PLK1 as a potential prognostic marker of gastric cancer through MEK-ERK pathway on PDTX models.

BACKGROUND: PLK1 has been identified as having a great effect on cell division and maintaining genomic stability in mitosis, spindle assembly, and DNA damage response by current studies.
MATERIALS AND METHODS: We assessed PLK1 expression in cervical cancer tissues and cells. We have also evaluated the effects of PLK1 on gastric cancer cell proliferation, migration, and apoptosis both in vitro and in vivo.
RESULTS: Our results show that PLK1 is overexpressed in gastric cancer tissues and cells. Inhibition of PLK1 contributes cell cycle G2-phase arrest and inhibits the proliferation, migration, and apoptosis of gastric cancer (GC) cells, whereas its overexpression promotes proliferation, migration, and apoptosis in these cells. Moreover, PLK1 inhibition reduces expression of pMEK and pERK. More importantly, in vivo by analyzing tumorigenesis in patient-derived tumor xenograft (PDTX) models, the inhibition of PLK1 activity by BI6727 significantly decreased the volume and weight of the tumors compared with control group (P<0.01).
CONCLUSION: Our results found that PLK1 has a significant impact on the survival of GC cells; it may become a prognostic judge, a potential therapeutic target, and a preventative biomarker of GC.

Introduction

Gastric cancer (GC) is one of the most common malignant tumors of the digestive system. Due to the complexity of its molecular mechanism and the clinical heterogeneity, gastric cancer death rate ranks third and its incidence worldwide has an increasing trend.1–4 Despite the current multimodal therapy for the treatment of gastric cancer, including surgery, radiotherapy and chemotherapy, and targeted therapies, the 5-year survival rate of gastric cancer patients is still low (25%–30%) and the prognosis is poor.5–7 Understanding the underlying molecular mechanisms of gastric cancer has important clinical implications for the treatment and prognosis of gastric cancer.8,9

The Polo-like kinase 1 (PLK1) is considered as a key mitotic regulator belonging to the human PLK family which is considered a group of highly conserved serine/threonine kinases that are often associated with mitosis and cell cycle progression10–13 and plays multiple roles in the cell cycle: regulates mitotic entry and the G2/M checkpoint, coordinates the centrosome and cell cycles, regulates spindle assembly and chromosome segregation, plays an important role in the spindle midzone and during abscission, facilitates DNA replication, and participates in cytokinesis and meiosis.14 PLK1 is essential for precisely regulating the cell division and maintaining genome stability in mitosis, spindle assembly, and DNA damage response.15,16 Previous studies have shown that overexpressed PLK1 is associated with poor survival rate in a variety of cancers including breast, melanoma, and non-small cell lung, suggesting PLK1 may be a promising therapeutic target for cancer management.17–19 However, the precise mechanism of PLK1 in gastric cancer has not been illuminated.

To address this, patient-derived tumor xenograft (PDTX) models were employed in this study. PDTXs are on the basis of the shift of primary tumors from the patient directly into a mouse with immune deficiency.20 Due to direct acquirement from human tumors, PDTX mice can inherit the human tumor microenvironment and keep interactions between tumor cells and the innate immune system. The two parts are closely associated with the development, proliferation, and metastasis of tumors.21 Therefore, they will benefit from the studies of tumor metastasis and genetic evolution. Furthermore, PDTX mice hold potential for development of anticancer therapies and personalized medicine for cancer patients.22 Here, we illustrate the regulatory mechanism of PLK1 in gastric cancer metastasis by regulating MEK-ERK pathway, which may provide evidence for potential combined therapy for gastric cancer.

Materials and methods

Tissue samples of gastric cancer

Fresh surgically resected cancer tissue samples and adjacent noncancerous tissues of gastric cancer were collected from 20 patients treated at the Department of General Surgery, the Affiliated Hospital of Jiangsu University, between January 2015 and September 2016. All patients recruited in this study were without any preoperative chemotherapy. After approval by the Ethics Committee of the Affiliated Hospital of Jiangsu University, these tissues were collected and used for the following studies. The patient consents of all patients participated in this study were written informed consents.

