1
CCL5/CCR5 axis induces vascular endothelial growth factor-mediated tumor
2
angiogenesis in human osteosarcoma microenvironment
3
4
Shih-Wei Wang1, Shih-Chia Liu2, Hui-Lung Sun3, Te-Yang Huang2, Chia-Han Chan2,
5
Chen-Yu Yang2, Hung-I Yeh1,4, Yuan-Li Huang5, Yu-Min Lin6,7*
6
and Chih-Hsin Tang8,9,5*
7
8
1
Department of Medicine, Mackay Medical College, New Taipei City, Taiwan
9
2
Department of Orthopaedics, Mackay Memorial Hospital, Taipei, Taiwan,
10
3
Department of Molecular Virology, Immunology and Mediccal Genetics, Ohio state
11
University, Columbus, OH, USA
12
4
Department of Internal Medicine, Mackay Memorial Hospital, Taipei, Taiwan
13
5
Department of Biotechnology, College of Health Science, Asia University, Taichung,
14
Taiwan
15
6
Institute of Medicine, Chung Shan Medical University, Taichung, Taiwan
16
7
Department of Orthopedic Surgery, Taichung Veterans General Hospital, Taichung,
17
Taiwan
18
8
19
Taiwan
20
9
21
Taiwan
Graduate Institute of Basic Medical Science, China Medical University, Taichung,
Department of Pharmacology, School of Medicine, China Medical University, Taichung,
22
23
*Address correspondence to:
24
Chih-Hsin Tang, PhD
25
Graduate Institute of Basic Medical Science, China Medical University, Taichung,
26
Taiwan. No. 91, Hsueh-Shih Road, Taichung, Taiwan
27
Tel: 886-4-22052121-7726; Fax: 886-4-22333641; E-mail: chtang@mail.cmu.edu.tw
28
Or
29
Yu-Min Lin, MD, PhD; E-mail: yuminlin@gmail.com
30
1
31
Abstract
32
Chemokines modulate angiogenesis and metastasis that dictate cancer
33
development in tumor microenvironment. Osteosarcoma is the most frequent bone
34
tumor and is characterized by a high metastatic potential. Chemokine CCL5
35
(previously called RANTES) has been reported to facilitate tumor progression and
36
metastasis. However, the crosstalk between chemokine CCL5 and vascular
37
endothelial growth factor (VEGF) as well as tumor angiogenesis in human
38
osteosarcoma microenvironment has not been well explored. In this study, we found
39
that CCL5 increased VEGF expression and production in human osteosarcoma cells.
40
The conditioned medium (CM) from CCL5-treated osteosarcoma cells significantly
41
induced tube formation and migration of human endothelial progenitor cells.
42
Pretreatment of cells with CCR5 antibody or transfection with CCR5 specific siRNA
43
blocked CCL5-induced VEGF expression and angiogenesis. CCL5/CCR5 axis
44
demonstrably activated protein kinase C (PKC), c-Src, and hypoxia-inducible
45
factor-1 alpha (HIF-1) signaling cascades to induce VEGF-dependent angiogenesis.
46
Furthermore, knockdown of CCL5 suppressed VEGF expression and attenuated
47
osteosarcoma CM-induced angiogenesis in vitro and in vivo. CCL5 knockdown
48
dramatically abolished tumor growth and angiogenesis in the osteosarcoma xenograft
49
animal model. Importantly, we demonstrated that the expression of CCL5 and VEGF
50
were correlated with tumor stage according the immunohistochemistry analysis of
51
human osteosarcoma tissues. Taken together, our findings provide evidence that
52
CCL5/CCR5 axis promotes VEGF-dependent tumor angiogenesis in human
53
osteosarcoma microenvironment through PKC/c-Src/HIF-1 signaling pathway.
54
CCL5 may represent a potential therapeutic target against human osteosarcoma.
55
2
56
Summary
57
This study reveals that CCL5/CCR5 axis promotes VEGF expression in human
58
osteosarcoma
59
PKC/c-Src/HIF-1 signaling pathway. This is the first indication that chemokine
60
CCL5 promotes tumor angiogenesis by VEGF production in human cancer cells.
cells,
and
contributes
to
tumor
angiogenesis
through
61
62
63
Running title: CCL5 promotes VEGF-dependent tumor angiogenesis
64
Keywords: CCL5; VEGF; Tumor Angiogenesis; Osteosarcoma
65
3
66
Introduction
67
Tumor growth and dissemination is the result of dynamic interactions between
68
tumor cells themselves, and also with components of the tumor microenvironment,
69
including endothelial cells, smooth muscle cells, stromal fibroblasts and
70
tumor-associated macrophages [1]. The interaction between tumor cells and their
71
microenvironment is dramatically facilitated by the soluble inflammatory mediators,
72
such as cytokines and chemokines [2]. Chemokines and chemokine receptors are
73
instrumental players in inflammation, immunosurveillance, and cancer progression.
74
The chemokine–chemokine receptor axis has been indicated to regulate tumor growth,
75
angiogenesis, invasion, and tissue-specific metastasis [3,4]. Regulated upon
76
Activation Normal T cell Expressed and Secreted (RANTES, CCL5) is an
77
inflammatory chemokine, primarily identified as potent inducers of leukocyte motility
78
[5]. CCL5 mediates its biological activities through activation of G protein–coupled
79
receptors, CCR1, CCR3, or CCR5, and binds to glycosaminoglycans. CCL5 is
80
associated with chronic inflammatory diseases such as rheumatoid arthritis,
81
inflammatory bowel disease, and cancer. CCL5 can be expressed and secreted either
82
by cancer cells or by nonmalignant stromal cells in the microenvironment [6].
83
Autocrine secretion of CCL5 controls the migration and invasion of cancer cells in
84
vitro [7]. CCL5 expression has been implicated in the correlation with a variety of
85
human tumors, including prostate, cervical, colon and lung cancers [8-11]. The most
86
striking findings thus far have been with breast cancer. Several investigations have
87
indicated that CCL5 was found to be highly expressed in breast cancer patients and
88
that expression levels correlated with advanced disease course [12,13].
89
Osteosarcoma is a high-grade malignant bone tumor that most frequently affects
90
children and young adolescents [14-16]. The chemotherapies regimens are not fully
91
effective, and osteosarcoma shows a predilection for metastasis to the distant organs.
4
92
Recurrence usually occurs as pulmonary metastases or, less frequently, metastases to
93
distant bones or as a local recurrence [1]. Angiogenesis has been demonstrated to
94
facilitate metastatic osteosarcoma within the tumor microenvironment [17].
95
Angiogenesis is a key event in the tumor growth and metastatic cascade of human
96
cancers. In angiogenic processes, endothelial cells must undergo migration,
97
proliferation, and tube formation to form tumor neovessel [18]. Subsequent studies
98
indicate that tumor angiogenesis is also supported by the mobilization and functional
99
incorporation of endothelial progenitor cells (EPCs) [19]. Recently, EPCs have been
100
proposed to mediate early tumor growth and late tumor metastasis by intervening with
101
the angiogenic switch [20]. Vascular endothelial growth factor (VEGF), one of the
102
major angiogenic factors, that is released by most types of cancer stimulates
103
angiogenesis to the tumour tissue. The induction of VEGF expression in tumors can
104
be caused by numerous environmental factors such as hypoxia, growth factors and
105
chemokines [21]. Compelling evidences indicate that chemokines exert the
106
angiogenesis directly or as a consequence of leukocyte trafficking, and/or the
107
induction of growth factor expression such as that of VEGF in the microenvironment
108
[3]. In recent years, VEGF antagonists significantly attenuate tumor angiogenesis and
109
successfully controls the disease progression of many cancers. Therefore, targeting
110
VEGF may represent a potential approach for preventing osteosarcoma angiogenesis
111
and metastasis [22].
