Biochemical Pharmacology 91 (2014) 522–533

Contents lists available at ScienceDirect

Biochemical Pharmacology journal homepage: www.elsevier.com/locate/biochempharm

Brain-derived neurotrophic factor increases vascular endothelial growth factor expression and enhances angiogenesis in human chondrosarcoma cells Chih-Yang Lin a,1, Shih-Ya Hung b,c,1, Hsien-Te Chen d,e,f, Hsi-Kai Tsou f,g,h, Yi-Chin Fong d,e, Shih-Wei Wang i, Chih-Hsin Tang a,j,k,* a

Graduate Institute of Basic Medical Science, China Medical University, No. 91, Hsueh-Shih Road, Taichung, Taiwan Epigenome Research Center, China Medical University Hospital, Taichung, Taiwan c Graduate Institute of Integrated Medicine, College of Chinese Medicine, China Medical University, Taichung, Taiwan d School of Chinese Medicine, College of Chinese Medicine, China Medical University, Taichung, Taiwan e Department of Orthopedic Surgery, China Medical University Hospital, Taichung, Taiwan f Department of Materials Science and Engineering, Feng Chia University, Taichung, Taiwan g Department of Neurosurgery, Taichung Veterans General Hospital, Taichung, Taiwan h Department of Early Childhood Care and Education, Jen-Teh Junior College of Medicine, Nursing and Management, Miaoli County, Taiwan i Department of Medicine, Mackay Medical College, New Taipei City, Taiwan j Department of Pharmacology, School of Medicine, China Medical University, Taichung, Taiwan k Department of Biotechnology, College of Health Science, Asia University, Taichung, Taiwan b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 18 March 2014 Accepted 11 August 2014 Available online 19 August 2014

Chondrosarcomas are a type of primary malignant bone cancer, with a potent capacity for local invasion and distant metastasis. Brain-derived neurotrophic factor (BDNF) is commonly upregulated during neurogenesis. The aim of the present study was to examine the mechanism involved in BDNF-mediated vascular endothelial growth factor (VEGF) expression and angiogenesis in human chondrosarcoma cells. Here, we knocked down BDNF expression in chondrosarcoma cells and assessed their capacity to control VEGF expression and angiogenesis in vitro and in vivo. We found knockdown of BDNF decreased VEGF expression and abolished chondrosarcoma conditional medium-mediated angiogenesis in vitro as well as angiogenesis effects in vivo in the chick chorioallantoic membrane and Matrigel plug nude mouse models. In addition, in the xenograft tumor angiogenesis model, the knockdown of BDNF significantly reduced tumor growth and tumor-associated angiogenesis. BDNF increased VEGF expression and angiogenesis through the TrkB receptor, PLCg, PKCa, and the HIF-1a signaling pathway. Finally, we analyzed samples from chondrosarcoma patients by immunohistochemical staining. The expression of BDNF and VEGF protein in 56 chondrosarcoma patients was significantly higher than in normal cartilage. In addition, the high level of BDNF expression correlated strongly with VEGF expression and tumor stage. Taken together, our results indicate that BDNF increases VEGF expression and enhances angiogenesis through a signal transduction pathway that involves the TrkB receptor, PLCg, PKCa, and the HIF-1a. Therefore, BDNF may represent a novel target for anti-angiogenic therapy for human chondrosarcoma. ß 2014 Elsevier Inc. All rights reserved.

Keywords: BDNF TrkB Chondrosarcoma VEGF Angiogenesis

1. Introduction Chondrosarcomas are formed by the abnormal proliferation of cartilage. It accounts for about 26% of bone cancer cases worldwide. Usually occurring in males between 10 and 80 years

* Corresponding author at: Graduate Institute of Basic Medical Science, China Medical University, No. 91, Hsueh-Shih Road, Taichung, Taiwan. Tel.: +886 4 22052121x7726; fax: +886 4 22333641. E-mail address: [email protected] (C.-H. Tang). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.bcp.2014.08.008 0006-2952/ß 2014 Elsevier Inc. All rights reserved.

of age, the tumors generally appear on the scapula, sternum, ribs, or pelvis [1]. Clinically, surgical resection remains the primary mode of therapy for chondrosarcoma. Chondrosarcoma can easily metastasize to other organs, such as the lung and liver [2–4]. When distant metastasis occurs, the patient has a poor prognosis. Because there is currently no effective adjuvant therapy, the development of novel molecular mechanisms is very important for the treatment of chondrosarcoma metastasis [5]. Tumor metastasis includes many functions, such as proliferation [6], migration [7], invasion [8], and angiogenesis [9]. Among these, the most important step is angiogenesis because when the

C.-Y. Lin et al. / Biochemical Pharmacology 91 (2014) 522–533

tumor reaches a certain size, it requires more nutrients and oxygen to support its growth. Unfortunately, the tumor can metastasize to remote organs through these new blood vessels. Vascular endothelial growth factor (VEGF), a protein that plays important roles in embryonic development, wound healing, and tumor angiogenesis [10]. In addition, VEGF has been associated with angiogenesis in breast [11], lung [12], pancreatic [13], and cervical carcinomas [14]. Many drugs reduce cancer growth and metastasis by inhibiting VEGF expression [15,16]. Therefore, VEGF is a critical target in cancer therapy and metastasis. Brain-derived neurotrophic factor (BDNF) is a small basic protein. It is a member of the neurotrophin family of growth factors, and it is a physiologically important nerve growth factor [17]. BDNF and its specific receptor, tropomyosin related kinase B (TrkB), play key roles in neural development, and some studies have suggested a role for BDNF in cancer cell proliferation [18], survival [19], metastasis [2], and angiogenesis [20]. BDNF promotes metastasis and angiogenesis through VEGF-dependent activation in several tumor types, including ovarian cancers [21], hepatocellular carcinoma [22], oral squamous cell carcinoma [23], multiple myeloma [24], and neuroblastoma [25]. Therefore, BDNF could be a novel target for tumor metastasis and angiogenesis. Because the prognosis of patients with distant metastasis of chondrosarcoma is very poor [26], the development of an antiangiogenic and anti-metastatic therapy could be useful for these patients. BDNF promotes chondrosarcoma metastasis through the upregulation of integrin b5 and matrix metalloproteinase-1 [2,27]. However, the role of BDNF in VEGF expression and angiogenesis in human chondrosarcoma is largely unknown. Here, we report that BDNF promotes VEGF expression in human chondrosarcoma via the activation of the TrkB receptor, PLCg, PKCa, and HIF-1a signaling pathways. In addition, the knockdown of BDNF significantly reduced VEGF expression and angiogenesis in vivo. Our results indicate that BDNF plays an important role in the metastasis and angiogenesis of human chondrosarcoma. Therefore, BDNF might be a new therapeutic target for the treatment of chondrosarcoma metastasis. 2. Experimental 2.1. Materials Rabbit polyclonal antibodies specific for p-PLCg and p-PKCa were purchased from Cell Signaling Technology (Danvers, MA, USA). Rabbit polyclonal antibodies specific for BDNF, TrkB, PLCg, PKCa, b-actin, and PCNA were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). HIF-1a, VEGF, and CD31 antibodies were purchased from Abcam (Cambridge, MA, USA). GF109203X and Go¨6976 were purchased from Enzo Biochem Inc (New York, NY, USA). Recombinant human BDNF and VEGF were purchased from R&D Systems (Minneapolis, MN, USA). DMEM, a-MEM, fetal bovine serum (FBS), and all the other cell culture reagents were purchased from Gibco-BRL life technologies (Grand Island, NY, USA). The PKCa dominant negative mutant was provided by Dr. V. Martin (Louis Pasteur de Strasbourg University, France). The HIF-1a dominant negative mutant and pHREluciferase construct were provided by Dr. W.M. Fu (National Taiwan University, Taiwan). The pSV-b-galactosidase vector and luciferase assay kit were purchased from Promega (Madison, WI, USA). All other chemicals were purchased from Sigma–Aldrich (St. Louis, MO, USA). 2.2. Cell culture The human chondrosarcoma cell line (JJ012) was kindly provided by Dr. Sean P. Scully (University of Miami School of