Immunohistochemistry (IHC) analysis

The gastric cancer tissues in different stages were analyzed by IHC as the previous description.23 In the evaluation of staining intensity, the intensity of PLK1 protein staining for each slide was scored and assessed according to the criterion: 0 and 1+ (negative, low staining), 2+ and 3+ (intense staining). PLK1 IHC data were analyzed using ImagePro Plus software (National Institutes of Health, Bethesda, MD, USA).

Quantitative real-time polymerase chain reaction (qRT-PCR) analysis

The total RNA was extracted from cancer tissue samples and adjacent noncancerous tissues; the following qRT-PCR was conducted on the basis of previous procedures. The level of PLK1 mRNA in the tissue samples and cells was determined by comparative Ct method including control group, BI6727 group, and BI6727+ PLK1 group. In this experiment, β-actin was the internal control. All results were conducted in triplicate. The primer sequences used in the present study were as follows: PLK1–forward 5′-GCAGCGTGCAGATCAACTTC-3′, reverse 5′-AGGAGACTCAGGCGGTATGT-3′; β-actin–forward 5′-GATGAGATTGGCATGGCTTT-3′, reverse 5′-GTCACCTTCACCGTTCCAGT-3′.

Cell lines and cell culture

The human gastric cancer cell lines MKN-45 and SGC-7901 were obtained from the American Type Culture Collection (Manassas, VA, USA). Cell lines were cultured in RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum (FBS) in a humidified incubator at 37°C with 5% CO2. The cells in exponential growth can be used for experiments.

Transfection and drug treatment

The PLK1 coding sequences were cloned from General Bio Co., Ltd (An-Hui, China). The sequences were blunt end ligated into the LV5-GFP plasmid. This vector has a cytome galovirusim mediate-early promoter upstream of the multiple cloning sites and a Puromycin resistance gene driven by a CMV early promoter. For transfection experiment, the LV5-PLK1-GFP plasmid was transfected using Lipofectamine2000™ (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA) according to protocols of the manufacturer.

BI6727, the inhibitor of PLK1, was purchased from SelleckChem (Houston, TX, USA; cat S2235). The cells were divided into three groups: in the control group, the cells were treated with DMSO; in the BI6727 group, 100 nM of BI6727 was incubated for 48 hours; in the BI6727+ PLK1 group, the cells were first administered with 100 nM of BI6727, then transfected with PLK plasmid constructed as the above description for another 48 hours incubation.

Cell proliferation assay

MTT assay with a minor modification was used to determine the cell viability. MKN-45 and SGC-7901 cells in the three different groups were seeded in 96-well micro-titer plates at the density of 1×104 cells per well. MTT dye (10 µL of 2.5 mg/mL in PBS) was added to each well 4 hours prior to experiment termination in a 37°C incubator containing 5% CO2. Centrifuge the plates for 15 minutes at 1,500 rpm and discard the supernatant without disturbing the formazan crystals and cells in the wells. At the same time, the formazan crystals were dissolved in 150 µL of DMSO and the plates were stirred for 5 minutes on a plate shaker. The absorbance at 490 nm was read on a microplate reader (Thermo Fisher Scientific). The IC50 values had been calculated using GraphPad Prism 5.0 software (GraphPad Software Inc., La Jolla, CA, USA) from the dose–response curves.

Cell migration assay

To determine the migratory abilities of MKN-45 and SGC-7901 cells affected by PLK1, we carried out a cell migration assay. Millipore membranes with a pore size of 8 µm (Millipore Corporation, Billerica, MA, USA) were employed to divide the upper and lower chambers of the Transwell® plates (Costar®; Corning Incorporated, Corning, NY, USA). The migratory activity is defined as the course of cells shifted from the upper to the lower surface of the membrane. After 24 h pretreatment of MKN-45 and SGC-7901 cells in different groups, 5×104 cells were cultured in 100 µL 1640 medium with 0.1% BSA in each upper chamber. The culture condition of the low chamber was 1640 medium with 20% FBS. After 24 hours of incubation, the migrating cells on the lower surface were stained with Coomassie brilliant blue and manually counted under an optical microscope in three randomly selected fields (Olympus Corporation, Tokyo, Japan). All experiments were done in triplicate.

