Pterostilbene enhances sorafenib’s anticancer effects on gastric adenocarcinoma

Abstract Sorafenib has been approved for the treatment of certain cancers in clinic. However, the effects of sorafenib on gastric adenocarcinoma (GAC) were still limited. This study aimed to evaluate both in vitro and in vivo efficacy of sorafenib in combination with pterostilbene (PTE) on the treatment of GAC. Here, the morphological changes and cell viability were recorded in both N87 and MKN45 cells. The cell cycle profile and apoptosis were assessed by flow cytometry. Subcutaneous tumour xenografts were constructed in nude mice, and IHC staining of the dissected tumour tissues was conducted. Our results showed that PTE enhanced sorafenib's inhibitory effects on cell viability. The obvious down‐regulation of cyclin D1, Cdk‐2, Cdk‐4, Cdk‐6 and p62 and the up‐regulation of LC3II, caspase‐9, caspase‐3 and PARP cleavages were observed for the combination treatment with PTE and sorafenib than monotherapy. The combination treatment resulted in a higher level of cell cycle arrest at G1 phase and apoptosis than either drug. Besides, drug combination significantly enhanced the inhibition of tumour growth than sorafenib or PET alone in nude mice. The percentage of Ki‐67‐ and PCNA‐positive cells was distinctly reduced, and the apoptotic cells was obviously increased when compared with single drug therapy. Altogether, PET obviously enhanced sorafenib's antitumour effects against GAC through inhibiting cell proliferation, inducing autophagy and promoting apoptosis. The combination therapy with PET and sorafenib may serve as a novel therapeutic strategy for treating GAC and deserve further clinical trials.


| INTRODUC TI ON
As the fourth most common and the second most deadly cancer worldwide, gastric adenocarcinoma (GAC) remains one of the major public health problems worldwide. 1 Nearly two-thirds of patients recur after curative resection. Currently, chemotherapy followed by surgery is the first-line treatment for most GAC patients. 2 Due to drug resistance and severe adverse side effects, combination therapy may be a potential therapeutic approach for GAC patients. 3 Combination therapy could sensitize GAC cells to the cytotoxic effects induced by monotherapy, reducing the doses of either drug and improving the clinical effects. 4 Sorafenib, a multi-kinase inhibitor, has been shown to suppress tumour cell proliferation and induce apoptosis. 5 It has been approved for the clinical treatment of advanced renal cell carcinoma and unresectable hepatocellular carcinoma. 6,7 Several other trials against various solid tumours are currently in progress, including gastric cancer, lung cancer, breast cancer and prostate cancer. [8][9][10][11] Due to the numerous adverse side effects of sorafenib, combination therapies are encouraged in the future investigations to reduce the dosage and improve the clinical therapeutic effects.
Pterostilbene (trans-3,5-dimethoxy-4-hydroxystilbene, PTE), as a natural dimethylated analogue of resveratrol (RESV) extracted from blueberries, exhibits diverse pharmacologic activities including anticancer, anti-inflammation, antioxidant, anti-proliferative and analgesic activities. 12,13 Under most circumstances, PTE shows more potent antitumour activity than RESV, resulting from the substitution of a hydroxy group with a methoxy group. 14,15 Therefore, PTE could be more potentially developed for clinical applications.
A recent study showed that sorafenib alone failed to inhibit GAC tumour growth in vivo, while a marked inhibitory effect was induced when co-administered with non-toxic diclofenac, an multidrug resistance-associated protein (MRP) inhibitor. 16 However, whether pterostilbene would also sensitize the GAC response to sorafenib has never been investigated. In this study, we examined the efficacy of sorafenib and pterostilbene combination on GAC both in vitro and in vivo, and also investigated the underlying mechanism of the enhanced anticancer effects against GAC.

| Chemicals
Sorafenib was purchased from International Laboratory USA (#320790); resveratrol (RESV) was obtained from J&K Scientific Ltd (Woburn, MA, USA), and pterostilbene (PTE) was purchased from Sigma-Aldrich (St. Louis, MO) with purity over 97% (see structure in Figure 1A,B). All compounds were dissolved in dimethyl sulphoxide (DMSO, Sigma, USA) and further diluted in sterile culture medium immediately prior to the in vitro and in vivo experiments.

| Cell lines and cell culture
The study was carried out on two cell lines (N87 and MKN45) derived from human gastric adenocarcinoma. Cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% foetal bovine serum and 1% penicillin/streptomycin (10 000 units of penicillin/mL and 10 mg/mL streptomycin) in an incubator at 37°C with 5% CO 2 in air.

