Background
Synovial sarcoma (SS) is a malignant soft tissue sarcoma characterized by a recurrent chromosomal translocation [t(X;18)(p11;q11)] that forms the fusion protein, SS18-SSX [
1]. SS accounts for 5%–10% of all soft tissue sarcomas and mainly affects adolescents and young adults [
2]. This disease commonly presents in the extremities (80%) [
2] and distant metastases tend to be found mainly in the lungs [
3]. Although the mainstay of treatment comprises wide surgical excision, chemotherapy and radiotherapy, the 5-year overall survival rate of SS is only 30%–70% [
3‐
7]. Therefore, developing novel therapeutic approaches for SS is urgently needed.
We previously reported that c-MET or platelet-derived growth factor receptor α (PDGFRα) signalling pathway was relevant for SS progression, based upon the findings of phospho-receptor tyrosine kinase (RTK) arrays [
8]. c-MET, an RTK encoded by the
c-met proto-oncogene, is known to be a hepatocyte growth factor (HGF) receptor [
9]. Activation of the HGF/c-MET axis in cancer has been reported to be involved in cellular proliferation, survival, migration and angiogenesis [
10]. We have found that a selective c-MET inhibitor suppresses the growth of Yamato-SS cells, but fails to suppress the growth of SYO-1 or HS-SY-II cells [
11]. PDGFRα and PGDFRβ signalling indirectly promotes tumour development by activating the mesenchymal cells in the tumour microenvironment and directly stimulates the growth of malignant cells [
12]. Pazopanib, a PDGFR/ vascular endothelial growth factor receptor (VEGFR)/ c-kit (stem cell factor receptor) inhibitor [
13], is the only tyrosine kinase inhibitor approved for advanced soft tissue sarcomas in Japan. Hosaka et al. showed that pazopanib suppressed the growth of SYO-1 and HS-SY-II cells through inhibition of the PDGFRα and phosphatidylinositol 3-kinase (PI3K)/AKT pathways [
14]. Based upon these studies, we hypothesize that inhibition of the c-MET or PDGFRα signalling pathway would be a therapeutic strategy for the treatment of SS.
TAS-115, a novel c-MET/VEGFR-targeting tyrosine kinase inhibitor that exerts its effect via ATP antagonism, has been reported to inhibit multiple RTKs [
15]. Recently, it was reported that TAS-115 had a favourable tolerability profile and exhibited antitumour activity in human gastric cancer [
15,
16] and in human lung cancer [
17,
18] via inhibition of c-MET/VEGFR signalling. However, the efficacy of this drug for soft tissue sarcomas remains unclear.
In the present study, we first evaluated the phosphorylation status of RTKs in three human SS cell lines, Yamato-SS, SYO-1 and HS-SY-II, and then investigated which RTK was critical for the viability of each of these cell lines. Next, we tested the antitumour activity and the mechanism of action of TAS-115 in these SS cells. Finally, we compared the inhibitory activity of TAS-115 on the c-MET and PDGFRα pathways with that of pazopanib. On the basis of our observations, we discuss the potential clinical value of TAS-115 monotherapy, via c-MET and PDGFRα signal inhibition, in patients with SS.
Methods
Cell lines
The Yamato-SS cell line was established from surgically resected tumours in our laboratory, as previously described [
19]. SYO-1 was kindly supplied by Dr. Ozaki (Okayama University, Okayama, Japan) [
20]. HS-SY-II [
21] was provided by the RIKEN BRC (Tsukuba, Japan) through the National Bio-Resource Project of the MEXT, Japan. We authenticated Yamato-SS and HS-SY-II through short tandem repeat inspection. SYO-1 was confirmed by the expression of the
SS18-SSX2 fusion gene by reverse transcription polymerase chain reaction. Yamato-SS and SYO-1 cells originally derived from biphasic synovial sarcomas, while HS-SY-II originated from a monophasic synovial sarcoma [
19‐
21]. These cells were cultured in Dulbecco’s Modified Eagle’s Medium (Life Technologies, Carlsbad, CA, USA) containing 10% foetal bovine serum (FBS; Sigma-Aldrich, St. Louis, MO, USA) at 37 °C with 5% CO
2 and 100% humidity.