Cell cycle assay

The cell cycle is an important indicator for cancer cells. First, we collected the MKN-45 and SGC-7901 cells for trypsinization, and then rinsed them twice with ice-cold PBS. Second, 75% ice-cold ethanol was added to the single cells accompanied with vortexing, then for overnight incubation at −20°C. Third, the treated cells were fixed and washed with ice-cold PBS for another 30 minutes of incubation at 37°C, the incubation was conducted in 0.5 mL PBS solution with 20 mg/mL RNase A, 0.2 mM EDTA, 0.2% TritonX-100, and 20 mg/mL of propidium iodide. After that, the G0/G1, S, and G2/M phases’ cells were determined as the percentage.

Cell apoptosis assay by flow cytometry

Apoptosis was measured using a mixture of 100 µL Annexin V-FITC and propidiumiodide (PI). Briefly, 5×105 MKN-45 or SGC-7901 cells with or without different treatments for 24 hours were harvested and first rinsed twice with cold PBS, and then resuspended in Annexin-V binding buffer and stained simultaneously with FITC-conjugated Annexin-V and PI at room temperature for 15 minutes in the dark, before the addition of binding buffer. The apoptotic cells were determined by FACScan Flow Immunocytometry (Becton, Dickinson & Co., Franklin Lakes, NJ, USA).

Western blotting analysis

By reported procedures,18 protein extraction and Western blotting were performed. The subfamily of pERK is ERK1 (pT202/pY204)+ Erk2 (pT185/pY187) (cat. no ab5001, dilution, 1:10,000; Abcam, Cambridge, MA, USA), the subfamily of ERK is ERK1+ ERK2 (cat. no ab115799, dilution 1:1,000; Abcam). The subfamily of pMEK is MEK1 (pS298) (cat. no ab96379, dilution 1:3,500; Abcam), and the subfamily of pMEK is MEK1 (cat. no ab32091, dilution 1:5,000; Abcam). β-Actin was considered as the internal control. All results were conducted in triplicate.

Construction of patient-derived tumor xenografts and drug treatment

After approval by Jiangsu University Animal Care and Use Committee, we maintained and handled all the experimental mice, and followed China Laboratory Animal-Guideline for Ethical Review of Animal Welfare. The collected patient tumor material with informed consent was kept in culture medium on wet ice. The engraftment must be done within 2 hours after resection. First, the necrotic and supporting tissue were carefully removed with a scalpel. Tissues were then divided into three portions at about 30–50 mg of each, tissue sections were separated and rapidly frozen at −80°C for the next genomic analysis, and about 50–100 mg tissues were fixed in 10% neutral-buffered formalin and paraffin embedded for histopathological analysis. Finally, 20–30 mg tissue fragments were subcutaneously implanted into the flank region of athymic nude (nu/nu) male mice with a trocar. Finally, successful engrafted tumor models were obtained after passaging and banking for another three passages in mice. Tumors (−200 mm3) were enrolled on a rolling base and treated with agents or saline buffer; the dosage of BI6727 was 25 mg/kg/day with intravenous injection. The tumor size was evaluated every 5 days with caliper measurements. The volumes of the mass were calculated using the formula (l×w×w)/2, where l represents the main axis of the tumor, and w represents the small axis of the tumor.

Statistical analysis

The collected data were analyzed with GraphPad Prism 6 and presented as mean ± standard deviation (SD). The differences between the two groups were compared by two-tailed Student’s t-tests, while categorical data were evaluated by the chi-squared test. Survival analysis was evaluated by Kaplan–Meier plot and log-rank test. Independent prognostic factors were evaluated using univariate and multivariate Cox proportional hazard regression models. Differences between groups were considered significant at probability values of P<0.05.

