
fluids (6 out of 10 mice, 60%) in most of the co-implanted group,
whereas only a few control mice had ascites (3 out of 20 mice, 15%).
The luminescence signal in the ascites cells indicated
they originated from the injected CWR22Rv1 cells (Supplementary
Figure 2B).
BM-MSCs coculture/co-implantation increased PCa stem cell
population
Interestingly, when we cocultured PCa cells with BM-MSCs, we
found some PCa cells grew as floating spheres, which is known to
& 2013 Macmillan Publishers Limited
BM-MSCs increase stem cells PCa population and metastatic ability
J Luo et al
3
Figure 2. BM-MSCs coculture-mediated increase in metastatic ability of PCa cells. (a) The cartoon describing the invasion assay. For coculture
of two types of cells, transwell plates with 0.4 mm membrane were used, whereas for invasion assay transwell plates with 8 mm membrane, precoated with 20% Matrigel, were used. (b) Invasion assay result. The C4-2 cells (1 105) were cocultured with the mouse primary BM-MSCs for 3
days and used for invasion assays. The invaded cells after 72 h incubation were stained with toluidine blue, and positively stained cells were
counted from five random areas. Quantitation is shown at right. (c) Invasion assay results using the LNCaP, C81 or CWR22Rv1 PCa cells.
(d) Invasion assay of the C4-2 cells upon human primary BM-MSCs coculture. Quantitation is shown at right. (e) Metastases analyses in the
CWR22Rv1 xenograft mouse model. The luc-CWR22Rv1 cells were implanted into the APs of the 8-week-old nude mice, either alone (control
group) or as a mixture with the mouse primary BM-MSCs cells (10:1, CWR22Rv1: BM-MSCs). Luminescence was detected using the IVIS Imager.
Percentage of metastasis incidence is shown at right. (f ) The mice imaging shows dissected primary tumor (right) and metastatic tumors (left).
(g) Picture showing metastasized tumors in abdominal diaphragm and lymph nodes. Arrows indicate the metastasized tumors.
(h) Hematoxylin-eosin and luciferase immunohistochemical (IHC) staining of diaphragm. (i) MMP9 IHC staining of the tumor tissues of the
two groups of mice. Quantitation is shown at right. **Po0.01.
be the characteristic of stem cells.18 Therefore, we investigated the
influence of BM-MSCs on the PCa stem cell population increase by
flow cytometry and found that the BM-MSCs coculture led to
increased CD133 þ cell population19–21 (Figure 3a). The immunofluorescence staining results also demonstrated increased CD133 þ
cells in LNCaP and C4-2 cells upon BM-MSCs coculture (Figure 3b).
We also found increased expressions of stem cell markers, such as
CD133, OCT4 and SOX222,23 in PCa cells upon BM-MSCs coculture
(Figure 3c). We confirmed the above in vitro cell line studies
results in in vivo mice studies by showing tumor tissues of the
& 2013 Macmillan Publishers Limited
CWR22Rv1-/BM-MSCs-co-implanted group of mice had higher
numbers of CD133 þ and CD44 þ cells compared with control
mice (Figure 3d). We also performed the sphere formation assay,
which is the well-defined method to examine stem cell self-renewal
capacity.24–26 The sphere’s size and numbers were increased
when PCa cells were cocultured with mouse primary BM-MSCs
compared with control (PCa cells alone; Figure 3e), and with human
primary cultured BM-MSCs (Figure 3f). To test whether PCa
stem cells increase was due to the increase in their self-renewal
ability, we isolated the CD133 þ population of LNCaP and C4-2 cells
Oncogene (2013), 1 – 11
BM-MSCs increase stem cells PCa population and metastatic ability
J Luo et al
4
Figure 3. Stem cells were increased in PCa cells upon coculture with BM-MSCs. (a) Flow cytometric analysis of CD133 þ population in PCa cells
upon BM-MSCs coculture. The left panel shows the flow cytometric analysis and the right panel table shows analyzed percentages of the PCa
CD133 þ cells of the PCa cells after coculture with the primary BM-MSCs for 72 h. (b) CD133 immunofluorescence staining of the LNCaP and
C4-2 cells upon BM-MSCs coculture (media used as control). (c) Quantitative real-time PCR (qPCR) analysis results analyzing mRNA expressions
of the stem markers, CD133, OCT4 and SOX2 in PCa cells upon BM-MSCs coculture. (d) CD133 and CD44 immunohistochemical staining of the
tumor tissues obtained from the luc-CWR22Rv1-xenografted mice. Quantitation is shown at right. (e, f ) Sphere formation assay of (e) PCa cells,
with or without coculture with the mouse primary BM-MSCs (media used as control) for 5 days, (f ) PCa cells, with or without coculture with the
human primary BM-MSCs (media used as control) for 5 days. Cells were then mixed with Matrigel (1:1, v/v), plated in 24-well plates and
cultured for 7 days. Quantification is shown at right. (g) CD133 þ cells were isolated from LNCaP and C4-2 cell lines by magnetic sorting,
cocultured with either BM-MSCs or media (control). Sphere formation assay was then performed. (h) Invasion assay of the LNCaP, and C4-2
CD133 and CD133 þ cells. Quantification is shown at right. (i) qPCR analysis of the mRNA expressions of the metastasis-related genes in the
C4-2 CD133 and CD133 þ cells. (j) qPCR analysis of the metastasis-related genes expressions in C4-2 cells, with or without coculture with
the BM-MSCs (media as controls). *Po0.05, **Po0.01.
by magnetic sorting method using CD133 antibody, and sphereforming abilities were tested, with or without BM-MSCs coculture.
