Thrombosis Research 136 (2015) 118–124
Contents lists available at ScienceDirect
Thrombosis Research journal homepage: www.elsevier.com/locate/thromres
Regular Article
The cell-membrane prothrombinase, fibrinogen-like protein 2, promotes angiogenesis and tumor development Esther Rabizadeh a,b, Izhack Cherny a, Doron Lederfein a, Shany Sherman a, Natalia Binkovsky a, Yevgenia Rosenblat c, Aida Inbal a,d,⁎ a
Hemato-Oncology Laboratory, Felsenstein Medical Research Center, Petach Tikva, Israel 1 Hematology Laboratory, Rabin Medical Center, Beilinson Hospital, Petach Tikva, Israel c Pathology Institute, Rabin Medical Center, Beilinson Hospital, Petach Tikva, Israel d Thrombosis and Hemostasis Unit, Hematology Institute, Rabin Medical Center, Beilinson Hospital, Petach Tikva, Israel 1
b
a r t i c l e
i n f o
Article history: Received 7 July 2014 Received in revised form 9 November 2014 Accepted 30 November 2014 Available online 4 December 2014 Keywords: Angiogenesis Fibrinogen-like protein 2 Tumorigenesis
a b s t r a c t The aim of the study was to further investigate the role of fibrinogen-like protein 2 (FGL-2), a transmembrane prothrombinase that directly cleaves prothrombin to thrombin, in angiogenesis and tumor development and the mechanism(s) underlying these processes. To study angiogenesis HUVEC clones with decreased fgl-2 mRNA were generated by specific siRNA. To study tumorigenesis SCID mice were implanted with intact (wild type) and fgl-2-silenced PC-3 clones. IFN-γ treated HUVEC expressing increased fgl-2 mRNA exhibited significant capillary sprouting that was not inhibited by hirudin, whereas fgl-2 silencing completely inhibited blood-vessel formation. Tumors (poorly differentiated carcinoma) developed in all 12 mice injected with wild type PC-3 compared with 8/12 mice injected with the fgl-2-silenced PC-3 clone. The tumors developed by fgl-2-silenced PC-3 clones were smaller and less aggressive and contained significantly fewer blood vessels (p b 0.05). All tumors’ sections were negative for thrombin staining, indicating that FGL-2-induced tumorigenesis was not mediated by thrombin. In fgl-2-silenced tumors there was a decrease in fgl-2 mRNA (p = 0.02) and ERK1/2 phosphorylation (p b 0.05) by 80% and a 20%, respectively. The mechanism underlying these processes, studied in PC-3 clones, revealed that fgl-2 silencing was associated with a 65% decrease in FGF-2 mRNA (p b 0.01) and a 30% down regulation of ERK1/2 phosphorylation (p b 0.05). Together, these results suggest that FGL-2 mediates angiogenesis and tumorigenesis not by thrombin-mediated mechanism but rather through FGF-2/ERK signaling pathway. FGL-2 may serve as a valuable therapeutic target in the future. © 2014 Elsevier Ltd. All rights reserved.
Introduction The bidirectional relationship between cancer and thrombosis has been known for almost two centuries [1–3]. Thrombosis often precedes the diagnosis of cancer, and its presence is associated with a detrimental disease course [4,5]. These findings support the paradigm that coagulation and tumor growth form a vicious circle in which hypercoagulability facilitates the aggressive biology of cancer and vice versa. The mechanism
Abbreviations: ERK, extracellular-signal-regulated kinases; FGF-2, basic fibroblast growth factor; FGL-2, fibrinogen-like protein 2; HUVEC, human umbilical vein endothelial cells; IFN-γ, human interferon-gamma; MAPK, mitogen-activated protein kinases; SCID, Severe Combined Immunodeficiency ⁎ Corresponding author at: Thrombosis and Hemostasis Unit, Hematology Institute, Rabin Medical Center, Beilinson Hospital, Petach Tikva 49100, Israel. Tel.: + 972 3 9377912; fax: +972 3 920 1568. E-mail addresses: [email protected] (E. Rabizadeh), [email protected] (I. Cherny), [email protected] (D. Lederfein), [email protected] (S. Sherman), [email protected] (N. Binkovsky), [email protected] (Y. Rosenblat), [email protected] (A. Inbal). 1 Affiliated with Sackler Faculty of Medicine, Tel Aviv University, Israel.
http://dx.doi.org/10.1016/j.thromres.2014.11.023 0049-3848/© 2014 Elsevier Ltd. All rights reserved.
underlying these events is still unclear. Malignant cells are known to directly activate blood coagulation in three ways: by producing procoagulant, fibrinolytic, and proaggregating factors; by releasing proinflammatory and proangiogenic cytokines, such as tumor necrosis factor alpha (TNF-α) and interleukin (IL)-1ß; and by interacting directly with host endothelial cells, leukocytes, and platelets via adhesion molecules [5]. However, the precise procoagulant proteins that stimulate tumorigenesis have not been identified. One potential candidate is the cell-membrane-associated protein, fibrinogen-like protein 2 (FGL-2)/fibroleukin, which has been shown to induce sprouting in vascular endothelial cells [6] and to be overexpressed in tumor cells [7]. FGL-2, also known as FGL-2prothrombinase, is a member of the fibrinogen family of proteins [8]. It exerts serine protease activity and is capable of directly cleaving prothrombin to thrombin in the absence of factor VII or factor X. Like plasmatic prothrombinase, factor Xa, FGL-2 prothrombinase requires phospholipids, calcium, and factor Va for optimal catalytic activity [9]. However, unlike factor Xa, FGL-2 is a transmembrane protein which is not inhibited by antithrombin in the presence of heparin or by other protease inhibitors that inhibit factor Xa [9].
E. Rabizadeh et al. / Thrombosis Research 136 (2015) 118–124
The human gene encoding FGL-2, originally cloned from cytotoxic T lymphocytes, spans approximately 7 kb on chromosome 7 and contains 2 exons. The 70-KD protein is predicted to be 439 amino acids (aa) long, with the N-terminus including a 2-aa-long cytoplasmic domain and a 21-aa-long transmembrane domain. The remaining 416 aa constitute the extracellular domain. FGL-2 shares 36% sequence homology with the fibrinogen β and γ chains and 40% homology with the fibrinogen-related domain (FRED) of tenascin [10]. The murine and human proteins share 78% overall identity, with greater conservation at the C terminus [11,12]. FGL-2 is expressed on the surface of activated macrophages and endothelial cells and also secreted by peripheral blood CD4+ and CD8+ T cells. The secreted protein is devoid of coagulation activity. It has potent modulatory effects on the adaptive immune system and was reported to inhibit the maturation of dendritic cells [13,14]. The prothrombinase and immune activities of FGL-2 are located on distinct domains on the FGL-2 molecule [9,14]. The prothrombinase activity of FGL-2, first described in a murine fulminant hepatitis model [15], is exhibited when FGL-2 is expressed on activated macrophages and endothelial cells in the form of a membrane-associated protein. It has shown to be associated with both experimental and human allograft rejection that was abrogated following neutralization of FGL-2 by antibodies or in FGL-2 knockout mice [16]. Macrophage and endothelial cell induction of FGL-2 occurs via interferon gamma (IFN –γ) [17]. FGL-2 also plays a role in tumor development. Overexpression of FGL-2 has been detected in tumor and interstitial inflammatory cells but not in the normal surrounding tissue [7]. A recent study found that knockdown of FGL-2 delayed tumor growth and angiogenesis in mice injected with the human hepatocellular carcinoma (HCC) cell line [18]. The authors hypothesized that the protumorigenic activity of FGL-2 is a result of FGL-2-induced generation of thrombin leading to thrombin-induced tumorigenesis [7,18]. However, this does not explain the upregulation of FGL-2 in tumor cells. The aim of the present study was to substantiate the role of FGL-2 in angiogenesis and tumor development and to uncover the mechanism underlying these processes. Using an in vitro pro-angiogenic assay and an in vivo mice model of tumor development, we established that FGL-2 exerts direct, non-thrombin-mediated, angiogenic and tumorigenic activity. Material and Methods Fgl-2 Expression The expression of fgl-2 was analyzed in human umbilical vein endothelial cells (HUVEC) in the presence or absence of IFN-γ (20 ng/ml) and in a wild type (WT) human prostate carcinoma cell line, PC-3, established from bone metastasis (ATCC, Beit-Haemek, Israel). Total RNA was isolated using RNAqueous™ (Ambion #AM1912, Invitrogen, Austin, TX, USA) and analyzed by real-time polymerase chain reaction (RT-PCR) using the Rotor-gene RG-3000 (Corbett, Australia). The difference in cycle time (ΔCT) was measured by comparing fgl-2 gene with abl-1 gene (housekeeping gene). The relative quantification was calculated with the formula RQ = 2-ΔΔCT. Generation of fgl-2-silenced Clones of HUVEC and PC-3 Cells Small interfering RNA (siRNA), which interferes with the expression of specific genes, was employed to evaluate the impact of inhibiting fgl2 expression on angiogenesis and tumorigenesis. The effect of siRNA was tested on HUVEC and PC-3 cells. The specific siRNA was purchased from Thermo Scientific (Pittsburg, PA, USA). Using pGIPZ lentiviral vector (Thermo Scientific), stable specific fgl-2-silenced clones and non-fgl-2-inhibited HUVEC and PC-3 clones (nonspecific siRNA clone)
119