112
Hypoxia-inducible factor 1 (HIF-1), a transcription factor that is critical for
113
tumor adaptation to microenvironmental stimuli, regulates the blood supply through
114
its effects on the expression of VEGF [23]. HIF-1 exists as a heterodimeric complex
115
consisting of HIF-1 and HIF-1. The availability of HIF-1 is determined primarily
116
by HIF-1, which is regulated at the protein level in an oxygen-sensitive manner, in
117
contrast to HIF-1, which is stably expressed. HIF-1α is expressed in many human
5
118
tumors and renders cells able to survive and grow under hypoxic condition [24].
119
During
120
Hippel-Lindau-dependent ubiquitin-proteasome pathway [25]. Under hypoxia,
121
however, HIF-1 is not degraded and accumulates to form transcriptionally active
122
complexes with HIF-1. The HIF-1 and HIF-1 complex can then bind to hypoxia
123
response elements (HREs) located in gene promoters to regulate transcription of
124
VEGF that enhance cellular adaptation to hypoxia [26]. In addition, a wide range of
125
growth-promoting stimuli, cytokines, and chemokines also induce modest HIF-1
126
accumulation [27]. Several signaling pathways such as phosphoinositide 3-kinase
127
(PI3K), mammalian target of rapamycin (mTOR), integrin-linked kinase (ILK), and
128
protein kinase C (PKC) have been shown to induce VEGF expression via
129
HIF-1-dependent mechanism [28-30].
normoxia,
HIF-1
is
efficiently
degraded
through
the
von
130
Previous studies have confirmed that CCL5 is associated with the disease status
131
and outcomes of cancers [6]. VEGF is the most potent angiogenic mediator that is
132
essential for angiogenesis and tumor growth [22]. However, the regulation of CCL5
133
on VEGF expression in human cancer cells is mostly unknown. The mechanism
134
underlying CCL5-mediated VEGF expression and tumor angiogenesis in human
135
osteosarcoma microenvironment is also unclear. In this study, we investigated the
136
relationship of CCL5 with VEGF expression and tumor angiogenesis, and further
137
elucidated its mechanism of action in human osteosarcoma.
138
6
139
Materials and Methods
140
Materials
141
The recombinant human CCL5 was purchased from PeproTech (Rocky Hill, NJ).
142
ON-TARGETplus siRNA of CCR5, PKC, c-Src, HIF-1, and control were
143
purchased from Dharmacon Research (Lafayette, CO). Anti-mouse and anti-rabbit
144
IgG-conjugated horseradish peroxidase, rabbit polyclonal antibodies specific for
145
HIF-1, HIF-1, β-actin, CD31, control shRNA plasmid, and CCL5 shRNA plasmid
146
were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Rabbit polyclonal
147
antibodies specific for CCL5 and VEGF antibodies were purchased from Abcam
148
(Cambridge, MA). Recombinant human VEGF, mouse monoclonal antibody specific
149
for CCR5 and VEGF were purchased from R&D Systems (Minneapolis, MN).
150
Rottlerin, PP2, and HIF-1 inhibitor were purchased from Calbiochem (San Diego,
151
CA). The pHRE-luciferase construct was provided from Dr. W.M. Fu (National
152
Taiwan University, Taiwan). pSV-β-galactosidase vector and luciferase assay kit were
153
purchased from Promega (Madison, WI). All other chemicals were purchased from
154
Sigma–Aldrich (St. Louis, MO).
155
Cell culture
156
The human osteosarcoma cell lines (U2OS and MG63) were purchased from the
157
American Type Cell Culture Collection (Manassas, VA). Cells were maintained in
158
RPMI-1640 medium containing 10% FBS, L-glutamine, penicillin and streptomycin
159
at 37°C with 5% CO2.
160
Preparation of conditioned medium (CM)
161
In the series of experiments, osteosarcoma cells were treated with CCL5 alone
162
for 24 h, or pretreated with pharmacological inhibitors, including CCR5 mAb,
163
rottlerin, PP2 and HIF-1 inhibitor (These inhibitor did not affect cell viability in
164
osteosarcoma cells; supplementary data S1), or transfected with siRNA of CCR5,
7
165
PKC, c-Src and HIF-1(Because the specificity of pharmacological inhibitors is
166
controversial, therefore, the specificity siRNAs were used) followed by stimulation
167
with CCL5 for 24 h. After treatment, cells were washed and changed to serum-free
168
medium. CM was then collected 2 days after the change of medium and stored at
169
−80°C until use.
170
Isolation and cultivation of endothelial progenitor cells (EPCs)
171
Ethical approval was granted by the Institutional Review Board of Mackay
172
Medical College, New Taipei City, Taiwan (reference number: P1000002). Informed
173
consent was obtained from healthy donors before the collection of peripheral blood
174
(80 mL). The peripheral blood mononuclear cells (PBMCs) were fractionated from
175
other blood components by centrifugation on Ficoll-Paque plus (Amersham
176
Biosciences, Uppala, Sweden) according to the manufacturer’s instructions.
177
CD34-positive progenitor cells were obtained from the isolated PBMCs using CD34
178
MicroBead kit and MACS Cell Separation System (Miltenyi Biotec, Bergisch
179
Gladbach, Germany). The maintenance and characterization of CD34-positive EPCs
180
were performed as described previously [31,32]. Briefly, human CD34-positive EPCs
181
were maintained and propagated in MV2 complete medium consisting of MV2 basal
182
medium and growth supplement (PromoCell, Heidelberg, Germany), supplied 20%
183
defined FBS (HyClone, Logan, UT). Cells were seeded onto 1% gelatin-coated plastic
184
ware and maintained in humidified air containing 5% CO2 at 37°C for further
185
treatment. Experiments were performed by using cells from passages 1 to 3.
186
Tube formation assay
187
Matrigel (BD Biosciences; Bedford, MA) was dissolved at 4°C overnight, and
188
48-well plates were prepared with 150 μL Matrigel in each well after coating and
189
incubating at 37°C for 30 min. EPCs (5 × 104 cells) in 100 μL cultured media which
190
included 50% MV2 complete medium and 50% osteosarcoma cells conditioned
8
191
medium. After 16 h of incubation at 37°C, EPCs tube formation was taken with the
192
inverted phase contrast microscope. Tube branches and total tube length were
193
calculated using MacBiophotonics Image J software.
194
Migration assay
195
Cell migration assay was performed using Transwell chambers with 8.0m pore
196
size (Coring, Coring, NY). EPCs (1 x 104 cells/well) were seeded onto the upper
197
chamber with MV2 complete medium, then incubated in the bottom chamber with
198
50% MV2 complete medium and 50% osteosarcoma cells conditioned medium. The
199
plates were incubated for 24 h at 37°C in 5% CO2, and cells were fixed in 4 %
200
formaldehyde solution for 15 min and stained with 0.05 % crystal violet in PBS for 15
201
min. Cells on the upper side of the filters were removed with cotton-tipped swabs, and
202
the filters were washed with PBS. Cell migration was quantified by counting the
203
number of stained cells in 10 random fields with the inverted phase contrast
204
microscope and photographed.
205
Quantitative real-time PCR
206
Total RNA was extracted from osteosarcoma cells as described previously [33].
207
Twog of total RNA was reverse transcribed into cDNA using oligo (dT) primer.
208
The quantitative real time PCR (qPCR) analysis was carried out using Taqman®
209
one-step PCR Master Mix (Applied Biosystems, CA). cDNA templates (2l) were
210
added per 25-μl reaction with sequence-specific primers and Taqman® probes.
