Myeloid ecotropic viral integration site 1 inhibits cell proliferation, invasion or migration in human gastric cancer

Myeloid ecotropic viral integration site 1 (MEIS1) has been identified to be a potential tumor suppressor in some cancers. However, the mechanisms underlying MEIS1-induced cancer development and progression were not clear. Here, we investigated the expression and role of MEIS1 in gastric cancer. In vivo, we analyzed tumor growth using nude mice model. In the present study, MEIS1 expression was obviously decreased in GC cell lines compared with that in normal gastric cell lines (all p<0.001). MEIS1 overexpression inhibited cell proliferation and G1/S transition accompanied by decreased Cyclin D1 and Cyclin A expression. Furthermore, MEIS1 overexpression decreased the expression of Survivin, and induced cell apoptosis (p<0.001). Transwell migration assay revealed that MEIS1 affects cell invasion and migration, and inhibited epithelial-mesenchymal transition (EMT). Finally, MEIS1 inhibits MKN28 cell growth in nude mice model. In conclusion, our study suggested that MEIS1 plays an important role in regulating cell survival, proliferation, anchorageindependent growth, cell cycle, apoptosis and metastasis. Thus, MEIS1 might be recommended as an effective target for GC patients.


INTRODUCTION
Gastric cancer (GC) is one of the most common solid malignant tumors [1]. Because locally advanced GC has high a frequency of micrometastasis and relapse, its prognosis is poor. The 5 years' survival after radical surgery was 30%~50%. Perioperative and adjuvant chemotherapy have been shown to be able to decrease distant metastasis and improve survival by 10-15% [2][3][4][5]. Chemoresistance likely has contributed to the relapse after chemotherapy. It is crucial to understand the underlying mechanism of chemoresistance and find novel therapeutic targets to further improve the outcome of GC.
MEIS1 is a kind of TALE (Triple amino acid loop extension) transcription factors, which has a homeodomain involved in cell proliferation, differentiation, and vertebrate embryogenesis [6][7][8][9]. MEIS1 also plays an important role in the development of stem cell regulation [10][11][12][13]. Some reports showed that MEIS1 can be identified as a negative regulator in some cancers, such as non-small-cell lung cancer, prostate cancer, and esophageal squamous cell carcinoma [14][15][16]. MEIS1 can inhibit cell proliferation and induce cell cycle arrest [14][15][16]. In addition, high MEIS1 expression was detected in normal prostate tissues compared with tumor tissues, suggesting that MEIS1 is a useful biomarker or a novel target of prostate carcinoma [16]. As reported by Rad et al, MEIS1 expression is inversely associated with tumor metastasis and stage of esophageal squamous cell carcinoma [17]. The previous study also demonstrated MEIS1 as a negative regulator of androgen receptor (AR) signaling. So, it is essential to uncover the role of MEIS1 in gastric cancer progression.
In the present study, we overexpressed MEIS1 using adenovirus vectors in GC cells. Western blot and real-time qPCR were conducted to detect the expression of MEIS1 protein and mRNA. Cell proliferation, survival, migration www.impactjournals.com/oncotarget/ Oncotarget, 2017, Vol. 8, (No. 52), pp: 90050-90060 Research Paper and invasion were investigated by MTT, colony formation, soft-agar, transwell assays. And then in vivo tumor growth was explored using nude mice model. Additionally, flow cytometry was applied to analyze cell cycle or cell apoptosis mediated by MEIS1.

Expression of MEIS1 in GC cell lines and tissues
To elucidate the expression of MEIS1 in GC cell lines and tissues, the expression level of MEIS1 in GC cells and normal cells was detected using Western blot. As shown in Figure 1A-1B, high expression of MEIS1 was detected in normal cell line RGM-1 and GES-1. By contrast, the expression of MEIS1 in GC cell line MKN28 and SGC7901 was markedly decreased. Afterwards, we further detected the expression of MEIS1 in 10 cases of GC cancer tissues and their paired non-tumor tissues ( Figure 1C). The expression of MEIS1 was obviously decreased in GC tissues compared with that in paired non-tumor tissues ( Figure 1C). Our results from cell lines and tissues indicated that MEIS1 may be involved in the development and progression of GC.
To investigate the molecular mechanisms of MEIS1 in the progression of GC, we used MEIS1-overexpressing adenoviral vectors, including Ad-control (the empty vector) and Ad-MEIS1 (full length coding sequence of Meis1 gene). Then, MKN28 or SGC7901 cells were transfected with Ad-control or Ad-MEIS1. As expected, we found that the expression of MEIS1 protein was obviously up-regulated due to Ad-MEIS1 transfection ( Figure 1D, 1E).