were generated. Stable transfectants were selected with 0.2 μg/ml puromycin. In vitro Angiogenesis Assay HUVEC (5 x 103/well) were cultured in F12 (HAM) medium complemented with 1% penicillin (10,000 UI/ml)/streptomycin (10,000 UI/ml) and 10% fetal calf serum (FCS). FCS was tested for prothrombin by one-stage prothrombin-time coagulation assay (ACL1000 Coagulometer). The cells were maintained at 37 °C in a humidified incubator (85%) in 5% CO2 atmosphere. The angiogenesis assay was performed with the Chemicon in vitro ECMatrix assay kit (ECM625, Millipore, Billerica, MA, USA) using HUVEC treated with 20 ng/ml IFNγ, fgl-2-silenced clone of HUVEC, and nonspecific siRNA clone of HUVEC, in the presence or absence of 10 u/ml of hirudin (H0393, Sigma, St. Louis, Mo, USA). The extent of tube formation was monitored by light microscopy, as described by the kit manufacturer. Pattern recognition was defined by photographing the cells at the end point of the assay. A numerical value representing the degree of angiogenesis progression was assigned to each pattern according to the manufacturer’s instructions. Mouse Model Ten-week-old male severe combined immunodeficiency (SCID) mice were purchased from Harlan Laboratories (Rehovot, Israel). The mice were maintained in laminar flow cabinets under specific pathogen-free conditions and a daily cycle of 12 h light/12 h dark. All food, water, and litter were sterilised prior to use, and temperature (20-21 °C) and humidity (50-60%) were controlled. Cages were changed fully once a week. Animals were manipulated under sterile conditions. The mice were managed in accordance with the NIH Guidelines on Laboratory Animal Welfare. Study and all protocols were approved and monitored by the Animal Care Committee of Rabin Medical Center. Twenty-four mice underwent subcutaneous cell implantation as follows: 10 unmanipulated (WT) PC-3 cells; 12 fgl-2 siRNA-silenced PC-3 clone; 2 PC-3 transfected with nonspecific siRNA. In each case, 2.5 x 106 tumor cells suspended in 100 μl of phosphate-buffered saline were injected without anesthesia using a 27-gauge needle and a 1-ml disposable syringe. Tumor progression was monitored by palpation 3 times a week. Subcutaneous tumor dimensions (length (L) and width (W)) were measured with a caliper, and volume was calculated according to the formula (L × W2)/2 [19]. Mice were sacrificed by cervical dislocation in week 6 after implantation. Autopsy was performed to assess the distribution of metastases. Tumors, spleen, liver, and lung were harvested. The tissues were fixed in 4% formaldehyde for histological analysis. Immunohistological Analysis and FGL-2 Expression in Mouse Tumor Tissues Histological sections of the tumors and other mouse tissues were stained with haematoxylin-eosin. Immunohistochemistry study was performed on formalin-fixed paraffin-embedded sections (4 microns thick) using anti-FGL-2 monoclonal antibody (Abnova, St. Louis, MO, USA) diluted 1:400; mouse monoclonal anti- thrombin antibody F-1 (Santa Cruz Biotechnology, Santa Cruz, CA, USA) diluted 1:200; and anti-basic fibroblast growth factor (FGF-2) monoclonal antibody diluted 1:100 (#SC-79, Santa Cruz Biotechnology). Normal hepatic tissue served as a positive control for the anti-thrombin antibody. Endogenous reactions were blocked with DAKO blocking kit. The slides were stained with the automatic Bench Mark XT kit (Ventana Medical Systems, Oro Valley, AZ, USA) consisting of labelled streptavidin-biotin reagents and counterstained with haematoxylin. For quantitation of blood vessels, the histological sections were stained with polyclonal anti-human von Willebrand factor (VWF) antibody diluted 1:300 (#A0082, DAKO,
120
E. Rabizadeh et al. / Thrombosis Research 136 (2015) 118–124
Denmark). Triplicates of each tumor section were stained and the number of VWF-positive blood vessels was counted in 10 fields per each stained slide by two independent examiners using light microscope. Proteomics of Angiogenesis and MAPK Pathway-related Proteins Induced by fgl-2 in PC-3 Clones or Tumors Induced by these Clones The effects of fgl-2 silencing on the expression of angiogenesisrelated proteins were studied using Proteome Profiler Human Angiogenesis Array kit (R&D Systems, Abingdon, UK). Proteins extracted from PC-3 clones before or after transfection with specific siRNA for fgl-2 silencing were hybridized with an array of antibodies against angiogenesis-related proteins according to manufacturer instructions. The MAPK phosphorylation profile was studied using Proteome Profiler Human Phospho-MAPK Array kit (R&D Systems, Abingdon, UK)). PC-3 cells before (WT) and after transfection with specific fgl-2 silencing siRNA as well as proteins extracted from the mice tumors induced by PC-3 clones were lysed in the presence of phosphatase inhibitor cocktails B and C (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Protein cell extracts were mixed with a specific cocktail of biotinylated detection antibodies and analyzed according to manufacturer’s instructions. The phosphorylation experiments were further validated with ERK1/2 phosphorylation ELISA kit (InstantOneTM. ELISA, eBioscience, San Diego, CA). Both total and phosphorylated ERK1/2 protein levels were determined. The normalized phosphorylation levels were calculated according to the total amount of ERK1/2. The chemiluminescence signal intensity was considered proportional to the amount of protein detected. Fgf-2 Expression in PC-3 Clones Total RNA was extracted from PC-3 clones, as follows: WT PC-3; fgl-2 silenced PC-3, and nonspecific siRNA PC-3. The levels of fgf-2 mRNA were analyzed by RT-PCR and compared to the house-keeping gene, abl-1. RT-PCR primers were obtained from Solaris (Thermo Fisher Scientific Inc.). cDNA was produced using Maxima First Strand cDNA Synthesis Kit for RT-qPCR (Thermo Fisher Scientific) according to the manufacturer’s instructions. qPCR was performed using GoTaq Green Master Mix (Promega, Madison WI, USA) according to the manufacturer’s instructions. Statistical Cnalysis Statistical analyses were performed with Student's t-test (independent sample or one sample) or Mann-Whitney U test, where appropriate, using GraphPad Prism (ver. 5) or SPSS (ver. 21) software. The difference in mean tumor growth rate under the various conditions was estimated by repeated measures ANOVA with post-hoc for each group of mice. A p-value of b 0.05 was considered statistically significant. Results Generation of HUVEC Clones with Different fgl-2 Expression HUVEC clones were generated with basal, increased, and decreased expression of fgl-2, as shown in Fig. 1. There was a 5-fold increase in basal fgl-2 mRNA after the addition of 20 ng/ml IFN-γ (Fig. 1). HUVEC transfection with specific fgl-2-siRNA led to 80% silencing of fgl2 mRNA whereas nonspecific siRNA decreased fgl-2 expression by only 10% (Fig. 1).
Fig. 1. Expression of fgl-2 in HUVEC. RT-PCR analysis of mRNA of fgl-2 from HUVEC incubated with IFN-γ before (WT), or after transfection with specific (S) or nonspecific (N) siRNA for fgl-2 silencing. Findings are expressed relative to the house-keeping gene, abl-1, and presented relative to untreated cells (left column). The results (mean ± SD) are duplicates of three separate experiments. *p b 0.05 for difference between IFN-γ-treated HUVEC with or without specific siRNA.
and nonspecific siRNA clone of HUVEC. Only traces of prothrombin were detected (0.05%) in the FCS used for the angiogenesis experiments. Nevertheless, to inhibit any traces of thrombin that may have been generated locally from the traces of prothrombin, 10 u/ml hirudin was added in some of the experiments. A representative picture of the tube arrays is shown in Fig. 2. The extent of tube arrays generated was similar for WT HUVEC (Fig. 2A) and HUVEC transfected with nonspecific siRNA (Fig. 2C). By contrast, tube array generation was almost completely inhibited by fgl-2 silencing (Fig. 2B, p b 0.001). Hirudin had no effect on the extent of tube formation (Fig. 2D). The number of patterns was expressed on a 5-point scale representing the degree of angiogenesis, from 0 (individual cells) to 5 (complex mesh-like structure) (Millipore). Mean ± SD values by test condition were as follows: WT HUVEC, 3.5 ± 0.5 (32 scored fields); fgl-2 siRNA-silenced HUVEC, 0.6 ± 0.80 (p b 0.001) (23 scored fields); and HUVEC transfected with nonspecific siRNA without hirudin, 3.1 ± 0.8 (30 scored fields), and with hirudin, 3.5 ± 0.5 (6 scored fields). Fgl-2 Expression in PC-3 To study the role of FGL-2 in tumor development we chose to use the human prostate cancer cell line PC-3 as a model. PC-3 efficiently develops subcutaneous tumors in mice and exhibits relatively high metastatic potential [20,21]. PC-3 clones with either intact or decreased fgl-2 expression were generated by transfection of PC-3 cell line with specific or nonspecific siRNA, respectively, and fgl-2 mRNA was measured using RT-PCR, as detailed in Methods. As shown in Fig. 3, fgl-2 mRNA was inhibited by more than 90% in the siRNA fgl-2silenced PC-3 clone relative to the WT PC-3 cells (8.0% versus 100% mRNA; p b 0.05). There was no significant difference in fgl-2 mRNA between PC-3 clone transfected with nonspecific siRNA and WT PC-3 (81.5% versus 100%, respectively; p = 0.1). Role of FGL-2 in Tumor Development
Role of FGL-2 in Angiogenesis The role of FGL-2 in angiogenesis was studied using different HUVEC clones grown in ECMatrix as specified in Methods: IFN-γ-treated HUVEC expressing increased fgl-2 (WT), fgl-2 siRNA-silenced HUVEC,
The role of FGL-2 in tumor development was studied in SCID mice injected with WT PC-3 (n = 10), siRNA-fgl-2 -silenced PC-3 clone (n = 12), or PC-3 clone transfected with nonspecific siRNA (n = 2). After 6 weeks of follow-up, tumors developed in all mice inoculated
E. Rabizadeh et al. / Thrombosis Research 136 (2015) 118–124
121
Fig. 2. In vitro angiogenesis assay. Representative picture of capillary tube formation induced by WT HUVEC (A), fgl-2 -silenced HUVEC (B), and HUVEC transfected with nonspecific siRNA in the absence (C) or presence (D) of hirudin. Magnification × 200.