211
Sequences for all target gene primers and probes were purchased commercially
212
(GAPDH was used as internal control) (Applied Biosystems, CA). The qPCR assays
213
were carried out in triplicate on a StepOnePlus sequence detection system. The
214
cycling conditions were 10-min polymerase activation at 95 °C followed by 40 cycles
215
at 95 °C for 15 s and 60 °C for 60 s. The threshold was set above the non-template
216
control background and within the linear phase of target gene amplification to
9
217
calculate the cycle number at which the transcript was detected (denoted CT).
218
Measurement of VEGF production
219
Human osteosarcoma cells were cultured in 24-well plates. After reaching
220
confluence, cells were changed to serum-free medium. Cells were then treated with
221
CCL5 alone for 24 h, or pretreated with pharmacological inhibitors or transfected
222
with specific siRNA followed by stimulation with CCL5 for 24 h. After treatment, the
223
medium was removed and stored at -80°C. Then, VEGF in the medium was
224
determined using VEGF ELISA kit (Cayman Chemical, Ann Arbor, MI) according to
225
the manufacturer's protocol.
226
Kinase activity assay
227
PKCδ and c-Src activities were assessed with a PKC Kinase Activity Assay Kit
228
(Assay Designs, Ann Arbor, MI) and a c-Src Kinase Activity Assay Kit (Abnova,
229
Taipei, Taiwan). The kinase activity kits are based on a solid phase ELISA that uses a
230
specific synthetic peptide as substrate for PKCδ or c-Src, and a polyclonal antibody
231
that recognizes the phosphorylated form of the substrate.
232
Western blot analysis
233
Cells were lysed with lysis buffer as described previously [34]. Cell
234
homogenates were diluted with loading buffer and boiled for 5 min for detecting
235
phosphorylation, and protein expression. Total protein was determined and equal
236
amounts of protein were separated by 8–12% SDS–PAGE and immunoblotted with
237
specific primary antibodies. Horseradish peroxidase-conjugated secondary antibodies
238
(Santa Cruz Biotechnology) were used, and the signal detected using an enhanced
239
chemiluminescence detection kit (Amersham, Buckinghamshire, UK).
240
Chromatin immunoprecipitation assay
241
Chromatin immunoprecipitation (ChIP) analysis was performed as described
242
previously [35]. DNA immunoprecipitated by anti-HIF-1 Ab was purified. The DNA
10
243
was then extracted with phenol-chloroform. The purified DNA pellet was subjected to
244
PCR. PCR products were then resolved by 1.5% agarose gel electrophoresis and
245
visualized by UV light. The primers 5’-CCTTTGGGTTTTGCCAGA-3’ and
246
5’-CCAAGTTTGTGGAGCTGA-3’ were utilized to amplify across the VEGF
247
promoter region.
248
Transfection and reporter assay
249
Cells were transfected with siRNA or HRE-Luc reporter plasmid using
250
Lipofectamine 2000 (Invitrogen, Carlsbad, CA) according to the manufacturer's
251
recommendations. After 24 h transfection, cells were pretreated with inhibitors for 30
252
min and then CCL5 or vehicle was added for 24 h. Cell extracts were then prepared,
253
luciferase and β-galactosidase activities were measured. The establishment of CCL5
254
shRNA stably transfected cells, see Supplementary material.
255
Establishment of CCL5 shRNA stably transfected cells
256
CCL5 shRNA or control shRNA plasmids are transfected into cancer cells with
257
Lipofetamine 2000 transfection reagent. Twenty-four hours after transfection, stable
258
transfectants are selected in puromycin (Life Technologies) at a concentration of 10
259
μg/mL. Thereafter, the selection medium is replaced every 3 days. After 2 weeks of
260
selection in puromycin, clones of resistant cells are isolated.
261
Chick chorioallantoic membrane (CAM) assay
262
Angiogenic activity was determined using a CAM assay as described previously
263
[36]. Fertilized chicken eggs were incubated at 38°C in an 80% humidified
264
atmosphere. On day 8, CM from U2OS/control-shRNA or U2OS/CCL5-shRNA cells
265
(2 × 104 cells) deposited in the center of the chorioallantoic. CAM results were
266
analyzed on the fourth day. Chorioallantoid membranes were collected for microscopy
267
and photographic documentation. Angiogenesis was quantified by counting the
268
number of blood vessel branch; at least 10 viable embryos were tested for each
11
269
treatment. All animal works were done in accordance with a protocol approved by the
270
China Medical University (Taichung, Taiwan) Institutional Animal Care and Use
271
Committee.
272
In vivo Matrigel plug assay
273
Thirty male nude mice (4-week of age) were used and randomized into three
274
groups: PBS (control), U2OS/control-shRNA, or U2OS/CCL5-shRNA resuspended
275
with Matrigel. Mice were subcutaneously injected with 0.4 ml matrigel containing 2 ×
276
105 cells. On day 7, Matrigel plugs were excised, partly were fixed with 4% formalin,
277
embedded in paraffin, and subsequently processed for CD31 staining using
278
immunohistochemistry. They were also partly used for measuring the extent of blood
279
vessel formation by hemoglobin assay.
280
In vivo tumor xenograft model
281
Twenty male nude mice (4-week of age) were used and randomized into two
282
groups. For experimental cells growing exponentially, each implanted into 10 nude
283
mice by subcutaneous injection, 1 × 106 cells (U2OS/control-shRNA or
284
U2OS/CCL5-shRNA) were resuspended in 0.1 ml of serum-free RPMI-1640 and
285
injected into the right flank. After 4 weeks, the mice were sacrificed and tumors were
286
excised for CD31 staining or hemoglobin assay. The mice were observed daily and the
287
body weights were monitored for toxicity. The tumor volume and weight were also
288
measured during this month.
289
Hemoglobin assay
290
All the sponges (Matrigel plugs or tumors) were processed for measuring blood
291
vessel formation. Briefly, the amount concentration of hemoglobin in the vessels that
292
have invaded the Matrigel or tumor were determined with Drabkin’s reagent
293
(Sigma-Aldrich) according to manufacturer instructions. Homogenized in 1 ml of
294
RIPA lysis buffer, and after centrifuged at 1000 rpm, 20ul of supernatants were added
12
295
to 100 ul of Darkin’s solution. The mixture was allowed to stand 30 min at room
296
temperature, and then readings were taken at 540 nm in a spectrophotometer. The
297
results are expressed in milligrams per milliliter.
298
Immunohistochemistry (IHC)
299
The human osteosarcoma tissue array was purchased from Biomax (Rockville,
300
MD, 15 cases for normal cartilage, 13 cases for type IIb osteosarcoma, and 12 cases
301
for type IIIb osteosarcoma). The tissues were placed on glass slides, rehydrated and
302
incubated in 3% hydrogen peroxide to block the endogenous peroxidase activity. After
303
trypsinization, sections were blocked by incubation in 3% bovine serum albumin in
304
PBS. The primary antibody anti-human CCL5 or VEGF was applied to the slides at a
305
dilution of 1:50 and incubated at 4°C overnight. After being washed three times in
306
PBS, the samples were treated with goat anti-mouse IgG biotin-labeled secondary
307
antibodies at a dilution of 1:50. Bound antibodies were detected with an ABC kit
308
(Vector Laboratories, Burlingame, CA). The slides were stained with chromogen
309
diaminobenzidine, washed, counterstained with Delafield's hematoxylin, dehydrated,
310
treated with xylene, and mounted. The sum of the intensity and percentage score was
311
used as the final staining scores (0 to 5).
312
Statistics
313
The values given are means ± S.E.M. The significance of difference between the
314
experimental groups and controls was assessed by Student’s t test. The difference is
315
significant if the p value is <0.05.
