Silencing MASTL Inhibits Cell Proliferation and Migration of Gastric Cancer MGC-803 Cells Via Suprressing AKT, mTOR, p38 Signal Pathways


 Background: Microtubule-associated serine/threonine kinase (MASTL) functions to regulate chromosome condensation and mitotic progression. Emerging reports showed that aberrant MASTL expression is commonly implicated in various human cancers and act as an oncogene. This study aimed to discover the potential significance of MASTL in gastric cancer, and to uncover relevant mechanisms. Methods: Lentivirus MASTL-shRNA was constructed and infected into MGC-803 cells to analysis its influences on cell proliferation by Green fluorescent protein (GFP)-based cellomics and colony formation assay, cell invasion and migration by transwell assay, apoptosis and cell cycle by flow cytometry detection, respectively. Nude mice and fluorescence imaging were used to characterize the regulation of tumor growth in vivo. Affymetrix mRNA microarray assay combined KEGG enrichment analysis were used to screen relevant molecules related to MASTL silencing. Finally，several aberrantly expressed genes were validated by quantitative reverse transcription PCR(RT-qPCR)and western blot detection. Results: Silencing MASTL significantly inhibited cell proliferation, migration and invasion, arrested cell cycle at G1 stage. Silencing MASTL reduced tumor growth in nude mice, and fluorescence imaging indicated that the total radiant efficiency of mice in the Lv-shMAST group was markedly reduced compared with in mice in the Lv-shCtrl group in vivo. Affymetrix mRNA microarray assay revealed that 124 genes upregulated, 167 genes downregulated. RT-qPCR and western blotting validation showed that cyclin dependent kinase 6(CDK6), bone morphogenetic protein 2(BMP2), snail family transcriptional repressor 2(SNAI2), phosphorylation-mechanistic target of rapamycin kinase (p-mTOR), phosphorylation-AKT serine/threonine kinase (p-AKT) and phosphorylation-p38 kinase (p-p38) are downregulated, and cyclin dependent kinase inhibitor 1A (CDKN1A) is upregulated. Conclusions: Silencing MASTL could significantly inhibit cell growth, migration ability, induce apoptosis, arrest cell cycle at G1 stage, and the mechanisms of which were mediated via inactivation of mTOR, AKT, p38 signal pathways.

Thus, nding novel biomarkers for early detection and prediction of treatment responses, as well as prognosis and identi cation of novel targets for effective control of gastric cancer, could help medical oncologists reduce the gastric cancer burden clinically.
To this end, microtubule-associated serine/threonine kinase-like enzyme (MASTL, also known as Greatwall, or Gwl) is a serine/threonine kinase that is active during mitotic division that regulates chromosome condensation and mitotic progression through the phosphorylation of cyclinB-Cdk1 [8].
Previous studies have demonstrated that aberrant MASTL expression occurs in various human cancers, including breast [9,10], gastric [11], and colorectal [12], and that MSATL overexpression is associated with poor patient survival [13], and resistance to chemotherapy [10,11] and tumor recurrence [13]. In contrast, knockdown of MASTL expression using RNAi induced tumor cell cycle arrests and inhibits tumor cell invasion and metastasis in colon cancer [11]. However, to date, there is still no in vitro study of MASTL shRNA in gastric cancer and therefore, in this study, We assessed the effects of MASTL knockdown on regulation of gastric cancer cell viability, proliferation, apoptosis, and gene expressions in vitro. We provide novel insights regarding MASTL in the development and progression of gastric cancer, and whether the targeting of MASTL expression using MASTL shRNA is useful as a potential therapeutic or detection strategy of MASTL expression as a novel potential biomarker for gastric cancer patients.

Cell lines and cell preparation
Gastric cancer cell lines MGC-803 were provided by the Type Culture Collection of Cancer Institute and Hospital, Chinese Academy of Medical Sciences (CAMS) (Beijing, China). MGC-803 cells were maintained in DMEM supplemented with 100 IU/mL Penicillin and 100 μg/mL Streptomycin, and 10% heatinactivated fetal bovine serum in a humidi ed cell incubator having an atmosphere of 5% CO 2 at 37˚C. All cells reaching to exponential growing stage were used for further experiments.

