Elevated microRNA-7 inhibits proliferation and tumor angiogenesis and promotes apoptosis of gastric cancer cells via repression of Raf-1

Collection of clinical data

From June 2015 to June 2016, 53 patients (28 males and 25 females) with GC who underwent neoadjuvant chemotherapy were collected. The patients were 38–75 years old with an average age of 57 years. Degree of differentiation: 37 cases of poor and moderate differentiation and 16 cases of high differentiation. Tumor node metastasis (TNM) staging: 34 patients in IIIb stage and 19 in stage IV. All patients were included if they met the following criteria: They were diagnosed by gastroscopy or cytology before surgery; GC patients were staged in advanced IIIb stage and IV stage with reference to TNM staging at the seventh edition of International Union Against Cancer/American Joint Committee on Cancer in 2010; None of patients had received radiotherapy, chemotherapy or other adjuvant therapy before admission; none of the patients was accompanied by contraindications to chemotherapy; Patients had measurable lesions according to Response Evaluation Criteria in Solid Tumors (RECIST1.1) [¹⁸]. Patients with distant metastases, other malignant tumors, severe heart disease or arrhythmia, or with a history of myocardial infarction within one year, or pregnant or lactating women were excluded. Another 53 healthy controls were selected from the medical examination center of The First Affiliate Hospital, School of Medicine, Shantou University.

GC patients were taken 10 mL of blood on an empty stomach in the morning before neoadjuvant chemotherapy. The blood was sent to the laboratory in 1 hour and the serum was extracted with 2.5 mL by the centrifuge, stored in cryotubes and put in a refrigerator at −80°C. After the neoadjuvant chemotherapy treatment (three cycles), the same method was used for collecting blood samples. The blood samples of 10 mL were obtained from healthy controls, centrifuged and stored at −80°C.

Neoadjuvant chemotherapy and efficacy assessment

Patients were intravenously injected with 5% glucose solution containing oxaliplatin (130 mg/m²) on the first day and orally administrated with capecitabine (1000 mg/m²) from the 1^st day to the 14^th day (21 days/cycle, for total 3 cycles). Neoadjuvant chemotherapy was assessed and re-operated after treatment. Antiemetic drugs were applied in treatment with oxaliplatin while oral liver protection, Baiyao and neurotrophic drugs in oral administration with capecitabine. The adverse reactions during chemotherapy were leukopenia (15 cases), nausea and vomiting (8 cases), abnormal liver function (7 cases), and diarrhea (3 cases). Abdominal CT or magnetic resonance imaging of the patients before and after chemotherapy was compared, and the efficacy assessment was performed according to the RECIST 1.1. Efficacy evaluation was classified into four grades including complete response (CR, all tumor lesions disappeared, lasting for 4 weeks), partial response (PR, the sum of the longest diameter of the tumor was reduced by more than 30%, lasting for 4 weeks), stable disease (SD, ranked between PR and PD), and progressive disease (PD, the sum of the longest diameter of the tumor was increased by 20% with the appearance of new lesions). The chemotherapy efficacy was calculated as (CR + PR)/total cases × 100% [¹⁹].

Reverse transcription quantitative polymerase chain reaction (RT-qPCR)

Total RNA in serum samples and cells was extracted by Trizol (Invitrogen, Carlsbad, CA, USA). The quality of RNA was assessed by ultraviolet and formaldehyde denaturation electrophoresis. miR-7, Raf-1, U6 and β-actin primers were designed via using Primer premier 5.0 software with reference to the gene sequences in GenBank, and synthesized by Sangon Biotech Co., Ltd. (Shanghai, China) (Table 1). Then, the cDNA was extracted by using a reverse transcription kit (Beyotime Biotechnology Co., Ltd., Shanghai, China) referring to the instructions. The reaction solution was taken for fluorescence quantitative PCR. The relative expression of miR-7 was expressed with U6 as the internal reference, and the relative expression of Raf-1, β-actin. The relative expression of the target gene was calculated by the relative quantitative method 2^−ΔΔCt.

Cell culture

RGM-1 normal gastric mucosal epithelial cells, human GC cell BGC823 (Shanghai Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences, Shanghai, China), human GC cell MGC-803 (Shanghai Institutes for Biological Sciences, Shanghai, China), and human GC cell SGC-7901 (Cell resource center, Chinese Academy of Sciences, Shanghai, China) were seeded in RPMI1640 medium with 100 ml/L fetal bovine serum (FBS) and 100 μg/mL penicillin, cultured with 5% CO₂ and saturated humidity at 37°C. After fully adherent to the wall, the cells were detached and sub-cultured in trypsin of concentration of 2.5 g/L. Cells in logarithmic growth phase were collected for subsequent experiments. The relative expression levels of the target genes in each group were determined by RT-qPCR. The two GC cell lines with largest or smallest differences of the relative expression of the target gene by contrast with RGM-1 cells were used for subsequent cell experiments.