Reagents and antibodies
TAS-115 [4-[2-fluoro-4-[[[(2-phenylacetyl)amino]thioxomethyl]amino]-phenoxy]-7-methoxy-N-methyl-6-quinolinecarboxamide] and pazopanib [5-[[4-[(2,3-dimethyl-2H-indazol-6-yl)methylamino]-2-pyrimidinyl]amino]-2-methylbenzenesulfonamide] were provided by Taiho Pharmaceutical Co., Ltd. (Tsukuba, Japan) and Novartis Pharma AG (Basel, Switzerland), respectively. According to the manufacturer’s instructions, TAS-115 and pazopanib were suspended in dimethyl sulfoxide (DMSO, Sigma-Aldrich) for in vitro experiments. TAS-115 and pazopanib were diluted to the appropriate concentrations for in vivo experiments, according to the manufacturer’s instruction. Recombinant human (rh) PDGF-BB was obtained from Sigma-Aldrich.
Antibodies against PDGFRα (#7074), p-PDGFRα (Tyr754; #2992, Tyr849; #3170, Tyr1018; #4547), c-MET (#8198), p-MET (Tyr1234/1235; #3077), AKT (#4691), p-AKT (Ser473; #4060), ERK (#4695), p-ERK (Thr202/Tyr204; #4370), PARP (#9542) and β-actin (#4970) were purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA). All of these antibodies were used at 1:1000 dilution for immunoblot analyses. An antibody against p-PDGFRα (Tyr762; AF21141) was purchased from R&D systems (Minneapolis, MN, USA). This antibody was used at a concentration of 0.5 μg/ml for immunoblot analyses. An antibody against PCNA (sc-56) was purchased from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA) and used at a concentration of 1:50 for immunohistochemistry. Horseradish peroxidase (HRP)-conjugated secondary antibody was obtained from GE Healthcare Life Sciences (Pittsburgh, PA, USA).
Immunoblot analysis
After washing with PBS, cells were lysed in RIPA buffer (Thermo Scientific, Waltham, MA, USA) supplemented with 1% protease/phosphatase inhibitor cocktail (Cell Signaling Technology). Protein concentrations were measured using the bicinchoninic acid method (Thermo Scientific). The cell lysates were separated on 4–12% Bis-Tris gels (Life Technologies) and transferred to polyvinylidene difluoride (PVDF) membranes (Nippon Genetics, Tokyo, Japan). After blocking with 5% skim milk in Tris-buffered saline supplemented with Tween20 (TBS-T) at room temperature, the membranes were incubated with primary antibodies in Can Get Signal solution 1 (Toyobo Life Science, Tokyo, Japan) at 4 °C overnight, followed by incubation with secondary antibodies in Can Get Signal solution 2 (Toyobo Life Science) at room temperature for 1 h. After washing with TBS-T, immunoreactive bands were visualized using chemiluminescent reagents (ECL prime; GE Healthcare Life Sciences and ImmunoStar LD; Wako, Osaka, Japan).
RNA interference
Lipofectamine 2000 (Life Technologies) was used to transfect cells with 20 nM siRNAs, according to the manufacturer’s instruction. Two kinds of siRNAs targeting c-MET (constructs I and II; #6618 and #6622) and a non-targeting siRNA (#6568) were purchased from Cell Signaling Technology, Inc. Two kinds of siRNAs targeting PDGFRα (Hs_PDGFRα_1109, 6393) and a non-targeting siRNA (SIC-001) were obtained from Sigma-Aldrich.
Cell proliferation assay
SS cells were plated in 96-well plates for cell proliferation assays. The cell proliferation rate was measured using the premixed WST-1 cell proliferation assay system (Takara Bio, Inc., Otsu, Japan). The relative cell proliferation rate was calculated by subtracting absorbance measurements obtained from a microplate reader at 690 nm from those obtained at 450 nm.