Lentiviral infection of MGC-803 cells
Human gastric cancer MGC-803 cells were plated on six-well plates at 5×10 4 cells/well and incubated at 37℃ at 5×10 4 cells/well and incubated at 37℃ in 50 mL/L CO 2 until 30% con uence was reached. The study was designed as both groups: negative control (shCtrl, transfected with GFP lentivirus) and shMASTL group (shMASTL, transfected with shMASTL GFP lentivirus). An appropriate amount of lentivirus was added according to the multiplicity of infection (MOI). The cells were repeatedly cultured in normal culture medium after 12 h. GFP-tagged gene expression was observed under a uorescence microscope at 3 d after transfection, and cells with a transfection e ciency > 80% using the Image J software (National Institute of Heath, Bethesda, MD, USA) were selected for subsequent analyses. Cells were harvested at 48h after post-transfection for further analysis.

Silencing effect detection by qPCR and western blot analysis
To detect silencing e ciency of MASTL knockdown in MGC-803 cells, RT-qPCR and western blot analysis were utilized. Brie y, Total RNA was extracted from tissues or cultured cells using RNAiso Plus reagent (Takara Bio, Dalian, China) and used to synthesize rst-strand cDNA with the Prime ScriptTM RT Reagent Kit (Takara Bio, Dalian, China). For mRNA analysis, RT-qPCR was performed using a SYBR Master Mixture (Yeasen, Shanghai, China) in Bio-Rad CFX96 Real-time PCR system. and the following protocol: denaturation at 95°C for 10 min, followed by 40 cycles of 95°C for 10 sec, 60°C for 30 sec, in which uorescence was acquired). GAPDH levels were measured as an internal control. The primer sequences (Takara Bio, Dalian, China) used were listed in Table1. At the end of all PCR cycles, melting curve were performed to analysis the speci city of PCR product. Each sample was run in triplicates for analysis. The level of MASTL mRNA was normalized to GAPDH mRNA using the 2-ΔΔCt method. The experiment was run in triplicates and repeated at least once.
Gastric cancer of shMASTL and shCtrl group were collected and washed twice with PBS and lysed in icecold lysis buffer for 5 min (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1% Triton X-100, 1 mM ethylene diamine tetraacetic acid, 1 mM PMSF, and 1% sodium deoxycholate). Total protein was isolated from the cell lysis buffer. Protein concentration was measured using a Bio-Rad protein assay kit (Bio-Rad Laboratories, Shanghai, China). Protein samples were separated by 10% SDS-PAGE and transferred to PVDF membranes. Blots were incubated with the appropriate primary antibody (Mouse Anti-Flag, Sigma, 1:2000, China headquarters, Shanghai, China) at room temperature. After washing in 5% non-fat milk in TBST (composed of Tris-Hcl, NaCl and tween20) saline at room temperature for 1 h, blots were incubated with the corresponding horseradish peroxidase (HRP)-conjugated secondary antibody (Goat Anti-Mouse IgG, 1:2000, Santa-Cruz Biotechnology, Dallas, Texas, USA) for 1.5 h. Bands were evaluated using chemiluminescence (ECL, Thermo Scienti c Pierce, Shanghai, China) and scanned images were quanti ed using ImageJ software (NIH Image for the Macintosh, USA). Western blotting experiments were performed in triplicate, with GAPDH (Mouse Anti-Flag, 1:2000, Santa-Cruz Biotechnology, Dallas, Texas, USA) used as a housekeeping control for normalization. The ratio of MASTL protein expression to GAPDH expression was used for semi-quanti cation and comparison between two groups.