We found that BM-MSCs coculture increased the self-renewal ability
Oncogene (2013), 1 – 11
of CD133 þ stem cells (Figure 3g). This data indicated the BM-MSCs
effect in increasing stem cell population might be due to increasing
the self-renewal ability of the original stem cell population within
& 2013 Macmillan Publishers Limited
BM-MSCs increase stem cells PCa population and metastatic ability
J Luo et al
5
the parental PCa cells. Alternatively, it is also possible that BM-MSCs
may influence the transition of non-stem cells into stem cells.
Higher population of stem cells in PCa led to more metastatic
potential
We found higher invasion ability of CD133 þ cells from the LNCaP
cell line (Figure 3h) and the C4-2 cell line (Figure 3h) than the
CD133 cells, consistent with the previous reports.6,27,28
Expressions of the metastasis-associated genes, including ZEB-1,
Snail, CXCR4 and MMP9,29–31 were also shown higher in the C4-2
CD133 þ cells than in CD133 cells (Figure 3i). When we
cocultured C4-2 cells with BM-MSCs, we found increased
expressions of these genes (Figure 3j). These results suggest that
the BM-MSCs-mediated increased PCa stem cell population may
be responsible for the enhanced metastatic ability of PCa cells.
Downregulation of AR signaling in PCa cells is essential for BMMSCs action
AR is the most important molecule in mediating PCa progression,
and even in castration-resistant PCa the AR expression
persists,32 and its linkage and impact on PCa progression
continues.33 However, recent studies suggested that AR might
have differential roles depending on the individual cell type
within the PCa TME and tumor stage.34,35 When we investigated
BM-MSCs effect on AR signaling in LNCaP and C4-2 cells, we
found little change in AR expression levels (Figure 4a), but
significant suppression of luciferase activity (Figure 4b) and AR
downstream genes expressions, including prostate-specific antigen (PSA), TMPRSS2 and FKBP5 (Figure 4c). The decreased PSA
level was also detected in tumor tissues obtained from
CWR22Rv1-/BM-MSCs-co-implanted xenografted mice (Supplementary Figure 3A). We further found significantly decreased AR
nuclear translocation in PCa cells upon BM-MSCs coculture
(Figure 4d), which might be the mechanism by which BM-MSCs
downregulate the AR signaling in PCa cells.
On the basis of the recent paper by Lee et al.,36 AR had a
suppressor role in self-renewal of PCa stem/progenitor cells. We
were interested to know whether the AR downregulation is
essential in exerting BM-MSCs’ action promoting PCa stem cell
population increase. We used the PC3 cell line (PCa cells lacking
AR) and PC3AR9 cell line (PC3 cells stably transfected with
functional human natural AR),37 and cocultured both with BMMSCs to test whether we could observe PCa stem cell increase
under both AR conditions. We found significant changes in
sphere-forming ability (Figure 4e) and CD133 expression
(Figure 4f, left panel) of the PC3AR9 cells upon BM-MSCs coculture,
but failed to detect stem cells increase in the AR-negative PC3
cells. Similar results were obtained when we tested C4-2-AR-small
interfering RNA (siRNA, AR knockdown by lentiviral infection is
shown in Supplementary Figure 3B) and C4-2-scrambled control
cells. However, we found no significant changes in sphere-forming
ability (Figure 4g) and CD133 expressions (Figure 4f, right panel) in
the AR knocked-down C4-2-AR-siRNA cells upon BM-MSCs
coculture.
We confirmed the above in vitro results with in vivo mice studies.
The PC3 cell-derived orthotopic xenograft mouse model was
developed. Mice groups were implanted with either PC3 cells alone
or the mixture of PC3 cells and BM-MSCs. As shown in Figure 4h, we
found little changes in CD133 þ cell numbers in prostate tumor
tissues obtained from the two mice groups, suggesting that BMMSCs failed to increase the stem cell population when AR is absent.
We then tested whether we could observe the BM-MSCs effect
in increasing metastasis in AR-negative PC3 cells. As expected, the
invasion ability (Figure 4i) and MMP9 expression (Figure 4j) of PC3
cells were no longer promoted upon BM-MSCs coculture,
compared with increases in PC3AR9 cells. Similarly, enhanced
MMP9 expression (Figure 4j) and PCa cells invasion (Figure 4k)
& 2013 Macmillan Publishers Limited
upon BM-MSCs coculture were observed in C4-2-scrambled
control cells, but not in C4-2-AR-siRNA cells. When we compared
metastases in the two PC3-xenografted groups, similar metastases
were observed (Figure 4l), indicating no significant effect of BMMSCs co-implantation in inducing metastases. Tissue staining
results also demonstrated similar numbers of MMP9-positively
stained cells in tumor tissues obtained from the two mice groups
(Figure 4m).
Together, results from Figures 4a–m conclude that downregulation of AR signaling is important in triggering infiltrated BMMSCs effect to increase stem cell numbers and metastatic ability
of PCa cells.