| Cell viability assay
Cell viability was evaluated using the MTT (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl-2-(4-sulfo-phenyl)-2Htetrazolium) (Sigma-Aldrich, France) assay. Cells at logarithmic growth period were plated in 96-well plate at a density of 5000 cells/well in a volume of 100 mL. Then, the cells were treated with target agents at the desired concentrations. Five replicates were conducted for each medication dose. After treated for 24 hours, 10 µL of MTT (10 mg/mL) was added to each well and incubated for another 4 hours. After abandoning the supernatants, 100 µL of DMSO (Sigma, USA) was added to each well to dissolve the crystals. Subsequently, the optical density (OD) of each well was measured using a microplate reader at a wavelength of 550 nm (BMG Labtech, Ortenberg, Germany). The 50% inhibitory concentration (IC 50 ) values were calculated using the GraphPad Prism ® 5 (ver-
Then, the protein lysates were denatured at 95°C for 5 minutes after mixing with 5x SDS-loading buffer. Subsequently, the cell extracts (30 µg protein) were separated on a sodium dodecyl sulphate-polyacrylamide electrophoretic gel (SDS-PAGE) and then transferred to nitrocellulose membranes. After blocked with 3% BSA for 2 hours, the membranes were incubated overnight at 4°C with the following primary antibodies at dilutions

| Immunohistochemistry
Firstly, the dissected tumour tissues were formalin-fixed, paraffinembedded and sectioned. Next, the sections were deparaffinized in xylene, rehydrated through the descending grades of alcohol and then washed with PBS. Antigen retrieval was performed with citrate buffer (pH 6.0) for 20 minutes in microwave, and then the activity of endogenous peroxidase was quenched with 3% H 2 O 2 for 30 minutes. After blocking with BSA, the slides were incubated with PCNA and Ki67 (Santa Cruz Biotechnology, sc-15402) overnight at 4°C at 1:50 dilution, and then incubated with the appropriate secondary antibody for 1 hour at room temperature. Besides, the TUNEL assay was conducted to detect apoptotic cells according to the protocol as described by the manufacturer (Beyotime Biotechnology). The IHC staining was visualized by 3,3′-diaminobenzidine (DAB), followed by counterstaining with haematoxylin (Sigma). Five random optical fields from each section were recorded at ×200 magnifications under Nikon Eclipse E400 microscope. And the number of positive staining cells was counted for each field.

| Statistical analyses
All analyses were performed with the software SPSS ver. 20.0 (SPSS Inc, Chicago, IL, USA). Data were presented as mean ± standard deviation (SD) in at least five independent experiments. t Test and twoway ANOVA was used for all the statistical analyses. A P value less than 0.05 was considered statistically significant.

| Cell viability and morphological changes
The differences in the chemical structures between RESV and PTE were the substitution of a hydroxy group with a methoxy group ( Figure 1A,B). First, N87 and MKN45 cells were treated with indicated concentration of PTE or RESV for 24 hours and then subjected to MTT assay for cell viability. As shown in Table 1, PTE showed a stronger cell growth inhibitory effect than RESV. The IC 50 value was 52.71 ± 1.23 μmol/L vs 116.68 ± 2.45 μmol/L for N87 cells and 65.63 ± 1.52 μmol/L vs 132.56 ± 2.38 μmol/L for MKN45 cells, respectively. Therefore, PTE was selected for combination studies. Figure 1C,D shows that PTE and sorafenib significantly inhibited cell viability in a dose-dependent manner. In N87 cells, the IC 50 of PET and sorafenib was 52.71 ± 1.23 μmol/L and 3.85 ± 0.23 μmol/L, respectively. In MKN45 cell, the IC 50 of PET and sorafenib was 65.63 ± 1.52 μmol/L and 6.27 ± 0.34 μmol/L, respectively (Table 1).
Next, to examine whether combination treatment displays better anticancer effects, PTE was used at a lower concentration of

| The enhanced effects on cell cycle arrest in G1 phase by PTE
As a marker of proliferation, PCNA expression was significantly decreased upon combined treatment when compared with PTE or sorafenib alone (Figure 2A,B). In order to reveal whether cell proliferation inhibited by PTE and sorafenib was caused by cell cycle arrest, we monitored the expression levels of the cell cycle regulatory proteins. Our results showed that the protein levels of cyclin D1, Cdk-2, Cdk-4 and Cdk-6 in both N87 and MKN45 cells were obviously down-regulated in the combination treatment groups than that in the monotherapy groups (Figure 2A,B). Similar to the cell viability, the effects on the expression of cell cycle regulatory proteins were also enhanced in a sorafenib dose-dependent manner (Figure 2A,B).
Besides, results from flow cytometry assays showed that treatment with either sorafenib (4 μmol/L for N87 and 6 μmol/L for MKN45) or PET (30 μmol/L) alone induced cells accumulated in G1 phase with a concomitant decrease of cells in S phase after 16 hours treatment.
As expected, the G1 phase accumulation was significantly higher in the combination treatment group than that in the monotherapy ones, confirming the enhanced growth inhibitory effects induced by cell cycle arrest (P < 0.01 Figure 2C-F).