316
13
317
Results
318
CCL5/CCR5 axis promotes VEGF expression and angiogenesis
319
Previous studies have shown that CCL5 directly promotes angiogenesis of
320
endothelial cells and chemotaxis of human EPCs [37,38]. However, it is still not
321
well-recognized whether CCL5 stimulates tumor angiogenesis by VEGF production
322
in human cancer cells, especially osteosarcoma cells. We first applied CCL5 to human
323
osteosarcoma cell line and determined the expression of VEGF. The results showed
324
that CCL5 concentration-dependently increased VEGF mRNA expression and
325
production in human osteosarcoma cells (Fig. 1A&B). The process of angiogenesis
326
mainly involves endothelial cells proliferation, migration, and tube formation to form
327
new blood vessels [18]. We then used an in vitro EPCs model to examine whether
328
CCL5-dependent VEGF expression induced angiogenesis. As shown in Fig. 1C, the
329
capillary tube like structure was facilitated by the conditioned medium (CM) from
330
CCL5-treated osteosarcoma cells. Furthermore, CM from CCL5-treated osteosarcoma
331
cells enhanced tube formation and migration of EPCs in a concentration-dependent
332
manner (Fig. 1D&E). To elucidate CCL5-dependent VEGF plays an important role in
333
angiogenesis, the VEGF antibody was used. We found that post-incubation of
334
CCL5-treated CM with VEGF antibody significantly prevented CCL5-induced
335
migration and tube formation of EPCs. These results indicate that CCL5-dependent
336
VEGF expression promotes angiogenesis in vitro.
337
It has been reported that CCL5 increases cell motility and angiogenesis through
338
interaction with its specific receptor CCR5 [34,37]. Our previous report indicated that
339
osteosarcoma cells expressed CCR5 receptor [34]. Pretreatment of cells with CCR5
340
mAb markedly inhibited CCL5-induced VEGF expression. In addition, transfection of
341
cells with CCR5 siRNA also abolished CCL5-induced VEGF expression (Fig. 1F&G).
342
Furthermore, CM from osteosarcoma cells demonstrated that CCR5 mAb and CCR5
14
343
siRNA both significantly reduced CCL5-mediated migration and tube formation of
344
EPCs (Fig. 1H&I). These data suggest that CCL5 and CCR5 interaction promotes
345
angiogenesis by VEGF expression.
346
PKC/c-Src signaling pathway is involved in CCL5-mediated VEGF expression and
347
angiogenesis
348
The activation of PKC has been reported to increase tumor angiogenesis and
349
VEGF production [30]. To examine whether PKC is involved in CCL5-mediated
350
VEGF expression and angiogenesis, the PKCinhibitor rottlerin was used.
351
Pretreatment of cells with rottlerin reduced CCL5-induced the expression of VEGF.
352
Additionally, transfection with PKC siRNA specifically reduced CCL5-induced
353
VEGF expression (Fig. 2A&B). Moreover, CCL5-mediated EPCs migration and tube
354
formation were also diminished by treatment with PKCinhibitor rottlerin and PKC
355
siRNA (Fig. 2C&D). Incubation of U2OS cells with CCL5 increased PKC kinase
356
activity in a time-dependent manner (Fig. 2E). Pretreatment of cells with CCR5 mAb
357
significantly blocked CCL5-induced PKC kinase activity (Fig. 2F). These results
358
demonstrate that CCL5 and CCR5 interaction promotes VEGF expression and
359
angiogenesis via PKC-dependent pathway.
360
Several studies have demonstrated that PKC mediated the activation of c-Src
361
kinase [39,40]. We next examined whether PKC/c-Src pathway is involved in the
362
CCL5-induced VEGF expression and angiogenesis. As shown in Fig. 3A&B,
363
CCL5-induced VEGF expression was markedly attenuated by pretreatment with c-Src
364
inhibitor PP2 and transfection with c-Src siRNA. Furthermore, CCL5-mediated EPCs
365
migration and tube formation were also profoundly suppressed by treatment with
366
c-Srcinhibitor
367
concentration-dependently increased the activity of c-Src kinase, and both CCR5
368
mAb and PKCinhibitor rottlerin significantly inhibited CCL5-activeted c-Src kinase
PP2
and
c-Src
siRNA (Fig.
3C&D).
Importantly,
CCL5
15
369
activity in U2OS cells (Fig. 3E&F). Taken together, CCL5/CCR5 axis appears to act
370
through PKC/c-Src signaling pathway to enhance angiogenesis and VEGF
371
expression in human osteosarcoma cells.
372
CCL5/CCR5 axis induces HIF-1 activation for VEGF expression and angiogenesis
373
Hypoxia-inducible factor 1 (HIF-1), a pivotal transcription factor that is critical
374
for VEGF expression in tumor microenvironment [27]. We therefore sought to
375
investigate whether HIF-1 activation was involved in CCL5-induced VEGF
376
expression in human osteosarcoma cells. We found that pretreatment with HIF-1
377
inhibitor and transfection with HIF-1 siRNA both markedly antagonized
378
CCL5-induced VEGF expression (Fig. 4A&B). CCL5-mediated EPCs migration and
379
tube formation were significantly suppressed by treatment with HIF-1 inhibitor and
380
HIF-1 siRNA (Fig. 4C&D). The results from Western blot indicated that CCL5
381
significantly increased protein level of HIF-1 time-dependently (Fig. 4E). However,
382
CCL5 did not affect the mRNA level of HIF-1 using q-PCR analysis (Fig. 4F).
383
These results suggest that CCL5 increases the accumulation of HIF-1, possibly by
384
enhancing HIF-1 protein stability, which subsequently promotes VEGF expression
385
and angiogenesis.
386
It has been reported that c-Src induced HIF-1α protein accumulation via a
387
general increase in cap-dependent translation [41]. We further explored whether
388
PKC/c-Src signaling pathway was involved in CCL5-induced HIF-1 activation in
389
human osteosarcoma cells. We performed ChIP assay to examine the DNA binding
390
activity of HIF-1 in CCL5-treated cells. As shown in Fig. 4G, the in vivo binding of
391
HIF-1 to the HRE element of the VEGF promoter occurred after CCL5 stimulation.
392
The binding of HIF-1 to the HRE element by CCL5 was markedly attenuated by
393
CCR5 mAb, rottlerin, and PP2. Moreover, HIF-1 activation was also evaluated
394
using HRE-luciferase assay. We found that CCL5-induced HRE-luciferase activity
16
395
was significantly reduced by pretreatment with CCR5 mAb, rottlerin, and PP2 (Fig.
396
4H). Based upon these finding, we suggest that CCR5/PKC/c-Src signaling pathway
397
is involved in CCL5-induced HIF-1activation.
398
Knockdown of CCL5 impairs angiogenesis in vitro and in vivo
399
To confirm the CCL5 mediated VEGF-dependent angiogenesis in human
400
osteosarcoma cells, the CCL5-shRNA expression cells was established. The
401
expression
402
U2OS/CCL5-shRNA cells (Fig. 5A). We found that CM from U2OS/control-shRNA
403
cells increased tube formation and migration of EPCs. Knockdown of CCL5
404
significantly suppressed CM-mediated EPCs migration and tube formation (Fig.
405
5B&C). In addition, the effect of CCL5 on angiogenesis in vivo was evaluated by
406
using the in vivo model of chick embryo CAM assay. CM from U2OS/control-shRNA
407
cells increased angiogenesis in CAM was clearly observed. In contrast, CCL5-shRNA
408
markedly reduced angiogenesis in CAM (Fig. 5D). We next performed the Matrigel
409
implant assay in mice to further confirm CCL5-mediated angiogenic response in vivo.
410
The results showed that Matrigel mixed with CM from U2OS/control-shRNA cells
411
increased microvessel formation. Accordingly, CM from U2OS/CCL5-shRNA cells
412
significantly abolished neovascularization (Fig. 5E). Knockdown of CCL5 also
413
reduced microvessel formation in the Matrigel plugs by analyzing the CD31 and
414
hemoglobin content (Fig. 5E&F). Therefore, these results indicate that CCL5 plays an
415
important role during osteosarcoma-mediated angiogenesis.