Secreted CCL5 levels in BM-MSCs/PCa cells were increased after
BM-MSCs infiltrated into PCa cells
We next investigated the mechanism by which BM-MSCs suppress
AR signaling in PCa cells. BM-MSCs secrete various cytokines/
growth factors to mediate their immune-modulator actions,38,39
and affect tumorigenicity and angiogenesis of cancer.40 Using the
arrays to compare the cytokine levels in the CM obtained from
coculture of C4-2 cells/BM-MSCs vs C4-2 cell culture, we found the
secretions of IP-10 and CCL5 were most dramatically increased in
BM-MSCs upon PCa coculture (Figure 5a), while decreases in some
cytokines including interleukin-6 and interleukin-8 were observed.
CCL5 increase was observed to mediate the BM-MSCs action in
promoting breast cancer metastasis,5 so we determined to select
this molecule for further studies. We validated significantly
increased CCL5 mRNA expression in BM-MSCs upon coculture
with various PCa cells (Figure 5b). We also detected dramatic
increase of CCL5 in PCa cells (Figure 5c), suggesting that BM-MSCs
and PCa cells interact to influence CCL5 secretion in both cell
types. We found significantly increased CCL5 staining in tumor
tissues obtained from CWR22Rv1-/BM-MSCs-co-implanted xenografted mice compared with tissues of CWR22Rv1 cells-injected
mice (Figure 5d).
In addition, adding recombinant CCL5 led to increased CD133
and MMP9 mRNA expressions (Figure 5e), and recombinant CCL5
effects on increases in stem cell population and invasion ability
were also demonstrated in sphere formation (Figure 5f) and
invasion (Figure 5g) assays. We then found suppression of AR
luciferase activity (Figure 5h) and PSA level decrease (Figure 5i)
when recombinant CCL5 was added into various PCa cells.
Furthermore, we showed that recombinant CCL5 suppressed AR
nuclear accumulation (Figure 5j) and induced expressions of the
metastasis-associated genes, ZEB-1, Snail and CXCR4, (Figure 5k).
We also performed neutralizing experiments adding the CCL5neutralizing antibody into PCa cells and found significantly
suppressed BM-MSCs effect in increasing stem cell population
(Figure 5l) and invasion ability (Figure 5m). Effects of the CCL5
antibody on blocking the metastasis-associated gene expressions
were also demonstrated (Figure 5n and Supplementary Figure 3C).
These results demonstrate that CCL5 is a strong candidate
molecule to downregulate AR signaling, which is essential in
increasing stem cell population and metastatic ability of PCa cells.
We tested in which cell type (PCA or MSCs) the CCL5 has the
central role in promoting metastasis. When we cocultured the
CCL5 knocked-down C4-2 (C4-2-siCCL5) cells with BM-MSCs, there
was no longer increased invasion of C4-2-siCCL5 cells. We then
cocultured C4-2 cells with the CCL5 knocked-down BM-MSCs cells
and observed no increase in C4-2 cells invasion, either. The further
mechanism dissection also showed that expressions of the target
genes did not change significantly after coculture of these two
types of cells no matter in which cell type CCL5 expression was
knocked down. So, we can conclude that CCL5 increase is
essential in both types of cells (PCa and MSCs; Supplementary
Figure 3D and Figure 5o).
Oncogene (2013), 1 – 11
BM-MSCs increase stem cells PCa population and metastatic ability
J Luo et al
6
DISCUSSION
The BM-MSCs contribution in tumor growth and metastasis of
other cancers and their roles in differentiation into stromal cells
Oncogene (2013), 1 – 11
has been studied,4–7,41 but relatively little has been understood
about their role in PCa progression, especially its linkage to AR
signaling. We found that BM-MSCs could be recruited into the
& 2013 Macmillan Publishers Limited
BM-MSCs increase stem cells PCa population and metastatic ability
J Luo et al
7
prostate tumor site and lead to increase metastatic ability, possibly
via increasing PCa stem cell population. Previous studies also
suggested that the increase in stem cell population might lead to
enhanced PCa metastasis,6,27,28,42 which matched well with our
results showing higher invasion ability (Figure 3h) and higher
expressions of the metastasis-associated genes, including ZEB-1,
Snail, CXCR4 and MMP9 (Figure 3i) in the isolated CD133 þ stem
cells than in the CD133 cells.
We performed in vitro studies using both BM-MSCs CM and the
coculture system. Under these conditions, BM-MSCs influences are
expected to be through the paracrine action. We also used a
culture system by placing PCa cells into the plates that already
had BM-MSCs growing, so the two types of cells could contact and
interact with each other. Under this condition, PCa cells were
growing on the top of BM-MSCs, but PCa cells were easily
removed by pipetting as their contact was not tight, and then the
BM-MSCs were collected for further analyses. We observed similar
effects with both of these coculture systems. Therefore, we
concluded that the BM-MSCs effect was through the paracrine
effect mediated by cytokines they secrete no matter what
condition we use. We also tested whether BM-MSCs can be
differentiated into cancer-associated fibroblasts and we found no
cancer-associated fibroblasts-specific marker expression increases
in BM-MSCs upon PCa CM treatment. Accordingly, we focused on
the paracrine effects of BM-MSCs in these studies.
We used 1:10 ratio of BM-MSCs/PCa cells in both in vitro coculture
and in vivo mice studies. BM-MSCs circulate and are recruited into
the tumor site and directly interact with PCa cells, so injecting as a
mixture of PCa cells and BM-MSCs is considered clinically relavant.