by WT PC-3. By contrast, in the group injected with fgl-2 siRNA-silenced PC-3, 4 mice did not develop tumors at all, and 8 had significantly smaller tumors compared to tumors developed following injection of WT PC3 (Fig. 4A and B). Two additional mice injected with PC-3 transfected with nonspecific siRNA acquired tumors similar to those inoculated
with WT PC-3. At each time-point of the follow-up the volume of the tumors in the WT PC-3 group was larger than that observed in the group injected with fgl-2 -silenced clone, approaching an 8-fold increase in volume at the end of the follow up (WT PC-3 clone - 413.9 mm3 versus fgl-2 silenced clone - 52.6 mm3; p = 0.002; Fig. 4C). Lung metastasis developed in 1 of the 10 mice injected with WT PC-3 and none of the mice injected with fgl-2-silenced clone. Histological study of tumour sections from the mice injected with WT PC-3 revealed poorly differentiated carcinoma penetrating muscle fibers. A representative histological section is shown in Fig. 4D. Most of the cells were large with prominent pleomorphism and a vesicular or hyperchromatic nucleus and nucleolus. The histology of the tumors from the fgl-2-silenced PC-3 group was similar, but there was no noticeable penetration into muscle fibers, and the number of blood vessels was significantly lower (Fig. 4E; p b 0.05). Immunohistochemical staining revealed clearly visible FGL-2 protein in the tumor cells in all mouse groups (Fig. 5A). In addition, almost all tumor cells were positive for FGF-2 (Fig. 5B). It should be noted here that the immunohistochemical staining technique is not a quantitative test, yet it is suitable to confirm or negate protein expression. Accordingly, staining of tumor sections from each group of mice with antithrombin antibody that recognizes human and mice thrombin was negative, strongly indicating a lack of thrombin in these sections. Tumors that developed after injection of fgl-2 silenced PC-3 clones showed an 80% reduction in fgl-2 mRNA (Fig. 5C; p b 0.05). Effect of fgl-2 Silencing on Angiogenesis-related Proteins
Fig. 3. Fgl-2 expression in different PC-3 clones. RT-PCR analysis of mRNA of fgl-2 extracted from WT PC-3, PC-3 transfected with nonspecific fgl-2 siRNA, or PC-3 transfected with specific fgl-2-silencing siRNA. Findings are expressed relative to the housekeeping gene abl-1 and presented relative to WT PC-3. The results (mean ± SD) are duplicates of three separate experiments.
To determine if FGL-2 affects the levels of proteins that participate in angiogenesis, proteins extracted from PC-3 clones before or after transfection with specific siRNA for fgl-2 silencing were hybridized with an array of antibodies against angiogenesis-related proteins. Fgl-2-
122
E. Rabizadeh et al. / Thrombosis Research 136 (2015) 118–124
Fig. 4. SCID mice model of PC-3-induced tumors. Macroscopic picture of the tumor (surrounded by arrows) in mice injected with (A) WT PC-3 or (B) fgl-2 -silenced PC-3. (C) Mean ± SEM volume of the tumors at each week in mice injected with WT PC-3 or fgl-2 -silenced PC-3. (D) Haematoxylin-eosin staining of the tumor generated by WT PC-3 showing vesicular nucleus and nucleolus (arrow) (magnification × 200). (E) Mean ± SEM of the number of blood vessels in the tumours induced by WT PC-3 or fgl-2 silenced PC-3 counted by light microscopy following staining with anti VWF antibody, as detailed in Methods. *p b 0.05.
silencing was associated with a decrease in the FGF-2 protein level (data not shown). This finding was further validated by fgf-2 mRNA analysis. In fgl-2 silenced PC3 clone, the expression of fgf-2 was inhibited by 65% (mean ± SD, 34.4% ± 1.7%; p b 0.01), whereas no effect was observed in WT PC3 or nonspecific siRNA clone (mean ± SD, 85.7% ± 7.1%; p = 0.19), (Fig. 6A). Effect of fgl-2 Silencing on MAPK Phosphorylation Since MAPK plays a key role in FGF-2-mediated signaling of cell proliferation and tumourigenesis [22], the effect of fgl-2 silencing on the MAPK phosphorylation profile was studied using an array of specific MAPK related proteins. Fgl-2-silencing was associated with a significant decrease in the ERK1 and ERK2 phosphorylation (data not shown). This finding was validated by ERK1/2 phosphorylation ELISA kit, which revealed that fgl-2-silencing was associated with downregulation of ERK1/2 phosphorylation by 30 % (p b 0.05; Fig. 6B). In line with this finding, there was a 20% decrease in ERK1/2 phosphorylation (Fig. 6C) in tumors generated by fgl-2 silenced clones (p b 0.05; Fig. 6C). Discussion This study shows that the transmembrane serine protease FGL-2, which has well-established prothrombinase coagulation activity, plays a substantial role in angiogenesis and tumorigenesis. It is known that coagulation and angiogenesis are closely interrelated. Following injury, proteins generated by the coagulation system
coordinate the spatial localization and stabilization of endothelial cells for the growth and repair of the damaged blood vessel. Once the clot is stabilized, angiogenesis is modulated by proteins and cryptic peptide fragments generated from the coagulation and fibrinolytic systems [23]. The well-characterized proangiogenic coagulation proteins are thrombin [24], factor VII, tissue factor [25], and factor XIII [26]. Angiogenesis is also a vital process during tumor development. The coagulation activation and tumor growth rely on the capacity of cancer cells to promote neoangiogenesis and metastasis [27]. Several components of the hemostatic system contribute to these processes, such as thrombin, tissue factor, factor VIIa, factor Xa, and fibrinogen, as clearly documented in both in vitro and in vivo tumor models [28] . Of these, tissue factor is the best characterized. Constitutively expressed on the malignant cell surface, tissue factor can lead to the formation of both localized and systemic procoagulant states. Another tumor procoagulant is cancer procoagulant (CP) that, unlike TF, directly activates FX independently of FVII. CP has been detected in various tumor cells [27,28]. Herein we present a new tumor procoagulant, FGL-2, which may serve a link between cancer and coagulation. Like the other coagulation proteins [24–28], FGL-2 exerts substantial pro-angiogenic activity, as indicated in the present study by the 5-fold increase in fgl-2 expression in IFN-γ-stimulated HUVEC. Accordingly, Su et al. [7] reported a 10-fold increase in fgl-2 mRNA following stimulation with IFN-γ. In our experiments, HUVEC stimulated with IFN-γ exhibited significant capillary tube formation on ECMatrix whereas fgl-2silenced HUVEC did not. The significant decrease in sprouting was
E. Rabizadeh et al. / Thrombosis Research 136 (2015) 118–124
123
Fig. 5. FGL-2 and FGF-2 immunohistochemistry and fgl-2 expression in tumor sections. (A) FGL-2 staining in malignant cells of tumour induced by WT PC-3. Tumour cells expressing FGL-2 protein were detected by monoclonal anti-FGL-2 antibody. (B) FGF-2 staining of tumour induced by WT PC-3 clone. FGF-2 was detected by monoclonal anti-FGF-2 antibody. Magnification × 400. (C) Fgl-2 mRNA extracted from the tumour of mice injected with WT PC-3 or fgl-2 -silenced PC-3 was analysed by RT-PCR and expressed relative to the housekeeping gene abl-1. Mean ± SEM of four experiments is presented. *p b 0.05.
apparently specific to fgl-2, as HUVEC transfected with nonspecific siRNA showed a pattern similar to that of WT HUVEC. Prompted by our finding on the effect of FGL-2 in angiogenesis, we tested its potential role in tumor development using an in vivo model of SCID mice. Injections of PC-3 with intact fgl-2 (WT) induced poorly
differentiated carcinoma in all mice. The malignant cells penetrated the muscle fibers and metastasized to the lung. By contrast, injection of fgl-2 -silenced PC-3 clone induced tumor development in only two-thirds of the mice, and the tumors were significantly smaller and less aggressive than in the WT PC-3 group, with significantly fewer blood vessels (Fig. 4E).
Fig. 6. Expression of fgf-2 in different PC-3 clones and ERK phosphorylation in PC-3 clones and mice tumors. (A) RT-PCR analysis of mRNA of fgf-2 extracted from WT PC-3, PC-3 with nonspecific fgl-2 silencing, or PC-3 with specific fgl-2-silencing. Findings are expressed relative to the housekeeping gene abl-1 are presented relative to WT PC-3. (B) ERK1/2 phosphorylation of fgl-2-silenced PC-3 relative to WT PC-3 (mean ± SD: 66.84% ± 8% and 100% ± 7.46%, respectively); (C) ERK1/2 phosphorylation of mouse tumor induced by fgl-2-silenced PC-3 relative to WT PC-3 (mean ± SD: 81.48% ± 7.1% and 100% ± 9.47%, respectively). The extent of phosphorylation was determined with ERK1/2 phosphorylation ELISA kit. Mean ± SD of three experiments is presented. *p b 0.05.