523

Medicine, Miami, FL, USA) [28]. We used lentiviral transfection to transfect BDNF-shRNA or control-shRNA plasmids into JJ012 cells. After 48 h, the medium was replaced. Subsequently, 10 mg/mL puromycin (Life Technologies) was used to select stable transfectants. Thereafter, the selection medium was replaced every 3 days. After 2 weeks of selection in puromycin, clones of resistant cells were isolated. All JJ012 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM)/a-MEM supplemented with 10% FBS. These cells were maintained at 37 8C in a humidified atmosphere of 5% CO2. The human endothelial progenitor cells (EPCs) used in this study were approved by the Institutional Review Board of Mackay Medical College, New Taipei City, Taiwan (reference number: P1000002), and all subjects gave informed written consent before enrollment in this study. After collected peripheral blood (80 ml) from healthy donors, the peripheral blood mononuclear cells (PBMCs) were fractionated from other blood components by centrifugation on Ficoll-Paque plus (Amersham Biosciences, Uppala, Sweden) according to the manufacturer’s instructions. CD34-positive progenitor cells were obtained from the isolated PBMCs using CD34 MicroBead kit and MACS Cell Separation System (Miltenyi Biotec, Bergisch Gladbach, Germany). The maintenance and characterization of CD34-positive EPCs were performed as described previously [29,30]. Briefly, human CD34positive EPCs were maintained and propagated in MV2 complete medium consisting of MV2 basal medium and growth supplement (PromoCell, Heidelberg, Germany) supplemented with 20% FBS (HyClone, Logan, UT, USA). The cultures were seeded onto 1% gelatin-coated plastic ware and maintained at 37 8C in a humidified atmosphere of 5% CO2 [30]. 2.3. Western blot analysis Cellular lysates were prepared as described previously [31]. Proteins were resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to Immobilon polyvinyl difluoride membranes (Immobilon P, Millipore). The blots were blocked with 4% non-fat milk for 1 h at room temperature and then probed with rabbit anti-human antibodies against p-PLCg, PLCg, p-PKCa, PKCa, HIF-1a, VEGF, b-actin, PCNA, BDNF, or TrkB (1:1000) for 1 h at room temperature. After 3 washes, the blots were subsequently incubated with a goat anti-rabbit or goat antimouse peroxidase-conjugated secondary antibody (1:1000) for 1 h at room temperature. The blots were visualized by enhanced chemiluminescence using Kodak X-OMAT LS film (Eastman Kodak, Rochester, NY, USA). 2.4. Quantitative real-time polymerase chain reaction Total RNA was extracted from chondrosarcoma cells by using a TRIzol kit (MDBio Inc., Taipei, Taiwan). The reverse transcription reaction was performed using 1 mg of total RNA that was reverse transcribed into cDNA using oligo(dT) primers [32]. Quantitative real-time PCR (qPCR) analysis was performed using Taqman1 One-Step RT-PCR Master Mix (Applied Biosystems, CA). Two microliters of cDNA template was added to each 25-mL reaction with sequence-specific primers and Taqman1 probes. The sequences for all target gene primers and probes were purchased commercially. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as an endogenous control to normalize expression data (Applied Biosystems). qPCR assays were carried out in triplicate in a StepOnePlus sequence detection system. The cycling conditions were as follows: initial 10-min polymerase activation at 95 8C, followed by 40 cycles at 95 8C for 15 s and 60 8C for 60 s. To calculate the cycle number at which the transcript was detected (CT), the threshold was set above the

524

C.-Y. Lin et al. / Biochemical Pharmacology 91 (2014) 522–533

non-template control background and within the linear phase of target gene amplification. 2.5. Transient transfection and reporter gene assay Human chondrosarcoma cells were plated in 12-well dishes. DNA and Lipofectamine 2000 (LF2000; Invitrogen) were premixed for 20 min and then applied to the cells. Twenty-four hours after transfection, the cells were incubated with the indicated agents. Cell extracts were prepared and used for measuring the luciferase and b-galactosidase activities. ON-TARGETplus siRNAs of PLCg, PKCa, HIF-1a, and control were purchased from Dharmacon Research (Lafayette, CO, USA). Transient transfection of siRNAs was carried out using DharmaFECT1 transfection reagent. The siRNA (100 nM) was formulated with DharmaFECT1 transfection reagent according to the manufacturer’s instruction. 2.6. Preparation of conditioned medium (CM) Human chondrosarcoma cell line (JJ012 cells) was plated in 6well dishes and grown to confluence. The culture medium was exchanged with serum-free DMEM/a-MEM medium. CM were collected 2 days after the change of media and stored at 20 8C until use. In the series of experiments, JJ012 cells were pretreated for 30 min with inhibitors, including K252a, U73122, U73343, GF109203X, Go¨6976, HIF-1 inhibitor, or vehicle control (0.1% DMSO), followed by treatment with BDNF for 24 h to prevent signaling via the BDNF pathway. 2.7. Migration assay Migration activity was measured using a Transwell assay (Costar, NY; pore size, 8-mm). Approximately 1  104 cells were added to the upper chamber in 200 mL of 10% FBS MV2 complete medium. The lower chamber contained 150 ml MV2 complete medium and 150 ml CM. The plates were incubated for 24 h at 37 8C in 5% CO2, and cells were fixed in 3.7% formaldehyde solution for 15 min and stained with 0.05% crystal violet in PBS for 15 min. Cells on the upper side of the filters were removed with cottontipped swabs, and the filters were washed with PBS. Cells on the underside of the filters were examined and counted under a microscope. Each clone was plated in triplicate for each experiment, and each experiment was repeated at least 3 times.