The number of BM-MSCs used in these studies may be higher than
the real clinical case and somewhat exaggerated, but we used this
ratio based on a previous report on breast cancer.5
It can be speculated that our in vitro and in vivo study
conditions might be artificial as properties of BM-MSCs might be
altered after they arrive at the PCa TME. We showed BM-MSCs
migration into the prostate tumor site (not to the normal
prostate), so we believe that the cross talk between BM-MSCs
and cancer cells does occur. We also found that our in vivo results
were consistent with the in vitro coculture experimental data. For
example, we observed CCL5 increase and stem cell increase in the
BM-MSCs-co-implanted group mice, suggesting that the in vitro
BM-MSCs effect was mimicked in the mice studies.
We elucidated mechanisms by which BM-MSCs increase
metastasis and found that BM-MSCs trigger upregulation of
CCL5, downregulation of AR, PCa stem cell increase and
upregulation of CXCR4/ZEB-1 in PCa cells. We showed that the
BM-MSCs function is through their paracrine action by secreting
cytokines, and the most critical cytokine molecule for this action
was CCL5. Increased secretion of CCL5 and its contribution to
breast cancer metastasis has been suggested.5 We investigated
whether AR downregulation is the earlier event and whether this
environment favoring stem cells expansion or stem cell increase
occurs as a consequence of this downregulation of AR level in PCa
cells. We found that BM-MSCs increased PCa stem cell population
in the AR-positive PCa cell lines, but not in the AR-negative cell
lines, such as PC3, indicating that AR downregulation may be the
earlier event and is essential to trigger stem cell increase.
Considering the suppressor role of AR in self-renewal of PCa
stem cells,36 it seems that these two processes may cooperate. This
newly found CCL5 -AR-CXCR4/ZEB-1 signaling axis could be used
to develop future therapeutic approaches to battle PCa (Figure 6).
Here we demonstrated that the AR signal downregulation is
essential in exerting BM-MSCs action to mediate increased metastatic
ability of PCa cells. The downregulation of AR-promoted metastasis
may be a new challenging concept as all recent therapeutic
approaches target AR signaling and may explain why the current
therapy fails and results in more malignant metastatic PCa.
In a previous study, we demonstrated the suppressor role of AR
in PCa metastasis,34 and our studies on the TRAMP derivative of the
prostate epithelial-specific ARKO mice also demonstrated the
differential AR roles in PCa initiation (promoter) vs metastasis
(suppressor).35 It was also reported that the AR downregulation
could promote epithelial-mesenchymal transition, suggesting that
AR might function as a suppressor during the metastasis process.43
Moreover, PCa stem cells have higher metastatic ability and these
cells also lack AR expression.36 We could not observe BM-MSCs
effect in increasing the invasion ability when AR signal was absent
or low (in PC3 and C4-2ARsi cells) and our in vivo results showed no
significant difference in metastasis in the PC3 cells-implanted mice
and the PC3-/BM-MSCs-co-implanted mice. These results contrast
with the recently published result showing increased invasion
ability of the PC3 cells upon incubation with human MSCs in vitro.41
So, we believe that the future therapeutic strategies should be
focused on simultaneous targeting of AR-mediated PCa growth
and AR downregulation-mediated metastasis.
As targeting the promoter role and suppressor role of AR at the
same time is impossible, revealing the signal paradigm, and
especially unveiling the downstream molecules of the suppressor
role of AR, such as CXCR4/ZEB-1, will be of great significance for
exploiting these signals for development of therapeutic
approaches to block metastasis.
In conclusion, we showed the importance of BM-MSCs as one
component in the PCa TME and as an important contributor in
increasing PCa metastasis. It should be noted that BM-MSCs may
Figure 4. Downregulation of AR signal was important in exerting the BM-MSCs effect to increase stem cell population and the metastatic
ability of PCa cells. (a) Western blot analysis of AR expression in the LNCaP and C4-2 cells upon BM-MSCs coculture. Total cell extracts were
obtained from PCa cells, with or without BM-MSCs coculture (media as control), and AR expression was analyzed. (b) Luciferase assay result.
The C4-2 cells were cocultured with the mouse primary BM-MSCs for 72 h, transfected with MMTV-Luc at different dihydrotestosterone (DHT)
conditions (0, 1 and 10 nM), and the luciferase activity was measured after 48 h. (c) Quantitative real-time PCR (qPCR) analysis of the AR
downstream genes mRNA levels. PCa cells were cocultured with the BM-MSCs (media as control) for 72 h, total RNAs were extracted and
mRNA expressions of the AR downstream genes, PSA, TMPRSS2 and FKBP5, were analyzed. (d) Western blot analysis analyzing nuclear
translocation of the AR protein. The C4-2 cells were cocultured with the BM-MSCs (media as control), the cytosolic and nuclear extracts were
obtained, and the AR expressions in the two compartments were analyzed. (e) Sphere formation assay result of PC3/PC3AR9 cells with or
without BM-MSCs coculture. (f) qPCR analysis of CD133 mRNA expressions in PC3/PC3AR9 (left) and C4-2-AR-small interfering RNA (siRNA)/
scramble control cells (right). Cells were cocultured with the BM-MSCs and the CD133 mRNA expressions were analyzed. (g) Sphere formation
assay result of C4-2-AR-siRNA/scramble control cells with or without BM-MSCs coculture. (h) CD133 immunohistochemical (IHC) staining of the
tumor tissues obtained in the two PC3-xenografted mouse models, with or without co-implantation with BM-MSCs. Quantification is shown at
right. (i) Invasion assays of PC3/PC3AR9 cells. PCa cells (1 105) that had been cocultured with the BM-MSCs (media as control) for 72 h were
used in invasion assays. (j) qPCR analysis of MMP9 expression in PC3/PC3AR9 (left) and C4-2-AR-siRNA/scramble control cells (right). Cells were
cocultured with the BM-MSCs and the mRNA level of the MMP9 was analyzed. (k) Invasion assays of C4-2-AR-siRNA/scramble control cells with
or without BM-MSCs coculture. (l) Comparison of the metastasis incidence in the PC3 cell-xenografted mice, with or without co-implantation
with the BM-MSCs (left panel). Comparison of metastasis with diaphragm in the two mouse models (right panel). (m) MMP9 IHC staining of the
tumor tissues obtained in PC3 cell-xenografted mouse models, with or without co-implantation with the BM-MSCs. Quantitation is shown at
right. **Po0.01.
