Inhibition of c‐MET increases the antitumour activity of PARP inhibitors in gastric cancer models

Abstract Gastric cancer is the fifth most common malignancy and the third leading cause of cancer‐related death worldwide. Activation of c‐MET increases tumour cell survival through the initiation of the DNA damage repair pathway. PARP is an essential key in the DNA damage repair pathway. The primary role of PARP is to detect and initiate an immediate cellular response to single‐strand DNA breaks. Tumours suppressor genes such as BRCA1/2 are closely associated with the DNA repair pathway. In BRCA1/2 mutations or deficiency status, cells are more likely to develop additional genetic alterations and chromosomal instability and can lead to cancer. In this study, we investigate the role of c‐MET and PARP inhibition in a gastric cancer model. We exploited functional in vitro and in vivo experiments to assess the antitumour potential of co‐inhibition of c‐MET (SU11274) and PARP (NU1025). This leads to a reduction of gastric cancer cells viability, especially after knockdown of BRCA1/2 through apoptosis and induction of γ‐Η2ΑΧ. Moreover, in AGS xenograft models, the combinatorial treatment of NU1025 plus SU11274 reduced tumour growth and triggers apoptosis. Collectively, our data may represent a new therapeutic approach for GC thought co‐inhibition of c‐MET and PARP, especially for patients with BRCA1/2 deficiency tumours.

models, may decrease the survival of cancer cells and promote a more effective antitumour therapy. 15 One crucial role of PARP is assisting in the repair of single-strand DNA breaks. As a result, PARP inhibition leads to DNA double-strand breaks (DSBs) that are the most deleterious form of DNA damage. 16 Clinical trials (NCT01063517 and GOLD, NCT01924533, respectively) use agents that focus on this DNA repair pathway mechanism.
In more detail, phase II/III clinical studies use PARP inhibitor in the chemotherapeutic scheme with paclitaxel. This co-treatment showed a beneficial effect on the survival rating of patients. [15][16][17][18] In light of the results from clinical studies, PARP inhibition in GC patients tries to improve our understanding of DSBs repair pathways and find new and more reliable predictive markers for this kind of cancer. 19,20 BRCA1/2 proteins are necessary for the HR progression as the cells are susceptible to PARP inhibition when the BRCA1/2 protein is deficient. 21,22 Many studies of BRCA1/2 mutations and GC are indirect and do not show the rate of BRCA1/2 mutations in patients with GC. 23 However, the link between BRCA1/2 mutation and increased risk of GC was verified in previous studies for families with hereditary breast and ovarian cancer. [24][25][26] In an analysis done in Israel, 5.7% of patients were detected with GC with specific BRCA2 mutations. 27 Zhang et al showed that loss of BRCA1 occurred in 21.4% of patients with GC. Patients with BRCA1 loss have reduced life expectancy due to higher tumour grade and advanced clinical stage. 28 Mutations in BRCA1/2 mutations increase the risk of developing CG around sixfold, especially between first-degree relatives. 29 It has been shown that c-MET stimulation is necessary to develop resistance to the DNA damaging agent. 30,31 Another study reports that inhibition of MET, in MET-overexpressing GC model, causes damage to the DNA, resulting in premature ageing. 32,33 In the current study, we try to explore the combination of c-met and PARP inhibition in GC cell lines models (AGS and HS746T). In more detail, co-treatment of GC cell lines with NU1025 and SU11274 (PARP and c-MET inhibitor, respectively) decreased cell viability through induction of apoptotic cell death in BRCA1/2 deficiency manner. Furthermore, in vivo experiment in AGS xenograft mouse model, co-inhibition of c-MET and PARP decreases tumour volume mass. Collectively, we proposed that co-treatment of PARP and c-MET inhibitors had a beneficial effect in the BRCA1/2 deficiency GC model and are a putative therapeutic approach for GC patients.

| Inhibitors and drugs
The c-MET inhibitor SU11274 (#S9820) and PARP inhibitor NU1025 (#N7287) were obtained from Sigma-Aldrich. Both inhibitors were dissolved in DMSO and stored at − 80°C.

| Cell culture
Hs746T and AGS GC cell lines were obtained from American Type Culture Collection (ATCC) and American Type European Collection of Authenticated Cell Cultures (ECACC).
All cell lines used in this study were grown in RPMI Medium 1640-GlutaMAX™ (#61870-010 Life Technologies Carlsbad) supplemented with 10% foetal bovine serum (FBS), penicillin and streptomycin antibiotics (all from Invitrogen). Cells were maintained at 37°C in a humidified incubator containing 5% CO 2 . The experiments were done with the approval of the Ethics Committee of our University.