MEIS1 repressed cell proliferation
To identify whether MEIS1 represses cell proliferation, we carried out MTT assays. MKN28 and SGC7901 cells transfected with Ad-MEIS1 showed much slower cell growth compared with those transfected with Ad-controls (Figure 2A, 2B). In addition, there were no significant differences between parental cells and cells with Ad-controls (Figure 2A, 2B). These results indicated that MEIS1 overexpression inhibits cell growth. Besides, colony formation assay revealed that the number of colonies was obviously decreased in MEIS1overexpressing MKN28 or SGC7901 cells than control cells ( Figure 2C-2F). These data demonstrated that MEIS1 affect GC cell proliferation and colony formation.
Subsequently, because normal human gastric GES-1 highly expressed endogenous MEIS1, we silenced MEIS1 expression in normal human gastric GES-1 cells using MEIS1 siRNA. As shown in Figure 3, MEIS1 siRNA prompts cell proliferation ( Figure 3A-3C), colony number and size ( Figure 3D-3F) of GES-1 cells. These findings identified that MEIS1 negatively regulates cell growth and proliferation. Next, we examined the role of MEIS1 in tumor activity, soft-agar assay revealed that MEIS1 overexpression inhibited anchorage-independent growth of MKN28 and SGC7901 cells ( Figure 4A, 4B).

MEIS1 induces cell apoptosis
Next, we investigated the involvement of MEIS1 in regulation of GC cell apoptosis. As shown in Figure 5, after MEIS1 overexpression, the proportion of apoptosis cells increased from 3.12% to 19.58% (MKN28), and from 4.10% to 21.59 (SGC7901), respectively ( Figure 5A, 5B). And then, we analyzed changes of apoptosis-related proteins. As shown in Figure 6, we found that MEIS1 overexpression in MKN28 cells inhibited the expression of cIAP-1, cIAP-2 and Survivin ( Figure 6A). MEIS1 also promoted the expression of BAX protein. Similar results were observed in SGC7901 cells ( Figure 6B). These findings validated that MEIS1 has an inhibitory role in GC cells proliferation and survival, which depends on its regulation in pro-apoptosis related proteins.

MEIS1 induces G1/S transition arrest
To further characterize the molecular mechanisms underlying MEIS1-regulated cell growth, we detected the impact of MEIS1 on cell cycle. Overexpression of MEIS1 in MKN28 cells reduced cell proportion in S phase (from 25.87% to 13.96%) and G2/M phase (from 15.02% to 4.11%) but increased cell proportion in G0/G1 phase (from 55.16% to 73.93%) ( Figure 7A). Similar data was also observed in SGC7901 cells ( Figure 7B). MEIS1 overexpression increased the proportion of SGC7901 cells in G0/G1 phase (from 54.55% to 82.32%), and decreased the proportion of SGC7901 cells in S phase (from 25.99% to 13.65%) or G2/M cells (from 18.17% to 8.09%) ( Figure  7C, 7D). These findings suggested that MEIS1 induced cell cycle arrest at G1/S transitions. Furthermore, we analyzed whether MEIS1 regulates the expression of Cyclin D1 and Cyclin A. As shown in Figure 8, MEIS1 overexpression down-regulated the expression of Cyclin A and Cyclin D1 in GC cells ( Figure  8A, 8B), suggesting that MEIS1 mediates cell cycle arrest at G1/S transitions via inhibiting the expression of cyclin A and cyclin D1.

MEIS1 affects cell invasion, migration, and EMT
In the present study, we used Transwell migration assay to assess the effects of MEIS1 on cell invasion and migration. Our findings suggested that MEIS1 overexpression inhibited MKN28 cell invasion and migration ( Figure 9A-9B). Mechanically, MEIS1 overexpression increased the expression of E-cadherin and decreased the expression of N-cadherin and Vimentin ( Figure 10A, 10B). These data indicated that MEIS1 inhibits cell invasion and migration by regulating EMT progression.

