microRNA-133b represses the progression of lung cancer through inhibiting SOX9/β-catenin signaling pathway

Abstract

MicroRNA-133b (miR-133b) has been shown to be down-regulated in lung cancer and functions as a tumor repressor. However, the underlying mechanisms of miR-133b in lung cancer are not clear. SOX9, a member of SOX family, serves as an oncogene in lung cancer by activating b-catenin signaling and was identified to be a direct target of miR-133b in breast cancer. Based on these data, the current study was performed to explore whether SOX9/b-catenin signaling is implicated in miR-133b-meditaed lung cancer repression. MiR-133b expression in lung cancer tissues and cells were detected by RT-PCR. CCK-8, colony formation, flow cytometry, transwell chamber and in vivo assays were carried out to determine cell proliferation, colony formation, apoptosis, cell cycle, invasion, and tumorigenesis. We found that miR-133b expression was decreased in lung cancer tissues and cells. Up-regulation of miR-133b reduced cell proliferation and colony formation, induced cell apoptosis and G0/G1 phase arrest, and decreased cell invasion. Besides, miR-133b up-regulation decreased the expression of b-catenin and SOX9. Cell viability inhibition and apoptosis promotion induced by miR-133b up-regulation were all impaired when SOX9 was up-regulated. Furthermore, miR-133b over-expression repressed the tumorigenesis of lung cancer cells with smaller tumor size and lower Ki-67 expression. Taken together, this study clarifies that miR-133b represses lung cancer progression by inhibiting SOX9/b-catenin signaling.

Introduction

Lung cancer ranks 2^th among the most frequent type of malignant tumors and is one of the leading causes of cancer-related deaths in the world [1]. Although advances have been made in lung cancer diagnose and treatment, the outcomes of patients remain poor, and only 15% of patients have more than 5 years’ survival after diagnosis [2]. Therefore, it’s essential to explore the mechanisms underlying lung cancer occurrence and development.

MicroRNAs (miRNAs) are small noncoding RNAs which modulate protein translation at the post-transcriptional level through inducing mRNA degradation or translation inhibition by binding to the 3’ untranslated region (3’UTR) of their targeting mRNAs [3]. Alterations of miRNA expression have been identified to be strongly related to the tumorigenesis of majority of cancers including lung cancer [4], indicating the vital roles of miRNAs in cancer treatment. Among these miRNAs, miR-133b has been found to be lowly expressed in non-small cell lung cancer (NSCLC) tissues and cells, and miR-133b up-regulation inhibited lung cancer cell migration and invasion by targeting fascin1 [5], as well as repressing cell growth through interaction with the 3’UTR of epidermal growth factor receptor (EGFR) [6], suggesting that miR-133b functions as a tumor suppressive gene in lung cancer. However, the mechanisms underlying miR-133b in lung cancer remains largely unknown.

SOX9 is a member of SOX family and serves as a stem cell-related transcription factor in holding stem cell properties [7]. Zhou et al. [8] have demonstrated that SOX9 is over-expressed in lung cancer cell lines and tissues, which predicts poor prognosis and advanced clinical progression of patients with lung cancer. Wang et al. [9] have revealed that up-regulation of SOX9 significantly promotes lung cancer A549 cell growth, migration and invasion. Moreover, Guo et al. [10] have illustrated that inhibition of Wnt/β-catenin signaling abolishes the role of SOX9 up-regulation in lung cancer A549 cell proliferation enhancement and apoptosis repression, indicating that SOX9 serves as an oncogene in lung cancer through activating Wnt/b-catenin signaling, which is of importance in both normal embryonic development and tumorigenesis including lung cancer [11,12]. Under normal conditions, β-catenin is wrapped in a destruction complex which is made up of adenomatous polyposis coli (APC), glycogen synthase kinase 3 (GSK3), axin 1, and casein kinase 1. This causes β-catenin to be phosphorylated and degraded through a ubiquitin-induced proteasomal pathway [13]. However, in human cancers, b-catenin separates from this complex and accumulates in the cytoplasm and nucleus, triggering the expression of its target genes such as cyclin D1 and c-Myc through activating TCF/LEF transcription factors [14]. Furthermore, miR-133b was reported to inhibit the tumorigenesis and metastasis of breast cancer through targeting SOX9 [15]. However, whether miR-133b represses the progression of lung cancer through the SOX9/β-catenin pathway remains unclear.