| The enhanced effects on the promotion of apoptotic and autophagy by PTE
As shown in Figure 3A   ( Figure 4C,D). We speculated that autophagy may also be involved in the anticancer enhancement. In summary, our results demonstrated PTE enhanced sorafenib's effects on GAC by the inhibition of cell cycle and the induction of apoptosis and autophagy.

| The enhanced antitumour effects by PTE in xenograft models
To evaluate the antitumour effects and the therapeutic safety of the PET and sorafenib combination in vivo, we constructed subcutaneous tumour xenografts using N87 cells in athymic nude mice. All treatment schemes were well tolerated and no apparent adverse effects (eg fatigue, mortality, significant weight loss, skin toxicity, hepatotoxicity, nephrotoxicity and cardiac toxicity) were observed ( Figure 5A; Table 2; Table S1). As shown in Figure 5B (Figure 2A,B). During cell division, cyclin D1 forms a complex with CDK4/CDK6 to promote DNA replication in cell cycle progression. 22 Oppositely, the decreased expression of cyclin D1/ Cdk4/Cdk6 inactivates pRb, causing cell cycle arrest in G1 phase and the inhibition of cell proliferation. 23,24 There are two major types of programmed cell death (PCD), that is apoptosis and autophagic death. 25,26 Our results showed that PET effectively sensitized GAC cells to sorafenib induced apoptosis via the intrinsic mitochondrial pathway. PTE combined with sorafenib dramatically promoted the cleavages of apoptotic-related proteins than monotherapy, including caspase-9, caspase-3 and PARP ( Figure 3A,B). Cytochrome c released from mitochondria to cytosol first forms apoptosome with procaspase-9 that further induces caspase-3 activation and PARP cleavage. [27][28][29] Caspase-3, an ultimate executioner of the caspase family, is responsible for the nuclear changes during apoptosis including chromatin condensation. 30 PARP, a highly conserved nuclear enzyme, plays significant roles in DNA repair, recombination, proliferation and genomic stability. 31,32 Besides, the permeabilization of cytochrome c is also regulated by anti-apoptotic proteins (Bcl-2, Bcl-xl, Mcl-1) and pro-apoptotic proteins (Bax, Bad, Bak). 33,34 Bcl-2 inhibits the oligomerization of Bax, leading to inhibition of cytochrome c release. 35 In our study, we demonstrated that the combination treatment strongly increased the Bax/Bcl-2 ratio than mono-treatment ( Figure 3A,B). So, the apoptosis induced by mitochondrial intrinsic pathway may be another explain for the enhanced anticancer effects by PTE, which were also been proven in vivo by the TUNEL staining in the tumour tissues ( Figure 6A,D).
Autophagy is characterized by the accumulation of autophagic vacuoles or autophagosomes, followed by fusion with lysosomes to form autophagolysosomes and subsequent degradation of their intracellular organelles. 36,37 We observed much more autophagic vacuoles in the cancer cells upon combination treatment ( Figure 4C,D).
The key event of autophagy is the conversion of LC3 from LC3-I (18 kD cytosolic free form) to LC3-II (16 kD autophagosomal membrane-bound form). 38 Meanwhile, P62/sequestosome1 ⁄SQSTM1, as a cargo protein binding to LC3-II and ubiquitinated proteins, is sequestered inside the autophagosome and degraded by autophagy. [39][40][41] Our results showed that the combination therapy with PET and sorafenib more strongly activated autophagy than monotherapy,

TA B L E 2
The antitumour effects of sorafenib with or without PET in N87 xenograft model evidenced by the up-regulation of LC3-II and down-regulation of p62, as compared with single-agent treatment alone ( Figure 4A,B).
In addition, the down-regulation of Bcl-2 could also apparently promote autophagy. 42 In conclusion, our results demonstrated that PET potently en-

CO N FLI C T O F I NTE R E S T
The authors declare that they have no competing interests.

E TH I C S A PPROVA L A N D CO N S E NT TO PA RTI CI PATE
Animal studies were performed in accordance with the criteria

DATA AVA I L A B I L I T Y S TAT E M E N T
The data sets used and/or analysed during the current study are available from the corresponding author on reasonable request.