& 2013 Macmillan Publishers Limited
Oncogene (2013), 1 – 11
BM-MSCs increase stem cells PCa population and metastatic ability
J Luo et al
8
not be the only contributor to the increased metastatic ability of
PCa cells, but our studies clearly demonstrated their importance.
This information will add to our current understanding of the TME
in influencing PCa progression.
MATERIALS AND METHODS
Cell culture
LNCaP, C4-2, C81 and CWR22Rv1 cell lines were purchased from the
American Type Culture Collection (Manassas, VA, USA) and cultured in
Oncogene (2013), 1 – 11
Roswell Park Memorial Institute media 1640 with 10% fetal
bovine serum. Mouse primary BM-MSCs were isolated from wild-type
C57BL/6 mice and C57BL/6-Tg (UBC-GFP) mice (The Jackson Laboratory,
Bar Harbor, ME, USA) and cultured in Dulbecco’s modified Eagle
media with 15% fetal bovine serum, 1 nonessential amino acids
and 10 mM hydroxyethyl piperazine ethane sulfonic acid. Human BM-MSCs
were purchased from StemCell Technologies Inc. (Vancouver, BC,
Canada) and cultured in Human MesenCult Proliferation Kit (StemCell
Technologies Inc.). All cells were maintained in a humidified 5% CO2
environment at 37 1C.
& 2013 Macmillan Publishers Limited
BM-MSCs increase stem cells PCa population and metastatic ability
J Luo et al
9
Cell invasion assay
Luciferase assay
Six- (0.4 mm pore size) or 24-well (8 mm pore size) transwell plates (Corning,
Lowell, MA, USA) were used for coculture and invasion assay, respectively.
PCa cells were cocultured with BM-MSCs in transwell plates for 36–48 h. For
in vitro invasion assays, transwell plate membranes were pre-coated with
diluted Matrigel (20%; BD Biosciences, Sparks, MD, USA) and PCa cells (105
cells in serum-free medium) were plated in the upper chamber while 10%
serum-containing media placed in the lower chamber. After 36–48 h
incubation, cells invaded into the lower chamber were stained with
toluidine blue, and positively stained cells were counted. The cell numbers
were counted in six random fields. Quantitation indicates means of
triplicate repeats±s.e.m.
PCa cells were plated in 24-well plates and transfected with MMTV-luc
containing ARE sequence using Lipofectamine (Invitrogen). After transfection, regular media were added with various dihydrotestosterone
concentrations, 0 (ethanol as vehicle control), 1 and 10 nM, and incubated
for 48 h. pRL-TK was used as the internal control. Luciferase activity was
measured by Dual-Luciferase Assay (Promega, Madison, WI, USA) according
to the manufacturer’s manual.
Cytokine array
In vivo BM-MSCs recruitment assay
Male nude mice (6–8 weeks) were used in the studies. One group (six mice)
remained untreated, another group (six mice) were injected into the
anterior prostates with 1 106 CWR22Rv1 cells. After 3 weeks, when
CM was collected from PCa cells, BM-MSCs and BM-MSCs-PCa cell
coculture and used for cytokine array analyses. The levels of selected
panel of cytokines were determined using the Human Antibody Array kit
(Affymetrix, Santa Clara, CA, USA; MA6120), following the manufacturer’s
instructions.
RNA extraction and quantitative real-time PCR analysis
Total RNAs were isolated using Trizol reagent (Invitrogen, Grand Island, NY,
USA). One microgram of total RNA was subjected to reverse transcription
using Superscript III transcriptase (Invitrogen). Quantitative real-time PCR
was conducted using a Bio-Rad CFX96 system (Hercules, CA, USA) with SYBR
Green to determine the mRNA expression level of a gene of interest.
Expression levels were normalized to glyceraldehyde 3-phosphate dehydrogenase level.
Western blot analysis
Cells were lysed in radioimmunoprecipitation assay buffer and proteins
(20–40 mg) were separated on 8–10% SDS-polyacrylamide gel electrophoresis, and then transferred onto polyvinylidene difluoride membranes
(Millipore, Billerica, MA, USA). After blocking membranes, they were
incubated with primary antibodies, horseradish peroxidase-conjugated
secondary antibodies and visualized using ECL system (Thermo Fisher
Scientific, Rochester, NY, USA). AR, glyceraldehyde 3-phosphate dehydrogenase, tubulin, poly (ADP-ribose) polymerase and GFP antibodies were
from Santa Cruz Biotechnology, Inc (Santa Cruz, CA, USA). MMP9 and
CD133 antibodies were from Abcam (Cambridge, MA, USA).