| Cell viability assay
Cell growth and viability were confirmed by MTT assay.
Approximately 3,000 cells were placed in a 96-well plate with 200 μL culture medium. At the end of treatment time, cells were incubated for 4 hours with 0.8 mg/mL of MTT, dissolved in a serum-free medium followed by DMSO (1 mL) and gentle shaking for 10 minutes to achieve the complete dissolution. Finally, the absorbance was measured at 560 nm using the microplate spectrophotometer system (SpectraMax 190-Molecular Devices). Results are presented as percentage of the control values.

| Western blot assay
As described in detail previously, 34

| Two-dimensional culture and confocal microscopy
For the 2D culture experiments, cells (5000 cells/well) were grown on coverslips in 24-well plates in medium, at 37°C. After knockdown of BRCA1/2 and treatment with NU1025 or SU11274, alone or in combination, for 24 hours cells were fixed in 4% paraformaldehyde, then permeabilized and then blocked with 0.5% BSA/PBS-5% 22 Triton X-100. Next, cells were treated with the primary H2AX antibody and then incubated with an antimouse fluorescence-labelled secondary antibody (#20014). Cells were examined using an Olympus FV1000 confocal microscope with an Olympus digital camera. The nuclei were stained with Dapi No. 33 342.

| Wound healing assay
HS746T GC cell line (40 000 cells/well) was grown in a 12-well plate and after knockdown of BRCA1/2 with siRNA transfection, cells were incubated with 5 μmol/L NU1025 and/or SU11274, alone or in combination, for 24 hours. At day 0, we formed the wound with a yellow pipette tip. After 24 hours of incubation, cells were photographed utilizing computer-assisted microscopy. We measured the gap distance of the wound on day 0 and after 24 hours using Image-Pro Plus software.

| Clonogenic cell survival assay
HS746T GC cell lines were plated into a 6-well plate. After knockdown of BRCA1/2 with siRNA transfection, cells were incubated with 5 μmol/L NU1025 and/or SU11274, alone or in combination, for 14 hours. We renew the inhibitors every 2 days. Following 14 days of incubation, colonies were fixed with methanol: acetic acid (3:1) solution and stained with haematoxylin. Cells were subsequently washed with PBS, dried and imaged.

| Mouse xenograft models
For the in vivo experiments, we used NSG MICE. All scid mice, housed in micro isolator cages, were used between 6 and 8 weeks of age. All The mice were randomly divided into groups (n = 5 per group) for each treatment, control, NU1025 (1 mg/mouse), SU11274 (1 mg/mouse) and NU1025 + SU11274. Inhibitors were injected intraperitoneally every 4 days. The mice were killed, and solid tumours were measured and excised after 20 days of treatment. The tumour volume was calculated using the following formula: 1/2(length × width2).

| Animal care and use statement
The animal protocol was designed to minimize pain or discomfort to the animals. The animals were acclimatized to laboratory conditions (23°C, 12-h/12-h light/dark, 50% humidity, and libitum access to food and water) for 2 weeks prior to experimentation. Intragastric gavage administration was carried out with conscious animals, using straight gavage needles appropriate for the animal size (15-17 g body weight: 22 gauge, 1-inch length, 1.25 mm ball diameter). All animals were killed by barbiturate overdose (intravenous injection, 150 mg/ kg pentobarbital sodium) for tissue collection.

| Statistical analysis
The results are representative of three independent experiments and expressed as mean values ± SD (standard deviation). For sample size, we used G*Power software version 3.1. For the calculation of tumour volume, we used Microsoft Excel 10. The results were evaluated by t test. Error bars indicate ± SD. *P < .05, **P < .005, ***<0.0005.

| Steady levels in primary gastric cancer cell lines
Different gastric cancer cell lines were examined regarding their protein levels of BRCA1, BRCA2 and c-MET using Western blot analysis.
In detail, the protein levels of BRCA1, BRCA2 and c-MET were decreased in the AGS cell line as compared to HS746T ( Figure 1).