2.10. Tube formation assay Matrigel (BD Biosciences, Bedford, MA, USA) was dissolved overnight at 4 8C and 48-well plates were prepared with 100 mL Matrigel in each well, after coating and incubating at 37 8C overnight. EPCs (3  104) were cultured in 200 mL media which containing 50% MV2 medium and 50% CM. After 6 h of incubation at 37 8C, EPC tube formation was assessed with a photomicroscope, and each well was photographed at 200 magnification. Tube branches and total tube length were calculated using MacBiophotonics Image J software. 2.11. Chick chorioallantoic membrane (CAM) assay Fertilized chicken eggs were incubated at 38 8C with 80% humidity. A small window was made in the shell on day 3 of chick embryo development under aseptic conditions. The window was resealed with adhesive tape and eggs were returned to the incubator until day 3 of chick embryo development. On day 7, CM from JJ012/control-shRNA or JJ012/BDNF-shRNA cells (2  104 cells) deposited in the center of the CAM. At 11 days, CAMs were collected for microscopy and photographic documentation. Angiogenesis was quantified by counting the number of blood vessel branches; at least 10 viable embryos were tested for each treatment. All animal work was done in accordance with a protocol approved by the China Medical University (Taichung, Taiwan) Institutional Animal Care and Use Committees. 2.12. In vivo tumor xenograft study Four-week-old male CB17-SCID mice were randomized into 2 groups. Experimental cells growing exponentially were implanted into 10 SCID mice by subcutaneous (SC) injection of 2  106 cells (JJ012/control-shRNA or JJ012/BDNF-shRNA resuspended in 200 mL of containing 50% serum-free DMEN/a-MEM and 50% Matrigel) injected into the right flank. Mice were observed at 5 weeks by the Xenogen IVIS system and images were captured 10 min after D-luciferin injection with a 60-s exposure using a CCD camera. Mice were euthanized by subjecting them to CO2 inhalation and the tumors were removed and fixed in 10% formalin, and their volume and weight were measured. The hemoglobin content was assayed using the Drabkin’s reagent kit (Sigma).

2.8. ELISA assay

2.13. Matrigel plug assay

Human chondrosarcoma cells were cultured in 24-well plates. Cells were incubated in a humidified incubator at 37 8C for 24 h. To examine the downstream signaling pathways involved in BDNF treatment, cells were pretreated with various inhibitors for 30 min before the addition of BDNF (50 ng/mL). After incubation, the medium was removed and stored at 80 8C until the assay was performed. VEGF in the medium was assayed using a VEGF enzyme immunoassay kit (Peprotech, Offenbach, Germany), according to the procedure described by the manufacturer.

Matrigel plug angiogenesis assay was adapted from a previously described assay [34]. Four-week-old male nude mice were randomized into 3 groups: serum-free medium, JJ012/controlshRNA CM, or JJ012/BDNF-shRNA CM resuspended with Matrigel. Mice were subcutaneously injected with 300 mL of Matrigel. After 7 days, Matrigel pellets were harvested, partially fixed with 4% formalin, embedded in paraffin, and subsequently processed for immunohistochemistry staining for vessel marker CD31. 2.14. Immunohistochemistry (IHC)

2.9. Chromatin immunoprecipitation assay Chromatin immunoprecipitation analysis was performed as described previously [32]. DNA immunoprecipitated with an antiHIF-1a Ab was purified and extracted with phenol–chloroform. The purified DNA pellet was subjected to PCR. PCR products were resolved by 2% agarose gel electrophoresis and visualized under UV light. The primers 50 -CCTTTGGGTTTTGCCAGA-30 and 50 -CCAAGTTTGTGGAGCTGA-30 were utilized for amplification across the VEGF promoter region [33].

The human chondrosarcoma tissue array was purchased from Cybrdi (Rockville, MD, USA) and Biomax (Rockville, MD, USA). The 8 cases for normal cartilage, 26 cases for grade I chondrosarcoma, 12 cases for grade II chondrosarcoma, and 18 cases for grade III chondrosarcoma); (Grade I chondrosarcoma grows relatively slowly, has cells whose histological appearance is quite similar to cells of normal cartilage, and have much less aggressive invasive. Grades 2 and 3 are increasingly faster-growing cancers, with more varied and abnormal-looking cells, and are much more likely to

C.-Y. Lin et al. / Biochemical Pharmacology 91 (2014) 522–533

infiltrate surrounding tissues). The tissues were placed on glass slides, rehydrated, and incubated in 3% hydrogen peroxide to block the endogenous peroxidase activity. After trypsinization, the sections were blocked by incubation in 3% bovine serum albumin in PBS. The primary polyclonal antibodies, rabbit anti-human BDNF and VEGF, were applied to the slides at a dilution of 1:200 and incubated at 4 8C overnight. After being washed 3 times in PBST, the samples were treated with goat anti-rabbit IgG biotin-labeled secondary antibodies at a dilution of 1:50. Bound antibodies were detected with an ABC kit (Vector Laboratories). The slides were stained with chromogen diaminobenzidine, washed, counterstained with Delafield’s hematoxylin, dehydrated, treated with xylene, and mounted.

525

of angiogenesis by determining the hemoglobin content of the tumor and found that reducing BDNF expression diminished chondrosarcoma-induced angiogenesis in vivo (Fig. 2E). The hemoglobin content also correlated with the tumor volume in these tumors (Fig. 2F). Finally, ex vivo analysis of tumors excised from the mice showed decreased BDNF and VEGF protein level (55% and 60%, respectively), by using qPCR, Western blotting, and ELISA assays, and CD31 expression in the JJ012/BDNF-shRNA group (Fig. 2G–I). Overall, these results suggest that BDNF promotes angiogenesis and tumor growth in vivo. 3.3. BDNF induces VEGF expression and angiogenesis through the TrkB receptor