Histology and immunohistochemistry
Tissues obtained were fixed in 10% (v/v) formaldehyde in phosphatebuffered saline, embedded in paraffin and cut into 5-mm sections. Prostate
tissue sections were deparaffinized in xylene solution and rehydrated and
immunostaining was performed. CCL5 antibody was from R&D (Minneapolis, MN, USA). MMP9, luciferase and CD133 antibodies were from Abcam.
Figure 6.
A cartoon summarizing the results.
Figure 5. Elevated CCL5 secretion in BM-MSCs was responsible for the downregulation of AR signal, the stem cell increase and the metastatic
ability increase of PCa cells. (a) The cytokine array analysis. The conditioned media (CM) collected from C4-2 cell culture, human primary BMMSCs culture and human primary BM-MSCs/C4-2 cell coculture were used for the analysis, according to the manufacturer’s instructions.
(b) Quantitative real-time PCR (qPCR) analysis of CCL5 mRNA expressions in BM-MSCs with or without coculture with PCa cells. (c) qPCR
analysis of CCL5 expression in PCa cells, with or without coculture with BM-MSCs. (d) CCL5 immunohistochemical staining in the tumor tissues
obtained from the CWR22Rv1-xenografted mice, with or without co-implantation with BM-MSCs. Quantitation is shown at right. (e) qPCR
analysis of CD133 and MMP9 mRNA expressions in PCa cells after 48 hr treatment with 10 ng/ml recombinant CCL5 (rCCL5). (f ) Sphere
formation assay of PCa cells, after treatment with rCCL5 as in e. (g) Invasion assay of PCa cells, after treatment with rCCL5 as in e. (h) Luciferase
assay result. C4-2 cells were transfected with MMTV-luc, treated with rCCL5 as in e at different dihydrotestosterone (DHT) concentrations, and
luciferase activity was measured after 48 h incubation. (i) qPCR analysis result of prostate-specific antigen (PSA) mRNA expression. The LNCaP
and C4-2 cells were treated with rCCL5 as in e and PSA mRNA expression level was analyzed. (j) After treating cells with rCCL5, cytosolic and
nuclear extracts were obtained, and AR expressions in cytosolic and nuclear compartments were examined. Tubulin and poly (ADP-ribose)
polymerase (PARP) were used to show purity of the cytosolic and nuclear extracts, respectively. (k) qPCR analysis of ZEB-1, Snail and CXCR4
expressions. C4-2 cells were treated with rCCL5 as in e and mRNA expressions of the three genes were analyzed. (l) The sphere formation assay
of C4-2 cells with or without coculture with BM-MSCs, in the absence and presence of 1 mg/ml CCL5 neutralization antibody (a-CCL5).
(m) The invasion assay of C4-2 cells, with or without coculture with BM-MSCs, in the absence and presence of 1 mg/ml a-CCL5. (n) Effect of aCCL5 on expressions of the stem cells/metastasis-related genes. The expressions of CD133, MMP9, PSA, ZEB-1, Snail and CXCR4 genes in C4-2
cells, with or without coculture with BM-MSCs, in the absence and presence of 1 mg/ml a-CCL5, were analyzed by qPCR analysis. (o) Invasion
assay using CCL5 knocked-down BM-MSCs and PCa cells. *Po0.05, **Po0.01.
& 2013 Macmillan Publishers Limited
Oncogene (2013), 1 – 11
BM-MSCs increase stem cells PCa population and metastatic ability
J Luo et al
10
tumors grew to palpable size, the primary GFP-labeled BM-MSCs (1 105)
that had been isolated from GFP mice were tail vein injected. Three days
later, mice were killed, and GFP signals in normal and tumor prostate
tissues were detected by IVIS (Caliper Life Sciences, Hopkinton, MA, USA)
and immunohistochemical staining performed. All animal studies were
performed under the supervision and guidelines of the University of
Rochester Medical Center Animal Care and Use Committee.
In vivo metastasis studies
The nude mice were from the Jackson Laboratory. CWR22Rv1 cells
were engineered to express luciferase reporter gene (PCDNA3.0-luciferase)
by stable transfection and the positive stable clones (luc-CWR22Rv1) were
selected and expanded in culture. Twenty control group mice (6–8 weeks)
were injected with luc-CWR22Rv1 cells (1 106, mixed with Matrigel, 1:1)
and 10 test group mice were co-injected with PCa cells combined with
mouse primary BM-MSCs (1 105). After 3 weeks, the mice were injected
with 150 mg/kg D-Luciferin, and metastasis in live mice was monitored
using IVIS at four different time points (3, 4, 5 and 6 weeks after injection).
After killing mice at 6 weeks, metastases were analyzed by staining and
IVIS. The peritoneal ascites were collected at the killing of mice and
luminescence was examined by IVIS.
Statistics
The data values were presented as the mean±s.d. Differences in mean
values between two groups were analyzed by two-tailed Student’s t-test.
Pp0.05 was considered statistically significant.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
ACKNOWLEDGEMENTS
We thank Karen Wolf for help with the manuscript preparation. This work was
supported by NIH Grants (CA122840 and CA256700), and Taiwan Department of
Health Clinical Trial and Research Center of Excellence (DOH102-TD-B-111-004) to
China Medical University, Taiwan.
REFERENCES
1 Zhu P, Baek SH, Bourk EM, Ohgi KA, Garcia-Bassets I, Sanjo H et al. Macrophage/
cancer cell interactions mediate hormone resistance by a nuclear receptor
derepression pathway. Cell 2006; 124: 615–629.