| The role of c-MET in PARP inhibition response in GC cell lines
We identified the effect of c-MET activation on cell viability by PARP inhibition (NU1025) in an increasing dose-dependent manner

| The impact of BRCA and c-MET on PARP inhibition in GC cells
In order to identify whether the expression of BRCA has differential drug sensitivity, we continue with the silence of BRCA1/2 by siRNA in the HS746T cell line ( Figure 2B). Then, HS746T cells were treated with increasing concentrations (0-40 μmol/L) of PARP inhibitor (NU1025) for 48 hours. We evaluate that the silence of BRCA1/2 increases the sensitivity of cell after 5 μmol/L of NU1025 ( Figure 2B). Effectively silence of BRCA1/2 was confirmed by the reduction of protein levels of BRCA1 or BRCA2 by Western blot analysis ( Figure 2B). siRNA of BRCA1 or BRCA2 displayed a slight difference between controls. These data support that BRCA1 or 2 siRNA-mediated knockdowns affect cell viability of HS746T cell lines.
Additionally, in order to identify the primary mechanism of expression of BRCA impact on c-MET, we continued with the silence of c-MET. The knockdown of c-MET increases the sensitivity of BRCA-deficient HS746T cells to NU1025 after 48 hours ( Figure 2C). BRCA1/2-deficient Hs746T cells revealed a slightly higher growth-inhibitory impact in comparison with the BRCA1/2-proficient cells after 48 hours ( Figure 2B).

| Inhibition of c-MET (SU11274) sensitizes and triggers apoptosis in GC cells with BRCA1/2 deficient to PARP inhibition (NU1025)
In the light of previous results that silence of c-MET and/or BRCA1/2 sensitizes AGS and Hs746T cells to NU1025 treatment, we continue with the investigation of an additive effect in co-inhibition of c-MET

| The NU1025 plus SU11274 combinatorial treatment reduces the tumour and triggers apoptotic cell death growth in AGS xenograft models
Collectively of our in vitro experiments, we support the hypothesis that PARP and c-MET inhibition decrease the viability of GC cell lines. In light of these results, we try to evaluate our in vitro results in xenograft mouse models. For xenograft, we used the AGS cell line, which expressed low BRCA1, BRCA2 and c-MET compared with HS746T ( Figure 1). Thirty days after subcutaneous inoculation of AGS (106 cells/mouse), SCID mice treated intraperitoneally with 1 mg/mouse of SU11274 or NU1025 for 20 days alone or in combinatorial treatment. AGS xenografts were not very sensitive to PARP inhibitor alone (NU1025; Figure 5A). In contrast, SU11274 was slightly more effective in AGS xenograft models. The combinatorial scheme of NU1025 and SU11274 effectively decreases the tumour volume as shown in Figure 5A left panel. Furthermore, the tumour growth curves evaluated that co-treatment of SU11274 plus NU1025 was more effective in AGS xenograft models compared to either agent alone ( Figure 5A right panel).
Following the tumour growth results, the co-treatment of NU1025 plus SU11274 on AGS xenograft models triggers apoptotic cell death as it was measured through PARP and cleaved caspase-3 by Western

| D ISCUSS I ON
In the current study, we support that GC tumours with low levels overexpression as a resistance mechanism against PARP inhibitor.
It is well known that c-MET activation increases the DNA repair function of PARP1. 46,47 Another study shows that MET inhibition in GC tumours induces the ability of cancer cells to fix DNA damage and increases the effectiveness of the undergoing radiotherapy. 33 In the CRC model, the combination of crizotinib with mitomycin C (MMC) appeared to synergist and has an anti-proliferative effect regardless of MSI or BRCA2 status. 48    Based on these findings, further clinical testing of this combinatorial scheme is suggested in patients with locally advanced and/or metastatic gastric cancer.

ACK N OWLED G EM ENTS
The authors thank Dr S. Gagos for Hs746T cell line authentication (Biomedical Research Foundation of the Academy of Athens, Greece).

CO N FLI C T S O F I NTE R E S T
The authors declare no potential conflicts of interest. Formal analysis (equal).

I N S TITUTI O N A L A N I M A L C A R E A N D US E CO M M IT TE E S TAT E M E N T
All procedures were carried out in accordance with the guidelines for animal experimentation of the National and Kapodistrian University of Athens, Medical School Bioethics Committee in agreement with the European Union (approval no. 3233/26-06-2018).

E TH I C S A PPROVA L A N D CO N S E NT TO PA RTI CI PATE
All authors confirm that any aspect of the work covered in this manuscript that has involved either cell lines or animal models has been conducted with the ethical approval of all relevant bodies and that such approvals are acknowledged within the manuscript.

CO N S E NT FO R PU B LI C ATI O N
All authors consent for the publication of the manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
All authors declare that the data are available upon request.