Cell cycle analysis
SS cells (5 × 105 per dish) were seeded in 10-cm culture dishes and incubated overnight and then treated with TAS-115 or control (DMSO) for 24 h. The cells were harvested and stained with Propidium Iodide (PI) solution (25 μg/ml PI, 0.03% NP-40, 0.02 mg/ml RNase A, 0.1% sodium citrate) for 30 min at room temperature. We analysed the cell cycle using a BD FACSCanto II flow cytometer (Becton Dickinson (BD) Biosciences, San Jose, CA, USA).
In vivo xenograft experiments
The animal studies were performed in accordance with a guideline approved by the Institutional Animal Care and Use Committee of the Osaka University Graduate School of Medicine. We used Yamato-SS and SYO-1 cells because of their consistency in producing tumours in xenograft models. Yamato-SS cells (3 × 107) or SYO-1 cells (1 × 107) were inoculated subcutaneously into the flank of 5-week-old male BALB/c nu/nu mice (SLC, Shizuoka, Japan). Tumour volume (mm3) was defined as (A × B2)/2, where A and B were the longest and the shortest diameter of the tumour, respectively. Oral administration of TAS-115 was initiated after the average size of the established tumours reached around 100 mm3. TAS-115 was administered once daily at a dose of 50 or 200 mg/kg for 4 weeks. Xenograft tumour volume and the body weight of mice were measured once a week. After 4 weeks of treatment, the mice were euthanized and the tumour weight was measured. Resected tumours were used for immunohistochemical studies.
For immunoblot analyses, mice bearing tumours were orally treated with TAS-115 (200 mg/kg) and with pazopanib (100 mg/kg) for consecutive 3 days. Three hours after the last drug administration, the tumours were resected and extracted in Tissue Protein Extraction Reagent (Thermo Scientific). The dose of TAS-115 or pazopanib was determined based upon prior reports in which daily administration of TAS-115 (50–200 mg/kg) or pazopanib (100 mg/kg) resulted in significant growth inhibition of the xenograft tumours [
15,
16,
22,
23].
Immunohistochemistry
Xenograft tumours were fixed in 10% neutral-buffered formalin and embedded in paraffin. Sections (4 μm) were deparaffinized and dehydrated. After antigen retrieval in 10-mM citrate buffer at 95 °C for 30 min, endogenous peroxidase activity was blocked with methanol containing 3% H2O2 for 10 min. The sections were incubated with primary antibodies at 4 °C overnight, with secondary antibodies for 1 h on the next day, and then stained with 3,3′-diaminobenzidine tetrahydrochloride (DAB; Dako, Glostrup, Denmark). Finally, these sections were counterstained with hematoxylin.
Measurement of PCNA positive rate
PCNA positive rate was assessed by counting >500 cells from 3 random fields of each specimen under × 200 magnification in the best-stained tumour area of each section.
Statistical analysis
We used Student’s t-tests for experiments in vitro and the Mann–Whitney U test for experiments in vivo. Values of p < 0.05 were considered statistically significant.
Discussion
Aberrant activation of HGF/c-MET signalling by mutation, autocrine or paracrine HGF stimulation, or overexpression has been implicated in the oncogenesis of large number of cancers [
10,
25‐
28]. Additionally, upregulation or mutational activation of PDGF ligand or PDGFR expression has also been reported in many types of malignancies [
12,
29‐
31]. Co-expression of HGF and c-MET has been seen in 14% of SS clinical samples, correlating with poor prognosis [
32]. Likewise, we noted that 36% of SS specimens expressed both HGF and c-MET, which resulted in a significantly worse clinical course in SS patients [
11]. A recent study revealed that PDGF-AA expression and phosphorylation of PDGFRα without any gene alteration led to AKT activation through an autocrine/paracrine-mediated loop in a subset of clinical SS tumour samples [
33]. In the present study, we found c-MET activation in the Yamato-SS cells and PDGFRα phosphorylation in all three SS cell lines. Silencing of c-MET or PDGFRα expression revealed that the proliferation of the Yamato-SS cells sustained the high dependency upon c-MET signalling and that the viability of the SYO-1 or HS-SY-II cells was primarily driven by PDGFRα pathway. Based upon these observations, we could divide the SS cell lines tested in this study into two groups: c-MET-dependent SS cells (Yamato-SS) and PDGFRα-dependent SS cells (SYO-1 and HS-SY-II).