Tumorigenesis in nude mice and in vivo imaging
The male BALB/c nude mice (age: 4 weeks, weight 15-19g, maintenance conditions: temperature: 22-24℃, humidity: 40-70%, food/water access: arti cial feeding per 2-3 days, light/dark cycle: 12h/12h) were purchased from Shanghai Lingchang Biotechnology Co., Ltd. (Shanghai, China). The eligible nude mice were inoculated with MGC-803 cells infected by LV-shMASTL and Lv-shCtrl. To be brief, a total number of 20 mice were classi ed into two groups with 10 mice in each group randomly. The inoculated MGC-803 cells were resuspended in physiological saline solution at a density of 5x10 7 cells/ml and a volume with 0.2 mL cell suspension was injected into the mice subcutaneously with a 6-gauge, 1 mL syringe. The mice were fed and survived until the tumors were visible, and tumor diameter and size were measured 8, 12, 15, 18, 21 and 24 days following inoculation. Tumor volume was monitored, recorded and calculated according to the following formula for hemi-ellipsoids routinely: Volume = length (cm) x width (cm) x height (cm) x 3.14/6. After the last measurement of tumor volume, the mice were injected with D-Luciferin 10 μL/g intraperitoneally (QianChen Biotechnology Co., Ltd, Shanghai, China). 15 min later, pentobarbital sodium with 70 mg/kg was injected intraperitoneally to anesthetize the mice for detecting the uorescence emitted from mice by a small animal Screening aberrantly genes by Affymetrix mRNA microarray assay Following successful transduction with Lv-shMASTL and Lv-shCtrl, MGC-803 cells of both groups were collected to perform mRNA microarray assay. Total RNA was isolated using TRIZOL reagent (Invitrogen, Carlsbad, CA, USA) in compliance with the manufacturer's instructions. RNA integrity was determined by gel electrophoresis, and the concentration and purity of total RNA were assessed by A 260 and A 280 , using a nanodrop spectrophotometer (ND-1000, Thermo Fisher Scienti c, Wilmington, DE, USA). Only the samples containing RNA with A260/A280 ratios >1.8 were used in this study. The mRNA microarray assay was performed using commercially available PrimeView™ Human Gene Expression Array (Genechem Corporation, Shanghai, China). The procedures of labeling, hybridization, washing, and scanning were performed according to the standard operating procedures provided by Affymetrix. Brie y, total RNA was used to synthesize cDNA in an in vitro transcription reaction, and the cDNA was labeled by biotin and T7 Enzyme Mix. After hybridized with gene expression array, non-speci cally bound molecules were removed from the microarray with two washing buffers. Subsequently, the arrays were scanned with a GeneChip® Scanner 3000 (Affymetrix, Cleveland, USA) to gather hybridization data, and the data were analyzed using Gene Matrix cloud service (http://gcloud.taogene.com).
RT-qPCR and Western blotting validation of aberrantly expressed genes screened by microarray Total RNA and protein was extracted from MGC-803 cells of shMASTL and shCtrl group. All procedures of RNA extraction, reverse transcription and ampli cation were same as the procedures mentioned above.
The primer sequences (Takara Bio, Dalian, China) were provided in Table 1. Likewise, all procedures of protein isolation, gel electrophoresis, transferring to PVDF and immune blots were same as the procedures mentioned above too. The blots were incubated with the appropriate primary antibody against CDK6 (1:1000

Statistical analysis
Data were expressed as the means ± standard deviation from at least 3 separate experiments. Statistical analyses were evaluated using Student's two-tailed t-test. Differences with P values of < 0.05 are considered statistically signi cant.

Results
E ciency of shRNA-mediated MASTL knockdown in gastric cancer cells.
MGC-803 cells infected with Lv-shCtrl or Lv-shMASTL were observed under a uorescence microscope to determine infection e ciency, and the infection rate was measured by monitoring GFP uorescence emitted by cells. The results indicated that the evaluated infection e ciency was >80% (Fig. 1A). As determined by RT-qPCR analysis, the mRNA and protein expression of MASTL were signi cantly decreased in the Lv-shMASTL group, with a knockdown e ciency of 78.9% (P<0.01; Fig.1B) at the level of MASTL mRNA and protein level (P< 0.01; Fig.1B) compared with that of the shCtrl group.
Effects of MASTL knockdown on reduction of MGC-803 cell proliferation by Cellomics Array Scan assay.
To assess the effects of MASTL on gastric cancer cells in vitro, we knocked down MASTL expression using a lentivirus carrying MASTL shRNA and Lv-shCtrl in MGC-803 cells. We then assessed the alteration in cell viability using the GFP-based Cellomics Array Scan imaging assay and found that knockdown of MASTL expression make the number and fold-change in proliferation of cells in the Lv-shMASTL group reduced when compared with the Lv-shCtrl group at 4 and 5 days following transduction of MGC-803 cells signi cantly (P<0.05; Fig.1C). The GFP-based Cellomics Array Scan imaging assay showed that knockdown of MASTL expression dramatically reduced MGC-803 cell proliferation. These ndings suggested that knockdown of MASTL may be associated with a reduction in cell proliferation.
MASTL silencing reduces gastric cancer cell colony formation.
A colony formation assay is used to assess the proliferative potential of cells. In the present study, the colony formation assay results demonstrated that the Lv-shMASTL group formed signi cantly fewer colonies in soft agar when compared with the Lv-shCtrl group (P<0.001; Fig. 1D). These results indicated that silencing MASTL may reduce the anchorage-independent proliferative potential of MGC-803 gastric cancer cells.