MEIS1 repressed tumor growth in nude mice model
In the present study, we examined the effect of MEIS1 overexpression on tumor growth using xenograft nude mice model ( Figure 11). We found that cells transfected with MKN28-Ad-MEIS1 could obviously and

DISCUSSION
In the present study, our study identifies that MEIS1 acts as a tumor suppressor in the development of GC. Firstly, the expression of MEIS1 is decreased in GC tissues or cell lines; Secondly, MEIS1 overexpression represses cell proliferation, colony formation and anchorage independent growth of GC cells; Furthermore, MEIS1 overexpression induces G1/S arrest and GC cell apoptosis. Mechanically, MEIS1 overexpression reduced the expression of cIAP-1, cIAP-2 and Survivin, and increased the expression of pro-apoptosis protein BAX to induce cell apoptosis. Moreover, MEIS1 overexpression can inhibit MKN28 cell migration and invasion to decrease EMT progression. These findings above indicate that MEIS1 might play a crucial role in the progression of GC.
It has been reported that cell cycle checkpoints are required for integrity of genomics [18]. Changes of cellcycle-corresponding molecules, such as Cyclin family or CDK family, would be useful and effective biomarkers in many types of human cancers [19,20]. The initiation and malignancies of human cancers were accompanied by the dysfunction of G1/S checkpoint of cell cycle [21,22]. In addition, cyclins and their related CDKs play an important role in cell cycle progression [23,24]. Of all cyclin/CDK complexes, cyclin D1/CDK4/6 and cyclin A/CDK1 are demonstrated to be implicated in the G1/S transition. In this work, we demonstrated that MEIS1 overexpression induces cell cycle arrest at G1/S transition, accompanied by decreased expression of cyclin D1 and cyclin A proteins. Previous reports revealed that MEIS1 dominates cells proliferation via regulating the expression of cyclinD1 and c-MYC proteins [25][26][27]. These findings suggested that MEIS1 might act as a cell cycle regulator to affect cell proliferation by controlling cell cycle. CDK inhibitors, such as P15, P21, and P27 have been reported to inhibit the activity of cyclin/CDK complexes [28]. Thus, deregulation of these inhibitors was closely correlated with poor prognosis of human cancers, and we assumed that MEIS1 may be involved in regulating the expression of CDK inhibitors.
Recent reports showed that MEIS1 might be involved in the progression of tumors, and serves as a tumor suppressor in some kinds of human cancers, involving leukemogenesis, lung cancer and prostate cancer [29]. MEIS1 induces cell differentiation and represses cell proliferation of normal epithelial cells of human prostates [30]. It should be noted that reduced MEIS1 expression predicted a worse overall survival, while increased MEIS1 expression means a better overall survival in prostate cancer. In our study, we found that MEIS1 overexpression could suppressed cell invasion and migration of high aggressive GC cell line MKN28 and decreased EMT processes, and affected the progression of in-vivo tumor growth. Because malignant cancers were characterized by cell invasion and migration, our results demonstrated that MEIS1 could be a potential target to improve the prognosis of GC patients.
In conclusion, MEIS1 expression is decreased in GC cells lines and tissues. MEIS1 overexpression obviously inhibits GC cell proliferation and apoptosis. Thus, MEIS1 serves as a tumor suppressor in the progression of gastric cancers. Our study will provide a novel way to treat gastric cancer patients.

Ethics statement
Ethical approval was obtained from the Institutional Ethics Committee at the Ethics Committee of Shandong Provincial Third Hospital. Patients or Animals enrolled in this study signed written informed consent. All procedures were subjected to the Declaration of Helsinki.

Plasmids and adenovirus vector
The MEIS1 expression vector or its small interfering RNA (siRNA) was used in this study as previously mentioned [16]. Adenovirus vectors of MEIS1 were applied according to the methods as described by Niu et al. [21]. In brief, full length cDNA of Meis1 gene was cloned into pShuttle-CMV vector. And then, pAdEasy-1 vector and pShuttle-Meis1 vector was co-transformed into cells to generate the recombinant adenovirus vector pAdcontrol or pAd-MEIS1. As for packaging processes, pAdcontrol or pAd-MEIS1 was transfected into cells and then purified with a cesium chloride gradient. All vectors were confirmed by Sanger sequencing.

Cell culture and reagents
Human GC cell lines SGC7901 or MKN28 (a high metastatic cell line), and non-tumor cell lines HEK293 or GES-1 were as previously described [18]. SGC7901, MKN28 and GES-1 cells were cultured in complete DMEM (Invitrogen, Carlsbad, CA, USA), and HEK293 was cultured in RPMI-1640 medium (Invitrogen, Carlsbad, CA, USA) in a sterile incubator maintained at 37 °C with 5% CO 2 .