2.15. Statistical analysis Data were presented as mean  standard error of the mean. Statistical analysis of comparisons between 2 samples was performed using Student’s t test. Statistical comparisons of more than 2 groups were performed using 1-way analysis of variance with Bonferroni’s post-hoc test. In all cases, p < 0.05 was considered significant. 3. Results 3.1. Knockdown of BDNF decreases VEGF expression in vitro and angiogenesis in vivo BDNF has cited to promote the metastasis of human chondrosarcoma cells [2,27]. However, the effects of BDNF on VEGF expression and angiogenesis in chondrosarcoma are largely unknown. Therefore, we established BDNF-shRNA-expressing chondrosarcoma cells (JJ012/BDNF-shRNA) to examine the role of BDNF in VEGF expression and angiogenesis. The overexpression of BDNF-shRNA reduced BDNF and VEGF expression, as shown in Fig. 1A–C, by using qPCR, Western blotting, and ELISA assays (35%, 55%, and 40%, respectively). However, knockdown of BDNF did not affect the cell proliferation of chondrosarcoma cells (Fig. 1D). In angiogenic processes, endothelial cells must undergo migration, proliferation, and tube formation to form tumor neovessel [35]. Subsequent study indicates that tumor angiogenesis is also supported by the mobilization and functional incorporation of EPCs [36]. Therefore, we next determined whether knockdown BDNF reduced angiogenesis in EPCs. Our results showed in Fig. 1E and F that knockdown of BDNF reduced chondrosarcoma CM-mediated EPC migration and tube formation (60% and 45%, respectively). On the other hand, the effect of BDNF on angiogenesis in vivo was evaluated using the in vivo model of the CAM assay. CM from JJ012/control-shRNA increased angiogenesis in CAM (Fig. 1G). However, BDNF shRNA completely reduced angiogenesis in CAM (Fig. 1G). We also analyzed Matrigel plug formation following subcutaneous implantation in mice. Knockdown of BDNF reduced vascular formation in the Matrigel and the expression of vessel marker CD31 (Fig. 1H and I). These results indicated that knockdown of BDNF decreases VEGF expression and angiogenesis in vitro and in vivo. 3.2. Knockdown of BDNF decreases tumor-associated angiogenesis in vivo In order to investigate whether the knockdown of BDNF was able to inhibit tumor-induced angiogenesis in vivo, a xenograft tumor-induced angiogenesis model was used. We used two chondrosarcoma cell lines (JJ012/control-shRNA and JJ012/ BDNF-shRNA) mixed with Matrigel and injected them into the right flanks of SCID mice. We observed the mice 5 weeks later by using the Xenogen IVIS system and found that the knockdown of BDNF reduced tumor growth (Fig. 2A–D). We quantified the level

Next, we directly applied BDNF to chondrosarcoma cells and examined VEGF expression. Stimulation of human chondrosarcoma cells with BDNF increased VEGF expression in a dose- and time-dependent manner (Fig. 3A–E) (2.5 to 4 fold). We then determined whether the BDNF-dependent VEGF expression induced angiogenesis in EPCs. We found that CM promoted EPCs cells migration and tube formation (Fig. 3F and G) (2.5 fold). To elucidate whether BDNF-dependent VEGF plays an important role in angiogenesis, the VEGF antibody was used. As shown in Fig. 3F and G, incubation of CM with VEGF antibody reduced BDNFinduced migration and tube formation in EPC. These data indicated that BDNF increases VEGF expression promotes angiogenesis in human chondrosarcoma cells. Previous studies have shown that BDNF exerts its effects by interacting with a specific TrkB receptor [37,38]. We therefore determined whether the TrkB receptor is involved in BDNF-induced VEGF expression and angiogenesis in chondrosarcoma. Pretreatment of cells with the TrkB receptor inhibitor K252a for 30 min or transfection of cells with TrkB shRNA for 24 h reduced BDNFinduced increases in VEGF expression (Fig. 3H–J). In addition, CM from chondrosarcoma demonstrated that the TrkB inhibitor or shRNA reduced BDNF-mediated migration and tube formation of EPCs (Fig. 3K and L). These data suggested that BDNF enhanced VEGF expression and promoted angiogenesis through the TrkB receptor. 3.4. PLCg and PKCa activation are involved in BDNF-induced VEGF expression and angiogenesis The PLCg/PKCa pathway is a common downstream molecule of the TrkB receptor [39,40]. We next examined whether the PLCg/ PKCa pathway is involved in BDNF-mediated VEGF expression and angiogenesis. The results show that pretreatment of cells for 30 min with PLCg inhibitor U7122 or PKCa inhibitors GF109203X and Go¨6976, or transfection of cells for 24 h with PLCg and PKCa siRNAs, or with a PKCa mutant abolished BDNF-induced VEGF expression, but this was not observed with PLCb inhibitor U73343 (Fig. 4A–F). In addition, BDNF-mediated EPC migration and tube formation were diminished by pre-treatment with PLCg and PKCa inhibitors or siRNAs (Fig. 4G–J). Moreover, incubation of JJ012 cells with BDNF increased PLCg and PKCa phosphorylation in a timedependent manner (Fig. 4K). Pretreatment of cells with K252a or U73122 markedly inhibited BDNF-induced PLCg and PKCa phosphorylation (Fig. 4L). Based on these results, it appears that BDNF acted through the TrkB, PLCg, and PKCa pathways to enhance VEGF expression and angiogenesis in human chondrosarcoma cells. 3.5. Involvement of HIF-1a in BDNF-induced angiogenesis and VEGF expression Recent studies have shown that HIF-1a activation is necessary for VEGF expression and angiogenesis of human chondrosarcoma

526

C.-Y. Lin et al. / Biochemical Pharmacology 91 (2014) 522–533

Fig. 1. Knockdown of BDNF decreases VEGF expression in vitro and angiogenesis in vivo. (A–C) The protein and mRNA levels of BDNF and VEGF in JJ012/control-shRNA and JJ012/BDNF-shRNA cells were measured using Western blotting, qPCR, and ELISA (n = 5). (D) The cell viability of BDNF-shRNA and control-shRNA cells was examined by MTT assay. (E) Cell migration of EPCs incubated in CM with JJ012/control-shRNA or JJ012/BDNF-shRNA for 24 h was examined using a Transwell assay (F) and tube formation was photographed under a microscope (n = 4). (G) Chick embryos were incubated for 4 days with serum-free medium, JJ012/control-shRNA CM, or JJ012/BDNF-shRNA CM and then resected, fixed, and photographed with a stereomicroscope (n = 5). (H and I) Mice were injected subcutaneously with Matrigel mixed with serum-free medium, JJ012/ control-shRNA CM, or JJ012/BDNF-shRNA CM for 7 days. The plugs were excised from mice and photographed or stained with CD31 (n = 6). Scale bar = 20 mm. Results are expressed as the mean  S.E. *, p < 0.05 compared with control; #, p < 0.05 compared with BDNF-treated group.

[41,42]. To examine whether HIF-1a activation is involved in the signal transduction pathway leading to VEGF expression caused by BDNF, we pretreated cells for 30 min with an HIF-1 inhibitor or transfected cells for 24 h with HIF-1a siRNA or a mutant, all of which markedly reduced VEGF expression (Fig. 5A and B). In addition, as shown in Fig. 5C and D, the HIF-1a inhibitor, siRNA, or mutant also reduced BDNF-mediated EPC migration and tube formation. Results from western blotting indicated that BDNF increased the protein level of HIF-1a (4 fold), but not the HIF-1b in a time-dependent manner (Fig. 5E). We then examined whether BDNF could up-regulate HIF-1a protein in chondrosarcoma via an

increase in mRNA level. We found that BDNF did not affect the mRNA level of HIF-1 a (Fig. 5F). Based on the above findings, we suggest that BDNF increases the stability of HIF-1 a protein. Regulation of HIF-1 a seems to occur principally at the protein level by HIF-1 a stabilization and activation [43]. Yen et al., has been reported that diosgenin increased HIF-1 a protein through enhancing HIF-1 a protein stability. However, the detail role of BDNF in stability of HIF-1 a protein is needs further examination. To examine whether the TrkB/PLCg/PKCa signaling pathway is involved in BDNF-induced HIF-1a activation, cells were pretreated