416
Essential role of CCL5 for tumor angiogenesis in human osteosarcoma
417
microenvironment
of
CCL5
and
VEGF
was
reduced
by
CCL5-shRNA
in
418
Herein, we investigated whether CCL5 promoted tumor angiogenesis and
419
progression in osteosarcoma. To determine the effect of CCL5-shRNA on tumor
420
angiogenesis, osteosarcoma xenograft-induced angiogenesis model was used. Human
17
421
osteosarcoma cells were mixed with Matrigel and injected into the flanks of nude
422
mice. As shown in Fig. 6A&B, knockdown of CCL5 profoundly suppressed tumor
423
growth and volume in mice. We also evaluated the level of angiogenesis in tumor
424
specimens from animals by determining the hemoglobin content. The results
425
demonstrated
426
angiogenesis in vivo (Fig. 6C). We next examined human osteosarcoma tissues for the
427
expression of CCL5 and VEGF using immunohistochemistry. The expression of
428
CCL5 and VEGF in osteosarcoma patients was significantly higher than that in
429
normal cartilage. In addition, the high level of CCL5 expression correlated strongly
430
with VEGF expression and tumor stage (Fig. 6D). The quantitative data also exhibited
431
the high positive relationship between the expression of CCL5 and VEGF in tissues
432
obtained from osteosarcoma patients (Fig. 6E). Overall, these results suggest that
433
CCL5 promotes VEGF-mediated tumor angiogenesis in human osteosarcoma
434
microenvironment (Fig. 6F).
that
CCL5-shRNA
markedly
inhibited
osteosarcoma-induced
435
436
437
438
439
18
440
Discussion
441
Increasing evidences suggest that chemokines are produced by tumor cells as
442
well as by cells of the tumor microenvironment. In this regard, chemokines are
443
emerging as key mediators not only in the homing of cancer cells to metastatic sites
444
but also in the recruitment of a number of different cell types to establish tumor
445
microenvironment, facilitating tumor-associated angiogenesis and metastasis [1,6,42].
446
Osteosarcoma is the most frequent primitive malignant tumor of the skeletal system
447
and is characterized by an aggressive clinical course and metastatic potential [14].
448
Previously we demonstrated that CCL5/CCR5 chemokine axis promoted cell motility
449
and expression of v3 integrin in human osteosarcoma [34]. In this study, we found
450
that CCL5 increased VEGF expression in human osteosarcoma cells, and
451
subsequently induced tube formation and migration in human EPCs, indicating CCL5
452
promoted angiogenesis by the induction of VEGF. Tumor-derived chemokines have
453
been shown to directly affect tumor cells in an autocrine manner [3]. Here, we showed
454
that knockdown of CCL5 suppressed VEGF expression and impaired angiogenesis in
455
vitro and in vivo. We also demonstrated that CCL5-shRNA significantly abolished
456
tumor growth and angiogenesis in human osteosarcoma. Compelling evidences
457
indicate that CCL5 released by cells of the tumor microenvironment and acting
458
through autocrine activities promotes proliferation, migration and invasion of tumor
459
cells [6]. Thus, we suggest that CCL5 released by osteosarcoma cells acts as an
460
autocrine factor to stimulate VEGF expression, and contribute to tumor angiogenesis
461
in osteosarcoma microenvironment. Furthermore, we found that the expression of
462
CCL5 and VEGF in osteosarcoma patients were correlated with tumor stage,
463
indicating CCL5 may be a potential predictive factor for disease progression of
464
human osteosarcoma. Taken together, our results suggest that CCL5 promotes
465
VEGF-dependent tumor angiogenesis in human osteosarcoma microenvironment.
19
466
This is also the first study to demonstrate that CCL5 induces tumor angiogenesis by
467
VEGF production in human cancer cells, especially osteosarcoma cells.
468
Within the tumor microenvironment, chemokines and their receptors play critical
469
roles in the regulation of angiogenesis, which enhances the progression and metastasis
470
of many cancers [3]. The first report of chemokine-dependent homing theory has
471
demonstrated that CXC chemokine ligand 12 (CXCL12, also known as SDF-1) and
472
its cell surface receptor CXC chemokine receptor 4 (CXCR4) coordinately modulate
473
the metastasis in breast cancer [4]. Accumulating evidences also suggest that
474
CXCL12/CXCR4 signaling axis induce angiogenesis and progression of tumors by
475
the activation of VEGF [43,44]. Likewise, CCL5/CCR5 axis was proposed to regulate
476
directional cancer migration, invasion, and metastasis in various types of tumor,
477
including osteosarcoma, breast, and lung cancers [11,13,34]. Recent studies have
478
shown that CCL5 directly induces angiogenesis of endothelial cells and chemotaxis of
479
human EPCs through the chemokine receptor CCR5 [37,38]. Furthermore,
480
CCL5-CCR5 interaction was highlighted and reported to promote breast cancer
481
metastasis in tumor microenvironment [45]. Current study found that CCL5-induced
482
VEGF expression in osteosarcoma cells was attenuated by CCR5 mAb and CCR5
483
siRNA. Additionally, treatment of CCR5 mAb and CCR5 siRNA also significantly
484
suppressed CCL5-indcued angiogenesis of EPCs. These results indicate that CCL5
485
and CCR5 interaction contributes to angiogenesis by increasing VEGF secretion from
486
cancer cells, suggesting CCL5/CCR5 chemokine axis mediates an additional indirect
487
effect on angiogenesis in tumor microenvironment.
488
The discovery of signaling pathway underlying CCL5/CCR5 axis-mediated
489
VEGF expression helps us to understand the mechanism of tumor angiogenesis in the
490
microenvironment and may lead us to develop effective therapy for cancer treatment.
491
The activation of PKC has been implicated in the growth of several epithelial
20
492
cancers [46]. Recent study further indicates that PKC activation promotes tumor
493
angiogenesis by increasing the levels of HIF-1 and VEGF in human prostate cancer
494
xenograft [30]. Here, we showed that CCL5-induced VEGF expression and
495
angiogenesis were both inhibited by the specific PKC inhibitor rottlerin and PKC
496
siRNA. Furthermore, CCL5-activated PKCactivity was significantly blocked by
497
CCR5 mAb in osteosarcoma cells. These data suggest that PKC activation is an
498
important signaling molecule in CCL5-induced VEGF expression and angiogenesis.
499
Src is a non-receptor tyrosine kinase that is deregulated in many types of cancer [47].
500
Several reports have also indicated that c-Src is a downstream effector of PKC, plays
501
a critical role in tumor progression and dissemination [48,49]. Thus, we examined the
502
potential role of c-Src in the signaling pathway for CCL5-induced VEGF expression.
503
The results of this study demonstrated that pretreatment with c-Src inhibitor PP2
504
antagonized CCL5-induced VEGF expression and angiogenesis. This pathway was
505
further confirmed by transfection with PKC siRNA markedly attenuated the
506
enhancement of VEGF and angiogenesis by CCL5 stimulation. Moreover,
507
CCL5-induced up-regulation of c-Srcactivity was profoundly suppressed by either
508
CCR5 mAb or PKC inhibitor rottlerin. Therefore, we suggest that CCL5/CCR5 axis
509
promotes VEGF-dependent angiogenesis through PKC/c-Src signaling pathway in
510
human osteosarcoma cells. HIF-1 is a key transcriptional factor that regulates gene
511
expression of VEGF [24]. Our data found that CCL5 significantly increased protein
512
level of HIF-1 time dependently but not at mRNA level in osteosarcoma cells.