2 Zeng Y, Opeskin K, Goad J, Williams ED. Tumor-induced activation of
lymphatic endothelial cells via vascular endothelial growth factor receptor-2 is
critical for prostate cancer lymphatic metastasis. Cancer Res 2006; 66:
9566–9575.
3 Augsten M, Hagglof C, Olsson E, Stolz C, Tsagozis P, Levchenko T et al. CXCL14 is
an autocrine growth factor for fibroblasts and acts as a multi-modal stimulator of
prostate tumor growth. Proc Natl Acad Sci USA 2009; 106: 3414–3419.
4 Quante M, Tu SP, Tomita H, Gonda T, Wang SS, Takashi S et al. Bone marrowderived myofibroblasts contribute to the mesenchymal stem cell niche and
promote tumor growth. Cancer Cell 2011; 19: 257–272.
5 Karnoub AE, Dash AB, Vo AP, Sullivan A, Brooks MW, Bell GW et al. Mesenchymal
stem cells within tumour stroma promote breast cancer metastasis. Nature 2007;
449: 557–563.
6 Kobayashi A, Okuda H, Xing F, Pandey PR, Watabe M, Hirota S et al. Bone morphogenetic protein 7 in dormancy and metastasis of prostate cancer stem-like
cells in bone. J Exp Med 2011; 208: 2641–2655.
7 Raaijmakers MH, Mukherjee S, Guo S, Zhang S, Kobayashi T, Schoonmaker JA et al.
Bone progenitor dysfunction induces myelodysplasia and secondary leukaemia.
Nature 2010; 464: 852–857.
8 Park D, Spencer JA, Koh BI, Kobayashi T, Fujisaki J, Clemens TL et al. Endogenous
bone marrow MSCs are dynamic, fate-restricted participants in bone maintenance
and regeneration. Cell Stem Cell 2012; 10: 259–272.
9 Chong JJ, Chandrakanthan V, Xaymardan M, Asli NS, Li J, Ahmed I et al.
Adult cardiac-resident MSC-like stem cells with a proepicardial origin. Cell Stem
Cell 2011; 9: 527–540.
10 Lee CH, Shah B, Moioli EK, Mao JJ. CTGF directs fibroblast differentiation from
human mesenchymal stem/stromal cells and defines connective tissue healing in
a rodent injury model. J Clin Invest 2010; 120: 3340–3349.
Oncogene (2013), 1 – 11
11 Gao Y, Yao A, Zhang W, Lu S, Yu Y, Deng L et al. Human mesenchymal stem
cells overexpressing pigment epithelium-derived factor inhibit hepatocellular
carcinoma in nude mice. Oncogene 2010; 29: 2784–2794.
12 Yong RL, Shinojima N, Fueyo J, Gumin J, Vecil GG, Marini FC et al. Human
bone marrow-derived mesenchymal stem cells for intravascular delivery
of oncolytic adenovirus Delta24-RGD to human gliomas. Cancer Res 2009; 69:
8932–8940.
13 Loebinger MR, Eddaoudi A, Davies D, Janes SM. Mesenchymal stem cell delivery
of TRAIL can eliminate metastatic cancer. Cancer Res 2009; 69: 4134–4142.
14 Xu S, Menu E, De Becker A, Van Camp B, Vanderkerken K, Van Riet I. Bone marrowderived mesenchymal stromal cells are attracted by multiple myeloma cell-produced chemokine CCL25 and favor myeloma cell growth in vitro and in vivo. Stem
Cells 2012; 30: 266–279.
15 Goldstein RH, Reagan MR, Anderson K, Kaplan DL, Rosenblatt M. Human bone
marrow-derived MSCs can home to orthotopic breast cancer tumors and promote
bone metastasis. Cancer Res 2010; 70: 10044–10050.
16 Jodele S, Chantrain CF, Blavier L, Lutzko C, Crooks GM, Shimada H et al.
The contribution of bone marrow-derived cells to the tumor vasculature in
neuroblastoma is matrix metalloproteinase-9 dependent. Cancer Res 2005; 65:
3200–3208.
17 Charoenfuprasert S, Yang YY, Lee YC, Chao KC, Chu PY, Lai CR et al. Identification
of salt-inducible kinase 3 as a novel tumor antigen associated with tumorigenesis
of ovarian cancer. Oncogene 2011; 30: 3570–3584.
18 Rossi DJ, Jamieson CH, Weissman IL. Stems cells and the pathways to aging and
cancer. Cell 2008; 132: 681–696.
19 Richardson GD, Robson CN, Lang SH, Neal DE, Maitland NJ, Collins AT. CD133, a
novel marker for human prostatic epithelial stem cells. J Cell Sci 2004; 117(Pt 16):
3539–3545.
20 Trerotola M, Rathore S, Goel HL, Li J, Alberti S, Piantelli M et al. CD133, Trop-2 and
alpha2beta1 integrin surface receptors as markers of putative human prostate
cancer stem cells. Am J Transl Res 2010; 2: 135–144.
21 Yu X, Lin Y, Yan X, Tian Q, Li L, Lin EH. CD133, stem cells, and cancer stem cells:
myth or reality? Curr Colorectal Cancer Rep 2011; 7: 253–259.
22 Leis O, Eguiara A, Lopez-Arribillaga E, Alberdi MJ, Hernandez-Garcia S, Elorriaga K
et al. Sox2 expression in breast tumours and activation in breast cancer stem cells.
Oncogene 2012; 31: 1354–1365.