C.-Y. Lin et al. / Biochemical Pharmacology 91 (2014) 522–533

527

Fig. 2. Knockdown of BDNF reduces tumor-associated angiogenesis in vivo. (A–E) JJ012/control-shRNA or JJ012/BDNF-shRNA cells were mixed with Matrigel and injected into flank sites of mice, which were observed for 5 weeks and then euthanized prior to tumor resection. The tumors were photographed with a microscope, weight and volume were measured, and hemoglobin levels were quantified (n = 6). (F) Correlation between tumor volume and hemoglobin levels. (G–I) Immunofluorescence staining of BDNF and VEGF, Western blotting, or immunohistochemical detection of CD31 in tissues from chondrosarcoma cell xenografts (n = 6). Scale bar = 20 mm. Results are expressed as the mean  S.E. *, p < 0.05 compared with control.

528

C.-Y. Lin et al. / Biochemical Pharmacology 91 (2014) 522–533

Fig. 3. BDNF increases VEGF expression and angiogenesis via the TrkB receptor (A–C) JJ012 cells were incubated with various concentrations of BDNF (30–100 ng/mL) for 24 h, and VEGF expression was examined using qPCR, western blotting, and ELISA (n = 5). (D and E) JJ012 cells were incubated with BDNF (50 ng/ml) for indicated time intervals, the protein and mRNA expression of VEGF were examined by using qPCR and Western blot (n = 4). (F and G) JJ012 cells were incubated with various concentrations of BDNF for 24 h or incubated with BDNF (50 ng/mL) for 24 h followed by stimulation with VEGF antibody (5 mg/mL) for 30 min. The medium was collected as CM and applied to EPCs cells for 24 h. Cell migration and capillary-like structure formation in EPCs were examined by Transwell and tube formation assays (n = 4). (H–J) Cells were pretreated with K252a (100 nM) for 30 min or transfected with TrkB shRNA for 24 h, followed by stimulation with BDNF (50 ng/mL), and VEGF expression was measured by qPCR, ELISA, and western blotting (n = 4). (K and L) JJ012 cells were pretreated with K252a (100 nM) for 30 min or transfected with TrkB shRNA for 24 h followed by stimulation with BDNF (50 ng/mL). The medium was collected as CM and then applied to EPCs cells for 24 h. The cell migration and capillary-like structure formation in EPCs were examined by Transwell and tube formation assays (n = 4). Results are expressed as the mean  S.E. *, p < 0.05 compared with control; #, p < 0.05 compared with BDNF-treated group.

with K252a, U73122, GF109203X, and Go¨6976. Reduced BDNFinduced HIF-1a accumulation in the nucleus and HIF-1a binding to the HRE element on the VEGF promoter (Fig. 5G and H) was observed. HIF-1a activation was also evaluated by analyzing the HRE luciferase activity. BDNF-mediated HRE luciferase activity was

reduced by pre-treatment with K252a, U73122, GF109203X, Go¨6976, and HIF inhibitor or by co-transfection with TrkB, PLCg, PKCa, and HIF-1a siRNA (Fig. 5I and J). Therefore, TrkB receptor, PLCg, and PKCa signaling pathways are involved in BDNFmediated HIF-1a activation.

C.-Y. Lin et al. / Biochemical Pharmacology 91 (2014) 522–533

529

Fig. 4. The PLCg/PKCa pathway is involved in BDNF-mediated VEGF expression and angiogenesis. Cells were pretreated for 30 min with U7322 (10 mM), U73343 (10 mM), GF109203X (3 mM), and Go¨6976 (3 mM) or transfected with PLCg and PKCa siRNAs or a dominant negative (DN) mutant of PKCa for 24 h, followed by stimulation with BDNF for 24 h. (A–F) The VEGF mRNA and protein expression was examined using qPCR, ELISA, and western blotting (n = 5). (G–J) The medium was collected as CM and then applied to EPCs for 24 h. Cell migration and capillary-like structure formation in EPCs were examined by Transwell and tube formation assays (n = 4). (K) JJ012 cells were incubated with BDNF (50 ng/mL) for the indicated time intervals, and PLCg or PKCa phosphorylation was examined by western blotting (n = 4). (L) JJ012 cells were pretreated for 30 min with K252a or U73122 for 30 min, followed by stimulation with BDNF (50 ng/mL) for 15 min, and PLCg or PKCa phosphorylation was determined by western blotting (n = 4). Results are expressed as the mean  S.E. *, p < 0.05 compared with the control; #, p < 0.05 compared with the BDNF-treated group.

530

C.-Y. Lin et al. / Biochemical Pharmacology 91 (2014) 522–533

Fig. 5. HIF-1a activation is involved in BDNF-promoted VEGF expression and angiogenesis. Cells were pretreated for 30 min with HIF-1 inhibitor (10 mM) or transfected with a dominant negative (DN) mutant and siRNA of HIF-1a for 24 h, followed by stimulation with BDNF for 24 h. (A and B) The VEGF mRNA and protein expression was examined using qPCR, ELISA, and western blotting (n = 5). (C and D) The medium was collected as CM and then applied to EPCs for 24 h. The cell migration and capillary-like structure formation in EPCs were examined by Transwell and tube formation assays (n = 4). (E and F) Cells were incubated with BDNF for the indicated time intervals; HIF-1a protein and mRNA expression were examined by western blotting and qPCR, respectively. JJ012 cells were pretreated with K252a, U73122, GF109203X, or Go¨6976 for 30 min, and then stimulated with BDNF for 240 min. (G) HIF-1a expression in the nucleus was determined by western blotting (n = 4). (H) HIF-1a binding to HRE was determined by the

