NDV-D90 suppresses growth of gastric cancer and cancer-related vascularization

Recent reports suggest promises on using oncolytic Newcastle disease viruses (NDV) to treat different cancers, while the effects of a NDV-D90 strain on gastric cancer remain unknown. Here we showed that NDV-D90 induced gastric cancer cell apoptosis in a dose-dependent manner in 3 gastric cancer cell lines BGC-823, SGC-7901 and MKN-28. Pronounced reduction in cell invasion was detected in NDV-D90-treated BGC-823 and SGC-7901 cells, but not in MKN-28 cells. The increases in cell apoptosis and reduction in cell growth in NDV-D90-treated gastric cancer cells seemingly resulted from augmentation of p38 signaling and suppression of ERK1/2 and Akt signaling. In vivo, orthotopic injection of NDV-D90 impaired tumor growth and induced intratumoral necrosis. Tumor cells that had been pre-treated with NDV-D90 showed defect in development of implanted tumor. Moreover, NDV-D90 appeared to reduce gastric tumor vascularization, possibly through suppression of vascular endothelial growth factor A and Matrix Metallopeptidase 2. Together, our data suggest that NDV-D90 may have potential anti-cancer effects on gastric cancer.


INTRODUCTION
Gastric cancer is a prevalent cancer, in which malignant cells form in the lining of the stomach [1]. Nearly all gastric cancers are adenocarcinomas [2]. Although surgical removal of the part of stomach that bares tumor may be effective for early stage gastric cancer, the majority of gastric cancer patients are diagnosed at an advanced stage due to lack of early signs or symptoms of the disease [1]. Therefore, novel therapies are required for treatment of advanced gastric cancer that may be amendable by surgery.
The anti-cancer potential of Newcastle disease viruses (NDV) has been discovered since 1805s [3].
The selective targeting elimination of replicating tumor cells by NDV stem from the presence of defective Interferon signaling pathways in tumor cells, but an effective antiviral response to hamper viral replication in normal cells [3]. Recently, selection, modification and production of oncolytic NDV with minimal damage of the normal adjacent tissues have been achieved [4,5]. Genetic editing of NDV to express immune modulators (i.e. GM-CSF) has amplified the effects of NDV on targeted cancer tissues, rendering them one of the most effective multitarget cancer vaccines [5]. Since the clinical study of oncolytic NDV, three strains of NDV, MTH-68 [6], NDV-HUJ [7] and PV701 [8][9][10][11] have now been used in phase I/ II clinical trials for tumor treatment.

Research Paper
Oncotarget 34517 www.impactjournals.com/oncotarget Very recently, a NDV-D90 strain that was isolated from natural sources in China, and was shown to specifically induce apoptosis of human lung adenocarcinoma cell line A549 through regulating caspases [12]. Moreover, application of NDV-D90 appeared to reduce tumor growth in vivo [13]. Furthermore, the anti-cancer effects of NDV-D90 were just reported in oral squamous cell carcinoma [14]. However, whether NDV-D90 may have similar effects on gastric cancer is unknown.
Here, we addressed this question. We found that NDV-D90 induced gastric cancer cell apoptosis in a dosedependent manner in 3 gastric cancer cell lines BGC-823, SGC-7901 and MKN-28. Pronounced reduction in cell invasion was detected in NDV-D90-treated BGC-823 and SGC-7901 cells, but not in MKN-28 cells. The increases in cell apoptosis and reduction in cell growth in NDV-D90-treated gastric cancer cells seemingly resulted from augmentation of p38 signaling and suppression of ERK1/2 and Akt signaling. In vivo, orthotopic injection of NDV-D90 impaired tumor growth and induced intratumoral necrosis. Tumor cells that had been pretreated with NDV-D90 showed defect in development of implanted tumor. Moreover, NDV-D90 appeared to reduce gastric tumor vascularization, possibly through suppression of vascular endothelial growth factor A (VEGF-A) and Matrix Metallopeptidase 2 (MMP-2). Together, our data suggest that NDV-D90 may have potential anti-cancer effects on gastric cancer.