Dual luciferase reporter gene assay

The target site for the binding relationship of Raf-1 and the corresponding miR-7 was predicted by online prediction software (http://www.microrna.org/microrna/home.do). The primers were designed and synthesized with the 3ʹuntranslated region (UTR) sequence of Raf-1 gene. Forward and reverse primers were introduced into the restriction enzyme sites of restriction endonucleases Hind III and Spe I. The mutation sequence of the binding site was designed, and the target sequence fragment was synthesized by Genscript Biotechnology Co., Ltd. (Nanjing, China). The amplified target products and the pMIR-REPORT^TM Luciferase vector plasmid were digested via restriction endonuclease Hind III and Spe I. The enzyme-digested product was recycled, ligated with T4 DNA ligase, and transformed with DH5α competent cells. The single colony was chosen and shaken, and the plasmid was extracted. The correct recombinant wild type (WT) and mutant type (MUT) plasmids were received by enzyme digestion, sequencing and identification. In a 12-well plate, 1 × 10⁵ BGC-823 and SGC-7901 cells were seeded into each well and co-transfected with the recombinant WT/MUT plasmids and miR-7 mimics/mimics NC for 48 h, and the cell culture fluid was discarded. Each well was added with 100 μL cell lysate in luciferase kit for lysing of 30 min. LAR II with 100 μL was joined to 20 μL cell lysate, and the fluorescence (A) was measured. The fluorescence value (B) was measured with the supplemented of 100 μL Stop&Glo reagent. The luciferase activity value C = B/A, with A as an internal reference.

Cell transfection

BGC-823 and SGC-7901 cells were transfected with negative control (NC) of miR-7 mimics, miR-7 mimics, NC of si-Raf-1, si-Raf-1, miR-7 inhibitor + si-Raf-1 NC, or miR-7 inhibitor + si-Raf-1. MiR-7 mimics, mimics NC and miR-7 inhibitors were purchased from Shanghai GenePharma Co., Ltd. (Shanghai, China). siRNA-Raf-1 and siRNA-Raf-1 NC were synthesized by Shanghai Sangon Biotechnology Co. Ltd. (Shanghai, China). When the cell confluence was 70%-80%, the MGC-823 and SGC-7901 cells were transiently transfected by lipofectamine2000 (Invitrogen, Carlsbad, California, USA). After transfection for 6 hours, the culture medium was changed and then cell lines were continually cultured for 48 h.

Cell proliferation assay

The cells in logarithmic phase transfected for 48 h were detached with 2.5 g/L trypsin. The cells (1 × 10³ cells/100 μL/well) were cultured in a humidified incubator with 3 duplicate wells in each group. The CCK-8 method was used for detection at 24^th h, 48^th h and 72^nd h, respectively. CCK-8 solution (Dojindo Laboratories, Kumamoto, Japan) with 10 μL was joined to each well. After 4-h reaction under normal conditions, the optical density (OD, 450 nm) value of cells was measured by a microplate reader.

Cell migration and invasion detection

The cells at 48 h post transfection were trypsinized and re-suspended with RPMI-1640 medium to reach 1 × 10⁵ cells/mL. The cell suspension (200 μL) was added to each well. RPMI-1640 medium with 500 μL was joined to the lower chamber of the Transwell chamber (Millipore, MA, USA). The culture medium in the upper chamber was removed. The cells were fixed with 95% ethanol and the chamber membrane was taken out. After fixation and staining, the membrane was observed under an inverted microscope. Five fields were chosen for counting the number of trans-membrane cells. Three parallel reaction wells were set for each sample. Matrigel (Millipore) was thawed in a refrigerator at 4°C, and diluted with RPMI-1640 medium at a ratio of 1:8. The diluted solution with 100 μL was joined to the upper Transwell chamber, air-dried at 4°C, and coagulated overnight for later use. The remaining steps were the same as the Transwell migration experiment. Three parallel reaction wells were set for each sample.

Cell cycle and apoptosis detection

The transfected cells after 48 h were trypsinized, centrifuged, and added with pre-cooled ethanol with a concentration of 700 g/L ethanol and fixed at 4°C. After centrifugation, the cells were stained with propidium iodide (PI, BD Biosciences, CA, USA) in darkness, and the cell cycle was measured by a flow cytometer.

The transfected cells after 48 h were trypsinized and centrifuged. Cells with 1 mL were centrifuged and re-suspended by adding 200 μL of the binding buffer to attain 1 × 10⁴ cells/mL. Annexin V with 5 μL and 5 μL of PI (both from BD Biosciences) were joined for staining. The binding buffer (400 μL) was added into cells for testing cell apoptosis rate on a flow cytometer.