Overall, this study was performed to probe the effects of miR-133b/SOX9/β-catenin in lung cancer progression through both in vitro and in vivo assays.

Materials and methods

Tissue samples

Twenty paired lung cancer tissues and the adjacent normal lung tissues derived from lung cancer patients were used in this study. Each patient had undergone pneumonectomy without any form of chemoradiotherapy treatment and signed an informed consent.

Cell lines and culture

Normal lung epithelial cell line BEAS-2B and lung cancer cell lines A549, H1299, and SPC-A1 were all purchased from BeNa Culture Collection (Beijing, China). BEAS-2B cells were cultured in high glucose Dulbecco’s Modified Eagle Medium (DMEM-H, Gibco, CA, USA); A549 cells were cultured in F-12K medium (Gibco, CA, USA); H1299 and SPC-A1 cells were cultured in RPMI-1640 medium (Gibco, CA, USA); supplemented with 10% fetal bovine serum (FBS, Gibco, USA). Cells were incubated in a humidified atmosphere at 37°C with 5% CO₂.

Western blot analysis

Protein was extracted from cells and tissue samples with lysis buffer (Roche, Shanghai, China) containing protease and phosphatase inhibitors (Solarbio, Beijing, China). After centrifugation at 4°C for 25 min, the protein samples were quantified by using a Bicinchoninic acid Protein Assay Reagent (Thermo Fisher Scientific, MA, USA). Then, protein samples were loaded into 10% SDS-polyacrylamide gel and separated by electrophoresis, followed by transformation to polyvinylidene difluoride membranes (PVDF, Millipore, MA, USA). After incubation with 5% non-fat milk for 1 h at room temperature, the membranes were probed with the indicated primary antibodies including β-catenin (No. MA1-2001, Invitrogen, MA, USA), SOX9 (No. ab26414, Abcam, MA, USA) at 4°C overnight and the secondary antibodies for 1 h at room temperature in succession. After washing three times with phosphate buffered saline (PBS), the bands were measured by ECL regent (Millipore, MA, USA) in a ProfiBlot-48 (Tecan, Switzerland), and gray-scale value analysis was carried out by ImageJ software.

Quantitative real time PCR (qPCR)

Total RNA was exacted from cells and cancer tissue samples with GenElute™ Total RNA Purification Kit (Sigma-Aldrich, MO, USA). After quantification, a total of 1 μg RNA was reverse transcribed (RT) into cDNA prior to qPCR analysis. All kits used in this experiment were bought from Takara (Dalian, China). U6 was used as an internal control for miR-133b, respectively. Relative gene levels were calculated by the 2^-ΔΔCt method. Primer sequences used in this study were obtained from Invitrogen (CA, USA) and were as follows:

miR-133b: RT-5’-GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACTAGCTG-3’, sense-5’-GATTTGGTCCCCTTCAAC-3’, antisense-5’-GTGCAGGGTCCGAGGT-3’; U6: RT-5’-AACGCTTCACGAATTTGCGT-3’, sense-5’-CTCGCTTCGGCAGCACA-3’, anti-sense-5’-AACGCTTCACGAATTTGCGT-3’.

CCK-8 assay

H1299 cells (2×10³/well) in logarithmic growth period was suspended in 100 μL culture medium and added into 96-well plates. Next day, the cells were given different treatments. After 1, 2, 3, 4, or 5 days of treatment, 10 μL of cell counting kit-8 (CCK-8) regent (Sangon Biotech, Shanghai, China) were added into each well and incubated for another 4 h at 37°C. Then, the optical density (OD) levels at 450 nm were examined using microplate reader.