Effects of MASTL knockdown on induction of MGC-803 cell apoptosis
Knockdown of MASTL expression also signi cantly inhibited MGC-803 cell colony formation (P=0.0001, Figure.  Affymetrix mRNA microarray analysis of aberrantly expressed genes. To investigate the modulation mechanism of MASTL in the tumorigenesis of gastric cancer, Affymetrix mRNA microarray were used to screen and analysis aberrantly expressed genes between Lv-shCtrl group and Lv-shMASTL group in MGC-803 cells. Knockdown of MASTL signi cantly induced 391 mRNAs were aberrantly expressed including 124 up-regulated genes, and 267 down-regulated genes between Lv-shMASTL group cells and Lv-shCtrl group cells (cut-off ≥1.5, Figure 5A).

Pathway analysis of aberrantly expressed genes with bioinformatical method.
In order to explore the potential mechanisms of aberrantly expressed genes related to tumorigenesis characterics of gastric cancer, the KEGG pathway analysis were conducted as following steps, a total of 391 aberrantly expressed genes were submitted to DAVID 6.8 online software (https://david.ncifcrf.gov/tools.jsp) to analysis relevant pathways. The results showed that 291 aberrantly expressed genes were enriched in 14 pathways, among which, 9 pathways were closely related to tumorigenesis were closely related to tumorigenesis including Small cell lung cancer, p53 signaling pathway, MAPK signaling pathway, Transcriptional misregulation in cancer, Ras signaling pathway, Pathways in cancer, Viral carcinogenesis, Focal adhesion, PI3K-Akt signaling pathway. In the end, several genes including CDK6, BMP2, SNAI2 and CDKN1A in the pathway of hsa05200 were selected for RT-qPCR and western blotting validation, (Figure 5(B, C)). In summary, this kind of bioinformatical analysis facility subsequent molecular mechanisms to discover the effects of silencing MASTL in gastric cancer cells.
Any aberrantly expressed genes validation by RT-qPCR and Western Blotting.
In order to explore the molecular mechanisms of phenotype changes related to silencing MASTL, it is a wise choice to validate any aberrantly expressed genes screened by the microarray assay furtherly. Therefore, combing with bioinformatical analysis, we decided to choose several genes enriched in Pathways in cancer (Hsa05200) including CDK6, BMP2, SNAI2 and CDKN1A to validate their expression by RT-qPCR and western blotting method. The validation results discovered that CDK6, BMP2 and SNAI2 are downregulated, and CDKN1A are upregulated in Lv-shMASTL group cells ( Figure. 6(A, B, C)). In addition, Western blotting analysis showed that phosphorylation form of p-mTOR, p-AKT and p-p38 proteins were downregulated signi cantly ( Figure. 6(A, C)). This results showed that CDK6, BMP2, SNAI2, CDKN1A, p-mTOR, p-AKT and p-p38 which play important roles in the cell growth, migration and arresting cell cycle of MGC-803 cells following MASTL silencing which responsible for the mechanisms of MASTL in gastric cancer progression.