NDV-D90 reduces gastric cancer cell growth in vitro
First, we examined the effects of NDV-D90 on the growth of gastric cancer cells in vitro. We used 3 human gastric cancer cell lines, BGC-823, SGC-7901 and MKN-28. BGC-823 is low differentiated gastric carcinoma, SGC-7901 is medium differentiated gastric carcinoma, while MKN-28 is highly differentiated gastric carcinoma. We found that NDV-D90 dose-dependently reduced the cell viability in all 3 lines, and the effects of NDV-D90 on BGC-823 and SGC-7901 cells appeared to be more pronounced ( Figure 1A-1C). Next, virus amplification was measured in the 3 cell lines, showing that viruses proliferated more quickly in BGC-823 and SGC-7901 cells, compared to MKN-28 cells ( Figure 1D). Together, these data suggest that NDV-D90 reduces gastric cancer cell growth in vitro, and NDV-D90 may be more effective on low differentiated, highly proliferative gastric cancer cells.

NDV-D90 reduces gastric cancer cell invasion in vitro
We then examined the alteration of gastric cancer cell invasion after NDV-D90 treatment. Tumor cell invasion test was performed and the invasion index (here Oncotarget 34518 www.impactjournals.com/oncotarget termed as the ratio of the NDV-D90-treated tumor cells that pass through the chamber versus the control tumor cells that pass through the chamber) was determined. We found that NDV-D90 treatment reduced the invasion index by 83% in BGC-823 cells, by 85% in SGC-7901 cells, but only by 20% in MKN-28 cells, shown by representative images (Figure 2A), and by quantification ( Figure 2B). These data suggest that NDV-D90 reduces gastric cancer cell invasion in vitro and this effect may be more pronounced on low differentiated, highly proliferative gastric cancer cells.

NDV-D90 induces gastric cancer cell apoptosis in vitro
Next, we analyzed the effects of NDV-D90 on cell apoptosis. We found that NDV-D90 induced apoptosis in a dose-dependent manner in all 3 gastric cancer cell lines BGC-823 ( Figure

NDV-D90 enhances p38 signaling and inhibits ERK1/2 and Akt signaling in gastric cancer cells
In order to understand the underlying mechanisms, we analyzed phosphorylation of a key factor in apoptosis-associated signaling pathway, p38, phosphorylation of 2 key factors in proliferation-associated signaling pathway, ERK1/2 and AKT. We found that NDV-D90 treatment induced the phosphorylation of p38, but suppressed the phosphorylation of ERK1/2 and AKT on BGC-823 cells ( Figure 4). These data suggest that the increases in cell apoptosis and reduction in cell growth in NDV-D90-treated gastric cancer cells may result from augmentation of p38 signaling and suppression of ERK1/2 and Akt signaling.

NDV-D90 reduces gastric cancer cell growth in vivo
The next question is whether NDV-D90 may have similar anti-cancer effects against gastric cancer in vivo. In order to examine the implanted tumor cells in living animals, we transduced BGC-823 cells with lentivirus carrying a RFP reporter ( Figure 5A-5D). The transduced cells (termed as BGC-823-RFP) were validated under fluorescent microscopy in vitro ( Figure 5E). BGC-823-RFP cells were then implanted into nude mice, after which NDV-D90 was intratumorally injected. The tumor was monitored at different time points after viral injection, showing suppression of tumor growth by representative images ( Figure 5F). Significant necrosis was detected exclusively in the implanted BGC-823-RFP tumor treated with NDV-D90 ( Figure 5G). When BGC-823-RFP cells were pre-treated with NDV-D90 before implantation, we found Oncotarget 34519 www.impactjournals.com/oncotarget that 48 hours after transplantation, the signals from the implanted tumor cells were hardly detected ( Figure 5H). Together, these data support an anti-cancer role of NDV-D90 in vivo in gastric cancer.