Colony formation assay

Stable transfected H1299 cells were inoculated into a 5-cm dish at a density of 3×10³/100 μL and were incubated at 37°C. The medium was replaced every 2 d until 14 d. Then, after being washed with PBS for three times, the cells were fixed with 5% paraformaldehyde for 15 min, and stained with 0.1% crystal violet solution for 20 min. The visible colonies were counted for cell colony formation ability evaluation.

Cell cycle and apoptosis detection

Flow cytometry was used to analyze cell cycle and apoptosis. After 48 h of different treatments, H1299 cells were harvested and washed with PBS for one time. Then, Annexin V-FITC and PI solution were added to each sample for 15 min in the dark. The fluorescent signals reflecting cell apoptosis were evaluated by flow cytometry within 1 h. Cells in FITC^-/PI^- quadrant were identified as live cells, FITC⁺/PI^- were early apoptotic cells and FITC⁺/PI⁺ were late apoptotic cells.

For cell cycle detection, cells were first synchronized with serum-free medium for 12 h prior to different treatments. After 48 h of cell treatment, cells including floating and attached H1299 cells were collected and fixed in 70% ethanol at 4°C for 3 h. Then the cells were incubated with 20 μg/mL RNase for 30 min at 37°C and 5 μg/mL PI solution for 30 min in the dark at 4°C. Then, the cell cycle was evaluated by using flow cytometry.

Transwell assay

Cell invasion was analyzed on H1299 cells using the Transwell chamber pre-coated with material (8 μm pore size; Corning, NY, USA). 1×10⁵ H1299 cells were re-suspended in 200 μL FBS-free RPMI-1640 medium and were seeded into the upper chamber. At the same time, 600 μL RPMI-1640 medium containing 10% FBS was added into the lower chamber. After incubation at 37°C for 48 h, cells attached in the lower chamber membrane were fixed in paraformaldehyde, and stained with 1% crystal violet for 10 min at room temperature. Cell numbers in six randomly selected fields were counted under an upright light microscope at ×200 magnification.

In vivo xenograft assay

Twenty male BALB/c nude mice with 4-week-old were purchased from the animal center of air force medical university (Shanghai, China) and were housed under specific pathogen free (SPF) conditions with free water and food available for 1 week. H1299 cells with miR-133b stable over-expression or not were established through selecting with 2 μg/ml puromycin (Sigma-Aldrich, MO, USA) for 7 days. After 1 weeks of breeding, mice were injected 1×10⁶ H1299 stable transfected cells around the left side of the neck, with 10 mice in each group. Mice were sacrificed and tumors were removed 28 days post-injection.

Immunohistochemistry

For immunohistochemistry, tumor tissues cut into 4-μm slides were deparaffinized, rehydrated, and blocked with 5% goat serum. Then the slides were subjected to immunohistochemical staining with Ki-67 antibody (No. #12202, Cell Signaling Technology, CA, USA) and incubated for overnight at 4°C, followed with incubation with the secondary antibodies. Finally, the slides were visualized by staining with chromogen 3, 3’-diaminobenzidine tetrachloride. Protein expression was evaluated by grey values at 400× magnification under a microscope.

Statistical analysis

Data were analyzed by SPSS23.0 software and are presented as mean ± SD. Data comparison was performed by Student’s t test for two groups and two-way ANOVA with Bonferroni post-tests for multiple groups. A p value less than 0.05 was considered significant.