C.-Y. Lin et al. / Biochemical Pharmacology 91 (2014) 522–533

3.6. BDNF and VEGF expression correlates with the tumor stage of chondrosarcoma patients To determine the clinical significance of BDNF and VEGF expression in patients with chondrosarcoma, we analyzed samples from chondrosarcoma patients by immunohistochemical staining. The expression of BDNF and VEGF was significantly higher in chondrosarcoma patients than in healthy individuals (Fig. 6A–C). In addition, the high level of BDNF expression correlated strongly with VEGF expression and tumor stage (Fig. 6A–C). Quantitative data also indicated that the expression of BDNF correlated with the expression of VEGF in human chondrosarcoma patients (Fig. 6D). Taken together, these results indicate that BDNF and VEGF expression correlates with tumor stage in patients with chondrosarcoma. 4. Discussion Chondrosarcoma is a rare but deadly bone cancer [44]. It is the second most common type, accounting for nearly 26% of all bone cancers. Unlike other mesenchymal malignancies such as osteosarcoma and Ewing’s sarcoma, which cause dramatic increases in long-term patient survival with the advent of systemic chemotherapy, chondrosarcoma shows a predilection for metastasis to the lungs, resulting in a poor prognosis for the patient because of the lack of an effective adjuvant therapy. Tumor metastasis is a result of many processes. Among these, the most important is angiogenesis. Therefore, it is important to explore potential targets for preventing chondrosarcoma angiogenesis and metastasis. In the current study, we using immunohistochemistry analysis and found that the expression of BDNF and VEGF in chondrosarcoma patients correlated with tumor grade and significantly higher than normal cartilage. Chondrosarcoma CM-mediated tube formation and angiogenesis was abolished by BDNF shRNA. In addition, BDNF knockdown reduced angiogenesis and tumor growth in vivo. Our study suggests that BDNF increased VEGF expression and subsequently promoted angiogenesis in human chondrosarcoma cells. One of the mechanisms underlying the BDNF-induced VEGF production and angiogenesis appeared to be activation of the TrkB receptor, as well as the PLCg, PKCa, and HIF-1a pathways. Therefore, BDNF may a novel target for chondrosarcoma angiogenesis and metastasis. BDNF has been report to increase tube formation in endothelial cells [45]. Direct application of BDNF (50 ng/ml) also increased 1.5 fold tube formation in EPCs (data not shown). In this study, we didn’t exchange the medium after BDNF stimulation (that contained the BDNF). However, VEGF antibody abolished CM-mediated tube formation in EPCs. Therefore, BDNF-induced VEGF plays key role in BDNF-increased tube formation. Nevertheless, in the pathological condition, BDNF may direct or indirect (through increasing VEGF) promotes angiogenesis in EPCs. It have been reported that BDNF induces neuronal survival, differentiation, and proliferation via the specific receptor, TrkB [18,19,46]. The TrkB receptor has been reported to mediate BDNFinduced metastasis in human chondrosarcoma cells [2,27]. In this study, we found that pretreatment of chondrosarcoma cells with a TrkB inhibitor or with shRNA blocked BDNF-induced VEGF expression. Furthermore, TrkB inhibitor or shRNA pretreatment also diminished BDNF-mediated migration and tube formation of EPCs. Therefore, the BDNF-TrkB interaction mediated VEGF expression and angiogenesis in human chondrosarcoma cells.

531

The key oncogenic signaling pathway has been linked to TrkB receptor activation, including the regulation of downstream signaling pathways, such as the PLC/PKC cascade [47]. PLCg is a member of the PLC serine/threonine family; phosphorylation of tyrosine783 activates its enzyme activity, transmits signals, and regulates corresponding cellular effects. The PKCa protein is a common downstream molecule of PLCg. PKCa has been characterized at the molecular level and has been found to mediate several cellular molecular responses, such as induction of VEGF expression [48,49]. Here, we demonstrated that PLCg and PKCa inhibitors or siRNAs antagonized BDNF-mediated VEGF expression and angiogenesis, suggesting that PLCg and PKCa activation is an obligatory event in BDNF-induced VEGF expression and angiogenesis in human chondrosarcoma cells. In addition, stimulating cells with BDNF also increased PLCg and PKCa phosphorylation. Furthermore, pretreatment of cells with K252a or U73122 reduced BDNF-promoted PLCg and PKCa phosphorylation. Thus, our results provide evidence that BDNF upregulates VEGF expression and angiogenesis in human chondrosarcoma cells via the TrkB receptor, and the PLCg and PKCa signaling pathways. HIF-1 is thought to play a major role in VEGF expression [50]. HIF-1a activates VEGF expression by binding to the HRE site within the VEGF promoter in response to hypoxia [51]. Similarly, the activation of HIF-1a by BDNF also resulted in the induction of VEGF transcription activity through the HRE site, although many other possible response elements, including activator protein-2, nuclear factor-kB, and simian virus 40 promoter factor 1, are located within the VEGF promoter. In the current study, a HIF-1a inhibitor, siRNA, or mutant reduced BDNF-induced VEGF expression and angiogenesis. Incubation of cells with BDNF promoted HIF-1a, but not HIF-1b, expression in a time-dependent manner. However, BDNF did not affect the mRNA expression of HIF-1a. It is apparent that BDNF-induced VEGF expression is substantially mediated by the HRE. In addition, TrkB receptor inhibitor reduced BDNF-induced HIF-1a accumulation in the nucleus and HIF-1a binding to HER element. Therefore, BDNF/TrkB/HIF-1a axis plays key role in BDNF-induced angiogenesis. In addition to cancer metastasis, a similar signal pathway has also been reported in the BDNF induced neuroblastoma migration, which through BDNF/ TrkB/HIF-1a axis [25]. Therefore, BDNF/TrkB/HIF-1a axis may have a role in treatment of cancer angiogenesis and metastasis. The rate-limiting step in metastasis and a critical stage in cancer progression is the acquisition of motility by a tumor cell. Lung metastasis is the major cause of mortality for patients with chondrosarcoma [52,53]. In addition, angiogenesis plays an important step in tumor metastasis. Thus, development of an anti-angiogenic and anti-metastatic therapy could conceivably be useful in these patients. In the current study, using immunohistochemistry analysis, we found that BDNF and VEGF expression in specimens from patients with chondrosarcoma correlated with tumor grade and was significantly higher than that of normal cartilage specimens. To the best our knowledge, the present study is the first to demonstrate a correlation between BDNF and VEGF expression in patients with chondrosarcoma. Further studies are required to determine whether increased BDNF and VEGF expression predicts relapse-free and progression-free survival. We also found that BDNF induced VEGF expression and subsequently promoted angiogenesis and tumor growth in human chondrosarcoma cells through activation of the TrkB receptor and the PLCg, PKCa, and HIF-1a signaling pathways (Fig. 6E). These findings may provide a better understanding of the mechanisms

chromatin immunoprecipitation assay (n = 4). (I and J) Cells were pretreated with K52a, U73122, GF109203X, Go¨6976, and HIF-1 inhibitor for 30 min or transfected with mutants or siRNAs against TrkB, PLCg, PKCa, and HIF-1a before exposure to BDNF. HRE luciferase activity was measured, and the results were normalized to b-galactosidase activity (n = 5). Results are expressed as the mean  S.E. *, p < 0.05 compared with the control; #, p < 0.05 compared with the BDNF-treated group.

532

C.-Y. Lin et al. / Biochemical Pharmacology 91 (2014) 522–533

Fig. 6. BDNF and VEGF expression correlates with tumor stage of patients with chondrosarcoma (A–C) Immunohistochemistry of BDNF and VEGF expressions in normal cartilage and chondrosarcoma tissue. The correlation and quantitative data are shown in (D). (E) Schematic diagram of the signaling pathways involved in BDNF, which regulates VEGF production and subsequently promotes angiogenesis in human chondrosarcoma cells through activation of the TrkB, PLCg, PKCa, and HIF-1a pathways. Scale bar = 20 mm.

of metastasis and may lead to the development of effective therapies of chondrosarcoma.