23 Kumar SM, Liu S, Lu H, Zhang H, Zhang PJ, Gimotty PA et al. Acquired cancer stem
cell phenotypes through Oct4-mediated dedifferentiation. Oncogene 2012; 31:
4898–4911.
24 Rybak AP, He L, Kapoor A, Cutz JC, Tang D. Characterization of sphere-propagating cells with stem-like properties from DU145 prostate cancer cells. Biochim
Biophys Acta 2011; 1813: 683–694.
25 Dubrovska A, Kim S, Salamone RJ, Walker JR, Maira SM, Garcia-Echeverria C et al.
The role of PTEN/Akt/PI3K signaling in the maintenance and viability of
prostate cancer stem-like cell populations. Proc Natl Acad Sci USA 2009; 106:
268–273.
26 Shi X, Gipp J, Bushman W. Anchorage-independent culture maintains prostate
stem cells. Dev Biol 2007; 312: 396–406.
27 Malanchi I, Santamaria-Martinez A, Susanto E, Peng H, Lehr HA, Delaloye JF et al.
Interactions between cancer stem cells and their niche govern metastatic colonization. Nature 2012; 481: 85–89.
28 Kleeberger W, Bova GS, Nielsen ME, Herawi M, Chuang AY, Epstein JI et al. Roles
for the stem cell associated intermediate filament Nestin in prostate cancer
migration and metastasis. Cancer Res 2007; 67: 9199–9206.
29 Mak P, Leav I, Pursell B, Bae D, Yang X, Taglienti CA et al. ERbeta impedes
prostate cancer EMT by destabilizing HIF-1alpha and inhibiting VEGF-mediated
snail nuclear localization: implications for Gleason grading. Cancer Cell 2010; 17:
319–332.
30 Li YM, Pan Y, Wei Y, Cheng X, Zhou BP, Tan M et al. Upregulation of CXCR4 is
essential for HER2-mediated tumor metastasis. Cancer Cell 2004; 6: 459–469.
31 Wegiel B, Bjartell A, Tuomela J, Dizeyi N, Tinzl M, Helczynski L et al.
Multiple cellular mechanisms related to cyclin A1 in prostate cancer invasion and
metastasis. J Natl Cancer Inst 2008; 100: 1022–1036.
32 Cai C, He HH, Chen S, Coleman I, Wang H, Fang Z et al. Androgen receptor
gene expression in prostate cancer is directly suppressed by the androgen
receptor through recruitment of lysine-specific demethylase 1. Cancer cell 2011;
20: 457–471.
33 Dayyani F, Gallick GE, Logothetis CJ, Corn PG. Novel therapies for metastatic
castrate-resistant prostate cancer. J Natl Cancer Inst 2011; 103: 1665–1675.
34 Niu Y, Altuwaijri S, Lai KP, Wu CT, Ricke WA, Messing EM et al. Androgen receptor
is a tumor suppressor and proliferator in prostate cancer. Proc Natl Acad Sci USA
2008; 105: 12182–12187.
35 Niu Y, Altuwaijri S, Yeh S, Lai KP, Yu S, Chuang KH et al. Targeting the stromal
androgen receptor in primary prostate tumors at earlier stages. Proc Natl Acad Sci
USA 2008; 105: 12188–12193.
& 2013 Macmillan Publishers Limited
BM-MSCs increase stem cells PCa population and metastatic ability
J Luo et al
11
36 Lee SO, Ma Z, Yeh CR, Luo J, Lin TH, Lai KP et al. New therapy targeting differential
androgen receptor signaling in prostate cancer stem/progenitor vs. non-stem/
progenitor cells. J Mol Cell Biol 2012; 5: 14–26.
37 Altuwaijri S, Wu CC, Niu YJ, Mizokami A, Chang HC, Chang C. Expression of human
AR cDNA driven by its own promoter results in mild promotion, but not suppression,
of growth in human prostate cancer PC-3 cells. Asian J Androl 2007; 9: 181–188.
38 Huang XP, Sun Z, Miyagi Y, McDonald Kinkaid H, Zhang L, Weisel RD et al. Differentiation of allogeneic mesenchymal stem cells induces immunogenicity and limits
their long-term benefits for myocardial repair. Circulation 2010; 122: 2419–2429.
39 Auletta JJ, Deans RJ, Bartholomew AM. Emerging roles for multipotent, bone
marrow-derived stromal cells in host defense. Blood 2012; 119: 1801–1809.
40 Liu S, Ginestier C, Ou SJ, Clouthier SG, Patel SH, Monville F et al. Breast cancer
stem cells are regulated by mesenchymal stem cells through cytokine networks.
Cancer Res 2011; 71: 614–624.
41 Ye H, Cheng J, Tang Y, Liu Z, Xu C, Liu Y et al. Human bone marrow-derived
mesenchymal stem cells produced TGFbeta contributes to progression and
metastasis of prostate cancer. Cancer Invest 2012; 30: 513–518.
42 Delude C. Tumorigenesis: testing ground for cancer stem cells. Nature 2011; 480:
S43–S45.
43 Sun Y, Wang BE, Leong KG, Yue P, Li L, Jhunjhunwala S et al. Androgen deprivation causes epithelial-mesenchymal transition in the prostate: implications for
androgen-deprivation therapy. Cancer Res 2012; 72: 527–536.
Supplementary Information accompanies this paper on the Oncogene website (http://www.nature.com/onc)
& 2013 Macmillan Publishers Limited
Oncogene (2013), 1 – 11