Western blot analysis

Total protein of the cells was extracted by total cell protein extraction kit (Beyotime Biotechnology Co., Ltd., Shanghai, China). The concentration and purity of the extracted total protein were detected by bicinchoninic acid method. A sample with 30 μg of the protein was applied into sodium dodecyl sulfate polyacrylamide gel electrophoresis (120 g/L). Then the protein was transferred and blocked. The membrane was joined with primary antibody Raf-1 (1:500), β-actin (1:500) and with the horseradish peroxidase-labeled secondary antibody (1:2000, all from Cell Signaling Technology, Beverly, MA, USA). The protein was exposed by chemical coloration, and analyzed by gel electrophoresis imager and the Quantity One software. Target protein expression was calculated with β-actin as the internal reference.

Tumor xenograft in nude mice

Forty-two BALB/c nude mice (4 ~ 5 weeks old) were purchased from Hunan SLAC Laboratory Animal Co., Ltd. (Changsha, China). The nude mice were raised under specific pathogen-free conditions, and acclimated to the new environment in the animal room for one week. The state of the animals was observed every day with disinfected food and water supply. The transfected GC BGC-823 cells and SGC-7901 cells were made for a single cell suspension. The PBS and Matrigel were mixed at 1: 1. The mixture was employed for cell re-suspension to 2 × 10⁷ cells/mL. The nude mice were fixed and subcutaneously injected with the transfected cells in the back of hind legs under aseptic conditions. Each group of nude mice was inoculated with a kind of cell. The changes in mental status, diet and body weight of the nude mice were observed regularly. From the day of inoculation, the subcutaneous tumors of nude mice were observed every other day, and the length (L), width (W) and height (H) of the tumor nodules were measured every 3 days. The formula for calculating the tumor volume was: V = π/6 (L × W × H). After 30 days, the nude mice were euthanized and the tumor was exfoliated. The specimen was fixed with 40 g/L formaldehyde, embedded in paraffin, and sectioned (about 4 μm).

Immunohistochemical staining

Paraffin section was dewaxed by conventional xylene, and dehydrated with gradient ethanol. Endogenous peroxidase activity was blocked via 30 mL H₂O₂. The sections were incubated with Histostain^TM-Plus reagent A (goat serum blocking solution, Beijing Zhongshang Golden Bridge Biotechnology Co., Ltd., Beijing, China). Then the sections were incubated with primary mouse CD34 monoclonal antibody (1:50, Cell Signaling Technology). The sections were incubated with Histostain^TM-Plus Reagent B (biotinylated secondary antibody, Beijing Zhongshang Golden Bridge Biotechnology Co., Ltd., Beijing, China), and then Histostain^TM-Plus Reagent C (horseradish peroxidase-labeled third antibody, Beijing Zhongshang Golden Bridge Biotechnology Co., Ltd., Beijing, China). The sections were treated with diaminobenzidine color development, hematoxylin counterstaining, gradient ethanol dehydration, permeabilization with xylene, and sealing with neutral gum. After that, five randomly selected high-power fields were utilized to count the average proportion of positive cells in each field via the true color multi-functional cell image analysis management system (Image-Pro Plus, Media Cybernetics, USA). PBS was used instead of the primary antibody as an NC.

miR-7 is down-regulated and Raf-1 is up-regulated in GC cells; Raf-1 is a target gene of miR-7

A study has revealed that miR-7 expression is reduced in GC cell lines and negatively connected with invasion and metastasis [²¹]. The relative expression of miR-7 in human normal gastric mucosal epithelial cell line RGM-1 and human GC cell lines (BGC-823, MGC-803 and SGC-7901) were tested by RT-qPCR, and highly expressed cell lines were screened. The results indicated that by contrast with RGM-1 cells, decreased miR-7 and raised Raf-1expression were manifested in GC cell lines (all P < 0.05) (Figure 2a–c). Among them, miR-7 had the highest relative expression in MGC-823 cells, and relatively low expression in SGC-7901 cells. Therefore, these two GC cells were selected for subsequent cell experiments and in vivo experiments in nude mice.

Open in a separate window

Figure 2.

miR-7 is down-regulated and Raf-1 is up-regulated in GC cells; Raf-1 is a target gene of miR-7. (a). The relative expression of miR-7 in human normal gastric mucosal epithelial cell line RGM-1 and GC cell lines; (b). The relative expression of Raf-1 mRNA in human normal gastric mucosal epithelial cell line RGM-1 and GC cell lines; (c). Raf-1 protein bands and protein expression in human normal gastric mucosal epithelial cell line RGM-1 and GC cell lines; (d)&(f). The binding site of Raf-1 and miR-7 predicted by bioinformatics website; (e)&(g). The binding relationship between miR-7 and Raf-1 analyzed by dual luciferase reporter gene assay; (h). The relative expression of miR-7 after transfection in each group of cells; (i). The relative expression of Raf-1 mRNA after transfection in each group of cells; (j). The protein bands and protein expression of Raf-1 after transfection in each group of cells; $ vs the normal gastric mucosal epithelial cells RGM-1, P < 0.05; * vs the mimics NC group, P < 0.05; & vs the si-NC group, P < 0.05; # vs the miR-7 inhibitors + si-NC group, P < 0.05; Measurement data were mean ± standard deviation of three independent experiments. One-way ANOVA was utilized for data analysis among multiple groups, followed by Tukey’s multiple comparisons test.