Ministry of Science and Technology of Taiwan (NSC102-2632-B039-001-MY3; 102-2320-B-039-030-MY3; 103-2628-B-039-002MY3) and China Medical University Hospital (DMR-103-047).

Conflict of interest statement All authors have no financial or personal relationships with other people or organizations that could inappropriately influence our work. Acknowledgments We thank Dr. W.M. Fu for providing HIF-1a mutant and HRE promoter construct. This work was supported by grants from the

References [1] Pescador D, Blanco J, Corchado C, Jimenez M, Varela G, Borobio G, et al. Chondrosarcoma of the scapula secondary to radiodermatitis. Int J Surg Case Rep 2012;3:134–6. [2] Lin CY, Chen HJ, Li TM, Fong YC, Liu SC, Chen PC, et al. beta5 Integrin upregulation in brain-derived neurotrophic factor promotes cell motility in human chondrosarcoma. PloS One 2013;8:e67990. [3] Wu MH, Chen LM, Hsu HH, Lin JA, Lin YM, Tsai FJ, et al. Endothelin-1 enhances cell migration through COX-2 up-regulation in human chondrosarcoma. Biochim Biophys Acta 2013;1830:3355–64.

C.-Y. Lin et al. / Biochemical Pharmacology 91 (2014) 522–533 [4] Abbas S, Vincourt JB, Habib L, Netter P, Greige-Gerges H, Magdalou J. The cucurbitacins E, D and I: investigation of their cytotoxicity toward human chondrosarcoma SW 1353 cell line and their biotransformation in man liver. Toxicol Lett 2013;216:189–99. [5] Yin MC. Humans and cancer. An ongoing fight. BioMedicine 2012;2:83. [6] Ling N, Gu J, Lei Z, Li M, Zhao J, Zhang HT, et al. microRNA-155 regulates cell proliferation and invasion by targeting FOXO3a in glioma. Oncol Rep 2013;30: 2111–8. [7] Hsieh JY, Huang TS, Cheng SM, Lin WS, Tsai TN, Lee OK, et al. miR-146a-5p circuitry uncouples cell proliferation and migration, but not differentiation, in human mesenchymal stem cells. Nucl Acids Res 2013;41:9753–63. [8] Li WH, Qiu Y, Zhang HQ, Liu Y, You JF, Tian XX, et al. P2Y2 receptor promotes cell invasion and metastasis in prostate cancer cells. Br J Cancer 2013;109: 1666–75. [9] Bhat TA, Nambiar D, Tailor D, Pal A, Agarwal R, Singh RP. Acacetin inhibits in vitro and in vivo angiogenesis and down-regulates Stat signaling and VEGF expression. Cancer Prev Res 2013;6:1128–39. [10] Keating AM, Jacobs DS. Anti-VEGF treatment of corneal neovascularization. Ocul Surf 2011;9:227–37. [11] De Francesco EM, Lappano R, Santolla MF, Marsico S, Caruso A, Maggiolini M. HIF-1alpha/GPER signaling mediates the expression of VEGF induced by hypoxia in breast cancer associated fibroblasts (CAFs). Breast Cancer Res 2013;15:R64. [12] Honguero Martinez AF, Arnau Obrer A, Figueroa Almazan S, Martinez Hernandez N, Guijarro Jorge R. Prognostic value of the expression of vascular endothelial growth factor A and hypoxia-inducible factor 1alpha in patients undergoing surgery for non-small cell lung cancer. Med Clin (Barc) 2013;142: 432–7. [13] Ma JX, Sun YL, Wang YQ, Wu HY, Jin J, Yu XF. Triptolide induces apoptosis and inhibits the growth and angiogenesis of human pancreatic cancer cells by downregulating COX-2 and VEGF. Oncol Res 2013;20:359–68. [14] Shin JM, Jeong YJ, Cho HJ, Park KK, Chung IK, Lee IK, et al. Melittin suppresses HIF-1alpha/VEGF expression through inhibition of ERK and mTOR/p70S6K pathway in human cervical carcinoma cells. PloS One 2013;8:e69380. [15] Schwartzberg LS, Tauer KW, Hermann RC, Makari-Judson G, Isaacs C, Beck JT, et al. Sorafenib or placebo with either gemcitabine or capecitabine in patients with HER-2-negative advanced breast cancer that progressed during or after bevacizumab. Clin Cancer Res 2013;19:2745–54. [16] Liu LP, Ho RL, Chen GG, Lai PB. Sorafenib inhibits hypoxia-inducible factor1alpha synthesis: implications for antiangiogenic activity in hepatocellular carcinoma. Clin Cancer Res 2012;18:5662–71. [17] Chen A, Xiong LJ, Tong Y, Mao M. Neuroprotective effect of brain-derived neurotrophic factor mediated by autophagy through the PI3K/Akt/mTOR pathway. Mol Med Rep 2013;8:1011–6. [18] Bechara RG, Kelly AM. Exercise improves object recognition memory and induces BDNF expression and cell proliferation in cognitively enriched rats. Behav Brain Res 2013;245:96–100. [19] Akil H, Perraud A, Melin C, Jauberteau MO, Mathonnet M. Fine-tuning roles of endogenous brain-derived neurotrophic factor, TrkB and sortilin in colorectal cancer cell survival. PloS One 2011;6:e25097. [20] Chu ZB, Sun CY, Yang D, Chen L, Hu Y. Role of brain-derived neurotrophic factor in bone marrow angiogenesis in multiple myeloma. J Huazhong Univ Sci Technol Med Sci 2013;33:485–90. [21] Au CW, Siu MK, Liao X, Wong ES, Ngan HY, Tam KF, et al. Tyrosine kinase B receptor and BDNF expression in ovarian cancers – effect on cell migration, angiogenesis and clinical outcome. Cancer Lett 2009;281:151–61. [22] Lam CT, Yang ZF, Lau CK, Tam KH, Fan ST, Poon RT. Brain-derived neurotrophic factor promotes tumorigenesis via induction of neovascularization: implication in hepatocellular carcinoma. Clin Cancer Res 2011;17: 3123–33. [23] Sasahira T, Ueda N, Yamamoto K, Bhawal UK, Kurihara M, Kirita T, et al. Trks are novel oncogenes involved in the induction of neovascularization, tumor progression, and nodal metastasis in oral squamous cell carcinoma. Clin Exp Metastasis 2013;30:165–76. [24] Sun CY, Hu Y, Huang J, Chu ZB, Zhang L, She XM, et al. Brain-derived neurotrophic factor induces proliferation, migration, and VEGF secretion in human multiple myeloma cells via activation of MEK-ERK and PI3K/AKT signaling. Tumour Biol 2010;31:121–8. [25] Nakamura K, Martin KC, Jackson JK, Beppu K, Woo CW, Thiele CJ. Brain-derived neurotrophic factor activation of TrkB induces vascular endothelial growth factor expression via hypoxia-inducible factor-1alpha in neuroblastoma cells. Cancer Res 2006;66:4249–55. [26] Bovee JV, Hogendoorn PC, Wunder JS, Alman BA. Cartilage tumours and bone development: molecular pathology and possible therapeutic targets. Nat Rev Cancer 2010;10:481–8. [27] Lin CY, Chang SL, Fong YC, Hsu CJ, Tang CH. Apoptosis signal-regulating kinase 1 is involved in brain-derived neurotrophic factor (BDNF)-enhanced cell motility and matrix metalloproteinase 1 expression in human chondrosarcoma cells. Int J Mol Sci 2013;14:15459–78.