Discussion
MASTL played as a serine/threonine kinase during mitotic division actively, in this process, it has been founded that MASTL could regulates chromosome condensation and mitotic progression [8]. Emerged studies showed that MASTL participate multiple tumors progression [7,[9][10][11][12], however, fewer study was reported about the roe of MASTL in gastric cancer. Therefore, it is extremely needed to uncover the role and function of MASTL during gastric cancer tumorigenesis. In this study, Knocking down of MASTL by shRNA lentivirus infection in MGC-803 cells signi cantly suppressed cell proliferation, colony formation, inhibited cell migration ability, arrested cell cycle at G1 stage, and induced apoptosis signi cantly in gastric cancer cells. Tumorigenesis in nude mice and orescence imaging in vivo also con rmed that MASTL silencing could inhibit tumor growth in vivo.
In order to explore the potential mechanisms associated with silencing MASTL, Affymetrix mRNA microarray assay and subsequent enriched KEGG pathway analysis were used to screen and discover relevant molecules related to MASTL silencing. In this study, a total of 291 aberrantly expressed genes screened by mRNA microarray analysis make it possible that bioinformatical analysis will lead to deepening understand relevant mechanisms from those moleculars in a larger scale by high throughput screening. Bioinformatical analysis of KEGG pathway enrichment analysis showed that some of aberrantly expressed genes enriched in 14 pathways of were involved in gastric cancer tumorgenesis, among which, 9 pathways were closely related to tumorigenesis, which includes small cell lung cancer, p53 signaling pathway, MAPK signaling pathway, transcriptional misregulation in cancer, ras signaling pathway, pathways in cancer, viral carcinogenesis, focal adhesion, PI3K-Akt signaling pathway. Therefore, it is an apparent sign that MASTL could participate in gastric cancer tumorigenesis which responsible for tumor biological phenotype, such as cell proliferation, cell cycle, cell migration, and also a good sign of gastric cancer biomarker for diagnosis and prognosis evaluation. In another hand, aberrantly expressed genes screened by mRNA microarray assay also provided signi cantly changed genes from the trivial perspective.
After Affymetrix mRNA microarray assay, any of those molecules related to cell growth, migration and cell cycle should be chosen from 291 aberrantly expressed mRNAs for more detection at mRNA and protein levels. In the end, 4 genes, CDK6, BMP2, SNAI2 and CDKN1A enriched in pathways in cancer were veri ed by the means of RT-qPCR and western blotting. In addition, p-mTOR, p-AKT and p-P38 were also detected by western blotting, which would make it easy to understand phenotype changes result from silencing MASTL by searching the potential upstream moleculars. The validation results showed that CDK6, BMP2, SNAI2, p-mTOR, p-AKT and p-P38 are downregulated, and CDKN1A are upregulated. CDK6, BMP2, SNAI2, and CDKN1A were classical phenotype molecular which responsible for tumor growth, cell cycle and metastasis, and overexpression of them could promote tumor invasiveness and metastasis.
CDK6, as serine/threonine protein kinases controlled cell cycle G1 phase progression and G1/S transition.
The expression of CDK6 protein was found upregulated in gastric cancer [14]. BMP-2 signaling pathway enhances tumor metastasis in gastric cancer by sequential activation of the PI3K/AKT or MAPK pathway followed by the induction of NF-κB and MMP-9 activity, indicating that BMP-2 has the potential to be a therapeutic molecular target to decrease metastasis [15]. BMP2 can induce AKT and ERK phosphorylation in a dose-dependent, Notably BMP2 alone treatment can induce the up-regulation of vimentin, snail, and N-cadherin in AGS cells, the down-regulation of E-cadherin also occurred. On the contrary, BMPR-II siRNA signi cantly prohibited BMP2-induced AKT and ERK phosphorylation [16]. BMP-2 may inhibit the proliferation of both normal and malignant gastric epithelial cells, down-regulate CDK4 expression in gastric cancer cells and arrest gastric cancer cells in G1-phase in cell cycle [17]. ShRNA-mediated silencing of SNAI2 suppressed the activation of Snail/Slug, whereby gastric cancer cell proliferation, invasion and migration, EMT, tumor growth, and lymph node metastasis were inhibited. High expression of miR-33a was a protective factor in uencing the prognosis of gastric cancer. This study suggests that miR-33a inhibited EMT, invasion, and metastasis of gastric cancer through the Snail/Slug signaling pathway by modulating SNAI2 expression. MiR-33a targets and inhibits the expression of SNAI2, overexpression of SNAI2 activates the Snail/Slug signaling pathway, the Snail/Slug signaling pathway promotes gastric cancer cell proliferation, invasion, and metastasis [17]. P21WAF1/CIP1, a member of the cyclin dependent kinase inhibitor (CKIs) family, is a well-known cell cycle regulator. It can inhibit the activity of cyclin/cyclin dependent kinase complex. It can combine with almost every cyclin CDK complex and widely inhibit all kinds of cyclin CDK complexes. P21WAF1/CIP1 can go through many ways Participate in the regulation of cell cycle and induce cell apoptosis [13]. When the cells were exposed to DNA damage agent, p21WAF1/CIP1 gene expression was further induced by activating p53 to inhibit the activity of cyclin dependent kinase complex, so that the damaged cells were blocked in G1 phase and the cells had enough time for damage repair. Low expression of CDKN1A has independent prognostic signi cance indicative of tumor progression and poor survival in patients with RGA. Evaluation of CDKN1A expression may assist in determining prognosis in patients with resected gastric adenocarcinoma.
A numbers of previous reports demonstrated that phosphorylation protein form of p-AKT [18][19][20][21][22], p-mTOR [23][24][25][26], p-p38 [27][28][29][30] in mTOR, AKT and p38 signal pathways were involved in the regulation of cell proliferation, cell cycle and cell migration in gastric cancer. In this study, western blotting validation con rmed that phosphorylation protein form of p-mTOR, p-AKT and p-P38 were downregulated signi cantly caused by MASTL silencing in gastric cancer cells. Collectively, Knockdown of MASTL in MGC-803 could inhibit cell proliferation, cell migration ability, and arrest cell cycle at G2 stage signi cantly. The phenotype change of silencing MASTL were associated with downregulation of CDK6, BMP2 and SNAI2, and upregulation of CDKN1A. Furthermore, knockdown of MASTL could lead to inactivation of phosphorylation form of p-mTOR, p-AKT and p-p38.