Results

Clinical significance of PLK1 expression in GC tissues

In 20 pairs of GC specimens of GC tissues and matched normal tissues, the mRNA levels of PLK1 were determined with RT-PCR, indicating that PLK1 mRNA expression in GC was significantly increased (P<0.001, Figure 1A). After that, immunohistochemistry was employed to assess PLK1 expression in the GC specimens in different differentiation grades. In the majority of the GC cases, strong positive immune reactivity for PLK1 was found (17/20, 85.0%). Besides that, as indicated in Figure 1B, the poorer differentiation, the higher PLK1 expression. The density means were also indicated in Figure 1B, and the score in the poorly differentiated cases was 3+, significantly different from well differentiation.

Figure 1

Upregulation of PLK1 in clinical specimens.

Notes: (A) The expression of PLK1 mRNA in 20 paired GC and adjacent non-tumorous tissue specimens, as analyzed by qRT-PCR. (B) Representative results of the upregulation of PLK in GC specimens in well, moderately and poorly differentiated carcinoma as determined by immunohistochemistry analyses. *P<0.05, **P<0.01, ***P<0.001. #P<0.05 vs moderate group.

Abbreviation: GC, gastric cancer.

PLK1 inhibition prevents the growth of GC cells in vitro

GC cell lines SGC-7901 and MKN-45 were utilized for in vitro PLK1 inhibition experiments to analyze the potential oncogenic function of PLK1. First, the cells were incubated with different concentrations of PLK1 inhibitor BI6727 for 48 hours to determine the best assay concentration with Western blot. As shown in Figure 2A, BI6727 at the concentration of 100 nM was demonstrated to significantly suppress the PLK expression. Therefore, 100 nM was considered as the suitable concentration for the following experiments.

Figure 2

(A) The expression of PLK in SGC-7901 and MKN-45 cell lines treated with BI6727 at different concentrations (1, 5, 25, and 100 nM). (B) The effect of PLK1 on cellular proliferation of GC cells in the three groups as measured by MTT cell proliferation assay at 48 hours. *P<0.05 vs cells without BI6727 incubation. **P<0.01 vs control; ##P<0.01 vs BI6727 group.

Abbreviation: GC, gastric cancer.

To explore whether PLK1 affects GC cell proliferation, we performed MTT assay. The cells were divided into three groups: in the control group, the cells were treated with DMSO; in the BI6727 group, 100 nM of BI6727 was incubated for 48 hours; in the BI6727+ PLK1 group, the cells first were administered with 100 nM of BI6727 for 48 hours, then transfected with PLK plasmid constructed as the above description for another 48 hours of incubation. The SGC-7901 and MKN-45 cells in BI6727 group showed significant prevention of cellular growth compared to the control group (P<0.01) and the BI6727+ PLK1 group (P<0.01), while there was no significant difference between the control and the BI6727+ PLK1 groups (P>;0.05) (Figure 2B), indicating that the PLK1 inhibition by BI6727 could suppress the GC cell proliferation, while the overexpression of PLK1 after its inhibition promoted additionally the cell growth.

PLK1 inhibition suppressed GC cell migration

To further investigate the potential effects of PLK1 inhibition on GC metastasis, an in vitro transwell migration assay was performed. The results in Figure 3A demonstrated that there was a significantly reduced migration capability of SGC-7901 and MKN-45 with PLK1 suppression in the BI6727 group compared with the control and BI6727+ PLK1 group. The number of migratory cells was counted and shown in Figure 3A. These results indicate that PLK1 inhibition significantly abolished GC migration in vitro.