RTK cascades appear to be the principal activators of intracellular signalling through the PI3K/AKT and mitogen-activated protein kinase (MAPK) pathways. These downstream pathways have been reported as critical to SS viability [
34,
35]. TAS-115, a novel c-MET/VEGFR-targeting dual tyrosine kinase inhibitor, has been reported to inhibit multiple RTKs besides c-MET and VEGFR under cell-free conditions [
15]. TAS-115 significantly suppressed the proliferation of both c-MET-dependent and PDGFRα-dependent SS cells, mainly by inducing G0/G1 cell cycle arrest. Treatment with TAS-115 attenuated c-MET signalling in Yamato-SS cells, leading to subsequent suppression of AKT and ERK 1/2 phosphorylation. Similarly, blocking PDGFRα function with TAS-115 decreased intracellular signalling in SYO-1 and HS-SY-II cells. Interestingly, the TAS-115 concentration at which c-MET or PDGFRα signalling was inhibited in each cell line, using immunoblot analyses, corresponded to the IC
50 value for TAS-115 against c-MET-dependent cells (Yamato-SS) or PDGFRα-dependent cells (SYO-1 or HS-SY-II) in WST-1 cell proliferation assays. In agreement with our in vitro findings, TAS-115 exerted anti-c-MET or anti-PDGFRα pathway activity and inhibited their downstream effectors in the Yamato-SS (c-MET-dependent) and SYO-1 (PDGFRα-dependent) cells in in vivo xenograft models. These insights indicate that the c-MET and PDGFRα cascades are the primary regulators of both the PI3K/AKT and MAPK pathways in c-MET-dependent and PDGFRα-dependent SS cells, respectively. This difference of signal dependency in SS cell lines may be ascribed to the diversity among individual tumours. Another possible explanation is an artificial selection of cells with activated c-MET or PDGFRα from heterogeneous cell populations within a tumour during cell-line establishment.
Pazopanib, a PDGFRα and PDGFRβ/ VEGFR/ c-kit-targeting tyrosine kinase inhibitor [
13], has been approved by Food and Drug Administration (FDA) for soft tissue sarcoma and renal cell carcinoma [
36,
37]. In Japan, this drug is the only tyrosine kinase inhibitor available for treatment of advanced soft tissue sarcomas. Accumulating data suggested that high plasma concentrations of HGF were significantly correlated with shorter progression-free survival in patients treated with pazopanib for metastatic renal cell carcinoma [
38,
39]. These clinical studies give rise to an idea that the effect of pazopanib on cancers with active HGF/c-MET signalling might be limited. In the SS cell lines tested, pazopanib attenuated only PDGFRα signalling, while c-MET phosphorylation was not inhibited even at high concentration of pazopanib. By contrast, TAS-115 successfully suppressed both the c-MET and PDGFRα pathways. Moreover, TAS-115 treatment inhibited the phosphorylation of PDGFRα on at least four representative autophosphorylation sites of the receptor, as well as its downstream effectors, equivalently to pazopanib. These data suggest that TAS-115 will have therapeutic capability in SS with active c-MET or PDGFRα pathways, via inhibition of c-MET and PDGFRα signalling, indicating that it might be clinically applicable.
Recently, several investigators argued the possibility of personalized therapy by evaluating the specific signalling activation state of tyrosine kinases in different types of malignancies, including lung carcinoma [
40], colorectal carcinoma [
41] and ovarian carcinoma [
42]. Most recently, we have reported that c-MET phosphorylation in clinical samples is a potential biomarker predicting response to a selective c-MET inhibitor for SS patients [
11]. Treatment with a specific tyrosine kinase inhibitor selected by predicting therapeutic efficiency may prevent or delay non-specific deleterious side effects [
43]. On the other hand, the possible advantages of a multi-targeting tyrosine kinase inhibitor include simplicity and capability of its use in daily practice for tumours with multiple signalling pathways. The tumours found in patients may have inter-individual differences in the activation states of RTKs that are involved in tumour growth or progression, even if they belong to a pathologically identical group. Moreover, the individual tumour may be a heterogeneous mixture of cell populations driven by multiple RTK signals. Indeed, our results indicated that monotherapy with TAS-115 suppressed both c-MET and PDGFRα signalling, resulting in strong therapeutic effects against two types of SS cells both in vitro and in vivo without outstanding side effects
. These observations support the view that inhibition of both c-MET and PDGFRα with a single agent is a practical consideration for SS treatment.