Conclusion
Silencing MASTL could signi cantly inhibit cell growth, migration ability, induce apoptosis and arrest cell cycle at G1 stage in MGC-803 cells, and the mechanisms of which were mediated by downregulation of CDK6, BMP2, SNAI2, p-mTOR, p-AKT and p-p38, and upregulation of CDKN1A. Therefore, it is suggested MASTL take part in proliferation, migration and cell cycle via inactivation of mTOR, AKT, p38 signal pathways. In conclusion, it is suggested MASTL take part in proliferation, migration and invasion, and which may be considered as a valuable target for gene therapeutic strategies in gastric cancer. Additionally, lots of pathways and aberrantly expressed mRNAs were still needed to study further to understand the role and function of MASTL in gastric cancer tumorigenesis. All of these studies would contribute to prove MASTL as a novel biomarker molecular in diagnosis and gene therapy target.

Declarations
Ethics approval and consent to participate     Fig 3(A, B)). These results indicate that MASTL silencing could inhibit gastric cancer cell invasion and migration in vitro. Compared with Lv-shCtrl group, the uorescence expression of Lv-shMASTL group was decreased (P < 0.05, Fig 4A). After 6 weeks, uorescence imaging assay showed that there were signi cantly more the total radiant e ciency of region of interest (ROI) in xenografts of mice group developed from Lv-shMASTL-infected MGC-803 cells than that in the corresponding Lv-shCtrl-infected group (P < 0.05, , Fig   4B).  Fig 5A). KEGG pathway analysis showed that 291 aberrantly expressed genes were enriched in 14 pathways, among which, 9 pathways were closely related to tumorigenesis were closely related to tumorigenesis. 9 pathways were closely related to tumorigenesis were closely related to tumorigenesis including Small cell lung cancer, p53 signaling pathway, MAPK signaling pathway, Transcriptional misregulation in cancer, Ras signaling pathway, Pathways in cancer, Viral carcinogenesis, Focal adhesion, PI3K-Akt signaling pathway. In summary, this kind of bioinformatical analysis facility subsequent molecular mechanisms to discover the effects of silencing MASTL in gastric cancer cells (P < 0.05, Fig 5(B.C)). Effects of MASTL gene knockdown on several aberrantly expressed genes validation by RT-qPCR and western blotting. RT-qPCR and western blotting validation discovered that CDK6, BMP2 and SNAI2 are downregulated, and CDKN1A are upregulated ( Figure. 6(A, B, C)). In addition, Western blotting analysis showed that phosphorylation form of p-mTOR, p-AKT and p-P38 proteins were downregulated signi cantly, ( Figure. 6(A, B)). suggesting that CDK6, BMP2, SNAI2, CDKN1A, p-mTOR, p-AKT and p-P38 which play important roles in the cell growth, migration and arresting cell cycle of MGC-803 cells