124
E. Rabizadeh et al. / Thrombosis Research 136 (2015) 118–124
Previous reports [7,18] have suggested that FGL-2 contributes to angiogenesis through a thrombin-dependent pathway. However, in our in vitro study, the addition of hirudin, a potent thrombin inhibitor, had no effect on capillary tube formation. Furthermore, in our in vivo study, tumor cells from all mouse groups stained positive for FGL-2, and none stained positive for thrombin. Together, these findings suggest that the proangiogenic/protumorigenic activity of FGL-2 is mediated by a thrombin-independent mechanism. The difference from the earlier studies may be at least partly explained by their use of human HCC cells [18] which, unlike prostate cancer cells, contain prothrombin. Therefore, it is possible that in tumors of liver origin such as HCC, thrombin might be generated locally, serving as a substrate for the prothrombinase activity of FGL-2 and subsequent thrombin-mediated growth. However, in cells that do not synthesize prothrombin, FGL-2 mediates angiogenesis and tumorigenesis by a thrombin-independent pathway. By analogy, others reported that the plasmatic prothrombinase coagulation factor Xa directly induces the expression and release of angiogenic growth factors. Our results provide alternative insight into the mechanism underlying the proangiogenic/protumorigenic activities of FGL-2, implying that these activities are mediated by basic FGF. In the gene-array experiments using PC-3 clones, FGF-2 was significantly downregulated by fgl-2 silencing, and this process was associated with a decrease in activation/phosphorylation of ERK1/2. These findings are in accordance with the previously well-established role of the FGF pathway in angiogenesis, wound healing, and carcinogenesis [22,29,30]. Specifically, FGF binding to its receptor, FGFR, leads to FGFR autophosphorylation and activation [31], ultimately followed by signal transduction through multiple downstream pathways, including the ERK/MAP kinases [24]. Moreover, FGF is known to be a key messenger in prostate growth and development, and its aberrant signaling has been implicated in prostate cancer development and progression [32–34]. Previous reports suggested that activation of FGF/FGFR signaling was sufficient to induce the development of prostate cancer in mouse models [35] and its attenuation is known to have inhibitory effect on cell proliferation, migration and invasion [36,37]. Thus, our findings suggest that the effect of cell membrane FGL-2 prothrombinase on angiogenesis and tumor development is thrombinindependent and is mediated by FGF/ERK signaling, In the future, FGL-2 might serve as a potential target for anti-tumor therapies. Conflict of Interest Statement The authors declare no competing financial interests. Acknowledgements This work was supported by research funding from Nufar (grant # 44776), Chief Scientist, Ministry of Health, Israel. We would like to thank Eti Melai RN, MA for her assistance in the project. References [1] Trousseau A. Clinique Medical de l'Hotel-dieu de Paris, Phlegmasia Alba Dolens. Paris: JB Balliere et Fils; 1865 654–715. [2] Blom JW, Doggen CJ, Osanto S, Rosendaal FR. Malignancies, prothrombotic mutations, and the risk of venous thrombosis. JAMA 2005;293:715–22. [3] Blom JW, Vanderschoot JP, Oostindier MJ, Osanto S, van der Meer FJ, Rosendaal FR. Incidence of venous thrombosis in a large cohort of 66,329 cancer patients: results of a record linkage study. J Thromb Haemost 2006;4:529–35. [4] Sorensen HT, Mellemkjaer L, Olsen JH, Baron JA. Prognosis of cancers associated with venous thromboembolism. N Engl J Med 2000;343:1846–50. [5] Blom JW, Osanto S, Rosendaal FR. High risk of venous thrombosis in patients with pancreatic cancer: a cohort study of 202 patients. Eur J Cancer 2006;42:410–4. [6] Kim I, Moon SO, Koh KN, Kim H, Uhm CS, Kwak HJ, et al. Molecular cloning, expression, and characterization of angiopoietin-related protein. angiopoietin-related protein induces endothelial cell sprouting. J Biol Chem 1999;274:26523–8. [7] Su K, Chen F, Yan WM, Zeng QL, Xu L, Xi D, et al. Fibrinogen-like protein 2/ fibroleukin prothrombinase contributes to tumor hypercoagulability via IL-2 and IFN-gamma. World J Gastroenterol 2008;14:5980–9.
[8] Ning Q, Sun Y, Han M, Zhang L, Zhu C, Zhang W, et al. Role of fibrinogen-like protein 2 prothrombinase/fibroleukin in experimental and human allograft rejection. J Immunol 2005;174:7403–11. [9] Chan CW, Chan MW, Liu M, Fung L, Cole EH, Leibowitz JL, et al. Kinetic analysis of a unique direct prothrombinase, fgl2, and identification of a serine residue critical for the prothrombinase activity. J Immunol 2002;168:5170–7. [10] Koyama T, Hall LR, Haser WG, Tonegawa S, Saito H. Structure of a cytotoxic Tlymphocyte-specific gene shows a strong homology to fibrinogen beta and gamma chains. Proc Natl Acad Sci U S A 1987;84:1609–13. [11] Levy GA, Liu M, Ding J, Yuwaraj S, Leibowitz J, Marsden PA, et al. Molecular and functional analysis of the human prothrombinase gene (HFGL2) and its role in viral hepatitis. Am J Pathol 2000;156:1217–25. [12] Yuwaraj S, Ding J, Liu M, Marsden PA, Levy GA. Genomic characterization, localization, and functional expression of FGL2, the human gene encoding fibroleukin: a novel human procoagulant. Genomics 2001;71:330–8. [13] Marazzi S, Blum S, Hartmann R, Gundersen D, Schreyer M, Argraves S, et al. Characterization of human fibroleukin, a fibrinogen-like protein secreted by T lymphocytes. J Immunol 1998;161:138–47. [14] Chan CW, Kay LS, Khadaroo RG, Chan MW, Lakatoo S, Young KJ, et al. Soluble fibrinogen-like protein 2/fibroleukin exhibits immunosuppressive properties: suppressing T cell proliferation and inhibiting maturation of bone marrow-derived dendritic cells. J Immunol 2003;170:4036–44. [15] Ding JW, Ning Q, Liu MF, Lai A, Leibowitz J, Peltekian KM, et al. Fulminant hepatic failure in murine hepatitis virus strain 3 infection: tissue-specific expression of a novel fgl2 prothrombinase. J Virol 1997;71:9223–30. [16] Ghanekar A, Mendicino M, Liu H, He W, Liu M, Zhong R, et al. Endothelial induction of fgl2 contributes to thrombosis during acute vascular xenograft rejection. J Immunol 2004;172:5693–701. [17] Hancock WW, Szaba FM, Berggren KN, Parent MA, Mullarky IK, Pearl J, et al. Intact type 1 immunity and immune-associated coagulative responses in mice lacking IFN gamma-inducible fibrinogen-like protein 2. Proc Natl Acad Sci U S A 2004;101: 3005–10. [18] Liu Y, Xu L, Zeng Q, Wang J, Wang M, Xi D, et al. Downregulation of FGL2/ prothrombinase delays HCCLM6 xenograft tumour growth and decreases tumour angiogenesis. Liver Int 2012;32:1585–95. [19] Tarasenko N, Cutts SM, Phillips DR, Inbal A, Nudelman A, Kessler-Icekson G, et al. Disparate impact of butyroyloxymethyl diethylphosphate (AN-7), a histone deacetylase inhibitor, and doxorubicin in mice bearing a mammary tumor. PLoS One 2012;7:e31393. [20] Kaighn ME, Narayan KS, Ohnuki Y, Lechner JF, Jones LW. Establishment and characterization of a human prostatic carcinoma cell line (PC-3). Invest Urol 1979;17:16–23. [21] Pulukuri SM, Gondi CS, Lakka SS, Jutla A, Estes N, Gujrati M, et al. RNA interferencedirected knockdown of urokinase plasminogen activator and urokinase plasminogen activator receptor inhibits prostate cancer cell invasion, survival, and tumorigenicity in vivo. J Biol Chem 2005;280:36529–40. [22] Corn PG, Wang F, McKeehan WL, Navone N. Targeting fibroblast growth factor pathways in prostate cancer. Clin Cancer Res 2013;19:5856–66. [23] Browder T, Folkman J, Pirie-Shepherd S. The hemostatic system as a regulator of angiogenesis. J Biol Chem 2000;275:1521–4. [24] Tsopanoglou NE, Pipili-Synetos E, Maragoudakis ME. Thrombin promotes angiogenesis by a mechanism independent of fibrin formation. Am J Physiol 1993;264:C1302–7. [25] Hembrough TA, Swartz GM, Papathanassiu A, Vlasuk GP, Rote WE, Green SJ, et al. Tissue factor/factor VIIa inhibitors block angiogenesis and tumor growth through a nonhemostatic mechanism. Cancer Res 2003;63:2997–3000. [26] Dardik R, Loscalzo J, Inbal A. Factor XIII (FXIII) and angiogenesis. J Thromb Haemost 2006;4:19–25. [27] Falanga A, Panova-Noeva M, Russo L. Procoagulant mechanisms in tumour cells. Best Pract Res Clin Haematol 2009;22:49–60. [28] Falanga A, Marchetti M, Vignoli A. Coagulation and cancer: biological and clinical aspects. J Thromb Haemost 2013;11:223–33. [29] Selzner N, Liu H, Boehnert MU, Adeyi OA, Shalev I, Bartczak AM, et al. FGL2/ fibroleukin mediates hepatic reperfusion injury by induction of sinusoidal endothelial cell and hepatocyte apoptosis in mice. J Hepatol 2012;56:153–9. [30] Hollborn M, Kohen L, Werschnik C, Tietz L, Wiedemann P, Bringmann A. Activated blood coagulation factor X (FXa) induces angiogenic growth factor expression in human retinal pigment epithelial cells. Invest Ophthalmol Vis Sci 2012;53:5930–9. [31] Nugent MA, Iozzo RV. Fibroblast growth factor-2. Int J Biochem Cell Biol 2000;32: 115–20. [32] Eswarakumar VP, Lax I, Schlessinger J. Cellular signaling by fibroblast growth factor receptors. Cytokine Growth Factor Rev 2005;16:139–49. [33] Yan G, Fukabori Y, McBride G, Nikolaropolous S, McKeehan WL. Exon switching and activation of stromal and embryonic fibroblast growth factor (FGF)-FGF receptor genes in prostate epithelial cells accompany stromal independence and malignancy. Mol Cell Biol 1993;13:4513–22. [34] Feng S, Wang F, Matsubara A, Kan M, McKeehan WL. Fibroblast growth factor receptor 2 limits and receptor 1 accelerates tumorigenicity of prostate epithelial cells. Cancer Res 1997;57:5369–78. [35] Yang F, Zhang Y, Ressler SJ, Ittmann MM, Ayala GE, Dang TD, et al. FGFR1 is essential for prostate cancer progression and metastasis. Cancer Res 2013;73:3716–24. [36] Wesley UV, McGroarty M, Homoyouni A. Dipeptidyl peptidase inhibits malignant phenotype of prostate cancer cells by blocking basic fibroblast growth factor signaling pathway. Cancer Res 2005;65:1325–34. [37] Yang F, Strand DW, Rowley DR. Fibroblast growth factor-2 mediates transforming growth factor-beta action in prostate cancer reactive stroma. Oncogene 2008;27: 450–9.