The cellular origins of SS are still unknown. While several potential cellular origins of SS have been reported, such as cells of a myogenic or neurogenic lineage [
44‐
46], we have claimed that a multipotent mesenchymal stem cell may be a cell of origin for SS [
19]. In this study, we noted the activation of c-MET or PDGFRα and the importance of these signals for viability in SS cell lines. In some translocation-related sarcomas, fusion genes are assumed to contribute to the upregulation of RTK cascades. For instance, c-MET was identified as a direct target of
PAX3-FOXO1 in alveolar rhabdomyosarcoma and
EWS-ATF1 in clear cell sarcoma [
27,
47]. Additionally,
EWS-FLI1 directly upregulated
PDGF-C expression, which in turn activated the PDGFRα pathway in Ewing sarcoma [
48]. Correspondingly,
EWS-WT1, which underlay desmoplastic small round cell tumours, induced
PDGF-A expression [
49]. By sharp contrast, our parallel study showed that
SS18-SSX silencing did not elicit the downregulation or inactivation of either c-MET or PDGFRα [
11]. These data suggest that the activation of c-MET or PDGFRα might not be evoked by
SS18-SSX. Rather, it reflects a prerequisite cellular background to be permissive for
SS18-SSX. It has been reported that the window of permissive cells might be relatively narrow for
SS18-SSX [
44,
50]. Thus we speculated that SS might arise from a multipotent mesenchymal stem cell natively driven by HGF/c-MET or PDGF/PDGFR signalling [
8]. These findings raise the possibility that SS inherits phosphorylated RTKs from a multipotent mesenchymal stem cell at a certain phase of differentiation and that these RTKs are suitable therapeutic targets for SS. Taken together, RTKs innately present in cells of origin, in addition to those epigenetically altered by fusion genes, might be plausible target molecules for the treatment of translocation-related sarcomas.
Although the findings detailed above do seem clear, there are some limitations in this study. First, since an environment in which either cultured cells or mouse xenografts grow is significantly different from that of their originating tumour, those model systems cannot always provide useful information to generate clinically relevant treatments. Second, in the clinical setting, distant metastases have been reported to be strongly associated with worse outcome of patients with SS [
3,
5,
7]. However, experimental models of human SS with spontaneous metastatic potential have not been successfully developed. Thereby, whether TAS-115 monotherapy is sufficient to treat metastatic SS remains unknown. In October 2016, olaratumab, a monoclonal antibody of PDGFRα, received FDA approval for clinical use in combination with doxorubicin for unresectable and metastatic soft tissue sarcomas based upon the clinical data that olaratumab plus doxorubicin showed gains in median progression free and overall survival as compared to doxorubicin alone [
51]. Those results lead us to hypothesize that combined therapy of conventional chemotherapeutic agents and TAS-115 targeting PDGFRα and c-MET may be effective for metastatic SS and warrants further investigations. Third, though treatment of TAS-115, introduced as an angiogenesis inhibitor via VEGFR signalling [
15], decreased MVD in our in vivo experiments, the extent to which VEGFR inhibition by TAS-115 in the tumour microenvironment contributed to suppression of tumour growth was unclear in this study. Since HGF/c-MET as well as PDGF/PDGFR systems have been recognised as important mediators of angiogenesis [
10,
12,
52], it could not be excluded that the antiangiogenic mechanisms of TAS-115 in SS at least in partly encompassed targeting c-MET or PDGFRα. Further clarification would be required in order to make definitive answers to these challenging issues in the future.
Acknowledgements
We thank Prof. Toshifumi Ozaki for the gift of the SS cell line SYO-1. We also thank Mari Shinkawa, Asa Tada and Yukiko Eguchi for excellent technical assistance and Hidenori Fujita for insightful comments and suggestions.