513
Pretreatment with HIF-1 inhibitor and transfection with HIF-1siRNA both
514
markedly attenuated CCL5-induced VEGF expression and angiogenesis. HIF-1
515
nuclear translocation is necessary for the transcriptional activation of HIF-1-regulated
516
VEGF expression [26]. We subsequently demonstrated that CCL5 increased the
517
binding of HIF-1 to the HRE element on VEGF promoter by ChIP assay. Using
21
518
transient transfection with HRE-luciferase as an indicator of HIF-1 activity, we
519
found that CCL5 also dramatically increased HRE-luciferase activity in osteosarcoma
520
cells. Collectively, we suggest that CCL5 enhances the protein stability and DNA
521
binding activity of HIF-1 to promote VEGF expression and angiogenesis. We next
522
explored whether the CCR5/PKC/c-Src pathway is an upstream signal in
523
CCL5-mediated HIF-1 activation. Our results showed that CCL5-induced the
524
binding of HIF-1 to the HRE element and HRE-luciferase activity were both
525
markedly antagonized by CCR5 mAb, rottlerin, and PP2. Previous evidence implies
526
that PKC regulates the stability of HIF-1 in cervical adenocarcinoma cells under
527
hypoxia, and knockdown of PKC inhibits hypoxia-induced VEGF expression and
528
angiogenesis [50]. c-Src kinase also induces HIF-1α activation through the
529
cap-dependent translation [41]. Here, we present that signaling of PKC-dependent
530
HIF-1 activation also exists in chemokine CCL5-treated cancer cells. We also
531
provide promising evidences that PKC/c-Src/HIF-1 signaling pathway controls
532
CCL5/CCR5 axis-induced VEGF expression and angiogenesis.
533
The prognosis of patients with osteosarcoma distant metastasis is generally
534
considered as very poor. Angiogenesis facilitates metastasis formation and contributes
535
to disease progression of osteosarcoma. Therefore, it is important to explore the novel
536
target for preventing osteosarcoma angiogenesis and metastasis nowadays. This study
537
showed that CCL5 and CCR5 interaction activates PKC, c-Src, and HIF-1
538
pathways, leading to up-regulation of VEGF expression, and contributes to tumor
539
angiogenesis and progression in osteosarcoma microenvironment (Fig. 6F). Based on
540
the findings herein, we suggest that CCL5 may be a potential target worthy of further
541
development to treat human osteosarcoma.
542
543
22
544
Reference
545
546
547
548
549
550
551
552
553
1.
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Mueller, M.M. and Fusenig, N.E. (2004) Friends or foes - bipolar effects of the
tumour stroma in cancer. Nat Rev Cancer, 4, 839-49.
Ben-Baruch, A. (2006) Inflammation-associated immune suppression in
cancer: the roles played by cytokines, chemokines and additional mediators.
Semin Cancer Biol, 16, 38-52.
Lazennec, G. and Richmond, A. (2011) Chemokines and chemokine receptors:
new insights into cancer-related inflammation. Trends Mol Med, 16, 133-44.
Muller, A., Homey, B., Soto, H., Ge, N., Catron, D., Buchanan, M.E.,
McClanahan, T., Murphy, E., Yuan, W., Wagner, S.N., Barrera, J.L., Mohar, A.,
Verastegui, E. and Zlotnik, A. (2001) Involvement of chemokine receptors in
breast cancer metastasis. Nature, 410, 50-6.
Schall, T.J., Bacon, K., Toy, K.J. and Goeddel, D.V. (1990) Selective attraction of
monocytes and T lymphocytes of the memory phenotype by cytokine RANTES.
Nature, 347, 669-71.
Borsig, L., Wolf, M.J., Roblek, M., Lorentzen, A. and Heikenwalder, M. (2013)
Inflammatory chemokines and metastasis-tracing the accessory. Oncogene.
Stormes, K.A., Lemken, C.A., Lepre, J.V., Marinucci, M.N. and Kurt, R.A. (2005)
Inhibition of metastasis by inhibition of tumor-derived CCL5. Breast Cancer
Res Treat, 89, 209-12.
Niwa, Y., Akamatsu, H., Niwa, H., Sumi, H., Ozaki, Y. and Abe, A. (2001)
Correlation of tissue and plasma RANTES levels with disease course in
patients with breast or cervical cancer. Clin Cancer Res, 7, 285-9.
Vaday, G.G., Peehl, D.M., Kadam, P.A. and Lawrence, D.M. (2006) Expression
of CCL5 (RANTES) and CCR5 in prostate cancer. Prostate, 66, 124-34.
Cambien, B., Richard-Fiardo, P., Karimdjee, B.F., Martini, V., Ferrua, B., Pitard,
B., Schmid-Antomarchi, H. and Schmid-Alliana, A. (2011) CCL5 neutralization
restricts cancer growth and potentiates the targeting of PDGFRbeta in
colorectal carcinoma. PLoS One, 6, e28842.
Huang, C.Y., Fong, Y.C., Lee, C.Y., Chen, M.Y., Tsai, H.C., Hsu, H.C. and Tang, C.H.
(2009) CCL5 increases lung cancer migration via PI3K, Akt and NF-kappaB
pathways. Biochem Pharmacol, 77, 794-803.
Luboshits, G., Shina, S., Kaplan, O., Engelberg, S., Nass, D., Lifshitz-Mercer, B.,
Chaitchik, S., Keydar, I. and Ben-Baruch, A. (1999) Elevated expression of the
CC chemokine regulated on activation, normal T cell expressed and secreted
(RANTES) in advanced breast carcinoma. Cancer Res, 59, 4681-7.
Velasco-Velazquez, M., Jiao, X., De La Fuente, M., Pestell, T.G., Ertel, A., Lisanti,
M.P. and Pestell, R.G. (2012) CCR5 antagonist blocks metastasis of basal
23
582
breast cancer cells. Cancer Res, 72, 3839-50.
583
584
585
586
587
588
589
590
591
592
14.
593
594
595
18.
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
15.
16.
17.
19.
20.
21.
22.
23.
24.
25.
26.
Arndt, C.A. and Crist, W.M. (1999) Common musculoskeletal tumors of
childhood and adolescence. N Engl J Med, 341, 342-52.
Tang, C.H. (2012) Molecular mechanisms of chondrosarcoma metastasis.
BioMedicine, 2, 92-98.
Chen, P.C., Cheng, H.C., Yang, S.F., Lin, C.W. and Tang, C.H. (2014) The CCN
family proteins: modulators of bone development and novel targets in
bone-associated tumors. BioMed research international, 2014, 437096.
PosthumaDeBoer, J., Witlox, M.A., Kaspers, G.J. and van Royen, B.J. (2011)
Molecular alterations as target for therapy in metastatic osteosarcoma: a
review of literature. Clin Exp Metastasis, 28, 493-503.
Carmeliet, P. and Jain, R.K. (2000) Angiogenesis in cancer and other diseases.
Nature, 407, 249-57.
Peters, B.A., Diaz, L.A., Polyak, K., Meszler, L., Romans, K., Guinan, E.C., Antin,
J.H., Myerson, D., Hamilton, S.R., Vogelstein, B., Kinzler, K.W. and Lengauer, C.
(2005) Contribution of bone marrow-derived endothelial cells to human
tumor vasculature. Nat Med, 11, 261-2.
Gao, D., Nolan, D.J., Mellick, A.S., Bambino, K., McDonnell, K. and Mittal, V.
(2008) Endothelial progenitor cells control the angiogenic switch in mouse
lung metastasis. Science, 319, 195-8.
Weis, S.M. and Cheresh, D.A. (2011) Tumor angiogenesis: molecular pathways
and therapeutic targets. Nat Med, 17, 1359-70.
Ferrara, N. (2004) Vascular endothelial growth factor as a target for
anticancer therapy. Oncologist, 9 Suppl 1, 2-10.