Contents lists available at ScienceDirect
Thrombosis Research journal homepage: www.elsevier.com/locate/thromres
Regular Article
The cell-membrane prothrombinase, fibrinogen-like protein 2, promotes angiogenesis and tumor development Esther Rabizadeh a,b, Izhack Cherny a, Doron Lederfein a, Shany Sherman a, Natalia Binkovsky a, Yevgenia Rosenblat c, Aida Inbal a,d,⁎ a
Hemato-Oncology Laboratory, Felsenstein Medical Research Center, Petach Tikva, Israel 1 Hematology Laboratory, Rabin Medical Center, Beilinson Hospital, Petach Tikva, Israel c Pathology Institute, Rabin Medical Center, Beilinson Hospital, Petach Tikva, Israel d Thrombosis and Hemostasis Unit, Hematology Institute, Rabin Medical Center, Beilinson Hospital, Petach Tikva, Israel 1
b
a r t i c l e
i n f o
Article history: Received 7 July 2014 Received in revised form 9 November 2014 Accepted 30 November 2014 Available online 4 December 2014 Keywords: Angiogenesis Fibrinogen-like protein 2 Tumorigenesis
a b s t r a c t The aim of the study was to further investigate the role of fibrinogen-like protein 2 (FGL-2), a transmembrane prothrombinase that directly cleaves prothrombin to thrombin, in angiogenesis and tumor development and the mechanism(s) underlying these processes. To study angiogenesis HUVEC clones with decreased fgl-2 mRNA were generated by specific siRNA. To study tumorigenesis SCID mice were implanted with intact (wild type) and fgl-2-silenced PC-3 clones. IFN-γ treated HUVEC expressing increased fgl-2 mRNA exhibited significant capillary sprouting that was not inhibited by hirudin, whereas fgl-2 silencing completely inhibited blood-vessel formation. Tumors (poorly differentiated carcinoma) developed in all 12 mice injected with wild type PC-3 compared with 8/12 mice injected with the fgl-2-silenced PC-3 clone. The tumors developed by fgl-2-silenced PC-3 clones were smaller and less aggressive and contained significantly fewer blood vessels (p b 0.05). All tumors’ sections were negative for thrombin staining, indicating that FGL-2-induced tumorigenesis was not mediated by thrombin. In fgl-2-silenced tumors there was a decrease in fgl-2 mRNA (p = 0.02) and ERK1/2 phosphorylation (p b 0.05) by 80% and a 20%, respectively. The mechanism underlying these processes, studied in PC-3 clones, revealed that fgl-2 silencing was associated with a 65% decrease in FGF-2 mRNA (p b 0.01) and a 30% down regulation of ERK1/2 phosphorylation (p b 0.05). Together, these results suggest that FGL-2 mediates angiogenesis and tumorigenesis not by thrombin-mediated mechanism but rather through FGF-2/ERK signaling pathway. FGL-2 may serve as a valuable therapeutic target in the future. © 2014 Elsevier Ltd. All rights reserved.
Introduction The bidirectional relationship between cancer and thrombosis has been known for almost two centuries [1–3]. Thrombosis often precedes the diagnosis of cancer, and its presence is associated with a detrimental disease course [4,5]. These findings support the paradigm that coagulation and tumor growth form a vicious circle in which hypercoagulability facilitates the aggressive biology of cancer and vice versa. The mechanism
Abbreviations: ERK, extracellular-signal-regulated kinases; FGF-2, basic fibroblast growth factor; FGL-2, fibrinogen-like protein 2; HUVEC, human umbilical vein endothelial cells; IFN-γ, human interferon-gamma; MAPK, mitogen-activated protein kinases; SCID, Severe Combined Immunodeficiency ⁎ Corresponding author at: Thrombosis and Hemostasis Unit, Hematology Institute, Rabin Medical Center, Beilinson Hospital, Petach Tikva 49100, Israel. Tel.: + 972 3 9377912; fax: +972 3 920 1568. E-mail addresses: [email protected] (E. Rabizadeh), [email protected] (I. Cherny), [email protected] (D. Lederfein), [email protected] (S. Sherman), [email protected] (N. Binkovsky), [email protected] (Y. Rosenblat), [email protected] (A. Inbal). 1 Affiliated with Sackler Faculty of Medicine, Tel Aviv University, Israel.
http://dx.doi.org/10.1016/j.thromres.2014.11.023 0049-3848/© 2014 Elsevier Ltd. All rights reserved.
underlying these events is still unclear. Malignant cells are known to directly activate blood coagulation in three ways: by producing procoagulant, fibrinolytic, and proaggregating factors; by releasing proinflammatory and proangiogenic cytokines, such as tumor necrosis factor alpha (TNF-α) and interleukin (IL)-1ß; and by interacting directly with host endothelial cells, leukocytes, and platelets via adhesion molecules [5]. However, the precise procoagulant proteins that stimulate tumorigenesis have not been identified. One potential candidate is the cell-membrane-associated protein, fibrinogen-like protein 2 (FGL-2)/fibroleukin, which has been shown to induce sprouting in vascular endothelial cells [6] and to be overexpressed in tumor cells [7]. FGL-2, also known as FGL-2prothrombinase, is a member of the fibrinogen family of proteins [8]. It exerts serine protease activity and is capable of directly cleaving prothrombin to thrombin in the absence of factor VII or factor X. Like plasmatic prothrombinase, factor Xa, FGL-2 prothrombinase requires phospholipids, calcium, and factor Va for optimal catalytic activity [9]. However, unlike factor Xa, FGL-2 is a transmembrane protein which is not inhibited by antithrombin in the presence of heparin or by other protease inhibitors that inhibit factor Xa [9].
E. Rabizadeh et al. / Thrombosis Research 136 (2015) 118–124
The human gene encoding FGL-2, originally cloned from cytotoxic T lymphocytes, spans approximately 7 kb on chromosome 7 and contains 2 exons. The 70-KD protein is predicted to be 439 amino acids (aa) long, with the N-terminus including a 2-aa-long cytoplasmic domain and a 21-aa-long transmembrane domain. The remaining 416 aa constitute the extracellular domain. FGL-2 shares 36% sequence homology with the fibrinogen β and γ chains and 40% homology with the fibrinogen-related domain (FRED) of tenascin [10]. The murine and human proteins share 78% overall identity, with greater conservation at the C terminus [11,12]. FGL-2 is expressed on the surface of activated macrophages and endothelial cells and also secreted by peripheral blood CD4+ and CD8+ T cells. The secreted protein is devoid of coagulation activity. It has potent modulatory effects on the adaptive immune system and was reported to inhibit the maturation of dendritic cells [13,14]. The prothrombinase and immune activities of FGL-2 are located on distinct domains on the FGL-2 molecule [9,14]. The prothrombinase activity of FGL-2, first described in a murine fulminant hepatitis model [15], is exhibited when FGL-2 is expressed on activated macrophages and endothelial cells in the form of a membrane-associated protein. It has shown to be associated with both experimental and human allograft rejection that was abrogated following neutralization of FGL-2 by antibodies or in FGL-2 knockout mice [16]. Macrophage and endothelial cell induction of FGL-2 occurs via interferon gamma (IFN –γ) [17]. FGL-2 also plays a role in tumor development. Overexpression of FGL-2 has been detected in tumor and interstitial inflammatory cells but not in the normal surrounding tissue [7]. A recent study found that knockdown of FGL-2 delayed tumor growth and angiogenesis in mice injected with the human hepatocellular carcinoma (HCC) cell line [18]. The authors hypothesized that the protumorigenic activity of FGL-2 is a result of FGL-2-induced generation of thrombin leading to thrombin-induced tumorigenesis [7,18]. However, this does not explain the upregulation of FGL-2 in tumor cells. The aim of the present study was to substantiate the role of FGL-2 in angiogenesis and tumor development and to uncover the mechanism underlying these processes. Using an in vitro pro-angiogenic assay and an in vivo mice model of tumor development, we established that FGL-2 exerts direct, non-thrombin-mediated, angiogenic and tumorigenic activity. Material and Methods Fgl-2 Expression The expression of fgl-2 was analyzed in human umbilical vein endothelial cells (HUVEC) in the presence or absence of IFN-γ (20 ng/ml) and in a wild type (WT) human prostate carcinoma cell line, PC-3, established from bone metastasis (ATCC, Beit-Haemek, Israel). Total RNA was isolated using RNAqueous™ (Ambion #AM1912, Invitrogen, Austin, TX, USA) and analyzed by real-time polymerase chain reaction (RT-PCR) using the Rotor-gene RG-3000 (Corbett, Australia). The difference in cycle time (ΔCT) was measured by comparing fgl-2 gene with abl-1 gene (housekeeping gene). The relative quantification was calculated with the formula RQ = 2-ΔΔCT. Generation of fgl-2-silenced Clones of HUVEC and PC-3 Cells Small interfering RNA (siRNA), which interferes with the expression of specific genes, was employed to evaluate the impact of inhibiting fgl2 expression on angiogenesis and tumorigenesis. The effect of siRNA was tested on HUVEC and PC-3 cells. The specific siRNA was purchased from Thermo Scientific (Pittsburg, PA, USA). Using pGIPZ lentiviral vector (Thermo Scientific), stable specific fgl-2-silenced clones and non-fgl-2-inhibited HUVEC and PC-3 clones (nonspecific siRNA clone)
119