Cell growth and colony formation assays
For measuring proliferation, cells were infected with Ad-control or Ad-MEIS1, and then cells were seeded in 96-well plates (Corning, NY, USA), incubated for 1, 2, 3 and 4 days, and the cells were analyzed for MTT assays [22]. GES-1 cells were transfected with siRNA of MEIS1 and then harvested for MTT analysis.
For colony formation, infected cells were seeded in 6-well plates at 500 cells per well [23]. Two to four weeks later, colonies were fixed with 4% paraformaldehyde and stained with 0.5% (W/W) crystal violet (diluted in phosphate buffer saline, PBS) for 30 min. Next, cells were harvested and measured by a multifunctional micro-plate reader at 546 nm. The relative colony number (relative survival cell number) = O.D. 546 administration group / O.D. 546 control group. GES-1 cells, which were transfected with siRNA of MEIS1, were also measured by colony formation assays.

Cell cycle analysis
Cell cycle was carried out by flow-cytometry following the instructions as previously described by Chen et al [24]. Cells were infected with Ad-control or Ad-MEIS1, were fixed in 70% ethanol for 18-24 h. Next, cells were washed with pre-cold PBS for three times and incubated with RNase A (0.2 mg/mL) diluted in pre-cold PBS. Then, PI (propidium Iodide) was added. Samples were analyzed by FACScalibur Flow Cytometer (Becton Dickinson, Bioscience, USA).

Cell death analysis
Cells were infected with Ad-control or Ad-MEIS1, and then cells were harvested and labelled with PI and FITC-Annexin V according to the manufacturer's instructions (Becton Dickinson, Biosciences, USA). A minimum of 2000 events for each sample were collected and analyzed using a FACScalibur Flow Cytometer (Becton Dickinson, Biosciences, USA).

Anchorage-independent growth
Cells were infected with Ad-control or Ad-MEIS1, and then cells were plated into 6-well plates (1000 cells per well) (Corning, Corning, NY, USA), with a bottom layer of 1.0% low melting temperature agar in DMEM and a top layer of 0.3% agar in DMEM. Colony number was manifested as the mean ± SD of three independent experiments scored after 3-4 weeks of growth [26].

Transwell migration assays
Cells were infected with Ad-control or Ad-MEIS1, and then cells were analyzed by transwell migration assays performed in 24-well plates chamber (Corning, NY, USA) by fitted with a polyethylene terephthalate filter membrane with 8 μm pores. For invasion, the top chambers were coated with 30 μl ECM (Extracellular matrix) gel (Sigma, USA) diluted with serum free RPMI-1640 in 1:5 dilutions for more than 4 h at 37 °C. Then, chambers were filled with 0.2 ml of cells (5 × 10 5 cells per ml) in serum-free medium, and the bottom chambers were filled with 0.25 ml of medium with 10% FBS. The cells were incubated in the trans-wells at 37 °C in 5% CO 2 for 12 h. For migration, the cells were incubated in chambers without ECM coated. After 10-15 h (for invasion) or 4-6 h (for migration), chamber membrane was fixed with 4% paraformaldehyde and stained with crystal violet. The relative invading cells were measured according to the instructions provided [27,28]. Values were presented as the mean ± SD of triplicated experiments.

In vivo analysis
In vivo nude mice model was created via xenotransplantation of MKN28 cells into 4-5 week-old nude mice (6 animals per group) [29]. Cells were infected with Ad-control or AD-MEIS1, were injected with 5 × 10 5 cells re-suspended in PBS or normal/physiological saline. Tumor volumes were measured every week by measuring length and width. Volumes of tumor were calculated: (width 2 × length) / 2 [29]. Animal Experiment Committee approved all protocols for treating animals, and the in vivo studies were carried out in accordance with the U.K. Animals (Scientific Procedures) Act, 1986 and associated guidelines.

Statistical analysis
The relative expression level was calculated: (indicated group protein expression level/loading control expression level)/ (control group protein expression level/ loading control expression level). All statistical significance analyses were performed using SPSS statistical software. P-value of <0.05 was considered statistically significant. Statistical significance in cell growth assays was analyzed by Bonferroni correction with or without two-way ANOVA.