NDV-D90 impairs gastric cancer vascularization
Finally, we examined the effects of NDV-D90 on gastric cancer vascularization. Implanted tumors treated with/without NDV-D90 were dissociated and analyzed for the percentage of CD31+ endothelial cells inside the tumor. We found that NDV-D90 treatment significantly reduced the percentage of the CD31+ endothelial cells inside the tumor, shown by representative flow charts ( Figure 6A), and by quantification ( Figure 6B). We then examined the regulators of vascularization in the tumor and found that the levels of VEGF-A were significantly reduced in NDV-D90-treated tumor, compared to control, Oncotarget 34520 www.impactjournals.com/oncotarget by ELISA ( Figure 6C) and by immunohistochemistry ( Figure 6D). Similarly, we found that the levels of MMP-2 were significantly reduced in NDV-D90-treated tumor, compared to control, by ELISA ( Figure 6E) and by immunohistochemistry ( Figure 6F). Thus, NDV-D90 appeared to reduce gastric tumor vascularization, possibly through suppression of VEGF-A and MMP-2. Together, our data suggest that NDV-D90 may have potential anticancer effects on gastric cancer.

DISCUSSION
NDV is also known as avian paramyxovirus serotype 1, which belongs to a non-segmented, negativestrand RNA virus of the family Paramyxoviridae with a natural avian host range [3]. The genome of NDV is approximately 15 kb in length that encodes 6 structural proteins. Recent advance in application of virulence associated with human pathogens renders animal viruses as a promising therapeutic agent for tumor, since NDV does not have significant adverse effects in humans [3]. Interestingly, the selective targeting and the destruction of replicative tumor cells by NDV seemingly originates from the defective interferon signaling pathways in tumor cells, since interferon regulatory factor genes encode DNAbinding proteins that are involved in the innate immune response to viral infection [15].
Very recently, NDV-D90 strain was shown to induce apoptosis of human lung cancer cells and inhibit tumor growth in vitro [12], and in vivo [13]. In another independent study, the anti-cancer effects of NDV-D90 were detected in oral squamous cell carcinoma. However, whether NDV-D90 may have similar effects on gastric cancer is unknown. Moreover, the underlying mechanisms remain ill-defined. Here, we addressed these questions.
We found that NDV-D90 induced gastric cancer cell apoptosis and reduced cell invasion in a dose-dependent manner in 3 gastric cancer cell lines BGC-823, SGC-7901 and MKN-28. However, pronounced effects were detected in NDV-D90-treated BGC-823 and SGC-7901 cells, but not in MKN-28 cells. Since BGC-823, SGC-7901 and MKN-28 represent low differentiated, medium differentiated and highly differentiated gastric cancer cells respectively, these data suggest that NDV-D90 may be more effective on low differentiated, highly proliferative gastric cancer cells, which is supported by the high amplification of NDV-D90 in low differentiated, highly proliferative gastric cancer cells.
Finally, we approved our in vitro findings in vivo. Orthotopic injection of NDV-D90 impaired tumor growth and induced intratumoral necrosis. The effects Oncotarget 34521 www.impactjournals.com/oncotarget of NDV-D90 on tumor growth occurred earlier after treatment. In addition, the formation of the implanted tumor appeared to be affected by NDV-D90 treatment.
Although the current study investigated the effects of NDV-D90 on human gastric cancer cells, the tumor environment was in mice, but not in humans. The immunodeficiency state of nude mice may also have an effect on the interpretation of the data. Future studies may be performed in human patients to conquer these limitations. However, based on the collected evidence here, our data suggest that NDV-D90 may have potential anti-cancer effects on gastric cancer.

Protocol approval
All the experimental methods in the current study has been approved by the research committee at Harbin Medical University. All the experiments have been carried out in accordance with the guidelines from the research committee at Harbin Medical University. All animal experiments were approved by the Institutional Animal Care and Use Committee at Harbin Medical University. Surgeries were performed in accordance with the

Reagents and cell line culture
NDV-D90 was obtained from National Key Laboratory of Veterinary Biotechnology of Harbin Veterinary Medicine. Gastric cancer lines BGC-823, SGC-7901 and MKN-28 were purchased from Chinese Academy of Sciences Culture Collection (Shanghai, China), and maintained in RPMI 1640 medium (Hyclone, Shanghai, China) supplemented with 10% fetal bovine serum (Hyclone) in a humidified chamber with 5% CO 2 at 37°C.