Results

Down-regulation of miR-133b in lung cancer tissues and cell lines

To explore the function of miR-133b in the occurrence and development of lung cancer, we first detected the expression level of miR-133b in the lung cancer tissues and cell lines. Compared with the normal tissues, the expression of miR-133b was obviously decreased in the lung cancer tissues (Figure 1A). Similarly, miR-133b was lowly expressed in lung cancer cell lines A549, H1299, and SPC-A1 when compared to that of the normal lung cell line BEAS-2B (Figure 1B). These results illustrated that the down-regulation of miR-133b might play a crucial role in lung cancer progression.

Figure 1.

Expression of miR-133b was decreased in lung cancer tissues and cell lines. A. QPCR analysis of the levels of miR-133b in 20 paired lung cancer tissues and adjacent normal lung tissues. B. QPCR analysis of the levels of miR-133b in normal lung cell line BEAS-2B and lung cancer cell lines A549, H1299, and SPC-A1. (**P<0.01).

Up-regulation of miR-133b inhibits the malignant phenotype transformation of lung cancer cells

Then, we explored the function of miR-133b in lung cancer progression through a gain-of-function assay in vitro. As the expression of miR-133b was decreased in A549, H1299, and SPC-A1 cells, we chose H1299 cells for further study. Results confirmed that transfection with the mimic of miR-133b significantly increased miR-133b expression in H1299 cells (Figure 2A). miR-133b up-regulation decreased cell colony formation (Figure 2B) and proliferation abilities (Figure 2C) and reduced cell apoptosis (Figure 2D) in H1299 cells. Moreover, miR-133b up-regulation induced G0/G1 phase increase and S phase reduction (Figure 2E), and impaired the invasion ability of H1299 cells (Figure 2F). Taken above, these findings illustrated that miR-133b functioned as a tumor suppressor in lung cancer progression.

Figure 2.

miR-133b inhibited the malignant phenotypic switching of lung cancer cells. H1299 cells were infected with mimic-miR-133b or mimic-NC, then: A. The expression of miR-133b was detected by QPCR. B. Cell colony formation ability was determined by colony formation assay. C. Cell viability was determined by CCK-8 assay. D, E. Cell cycle and apoptosis were evaluated by flow cytometry with PI staining and Annexin V/PI double staining, respectively. F. Cell invasion was assessed by transwell chamber with Matrigel coated. (n=3, *P<0.05, **P<0.01).

miR-133b inhibits lung cancer progression through negative regulation of SOX9/β-catenin signaling

Next, we explored whether SOX9/β-catenin signaling was implicated in lung cancer progression in which miR-133b functioned as a tumor suppressor. Western blot results showed that up-regulation of miR-133b with mimic infection obviously decreased both β-catenin and SOX9 expression in H1299 cells (Figure 3A). Subsequently, we overexpressed SOX9 with OE-SOX9 plasmid transfection. Figure 3B showed its transfection efficiency. However, miR-133b-induced cell colony formation repression and cell apoptosis promotion were all abolished when SOX9 was up-regulated in H1299 cells (Figure 3C, 3D), suggesting that miR-133b repressed lung cancer progression through negative regulation of SOX9/β-catenin signaling.

Figure 3.

miR-133b repressed the progression of lung cancer through down-regulating β-catenin and SOX9 expression in H1299 cells. A. Western blot analysis of the protein expression of β-catenin and SOX9 after 48 h of H1299 cells infection with mimic-NC or mimic-miR-133b (n=3, *P<0.05, **P<0.01). B. Western blot analysis of the transfection efficiency of OE-SOX9 (n=3, **P<0.01). C. Cell colony formation ability was determined by colony formation assay after H1299 cells were treated with mimic-miR-133b or mimic-miR-133b + OE-SOX9. D. Cell apoptosis was evaluated by flow cytometry after H1299 cells were treated with mimic-miR-133b or mimic-miR-133b + OE-SOX9. (n=3, **P<0.01, compared with control group; #P<0.05, compared with mimic-miR-133b group).

miR-133b up-regulation represses the tumorigenesis of lung cells in vivo

Subsequently, we used a mice tumor-bearing assay to investigate the role of miR-133b in lung cancer. Compared with the control group, up-regulation of miR-133b decreased the tumorigenesis of H1299 cells with smaller tumor size and weight (Figure 4A) and decreased expression of Ki-67 in tumor tissues (Figure 4B).