were generated. Stable transfectants were selected with 0.2 μg/ml puromycin. In vitro Angiogenesis Assay HUVEC (5 x 103/well) were cultured in F12 (HAM) medium complemented with 1% penicillin (10,000 UI/ml)/streptomycin (10,000 UI/ml) and 10% fetal calf serum (FCS). FCS was tested for prothrombin by one-stage prothrombin-time coagulation assay (ACL1000 Coagulometer). The cells were maintained at 37 °C in a humidified incubator (85%) in 5% CO2 atmosphere. The angiogenesis assay was performed with the Chemicon in vitro ECMatrix assay kit (ECM625, Millipore, Billerica, MA, USA) using HUVEC treated with 20 ng/ml IFNγ, fgl-2-silenced clone of HUVEC, and nonspecific siRNA clone of HUVEC, in the presence or absence of 10 u/ml of hirudin (H0393, Sigma, St. Louis, Mo, USA). The extent of tube formation was monitored by light microscopy, as described by the kit manufacturer. Pattern recognition was defined by photographing the cells at the end point of the assay. A numerical value representing the degree of angiogenesis progression was assigned to each pattern according to the manufacturer’s instructions. Mouse Model Ten-week-old male severe combined immunodeficiency (SCID) mice were purchased from Harlan Laboratories (Rehovot, Israel). The mice were maintained in laminar flow cabinets under specific pathogen-free conditions and a daily cycle of 12 h light/12 h dark. All food, water, and litter were sterilised prior to use, and temperature (20-21 °C) and humidity (50-60%) were controlled. Cages were changed fully once a week. Animals were manipulated under sterile conditions. The mice were managed in accordance with the NIH Guidelines on Laboratory Animal Welfare. Study and all protocols were approved and monitored by the Animal Care Committee of Rabin Medical Center. Twenty-four mice underwent subcutaneous cell implantation as follows: 10 unmanipulated (WT) PC-3 cells; 12 fgl-2 siRNA-silenced PC-3 clone; 2 PC-3 transfected with nonspecific siRNA. In each case, 2.5 x 106 tumor cells suspended in 100 μl of phosphate-buffered saline were injected without anesthesia using a 27-gauge needle and a 1-ml disposable syringe. Tumor progression was monitored by palpation 3 times a week. Subcutaneous tumor dimensions (length (L) and width (W)) were measured with a caliper, and volume was calculated according to the formula (L × W2)/2 [19]. Mice were sacrificed by cervical dislocation in week 6 after implantation. Autopsy was performed to assess the distribution of metastases. Tumors, spleen, liver, and lung were harvested. The tissues were fixed in 4% formaldehyde for histological analysis. Immunohistological Analysis and FGL-2 Expression in Mouse Tumor Tissues Histological sections of the tumors and other mouse tissues were stained with haematoxylin-eosin. Immunohistochemistry study was performed on formalin-fixed paraffin-embedded sections (4 microns thick) using anti-FGL-2 monoclonal antibody (Abnova, St. Louis, MO, USA) diluted 1:400; mouse monoclonal anti- thrombin antibody F-1 (Santa Cruz Biotechnology, Santa Cruz, CA, USA) diluted 1:200; and anti-basic fibroblast growth factor (FGF-2) monoclonal antibody diluted 1:100 (#SC-79, Santa Cruz Biotechnology). Normal hepatic tissue served as a positive control for the anti-thrombin antibody. Endogenous reactions were blocked with DAKO blocking kit. The slides were stained with the automatic Bench Mark XT kit (Ventana Medical Systems, Oro Valley, AZ, USA) consisting of labelled streptavidin-biotin reagents and counterstained with haematoxylin. For quantitation of blood vessels, the histological sections were stained with polyclonal anti-human von Willebrand factor (VWF) antibody diluted 1:300 (#A0082, DAKO,
120
E. Rabizadeh et al. / Thrombosis Research 136 (2015) 118–124
Denmark). Triplicates of each tumor section were stained and the number of VWF-positive blood vessels was counted in 10 fields per each stained slide by two independent examiners using light microscope. Proteomics of Angiogenesis and MAPK Pathway-related Proteins Induced by fgl-2 in PC-3 Clones or Tumors Induced by these Clones The effects of fgl-2 silencing on the expression of angiogenesisrelated proteins were studied using Proteome Profiler Human Angiogenesis Array kit (R&D Systems, Abingdon, UK). Proteins extracted from PC-3 clones before or after transfection with specific siRNA for fgl-2 silencing were hybridized with an array of antibodies against angiogenesis-related proteins according to manufacturer instructions. The MAPK phosphorylation profile was studied using Proteome Profiler Human Phospho-MAPK Array kit (R&D Systems, Abingdon, UK)). PC-3 cells before (WT) and after transfection with specific fgl-2 silencing siRNA as well as proteins extracted from the mice tumors induced by PC-3 clones were lysed in the presence of phosphatase inhibitor cocktails B and C (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Protein cell extracts were mixed with a specific cocktail of biotinylated detection antibodies and analyzed according to manufacturer’s instructions. The phosphorylation experiments were further validated with ERK1/2 phosphorylation ELISA kit (InstantOneTM. ELISA, eBioscience, San Diego, CA). Both total and phosphorylated ERK1/2 protein levels were determined. The normalized phosphorylation levels were calculated according to the total amount of ERK1/2. The chemiluminescence signal intensity was considered proportional to the amount of protein detected. Fgf-2 Expression in PC-3 Clones Total RNA was extracted from PC-3 clones, as follows: WT PC-3; fgl-2 silenced PC-3, and nonspecific siRNA PC-3. The levels of fgf-2 mRNA were analyzed by RT-PCR and compared to the house-keeping gene, abl-1. RT-PCR primers were obtained from Solaris (Thermo Fisher Scientific Inc.). cDNA was produced using Maxima First Strand cDNA Synthesis Kit for RT-qPCR (Thermo Fisher Scientific) according to the manufacturer’s instructions. qPCR was performed using GoTaq Green Master Mix (Promega, Madison WI, USA) according to the manufacturer’s instructions. Statistical Cnalysis Statistical analyses were performed with Student's t-test (independent sample or one sample) or Mann-Whitney U test, where appropriate, using GraphPad Prism (ver. 5) or SPSS (ver. 21) software. The difference in mean tumor growth rate under the various conditions was estimated by repeated measures ANOVA with post-hoc for each group of mice. A p-value of b 0.05 was considered statistically significant. Results Generation of HUVEC Clones with Different fgl-2 Expression HUVEC clones were generated with basal, increased, and decreased expression of fgl-2, as shown in Fig. 1. There was a 5-fold increase in basal fgl-2 mRNA after the addition of 20 ng/ml IFN-γ (Fig. 1). HUVEC transfection with specific fgl-2-siRNA led to 80% silencing of fgl2 mRNA whereas nonspecific siRNA decreased fgl-2 expression by only 10% (Fig. 1).
Fig. 1. Expression of fgl-2 in HUVEC. RT-PCR analysis of mRNA of fgl-2 from HUVEC incubated with IFN-γ before (WT), or after transfection with specific (S) or nonspecific (N) siRNA for fgl-2 silencing. Findings are expressed relative to the house-keeping gene, abl-1, and presented relative to untreated cells (left column). The results (mean ± SD) are duplicates of three separate experiments. *p b 0.05 for difference between IFN-γ-treated HUVEC with or without specific siRNA.
and nonspecific siRNA clone of HUVEC. Only traces of prothrombin were detected (0.05%) in the FCS used for the angiogenesis experiments. Nevertheless, to inhibit any traces of thrombin that may have been generated locally from the traces of prothrombin, 10 u/ml hirudin was added in some of the experiments. A representative picture of the tube arrays is shown in Fig. 2. The extent of tube arrays generated was similar for WT HUVEC (Fig. 2A) and HUVEC transfected with nonspecific siRNA (Fig. 2C). By contrast, tube array generation was almost completely inhibited by fgl-2 silencing (Fig. 2B, p b 0.001). Hirudin had no effect on the extent of tube formation (Fig. 2D). The number of patterns was expressed on a 5-point scale representing the degree of angiogenesis, from 0 (individual cells) to 5 (complex mesh-like structure) (Millipore). Mean ± SD values by test condition were as follows: WT HUVEC, 3.5 ± 0.5 (32 scored fields); fgl-2 siRNA-silenced HUVEC, 0.6 ± 0.80 (p b 0.001) (23 scored fields); and HUVEC transfected with nonspecific siRNA without hirudin, 3.1 ± 0.8 (30 scored fields), and with hirudin, 3.5 ± 0.5 (6 scored fields). Fgl-2 Expression in PC-3 To study the role of FGL-2 in tumor development we chose to use the human prostate cancer cell line PC-3 as a model. PC-3 efficiently develops subcutaneous tumors in mice and exhibits relatively high metastatic potential [20,21]. PC-3 clones with either intact or decreased fgl-2 expression were generated by transfection of PC-3 cell line with specific or nonspecific siRNA, respectively, and fgl-2 mRNA was measured using RT-PCR, as detailed in Methods. As shown in Fig. 3, fgl-2 mRNA was inhibited by more than 90% in the siRNA fgl-2silenced PC-3 clone relative to the WT PC-3 cells (8.0% versus 100% mRNA; p b 0.05). There was no significant difference in fgl-2 mRNA between PC-3 clone transfected with nonspecific siRNA and WT PC-3 (81.5% versus 100%, respectively; p = 0.1). Role of FGL-2 in Tumor Development
Role of FGL-2 in Angiogenesis The role of FGL-2 in angiogenesis was studied using different HUVEC clones grown in ECMatrix as specified in Methods: IFN-γ-treated HUVEC expressing increased fgl-2 (WT), fgl-2 siRNA-silenced HUVEC,
The role of FGL-2 in tumor development was studied in SCID mice injected with WT PC-3 (n = 10), siRNA-fgl-2 -silenced PC-3 clone (n = 12), or PC-3 clone transfected with nonspecific siRNA (n = 2). After 6 weeks of follow-up, tumors developed in all mice inoculated
E. Rabizadeh et al. / Thrombosis Research 136 (2015) 118–124
121
Fig. 2. In vitro angiogenesis assay. Representative picture of capillary tube formation induced by WT HUVEC (A), fgl-2 -silenced HUVEC (B), and HUVEC transfected with nonspecific siRNA in the absence (C) or presence (D) of hirudin. Magnification × 200.