Figure 3

(A) The migratory condition of treated SGC-7901 and MKN-45 cells in the three groups as measured by the transwell migration assay. (B) The effects of PLK1 on the cell cycle. The proportion of cells in the G0–G1, S, and G2/M phases of the cell cycle in the three groups are presented. (C) Cell apoptosis in treated SGC-7901 and MKN-45 cells in the three groups, as determined by Annexin V-FITC and PI staining using flow cytometry. The representative apoptosis pattern is shown, and the apoptotic cells are indicated in the upper right and lower right quadrants. **P<0.01 vs control; ***P<0.001 vs group; ##P<0.01 vs BI6727 group; #P<0.05 vs B16727 group.

GC cells arrested at G2/M phase by PLK1 inhibition

Cell cycle progression is responsible for cell proliferation to some extent. The inhibition of PLK1 against proliferation of carcinoma cells by arresting cells in G2/M phase was demonstrated in the prior study.8 Some carcinogenic factors play a key role in tumor progression and affect the cell cycle. As a result, the GC cells treated with BI6727 or full-length PLK1 DNA were analyzed by flow cytometry. In comparison with the control group, inhibition of PLK1 in treated GC cells gave rise to the cells arrested at the G2/M phase (Figure 3B), suggesting that PLK1 inhibition can induce cell cycle arrest at the G2/M phase in GC cells.

PLK1 inhibition promotes GC cell apoptosis

To further characterize the effect of PLK1 on the proliferation and metastasis of GC cells, analysis of DNA content with Annexin V/FITC and PI staining was performed to evaluate early apoptosis and necrosis/late apoptosis. Treatment with BI6727 for inhibiting PLK1 progressively reduced the proportion of intact cells (Figure 3C), consisting of both the early and late apoptotic cells. The apoptotic cells were significantly increased in the BI6727 group compared with the control group (18.93% vs 5.80% for MKN-45 cells (P<0.01); 14.63% vs 3.86% for SGC-7901 cells (P<0.01)), also significantly higher than the BI6727+ PLK1 group (18.93% vs 8.75% for MKN-45 cells (P<0.01); 14.63% vs 7.93% for SGC-7901 cells (P<0.01)).

PLK1 inhibition affects MEK-ERK signaling pathway

To determine the underlying molecular mechanism(s) by which PLK1 is involved in tumor growth, progression, and metastasis, the potential molecules regulated by PLK1 were analyzed. The Western blotting assay revealed that MEK-ERK signaling was disrupted. As shown in Figure 4A, inhibition of PLK1 activity with BI6727 significantly blocked MEK/ERK phosphorylation and activation in both SGC-7901 and MKN-45 cell lines compared with the control and BI6727+ PLK1 groups (P<0.01).

Figure 4

(A) MEK/ERK signaling in treated SGC-7901 and MKN-45 cells in the three groups. (B) GC tumor sizes and weight in PDTX models measured every 5 days. All data shown are mean ± SD. **P<0.01. β-Actin was used as the loading control.

Differential expression of PLK1 affects tumorigenesis and tumor burden on PDTX

After demonstrating the role of PLK1 in GC cells in vitro, the potential benefit of PLK1 to the tumorigenic potential of GC cell lines was assessed using PDTX models in vivo, which capture the diverse heterogeneity of human malignancies and clinical predictive power. The tumors with the same cancer grade and stage were selected to maintain consistency and to promote statistical analysis. None of the PDTX tumors received any chemotherapy before surgery or following implantation.

Compared with the mice intraperitoneally injected with vehicle or BI6727, the mice in the BI6727 group displayed a significant decrease in tumor volume at the end of this study (tumor volume variation of 25 mg/kg BI6727 treatment group vs control: −67.3%); besides, a significant difference in tumor weights occurred between the two xenograft groups treated or not treated with BI6727 (Figure 4B). All these in vivo data indicated that PLK1 could function as a promoter of cancer cell growth and tumorigenicity in GC tumors.