533

[28] Block JA, Inerot SE, Gitelis S, Kimura JH. Synthesis of chondrocytic keratan sulphate-containing proteoglycans by human chondrosarcoma cells in longterm cell culture. J Bone Joint Surg Am 1991;73:647–58. [29] Wang HH, Lin CA, Lee CH, Lin YC, Tseng YM, Hsieh CL, et al. Fluorescent gold nanoclusters as a biocompatible marker for in vitro and in vivo tracking of endothelial cells. ACS Nano 2011;5:4337–44. [30] Wu MH, Huang CY, Lin JA, Wang SW, Peng CY, Cheng HC, et al. Endothelin-1 promotes vascular endothelial growth factor-dependent angiogenesis in human chondrosarcoma cells. Oncogene 2014;33:1725–35. [31] Huang CY, Chen SY, Tsai HC, Hsu HC, Tang CH. Thrombin induces epidermal growth factor receptor transactivation and CCL2 expression in human osteoblasts. Arthritis Rheum 2012;64:3344–54. [32] Tang CH, Hsu CJ, Fong YC. The CCL5/CCR5 axis promotes interleukin-6 production in human synovial fibroblasts. Arthritis Rheum 2010;62:3615–24. [33] Nickols NG, Jacobs CS, Farkas ME, Dervan PB. Modulating hypoxia-inducible transcription by disrupting the HIF-1-DNA interface. ACS Chem Biol 2007;2:561–71. [34] Passaniti A, Taylor RM, Pili R, Guo Y, Long PV, Haney JA, et al. A simple, quantitative method for assessing angiogenesis and antiangiogenic agents using reconstituted basement membrane, heparin, and fibroblast growth factor. Lab Invest 1992;67:519–28. [35] Bergers G, Benjamin LE. Tumorigenesis and the angiogenic switch. Nature Rev Cancer 2003;3:401–10. [36] Peters BA, Diaz LA, Polyak K, Meszler L, Romans K, Guinan EC, et al. Contribution of bone marrow-derived endothelial cells to human tumor vasculature. Nat Med 2005;11:261–2. [37] Besusso D, Geibel M, Kramer D, Schneider T, Pendolino V, Picconi B, et al. BDNF-TrkB signaling in striatopallidal neurons controls inhibition of locomotor behavior. Nat Commun 2013;4:2031. [38] Girardet C, Lebrun B, Cabirol-Pol MJ, Tardivel C, Francois-Bellan AM, Becquet D, et al. Brain-derived neurotrophic factor/TrkB signaling regulates daily astroglial plasticity in the suprachiasmatic nucleus: electron-microscopic evidence in mouse. Glia 2013;61:1172–7. [39] Berghuis P, Agerman K, Dobszay MB, Minichiello L, Harkany T, Ernfors P. Brainderived neurotrophic factor selectively regulates dendritogenesis of parvalbumin-containing interneurons in the main olfactory bulb through the PLCgamma pathway. J Neurobiol 2006;66:1437–51. [40] Corbett GT, Roy A, Pahan K. Sodium phenylbutyrate enhances astrocytic neurotrophin synthesis via protein kinase C (PKC)-mediated activation of cAMP-response element-binding protein (CREB): implications for Alzheimer disease therapy. J Biol Chem 2013;288:8299–312. [41] Fang J, Yan L, Shing Y, Moses MA. HIF-1alpha-mediated up-regulation of vascular endothelial growth factor, independent of basic fibroblast growth factor, is important in the switch to the angiogenic phenotype during early tumorigenesis. Cancer Res 2001;61:5731–5. [42] Lin C, McGough R, Aswad B, Block JA, Terek R. Hypoxia induces HIF-1alpha and VEGF expression in chondrosarcoma cells and chondrocytes. J Orthop Res 2004;22:1175–81. [43] Jiang BH, Rue E, Wang GL, Roe R, Semenza GL, Dimerization. DNA binding, and transactivation properties of hypoxia-inducible factor 1. J Biol Chem 1996;271:17771–78. [44] Wu MH, Huang CY, Lin JA, Wang SW, Peng CY, Cheng HC, et al. Endothelin-1 promotes vascular endothelial growth factor-dependent angiogenesis in human chondrosarcoma cells. Oncogene 2013;33:1725–35. [45] Long BL, Rekhi R, Abrego A, Jung J, Qutub AA. Cells as state machines: cell behavior patterns arise during capillary formation as a function of BDNF and VEGF. J Theor Biol 2013;326:43–57. [46] Perez-Gomez A, Tasker RA. Transient domoic acid excitotoxicity increases BDNF expression and activates both MEK- and PKA-dependent neurogenesis in organotypic hippocampal slices. BMC Neurosci 2013;14:72. [47] Fenner BM. Truncated TrkB: beyond a dominant negative receptor. Cytokine Growth Factor Rev 2012;23:15–24. [48] Berk BC, Taubman MB, Cragoe Jr EJ, Fenton 2nd JW, Griendling KK. Thrombin signal transduction mechanisms in rat vascular smooth muscle cells. Calcium and protein kinase C-dependent and -independent pathways. J Biol Chem 1990;265:17334–40. [49] Bekhite MM, Finkensieper A, Binas S, Muller J, Wetzker R, Figulla HR, et al. VEGF-mediated PI3K class IA and PKC signaling in cardiomyogenesis and vasculogenesis of mouse embryonic stem cells. J Cell Sci 2011;124:1819–30. [50] Ahluwalia A, Tarnawski AS. Critical role of hypoxia sensor – HIF-1alpha in VEGF gene activation. Implications for angiogenesis and tissue injury healing. Curr Med Chem 2012;19:90–7. [51] Shima DT, Kuroki M, Deutsch U, Ng YS, Adamis AP, D’Amore PA. The mouse gene for vascular endothelial growth factor. Genomic structure, definition of the transcriptional unit, and characterization of transcriptional and posttranscriptional regulatory sequences. J Biol Chem 1996;271:3877–83. [52] Tang CH. Molecular mechanisms of chondrosarcoma metastasis. BioMedicine 2012;2:92–8. [53] Chen JC, Fong YC, Tang CH. Novel strategies for the treatment of chondrosarcomas: targeting integrins. BioMed Res Int 2013;2013:1–11.