by WT PC-3. By contrast, in the group injected with fgl-2 siRNA-silenced PC-3, 4 mice did not develop tumors at all, and 8 had significantly smaller tumors compared to tumors developed following injection of WT PC3 (Fig. 4A and B). Two additional mice injected with PC-3 transfected with nonspecific siRNA acquired tumors similar to those inoculated
with WT PC-3. At each time-point of the follow-up the volume of the tumors in the WT PC-3 group was larger than that observed in the group injected with fgl-2 -silenced clone, approaching an 8-fold increase in volume at the end of the follow up (WT PC-3 clone - 413.9 mm3 versus fgl-2 silenced clone - 52.6 mm3; p = 0.002; Fig. 4C). Lung metastasis developed in 1 of the 10 mice injected with WT PC-3 and none of the mice injected with fgl-2-silenced clone. Histological study of tumour sections from the mice injected with WT PC-3 revealed poorly differentiated carcinoma penetrating muscle fibers. A representative histological section is shown in Fig. 4D. Most of the cells were large with prominent pleomorphism and a vesicular or hyperchromatic nucleus and nucleolus. The histology of the tumors from the fgl-2-silenced PC-3 group was similar, but there was no noticeable penetration into muscle fibers, and the number of blood vessels was significantly lower (Fig. 4E; p b 0.05). Immunohistochemical staining revealed clearly visible FGL-2 protein in the tumor cells in all mouse groups (Fig. 5A). In addition, almost all tumor cells were positive for FGF-2 (Fig. 5B). It should be noted here that the immunohistochemical staining technique is not a quantitative test, yet it is suitable to confirm or negate protein expression. Accordingly, staining of tumor sections from each group of mice with antithrombin antibody that recognizes human and mice thrombin was negative, strongly indicating a lack of thrombin in these sections. Tumors that developed after injection of fgl-2 silenced PC-3 clones showed an 80% reduction in fgl-2 mRNA (Fig. 5C; p b 0.05). Effect of fgl-2 Silencing on Angiogenesis-related Proteins
Fig. 3. Fgl-2 expression in different PC-3 clones. RT-PCR analysis of mRNA of fgl-2 extracted from WT PC-3, PC-3 transfected with nonspecific fgl-2 siRNA, or PC-3 transfected with specific fgl-2-silencing siRNA. Findings are expressed relative to the housekeeping gene abl-1 and presented relative to WT PC-3. The results (mean ± SD) are duplicates of three separate experiments.
To determine if FGL-2 affects the levels of proteins that participate in angiogenesis, proteins extracted from PC-3 clones before or after transfection with specific siRNA for fgl-2 silencing were hybridized with an array of antibodies against angiogenesis-related proteins. Fgl-2-
122
E. Rabizadeh et al. / Thrombosis Research 136 (2015) 118–124
Fig. 4. SCID mice model of PC-3-induced tumors. Macroscopic picture of the tumor (surrounded by arrows) in mice injected with (A) WT PC-3 or (B) fgl-2 -silenced PC-3. (C) Mean ± SEM volume of the tumors at each week in mice injected with WT PC-3 or fgl-2 -silenced PC-3. (D) Haematoxylin-eosin staining of the tumor generated by WT PC-3 showing vesicular nucleus and nucleolus (arrow) (magnification × 200). (E) Mean ± SEM of the number of blood vessels in the tumours induced by WT PC-3 or fgl-2 silenced PC-3 counted by light microscopy following staining with anti VWF antibody, as detailed in Methods. *p b 0.05.
silencing was associated with a decrease in the FGF-2 protein level (data not shown). This finding was further validated by fgf-2 mRNA analysis. In fgl-2 silenced PC3 clone, the expression of fgf-2 was inhibited by 65% (mean ± SD, 34.4% ± 1.7%; p b 0.01), whereas no effect was observed in WT PC3 or nonspecific siRNA clone (mean ± SD, 85.7% ± 7.1%; p = 0.19), (Fig. 6A). Effect of fgl-2 Silencing on MAPK Phosphorylation Since MAPK plays a key role in FGF-2-mediated signaling of cell proliferation and tumourigenesis [22], the effect of fgl-2 silencing on the MAPK phosphorylation profile was studied using an array of specific MAPK related proteins. Fgl-2-silencing was associated with a significant decrease in the ERK1 and ERK2 phosphorylation (data not shown). This finding was validated by ERK1/2 phosphorylation ELISA kit, which revealed that fgl-2-silencing was associated with downregulation of ERK1/2 phosphorylation by 30 % (p b 0.05; Fig. 6B). In line with this finding, there was a 20% decrease in ERK1/2 phosphorylation (Fig. 6C) in tumors generated by fgl-2 silenced clones (p b 0.05; Fig. 6C). Discussion This study shows that the transmembrane serine protease FGL-2, which has well-established prothrombinase coagulation activity, plays a substantial role in angiogenesis and tumorigenesis. It is known that coagulation and angiogenesis are closely interrelated. Following injury, proteins generated by the coagulation system
coordinate the spatial localization and stabilization of endothelial cells for the growth and repair of the damaged blood vessel. Once the clot is stabilized, angiogenesis is modulated by proteins and cryptic peptide fragments generated from the coagulation and fibrinolytic systems [23]. The well-characterized proangiogenic coagulation proteins are thrombin [24], factor VII, tissue factor [25], and factor XIII [26]. Angiogenesis is also a vital process during tumor development. The coagulation activation and tumor growth rely on the capacity of cancer cells to promote neoangiogenesis and metastasis [27]. Several components of the hemostatic system contribute to these processes, such as thrombin, tissue factor, factor VIIa, factor Xa, and fibrinogen, as clearly documented in both in vitro and in vivo tumor models [28] . Of these, tissue factor is the best characterized. Constitutively expressed on the malignant cell surface, tissue factor can lead to the formation of both localized and systemic procoagulant states. Another tumor procoagulant is cancer procoagulant (CP) that, unlike TF, directly activates FX independently of FVII. CP has been detected in various tumor cells [27,28]. Herein we present a new tumor procoagulant, FGL-2, which may serve a link between cancer and coagulation. Like the other coagulation proteins [24–28], FGL-2 exerts substantial pro-angiogenic activity, as indicated in the present study by the 5-fold increase in fgl-2 expression in IFN-γ-stimulated HUVEC. Accordingly, Su et al. [7] reported a 10-fold increase in fgl-2 mRNA following stimulation with IFN-γ. In our experiments, HUVEC stimulated with IFN-γ exhibited significant capillary tube formation on ECMatrix whereas fgl-2silenced HUVEC did not. The significant decrease in sprouting was
E. Rabizadeh et al. / Thrombosis Research 136 (2015) 118–124
123
Fig. 5. FGL-2 and FGF-2 immunohistochemistry and fgl-2 expression in tumor sections. (A) FGL-2 staining in malignant cells of tumour induced by WT PC-3. Tumour cells expressing FGL-2 protein were detected by monoclonal anti-FGL-2 antibody. (B) FGF-2 staining of tumour induced by WT PC-3 clone. FGF-2 was detected by monoclonal anti-FGF-2 antibody. Magnification × 400. (C) Fgl-2 mRNA extracted from the tumour of mice injected with WT PC-3 or fgl-2 -silenced PC-3 was analysed by RT-PCR and expressed relative to the housekeeping gene abl-1. Mean ± SEM of four experiments is presented. *p b 0.05.
apparently specific to fgl-2, as HUVEC transfected with nonspecific siRNA showed a pattern similar to that of WT HUVEC. Prompted by our finding on the effect of FGL-2 in angiogenesis, we tested its potential role in tumor development using an in vivo model of SCID mice. Injections of PC-3 with intact fgl-2 (WT) induced poorly
differentiated carcinoma in all mice. The malignant cells penetrated the muscle fibers and metastasized to the lung. By contrast, injection of fgl-2 -silenced PC-3 clone induced tumor development in only two-thirds of the mice, and the tumors were significantly smaller and less aggressive than in the WT PC-3 group, with significantly fewer blood vessels (Fig. 4E).
Fig. 6. Expression of fgf-2 in different PC-3 clones and ERK phosphorylation in PC-3 clones and mice tumors. (A) RT-PCR analysis of mRNA of fgf-2 extracted from WT PC-3, PC-3 with nonspecific fgl-2 silencing, or PC-3 with specific fgl-2-silencing. Findings are expressed relative to the housekeeping gene abl-1 are presented relative to WT PC-3. (B) ERK1/2 phosphorylation of fgl-2-silenced PC-3 relative to WT PC-3 (mean ± SD: 66.84% ± 8% and 100% ± 7.46%, respectively); (C) ERK1/2 phosphorylation of mouse tumor induced by fgl-2-silenced PC-3 relative to WT PC-3 (mean ± SD: 81.48% ± 7.1% and 100% ± 9.47%, respectively). The extent of phosphorylation was determined with ERK1/2 phosphorylation ELISA kit. Mean ± SD of three experiments is presented. *p b 0.05.