Discussion

Different cancers’ involvement in cell invasion and metastasis, and different genes and signal transduction pathways have been the main research directions recently.24,25 Although GC treatment has been greatly improved, patients still have a high mortality rate. With the progress of GC research, more and more biomarkers are being used to prevent, diagnose, and treat diseases. PLK1 is a highly conserved silk/threonine kinase that is highly expressed in a variety of human tumor cells, and its important role in mitotic progression has been elucidated and PLK1 is one of the most important members.26 Several important physiological courses such as cell cycle progression, forming spindle, and segregating chromosome during mitosis are closely associated with PLK1.27,28 Furthermore, the PLK1 expression is found to be increased in diverse human malignancies including breast cancer, non-small-cell lung cancer, melanomas, and colorectal cancer.29–32 Regarding the significance of PLK1 in tumors, the mechanism of abnormally expressed PLK1 should be urgently explored.

Our related study revealed that, in comparison with non-tumor counterparts, PLK1 was highly expressed in gastric cancer tissues, suggesting that PLK1 might play a role in GC development. In addition, immunohistochemistry exhibited that histological differentiation had a close effect on the degree of PLK1 overexpression. Therefore, overexpression of PLK1 in GC may act as a biological marker for early detection and accurate prognosis.

Based on the results of this study, it was confirmed that PLK1 was expressed in SGC-7901 and MKN-45 cells by Western blotting, and significantly inhibited by PLK1 inhibitor BI6727, providing a suitable comparison to analyze the role and mechanisms of PLK1 in GC. The growth of cancer cells and DNA replication were significantly reduced by inhibition of PLK1 in a series of assays in vitro and in vivo, suggesting the eminence of PLK1 in proliferation and migration of cancer cells and tumor growth. Furthermore, it was demonstrated that the migration of GC cells was induced by PLK1, which might have a close association with genetic alteration related to metastasis in GC cells. Metastasis, one of the most common causes of death in patients with GC, is defined as a multistep cellular process contributing to the expansion of cancer cells from a primary to a secondary site, implicating diverse molecular factors connected with adhesion, degradation of the basement membrane, proliferation and migration of cancer cells, invasion, and angiogenesis. Accompanied by consociation of stromal cells, the genetic and/or epigenetic alterations in cancer cells will contribute to cancer metastasis.33,34

In further exploration, the survival of PLK1-overexpressed tumor cells may be on account of the anti-apoptotic function. Inhibition of apoptosis is the primary way of tumor development, and proliferation of tumor cells is accumulated with uncontrolled proliferation, genetic instability, and mutations.35 As a result, impaired growth and metastasis of GC cells caused by PLK1 inhibition could be illuminated by suppression of the apoptosis pathway to some extent.

Abnormal activation of the MEK/ERK cascades is closely associated with tumorigenesis including GC. Some studies have demonstrated that PLK1 was closely associated with RAS/MEK/ERK pathways, including regulation of MEK1/2 and proliferation in airway smooth muscle cells,36 induction of epithelial-to-mesenchymal transition and promotion of epithelial cell motility by activating CRAF/ERK signaling,37 and activation by RAS-RAF signaling as a therapeutic target in HCC.38 As a result, we explore the relationship between PLK1 and RAS/MEK/ERK in GC. The Western blotting and in vivo assays demonstrated the inhibition of PLK1 activity with BI6727 significantly blocked MEK/ERK phosphorylation and activation, resulting in an obvious reduction in tumor volume and weight. However, the expression of PLK1, the different stages of cancer, the heterogeneity between tumor and normal or tumor tissue, different tumor microenvironment, and different signal pathways all may lead to the different effects of PLK1 in GC. A more comprehensive understanding of the oncogenic mechanisms of PLK1 during the trigger and evolvement of GC could provide hints for future better treatments for GC patients.

Conclusion

We have demonstrated that the abnormally high expression of PLK1 in GC tissues and cells in this study has a significant inhibitory effect on the proliferation and migration of GC cells. Moreover, PLK1 contributes to anti-apoptotic effects to promote tumor growth by activation of MEK/ERK phosphorylation. As a result of this, our findings show that PLK1 plays an important role in GC tumor cell function and that PLK1 can be used as a potential therapeutic and prophylactic biomarker for GC.