Figure 4.

miR-133b reduced the tumor formation ability of lung cancer cells in vivo. H1299 cells were stably transfected with mimic-miR-133b or mimic-NC, then these stably transfected cell lines were injected into nude mice to test their tumorigenesis. A. Tumor weights were measured. B. The expression of Ki-67 in tumor tissues was evaluated by immunohistochemistry (n=10 **P<0.01).

Discussion

miR-133b, a muscle-specific microRNA, appears to exert different roles in different kinds of cancer cells. On one hand, low expression of miR-133b is associated with shorter survival and metastasis rate and functions as a tumor repressor in colorectal cancer [16-18], gastric cancer [19,20], bladder cancer [21], ovarian cancer [22], breast cancer [15], as well as lung cancer [23,24]. However, Li et al. [25] revealed that miR-133b improved tumor-promoting properties in less aggressive prostate cancer LNCaP cells, whereas it acted as a tumor suppressor in more aggressive prostate cancer PC-3 cells. In the present study, we showed that miR-133b was down-regulated in lung cancer cell lines and tissues as compared with the normal lung epithelial cell line BEAS-2B and the para-carcinoma normal lung tissues. Up-regulation of miR-133b to lung cancer H1299 cells significantly reduced cell colony formation, growth, invasion, and tumorigenesis capacities while increasing the cell apoptosis rate and induced G0/G1 phase increase and S phase reduction, suggesting that miR-133b served as a tumor suppressor in lung cancer progression, which was consistent with previous studies [23,24]. Moreover, a study also indicated that high expression of miR-133b was also associated with high chemosensitivity of NSCLC patients to erlotinib in the second or third line [26].

To further explore the molecular mechanisms underlying miR-133b in lung cancer, we asked whether SOX9/β-catenin signaling was involved in miR-133b-mediated the inhibition of lung cancer progression. Results showed that the expression of β-catenin and SOX9 were all decreased when miR-133b was up-regulated in H1299 cells. Additionally, we found that SOX9 up-regulation abolished the roles of miR-133b up-regulation in the repression of cell colony formation and the enhancement of cell apoptosis, demonstrating that miR-133b repressed the progression of lung cancer in a SOX9 down-regulation-dependent manner. SOX9 has been identified to play an important role in lung development, as Schafer et al. [27] demonstrated that infants with mutant SOX9 usually die from respiratory distress shortly after birth and Li et al. [28] revealed that SOX9 was required for lung function recovery after acute lung injury. Until now, SOX9 has been shown to be deregulated and involved in the carcinogenesis of many kinds of cancers. In most cancers such as prostate [29], esophageal [30], colorectal [31], gastric [32] and lung cancers [9,33], SOX9 is characterized to be an oncogene. The high expression of SOX9 predicts a shorter overall survival of lung cancer patients [8]. However, SOX9 functions as a tumor suppressor in melanoma and endometrial carcinoma [34,35], suggesting that SOX9 roles in tumorigenesis are dependent on cell contents. Moreover, SOX9 positively regulates the expression of low density lipoprotein receptor-related protein 6 (LRP6) and T-cell factor 4 (TCF4), two major components of the Wnt/β-catenin pathway, then supports Wnt/β-catenin activity [36]. In lung cancer, SOX9 significantly enhanced A549 cell proliferation and reduced cell apoptosis through activating Wnt/β-catenin signaling [10], suggesting that SOX9/β-catenin signaling plays a key role in lung cancer progression.

Collectively, this study clarifies that miR-133b exerts a tumor-inhibiting role in lung cancer through inactivating SOX9/β-catenin signaling.

Disclosure of conflict of interest

None.