124
E. Rabizadeh et al. / Thrombosis Research 136 (2015) 118–124
Previous reports [7,18] have suggested that FGL-2 contributes to angiogenesis through a thrombin-dependent pathway. However, in our in vitro study, the addition of hirudin, a potent thrombin inhibitor, had no effect on capillary tube formation. Furthermore, in our in vivo study, tumor cells from all mouse groups stained positive for FGL-2, and none stained positive for thrombin. Together, these findings suggest that the proangiogenic/protumorigenic activity of FGL-2 is mediated by a thrombin-independent mechanism. The difference from the earlier studies may be at least partly explained by their use of human HCC cells [18] which, unlike prostate cancer cells, contain prothrombin. Therefore, it is possible that in tumors of liver origin such as HCC, thrombin might be generated locally, serving as a substrate for the prothrombinase activity of FGL-2 and subsequent thrombin-mediated growth. However, in cells that do not synthesize prothrombin, FGL-2 mediates angiogenesis and tumorigenesis by a thrombin-independent pathway. By analogy, others reported that the plasmatic prothrombinase coagulation factor Xa directly induces the expression and release of angiogenic growth factors. Our results provide alternative insight into the mechanism underlying the proangiogenic/protumorigenic activities of FGL-2, implying that these activities are mediated by basic FGF. In the gene-array experiments using PC-3 clones, FGF-2 was significantly downregulated by fgl-2 silencing, and this process was associated with a decrease in activation/phosphorylation of ERK1/2. These findings are in accordance with the previously well-established role of the FGF pathway in angiogenesis, wound healing, and carcinogenesis [22,29,30]. Specifically, FGF binding to its receptor, FGFR, leads to FGFR autophosphorylation and activation [31], ultimately followed by signal transduction through multiple downstream pathways, including the ERK/MAP kinases [24]. Moreover, FGF is known to be a key messenger in prostate growth and development, and its aberrant signaling has been implicated in prostate cancer development and progression [32–34]. Previous reports suggested that activation of FGF/FGFR signaling was sufficient to induce the development of prostate cancer in mouse models [35] and its attenuation is known to have inhibitory effect on cell proliferation, migration and invasion [36,37]. Thus, our findings suggest that the effect of cell membrane FGL-2 prothrombinase on angiogenesis and tumor development is thrombinindependent and is mediated by FGF/ERK signaling, In the future, FGL-2 might serve as a potential target for anti-tumor therapies. Conflict of Interest Statement The authors declare no competing financial interests. Acknowledgements This work was supported by research funding from Nufar (grant # 44776), Chief Scientist, Ministry of Health, Israel. We would like to thank Eti Melai RN, MA for her assistance in the project. References [1] Trousseau A. Clinique Medical de l'Hotel-dieu de Paris, Phlegmasia Alba Dolens. Paris: JB Balliere et Fils; 1865 654–715. [2] Blom JW, Doggen CJ, Osanto S, Rosendaal FR. Malignancies, prothrombotic mutations, and the risk of venous thrombosis. JAMA 2005;293:715–22. [3] Blom JW, Vanderschoot JP, Oostindier MJ, Osanto S, van der Meer FJ, Rosendaal FR. Incidence of venous thrombosis in a large cohort of 66,329 cancer patients: results of a record linkage study. J Thromb Haemost 2006;4:529–35. [4] Sorensen HT, Mellemkjaer L, Olsen JH, Baron JA. Prognosis of cancers associated with venous thromboembolism. N Engl J Med 2000;343:1846–50. [5] Blom JW, Osanto S, Rosendaal FR. High risk of venous thrombosis in patients with pancreatic cancer: a cohort study of 202 patients. Eur J Cancer 2006;42:410–4. [6] Kim I, Moon SO, Koh KN, Kim H, Uhm CS, Kwak HJ, et al. Molecular cloning, expression, and characterization of angiopoietin-related protein. angiopoietin-related protein induces endothelial cell sprouting. J Biol Chem 1999;274:26523–8. [7] Su K, Chen F, Yan WM, Zeng QL, Xu L, Xi D, et al. Fibrinogen-like protein 2/ fibroleukin prothrombinase contributes to tumor hypercoagulability via IL-2 and IFN-gamma. World J Gastroenterol 2008;14:5980–9.
[8] Ning Q, Sun Y, Han M, Zhang L, Zhu C, Zhang W, et al. Role of fibrinogen-like protein 2 prothrombinase/fibroleukin in experimental and human allograft rejection. J Immunol 2005;174:7403–11. [9] Chan CW, Chan MW, Liu M, Fung L, Cole EH, Leibowitz JL, et al. Kinetic analysis of a unique direct prothrombinase, fgl2, and identification of a serine residue critical for the prothrombinase activity. J Immunol 2002;168:5170–7. [10] Koyama T, Hall LR, Haser WG, Tonegawa S, Saito H. Structure of a cytotoxic Tlymphocyte-specific gene shows a strong homology to fibrinogen beta and gamma chains. Proc Natl Acad Sci U S A 1987;84:1609–13. [11] Levy GA, Liu M, Ding J, Yuwaraj S, Leibowitz J, Marsden PA, et al. Molecular and functional analysis of the human prothrombinase gene (HFGL2) and its role in viral hepatitis. Am J Pathol 2000;156:1217–25. [12] Yuwaraj S, Ding J, Liu M, Marsden PA, Levy GA. Genomic characterization, localization, and functional expression of FGL2, the human gene encoding fibroleukin: a novel human procoagulant. Genomics 2001;71:330–8. [13] Marazzi S, Blum S, Hartmann R, Gundersen D, Schreyer M, Argraves S, et al. Characterization of human fibroleukin, a fibrinogen-like protein secreted by T lymphocytes. J Immunol 1998;161:138–47. [14] Chan CW, Kay LS, Khadaroo RG, Chan MW, Lakatoo S, Young KJ, et al. Soluble fibrinogen-like protein 2/fibroleukin exhibits immunosuppressive properties: suppressing T cell proliferation and inhibiting maturation of bone marrow-derived dendritic cells. J Immunol 2003;170:4036–44. [15] Ding JW, Ning Q, Liu MF, Lai A, Leibowitz J, Peltekian KM, et al. Fulminant hepatic failure in murine hepatitis virus strain 3 infection: tissue-specific expression of a novel fgl2 prothrombinase. J Virol 1997;71:9223–30. [16] Ghanekar A, Mendicino M, Liu H, He W, Liu M, Zhong R, et al. Endothelial induction of fgl2 contributes to thrombosis during acute vascular xenograft rejection. J Immunol 2004;172:5693–701. [17] Hancock WW, Szaba FM, Berggren KN, Parent MA, Mullarky IK, Pearl J, et al. Intact type 1 immunity and immune-associated coagulative responses in mice lacking IFN gamma-inducible fibrinogen-like protein 2. Proc Natl Acad Sci U S A 2004;101: 3005–10. [18] Liu Y, Xu L, Zeng Q, Wang J, Wang M, Xi D, et al. Downregulation of FGL2/ prothrombinase delays HCCLM6 xenograft tumour growth and decreases tumour angiogenesis. Liver Int 2012;32:1585–95. [19] Tarasenko N, Cutts SM, Phillips DR, Inbal A, Nudelman A, Kessler-Icekson G, et al. Disparate impact of butyroyloxymethyl diethylphosphate (AN-7), a histone deacetylase inhibitor, and doxorubicin in mice bearing a mammary tumor. PLoS One 2012;7:e31393. [20] Kaighn ME, Narayan KS, Ohnuki Y, Lechner JF, Jones LW. Establishment and characterization of a human prostatic carcinoma cell line (PC-3). Invest Urol 1979;17:16–23. [21] Pulukuri SM, Gondi CS, Lakka SS, Jutla A, Estes N, Gujrati M, et al. RNA interferencedirected knockdown of urokinase plasminogen activator and urokinase plasminogen activator receptor inhibits prostate cancer cell invasion, survival, and tumorigenicity in vivo. J Biol Chem 2005;280:36529–40. [22] Corn PG, Wang F, McKeehan WL, Navone N. Targeting fibroblast growth factor pathways in prostate cancer. Clin Cancer Res 2013;19:5856–66. [23] Browder T, Folkman J, Pirie-Shepherd S. The hemostatic system as a regulator of angiogenesis. J Biol Chem 2000;275:1521–4. [24] Tsopanoglou NE, Pipili-Synetos E, Maragoudakis ME. Thrombin promotes angiogenesis by a mechanism independent of fibrin formation. Am J Physiol 1993;264:C1302–7. [25] Hembrough TA, Swartz GM, Papathanassiu A, Vlasuk GP, Rote WE, Green SJ, et al. Tissue factor/factor VIIa inhibitors block angiogenesis and tumor growth through a nonhemostatic mechanism. Cancer Res 2003;63:2997–3000. [26] Dardik R, Loscalzo J, Inbal A. Factor XIII (FXIII) and angiogenesis. J Thromb Haemost 2006;4:19–25. [27] Falanga A, Panova-Noeva M, Russo L. Procoagulant mechanisms in tumour cells. Best Pract Res Clin Haematol 2009;22:49–60. [28] Falanga A, Marchetti M, Vignoli A. Coagulation and cancer: biological and clinical aspects. J Thromb Haemost 2013;11:223–33. [29] Selzner N, Liu H, Boehnert MU, Adeyi OA, Shalev I, Bartczak AM, et al. FGL2/ fibroleukin mediates hepatic reperfusion injury by induction of sinusoidal endothelial cell and hepatocyte apoptosis in mice. J Hepatol 2012;56:153–9. [30] Hollborn M, Kohen L, Werschnik C, Tietz L, Wiedemann P, Bringmann A. Activated blood coagulation factor X (FXa) induces angiogenic growth factor expression in human retinal pigment epithelial cells. Invest Ophthalmol Vis Sci 2012;53:5930–9. [31] Nugent MA, Iozzo RV. Fibroblast growth factor-2. Int J Biochem Cell Biol 2000;32: 115–20. [32] Eswarakumar VP, Lax I, Schlessinger J. Cellular signaling by fibroblast growth factor receptors. Cytokine Growth Factor Rev 2005;16:139–49. [33] Yan G, Fukabori Y, McBride G, Nikolaropolous S, McKeehan WL. Exon switching and activation of stromal and embryonic fibroblast growth factor (FGF)-FGF receptor genes in prostate epithelial cells accompany stromal independence and malignancy. Mol Cell Biol 1993;13:4513–22. [34] Feng S, Wang F, Matsubara A, Kan M, McKeehan WL. Fibroblast growth factor receptor 2 limits and receptor 1 accelerates tumorigenicity of prostate epithelial cells. Cancer Res 1997;57:5369–78. [35] Yang F, Zhang Y, Ressler SJ, Ittmann MM, Ayala GE, Dang TD, et al. FGFR1 is essential for prostate cancer progression and metastasis. Cancer Res 2013;73:3716–24. [36] Wesley UV, McGroarty M, Homoyouni A. Dipeptidyl peptidase inhibits malignant phenotype of prostate cancer cells by blocking basic fibroblast growth factor signaling pathway. Cancer Res 2005;65:1325–34. [37] Yang F, Strand DW, Rowley DR. Fibroblast growth factor-2 mediates transforming growth factor-beta action in prostate cancer reactive stroma. Oncogene 2008;27: 450–9.