Brown, J.M. and Wilson, W.R. (2004) Exploiting tumour hypoxia in cancer
treatment. Nat Rev Cancer, 4, 437-47.
Maxwell, P.H., Dachs, G.U., Gleadle, J.M., Nicholls, L.G., Harris, A.L., Stratford,
I.J., Hankinson, O., Pugh, C.W. and Ratcliffe, P.J. (1997) Hypoxia-inducible
factor-1 modulates gene expression in solid tumors and influences both
angiogenesis and tumor growth. Proc Natl Acad Sci U S A, 94, 8104-9.
Jaakkola, P., Mole, D.R., Tian, Y.M., Wilson, M.I., Gielbert, J., Gaskell, S.J., von
Kriegsheim, A., Hebestreit, H.F., Mukherji, M., Schofield, C.J., Maxwell, P.H.,
Pugh, C.W. and Ratcliffe, P.J. (2001) Targeting of HIF-alpha to the von
Hippel-Lindau ubiquitylation complex by O2-regulated prolyl hydroxylation.
Science, 292, 468-72.
Forsythe, J.A., Jiang, B.H., Iyer, N.V., Agani, F., Leung, S.W., Koos, R.D. and
Semenza, G.L. (1996) Activation of vascular endothelial growth factor gene
transcription by hypoxia-inducible factor 1. Mol Cell Biol, 16, 4604-13.
24
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
Semenza, G.L. (2002) HIF-1 and tumor progression: pathophysiology and
therapeutics. Trends Mol Med, 8, S62-7.
Land, S.C. and Tee, A.R. (2007) Hypoxia-inducible factor 1alpha is regulated by
the mammalian target of rapamycin (mTOR) via an mTOR signaling motif. J
Biol Chem, 282, 20534-43.
Jiang, B.H., Jiang, G., Zheng, J.Z., Lu, Z., Hunter, T. and Vogt, P.K. (2001)
Phosphatidylinositol 3-kinase signaling controls levels of hypoxia-inducible
factor 1. Cell Growth Differ, 12, 363-9.
Kim, J., Koyanagi, T. and Mochly-Rosen, D. (2011) PKCdelta activation
mediates angiogenesis via NADPH oxidase activity in PC-3 prostate cancer
cells. Prostate, 71, 946-54.
Wu, M.H., Huang, C.Y., Lin, J.A., Wang, S.W., Peng, C.Y., Cheng, H.C. and Tang,
C.H. (2014) Endothelin-1 promotes vascular endothelial growth
factor-dependent angiogenesis in human chondrosarcoma cells. Oncogene,
33, 1725-35.
Chung, C.H., Chang, C.H., Chen, S.S., Wang, H.H., Yen, J.Y., Hsiao, C.J., Wu, N.L.,
Chen, Y.L., Huang, T.F., Wang, P.C., Yeh, H.I. and Wang, S.W. (2013) Butein
Inhibits Angiogenesis of Human Endothelial Progenitor Cells via the
Translation Dependent Signaling Pathway. Evid Based Complement Alternat
Med, 2013, 943187.
Tsai, H.C., Su, H.L., Huang, C.Y., Fong, Y.C., Hsu, C.J. and Tang, C.H. (2014) CTGF
increases matrix metalloproteinases expression and subsequently promotes
tumor metastasis in human osteosarcoma through down-regulating miR-519d.
Oncotarget, 5, 3800-12.
Wang, S.W., Wu, H.H., Liu, S.C., Wang, P.C., Ou, W.C., Chou, W.Y., Shen, Y.S. and
Tang, C.H. (2012) CCL5 and CCR5 interaction promotes cell motility in human
osteosarcoma. PLoS One, 7, e35101.
Wu, M.H., Huang, C.Y., Lin, J.A., Wang, S.W., Peng, C.Y., Cheng, H.C. and Tang,
C.H. (2013) Endothelin-1 promotes vascular endothelial growth
factor-dependent angiogenesis in human chondrosarcoma cells. Oncogene.
Storgard, C., Mikolon, D. and Stupack, D.G. (2005) Angiogenesis assays in the
chick CAM. Methods Mol Biol, 294, 123-36.
Suffee, N., Hlawaty, H., Meddahi-Pelle, A., Maillard, L., Louedec, L., Haddad,
O., Martin, L., Laguillier, C., Richard, B., Oudar, O., Letourneur, D., Charnaux, N.
and Sutton, A. (2012) RANTES/CCL5-induced pro-angiogenic effects depend
on CCR1, CCR5 and glycosaminoglycans. Angiogenesis, 15, 727-44.
Ishida, Y., Kimura, A., Kuninaka, Y., Inui, M., Matsushima, K., Mukaida, N. and
Kondo, T. (2012) Pivotal role of the CCL5/CCR5 interaction for recruitment of
25
658
endothelial progenitor cells in mouse wound healing. J Clin Invest, 122,
659
660
661
662
663
664
665
666
667
668
41.
711-21.
Hsieh, H.L., Sun, C.C., Wang, T.S. and Yang, C.M. (2008)
PKC-delta/c-Src-mediated EGF receptor transactivation regulates
thrombin-induced COX-2 expression and PGE(2) production in rat vascular
smooth muscle cells. Biochim Biophys Acta, 1783, 1563-75.
Brandt, D.T., Goerke, A., Heuer, M., Gimona, M., Leitges, M., Kremmer, E.,
Lammers, R., Haller, H. and Mischak, H. (2003) Protein kinase C delta induces
Src kinase activity via activation of the protein tyrosine phosphatase PTP
alpha. J Biol Chem, 278, 34073-8.
Karni, R., Dor, Y., Keshet, E., Meyuhas, O. and Levitzki, A. (2002) Activated
42.
pp60c-Src leads to elevated hypoxia-inducible factor (HIF)-1alpha expression
under normoxia. J Biol Chem, 277, 42919-25.
Yin, M.C. (2013) Development of natural antitumor agents. BioMedicine, 3,
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
39.
40.
43.
44.
45.
46.
47.
48.
49.
50.
105.
Liang, Z., Brooks, J., Willard, M., Liang, K., Yoon, Y., Kang, S. and Shim, H. (2007)
CXCR4/CXCL12 axis promotes VEGF-mediated tumor angiogenesis through
Akt signaling pathway. Biochem Biophys Res Commun, 359, 716-22.
Wang, J., Wang, J., Sun, Y., Song, W., Nor, J.E., Wang, C.Y. and Taichman, R.S.
(2005) Diverse signaling pathways through the SDF-1/CXCR4 chemokine axis
in prostate cancer cell lines leads to altered patterns of cytokine secretion
and angiogenesis. Cell Signal, 17, 1578-92.
Karnoub, A.E., Dash, A.B., Vo, A.P., Sullivan, A., Brooks, M.W., Bell, G.W.,
Richardson, A.L., Polyak, K., Tubo, R. and Weinberg, R.A. (2007) Mesenchymal
stem cells within tumour stroma promote breast cancer metastasis. Nature,
449, 557-63.
Griner, E.M. and Kazanietz, M.G. (2007) Protein kinase C and other
diacylglycerol effectors in cancer. Nat Rev Cancer, 7, 281-94.
Yeatman, T.J. (2004) A renaissance for SRC. Nat Rev Cancer, 4, 470-80.
Amos, S., Martin, P.M., Polar, G.A., Parsons, S.J. and Hussaini, I.M. (2005)
Phorbol 12-myristate 13-acetate induces epidermal growth factor receptor
transactivation via protein kinase Cdelta/c-Src pathways in glioblastoma cells.
J Biol Chem, 280, 7729-38.
Yang, S.F., Chen, M.K., Hsieh, Y.S., Chung, T.T., Hsieh, Y.H., Lin, C.W., Su, J.L.,
Tsai, M.H. and Tang, C.H. (2010) Prostaglandin E2/EP1 signaling pathway
enhances intercellular adhesion molecule 1 (ICAM-1) expression and cell
motility in oral cancer cells. J Biol Chem, 285, 29808-16.