The http://www.microrna.org/microrna/home.do website predicted that Raf-1 contained two miR-7 target sequences (Figure 2d,f). To confirm that the predicted binding site of miR-7 caused a change in luciferase activity, the MUT sequence and WT sequence of Raf-1 3ʹ-UTR were inserted into the reporter plasmid, respectively. Through dual luciferase activity assay, GC cells BGC-823 and SGC7901 were co-transfected with miR-7 mimics and WT-Raf-1 or MUT-Raf-1 recombinant plasmid, respectively. The results conveyed (Figure 2e,g), in BGC-823 cells, miR-7 mimics had no apparent effect on the luciferase activity of the MUT-Raf-1 plasmid (P > 0.05), but the luciferase activity of the WT-Raf-1 plasmid was distinctly reduced (P < 0.05). The same trend also existed in SGC-7901 cells.

For clarification of the role of miR-7 and Raf-1 in GC cells, miR-7 mimics/inhibitors and Raf-1 siRNA were transiently transfected into BCG-823 and SGC-7901 cells, respectively. It was determined that miR-7 up-regulation or Raf-1 down-regulation affected Raf-1 mRNA. Western blot analysis confirmed that miR-7 inhibited Raf-1 protein expression. Spontaneously down-regulating Raf-1 and miR-7 decreased Raf-1 expression versus to down-regulating miR-7 alone, indicating that Raf-1 down-regulation effectively mitigated the effects of miR-7 knockdown on Raf-1 expression in GC cells (Figure 2h–j)

GC cell proliferation is inhibited while apoptosis is promoted in vitro by elevated miR-7 and reduced Raf-1

For clarification of the functions of miR-7 and Raf-1 in GC cell proliferation and apoptosis, CCK-8 assay was utilized in testing OD values of cells at 24^th h, 48^th h, and 72^nd h (Figure 3a,b) while PI single staining and Annexin V/PI double staining were in testing cell cycle entry and apoptosis rate (Figure 3c–g). The results showed that transfection of miR-7 mimics or Raf-1 siRNA could reduce GC cell proliferation, increased the proportion of G₁ phase cells, decreased the proportion of S phase and G₂/M phase cells and increased apoptosis rate. However, inhibition of miR-7 had the opposite effects. The effect of down-regulated miR-7 on GC cells could be reversed by Raf-1 knockdown, as manifested by impaired cell proliferation activity and enhanced apoptosis rate. These results indicated that miR-7 could target Raf-1 to affect cell proliferation and apoptosis of GC cells.

Open in a separate window

Figure 3.

GC cell proliferation is depressed and apoptosis is promoted in vitro by elevated miR-7 and reduced Raf-1. (a)&(b). The change of BCG-823 and SGC-7901 cell proliferation ability after transfection; (c)&(d). The comparison of cell cycle distribution; (e–g). The comparison of cell apoptosis rate; * vs the mimics NC group, P < 0.05; & vs the si-NC group, P < 0.05; # vs the miR-7 inhibitors + si-NC group, P < 0.05; Measurement data were mean ± standard deviation of three independent experiments. One-way ANOVA was utilized for data analysis, followed by Tukey’s multiple comparisons test.

GC cell migration and invasion are repressed by increased miR-7 and decreased Raf-1

The effects of miR-7 and Raf-1 on GC cell migration and invasion were determined by Transwell assay (Figure 4a–d). It was recognizable that cell migration and invasion ability were weakened by miR-7 mimics or si-Raf-1 transfection. si-Raf-1 transfection reversed the promoting effects of miR-7 inhibitors treatment on cell migration and invasion ability. SGC7901 cells also had the same trend.

Open in a separate window

Figure 4.

GC cell migration and invasion are repressed by up-regulated miR-7 and reduced Raf-1. (a)&(b). The number of migrating and invasive BGC-823 cells; (c)&(d). The number of migrating and invasive SGC-7901 cells; * vs the mimics NC group, P < 0.05; & vs the si-NC group, P < 0.05; # vs the miR-7 inhibitors + si-NC group, P < 0.05; Measurement data were mean ± standard deviation of three independent experiments. One-way ANOVA was utilized for data analysis, followed by Tukey’s multiple comparisons test.

We would like to acknowledge the reviewers for their helpful comments on this paper.