Cell viability assay
The CCK-8 detection kit (DOJINDO, Shanghai, China) was used to measure cell viability. Briefly, gastric cancer cells were seeded in a 96-well microplate at a density of 10 4 /ml. After successful attachment, cells were treated with NDV-D90 at a multiplicities of infection (MOI) of 0.001, 0.01, 0.1 and 1, respectively, for a duration of 2 hours. The cells were cultured for 12 hours, 24 hours, 36 hours, 48 hours, 60 hours and 72 hours, after which 10 µl CCK-8 solution was added in each well and the plate was incubated at 37 o C for 4 hours before absorbance was measured with a monochromator microplate reader at a wavelength of 450 nm. The optical density value was reported as the percentage of cell viability in relation to the control group (set as 100%).

Measurement of virus amplification
Gastric cancer cells were seeded in a 6-well microplate at a density of 10 5 /ml. After successful attachment, cells were treated with NDV-D90 at a MOI of 0.01 for a duration of 2 hours. The cells were cultured for 12 hours, 24 hours, 36 hours, 48 hours, 60 hours and 72 hours, and the culture supernatant was collected at each time point for measurement of the TCID 50 .

Tumor cell invasion test
Gastric cancer cells were seeded in a 6-well microplate at a density of 10 5 /ml. After successful attachment, cells were treated with NDV-D90 at a MOI of 0.01 for a duration of 2 hours. The infected cells were collected and 500 µl was added into the upper chamber of Oncotarget 34523 www.impactjournals.com/oncotarget a BD BioCoat™ Matrigel™ Invasion Chamber (Becton-Dickinson Biosciences, San Jose, CA, USA) at a density of 5 × 10 4 /ml. The lower chamber was added with 750 µl culture medium containing 10% FBS. The incubation time was 14 hours, after which 50 µl Calcein-AM staining solution was added for 30 minutes to allow cells to be stained and visualized by fluorescent microscopy.

Apoptosis assay and flow cytometry
Cells were labeled with annexin V-FITC and propidium iodide (PI), using an apoptosis detecting kit (KeyGEN Biotech, Nanjing, China), and analyzed by flow cytometry using CellQuest software (Becton-Dickinson Biosciences, San Jose, CA, USA). For analyzing CD31+ cells in implanted tumor, the tumor was resected, minced into small pieces, and then digested in the digestion media containing 40 mg/dl collagenase (Sigma-aldrich, San Jose, CA, USA) and 0.05% trypsine (Sigma-Aldrich) at 37°C for 30 min. After the digestion, the cells that passed a 40 µm filter were subjected to flow cytometric analysis and sorting, using a FITC-conjugated rat-anti-mouse CD31 antibodiy (Becton-Dickinson Biosciences). Data were analyzed using FlowJo software (Flowjo LLC, Ashland, OR, USA).

ELISA
ELISA was performed using mouse VEGF-A or MMP-2 ELISA kit (R&D System, Los Angeles, CA, USA) according to manufacturer's instruction.

Immunohistochemistry
Immunohistochemistry was performed using a HRP/ DAB (ABC) Detection IHC kit (Abcam, Cambridge, MA, USA) according to manufacturer's instruction. Both rabbit anti-VEGF-A and anti-MMP2 were purchased from R&D System.

Nude mouse tumor model
BGC-823 cells were transduced with lenti-RFP viruses (ABM, Shanghai, China) to allow visualization of them in the implanted tumor by fluorescent microscopy. The male nude mice were purchased from SLAC Laboratory Animal Co. Ltd (Shanghai, China). These mice were used at 12 weeks of age, when they received subcutaneous transplantation with 108 gastric cancer cells on the back. After 3 weeks, the formed tumor received intratumoral injection of NDV-D90 or PBS as a control.

Statistical analysis
All statistical analyses were carried out using the SPSS 18.0 statistical software package. All data were statistically analyzed using one-way ANOVA with a Bonferroni correction, followed by Fisher's exact test to compare two groups. All values in cell and animal studies are depicted as mean ± standard deviation and are considered significant if p < 0.05.