Lee, J.W., Park, J.A., Kim, S.H., Seo, J.H., Lim, K.J., Jeong, J.W., Jeong, C.H.,
26
696
Chun, K.H., Lee, S.K., Kwon, Y.G. and Kim, K.W. (2007) Protein kinase C-delta
697
698
699
regulates the stability of hypoxia-inducible factor-1 alpha under hypoxia.
Cancer Sci, 98, 1476-81.
700
701
27
702
DISCLOSURES
703
None of the authors have any financial or personal relationships with other people or
704
organizations that could inappropriately influence this research.
705
706
ACKNOWLEDGMENTS
707
We thank Dr. W.M. Fu for providing HRE promoter construct. This work was
708
supported by grants from the National Science Council of Taiwan (NSC
709
101-2320-B-715-002-MY3; 103-2628-B-039-002-MY3), Mackay Medical College
710
(MMC-1011B01), Mackay Memorial Hospital (MMH-MM-10205), and China
711
Medical University (CMU102-ASIA-10).
712
713
28
714
FIGURE LEGENDS
715
Fig. 1 CCL5 and CCR5 interaction promotes VEGF expression and angiogenesis.
716
(A&B) Cells (U2OS and MG63 cells) were incubated with CCL5 (0-3 ng/ml) for 24 h,
717
and VEGF expression was examined by qPCR and ELISA. (C-E) Cells were
718
incubated with CCL5 (0-3 ng/ml) for 24 h. The medium was collected followed by
719
stimulation with VEGF antibody (5 μg/mL) for 30 min and then then applied to EPCs
720
for 24 h. The capillary-like structures formation and cell migration in EPCs were
721
examined by tube formation and Transwell assay. (F&G) Cells were pretreated with
722
the isotype control (5 g/ml) and CCR5 mAb (5 g/ml) for 30 min or transfected with
723
CCR5 siRNA for 24 h followed by stimulation with CCL5 (3 ng/ml) for 24 h, and
724
VEGF expression was examined by qPCR and ELISA. (H&I) In addition, the medium
725
was collected as CM and then applied to EPCs for tube formation and Transwell assay.
726
Results are expressed as the mean ± S.E. *, p < 0.05 compared with control; #, p <
727
0.05 compared with CCL5-treated group
728
729
Fig. 2 PKC is involved in CCL5–induced VEGF expression and angiogenesis.
730
(A&B) Cells were pretreated with the rottlerin (3 μM) for 30 min or transfected with
731
PKC siRNA for 24 h followed by stimulation with CCL5 (3 ng/ml) for 24 h, and
732
VEGF expression was examined by qPCR and ELISA. (C&D) In addition, the
733
medium was collected as CM and then applied to EPCs for 24 h. The capillary-like
734
structures formation and cell migration in EPCs were examined by tube formation and
735
Transwell assay. (E&F) U2OS cells were incubated with CCL5 (0-3 ng/ml) for the
736
indicated times, or pretreated with the CCR5 mAb (5 g/ml) for 30 min followed by
737
stimulation with CCL5 (3 ng/ml) for 60 min, and the activity of PKC was
738
determined by PKC kinase assay. Results are expressed as the mean ± S.E. *, p <
739
0.05 compared with control; #, p < 0.05 compared with CCL5-treated group.
29
740
Fig. 3 c-Src is involved in CCL5–induced VEGF expression and angiogenesis.
741
(A&B) Cells were pretreated with the PP2 (3 μM) for 30 min or transfected with
742
c-Src siRNA for 24 h followed by stimulation with CCL5 (3 ng/ml) for 24 h, and
743
VEGF expression was examined by qPCR and ELISA. (C&D) In addition, the
744
medium was collected as CM and then applied to EPCs for 24 h. The capillary-like
745
structures formation and cell migration in EPCs were examined by tube formation and
746
Transwell assay. (E&F) U2OS cells were incubated with CCL5 (0-3 ng/ml) for the
747
indicated times, or pretreated with the CCR5 mAb (5 g/ml) and rottlerin (3 μM) for
748
30 min followed by stimulation with CCL5 (3 ng/ml) for 60 min, and the activity of
749
c-Src was determined by c-Src kinase assay. Results are expressed as the mean ± S.E.
750
*, p < 0.05 compared with control; #, p < 0.05 compared with CCL5-treated group.
751
752
Fig. 4 HIF-1activation is involved in CCL5-induced VEGF expression and
753
angiogenesis.
754
(A&B) Cells were pretreated with the HIF-1inhibitor (10 μM) for 30 min or
755
transfected with HIF-1siRNA for 24 h followed by stimulation with CCL5 (3 ng/ml)
756
for 24 h, and VEGF expression was examined by qPCR and ELISA. (C&D) In
757
addition, the medium was collected as CM and then applied to EPCs for 24 h. The
758
capillary-like structures formation and cell migration in EPCs were examined by tube
759
formation and Transwell assay. (E&F) U2OS cells were incubated with CCL5
760
(3 ng/ml) for the indicated times and HIF-1expression was determined by Western
761
blotting and qPCR. (G) U2OS cells were pretreated for 30 min with CCR5 mAb,
762
rottlerin, or PP2 for 30 min followed by stimulation with CCL5 (3 ng/ml) for 120 min.
763
The HIF-1activation was examined by chromatin immunoprecipitation and HRE
764
luciferase activity. Results are expressed as the mean ± S.E. *, p < 0.05 compared
765
with control; #, p < 0.05 compared with CCL5-treated group.
30
766
Fig. 5 Knockdown of CCL5 suppresses angiogenic effects in both in vitro and
767
in vivo assays.
768
(A) The protein expression of CCL5 and VEGF in U2OS/control-shRNA and
769
U2OS/CCL5-shRNA cells was examined by western blotting. (B&C) EPCs were
770
incubated CM from U2OS/control-shRNA or U2OS/CCL5-shRNA for 24 h, and tube
771
formation or cell migration was photographed under microscope or examined by
772
Transwell. (D) Chick embryos were incubated with PBS, U2OS/control-shRNA CM,
773
or U2OS/CCL5-shRNA CM for 4 day, and then resected, fixed, and photographed
774
with a stereomicroscope. (E&F) Mice were injected subcutaneously with matrigel
775
mixed with PBS, U2OS/control-shRNA CM, or U2OS/CCL5-shRNA CM for 7 day.
776
The plugs were excised from mice and photographed, quantified of hemoglobin
777
content, and stained with CD31. Results are expressed as the mean ± S.E. *, p < 0.05
778
compared with control; #, p < 0.05 compared with CCL5-treated group.
779
780
Fig. 6 Role of CCL5 on tumor angiogenesis in human osteosarcoma
781
microenvironment.
782
(A-C) U2OS/control-shRNA or U2OS/CCL5-shRNA cells were mixed with Matrigel
783
and injected into flank sites of mice for 10 day, and then resected. The tumors were
784
measured weight and volume, and quantified the hemoglobin levels. Results are
785
expressed as the mean ± S.E. *, p < 0.05 compared with control. (D&E) The protein
786
expression
787
immunohistochemistry in normal bone and osteosarcoma tissue. The correlation and
788
quantitative data are shown in the (E). (F) Schematic diagram summarizes the
789
mechanism of CCL5-mediated tumor angiogenesis and progression in human
790
osteosarcoma microenvironment. CCL5/CCR5 axis induces tumor angiogenesis by
791
VEGF production in human osteosarcoma through the activation of PKC, c-Src, and
of
CCL5
and
VEGF
(arrowhead)
was
examined
using
31
792
HIF-1 signaling cascades.
32