Human HT-29 and SW-620 CRC cell lines were obtained from the Cell Center of Xiangya School of Medicine, Central South University (Hunan, China). The human CCD-18Co healthy colon cell line was obtained from the American Type Culture Collection (Manassas, VA, USA). RPMI-1640 was obtained from Invitrogen Life Technologies (Carlsbad, CA, USA). Fetal bovine serum (FBS) was purchased from Sijiqing Biological Engineering Materials Co., Ltd. (Hangzhou, China). Cells were cultured at 37°C in a humidified incubator with 5% CO₂ and maintained in RPMI-1640 supplemented with heat-inactivated 10% FBS (Invitrogen, Carlsbad, CA, USA), 100 U/ml penicillin and 100 μg/ml streptomycin (Sijiqing Biological Engineering Materials Co., Ltd). Serum-free Opti-MEM^® medium was obtained from Invitrogen Life Technologies. Protease inhibitor was obtained from AppliChem (Gatersleben, Germany). miR-133b mimics and the miR-133b inhibitor were synthesized by Shanghai GenePharma Co., Ltd (Shanghai, China). The transfection reagent Lipofectamine 2000 was obtained from Invitrogen Life Technologies. Primers for CTGF, miR-133b and U6-snRNA were synthesized by Changsha YRBIO Co., Ltd (Changsha, China). CTGF goat polyclonal antibody was obtained from Santa Cruz Biotechnology (Dallas, TX, USA). GAPDH rabbit polyclonal antibody was obtained from Proteintech Group, Inc. (Chicago, IL, USA). The horseradish peroxidase (HRP)-goat anti-rabbit antibody and the HRP-rabbit anti-goat antibody used for western blot analysis were obtained from Zhong Shan Goldenbridge Biotechnology Co., Ltd (Beijing, China). The PCR 2.1 vector was purchased from Invitrogen Life Technologies.

The miR expression vector pYr-pri-miRNA (YRBIO Co. Ltd) was used to construct a pYr-pri-Homo sapiens (hsa)-mir-133b plasmid. Human genome DNA was used as the template. A fragment of ~600 bp containing pri-miR-133b was cloned into the pYr vector. HT-29 and SW-620 cells were seeded at 3×10⁴ cells per well in 12-well plates and transfected the following day in Opti-MEM medium with pYr-pri-hsa-mir-133b plasmid (80 nmol/l; Shanghai GenePharma Co., Ltd.) using Lipofectamine 2000. Cells were harvested three days prior to transfection.

The CRC cell lines and transfected cell lines were sent to CapitalBio (Beijing, China) for microarray hybridization. Cells were centrifuged in RNase-free tubes treated with TRIzol^® reagent (Invitrogen Life Technologies) and stored at −80°C prior to RNA extraction, which was performed according to the manufacturer’s instructions (Invitrogen Life Technologies). The RNA was purified using the NucleoSpinH RNA clean up kit (Macherey-Nagel, Düren, Germany). RNA concentration was assessed by Nanodrop 2000 spectrophotometry (Thermo Fisher Scientific, Waltham, MA, USA). RNA quality was determined by formaldehyde denaturation electrophoresis, and only those samples showing no degradation (ratios approaching 2:1 for the 28S and 18S bands) were used to generate labeled targets. Each RNA sample was hybridized to one Roche NimbleGen Porcine Genome Expression Array (Roche, Basel, Switzerland). Briefly, double-stranded cDNA was synthesized from 6 mg total RNA using a T7-oligo (dT) primer. The cDNA was further purified and converted into cRNA using an in vitro transcription reaction. 5 mg cRNA was reverse transcribed to cDNA, fragmented and then labeled with Cy5-dCTP and Cy3-dCTP (GE Healthcare, Little Chalfont, UK) using Taq DNA enzyme (Carolina Biosystems, Beijing, China). These labeled cDNA fragments were hybridized for 16 h at 45°C using the 2X GE Hybridization Buffer (Roche Genome Expression Array kit; Roche). The GeneChips were then washed, stained and scanned with an Agilent G2565CA Microarray Scanner (Agilent Technologies, Santa Clara, CA, USA).

Agilent Feature Extraction software (version 11.0.1.1; Agilent Technologies) was used to analyze the acquired array images. Quantile normalization and subsequent data processing were conducted using the GeneSpring GX v12.0 software package (Agilent Technologies). Following quantile normalization of the raw data, genes with ≥1 out of 2 samples with raw data in the microarray were selected for further data analysis. Differential expression of genes was considered significant if the false discovery rate (FDR) was calculated to the correct P-value (controlling the expected FDR to no more than 5%) and the fold change >|1.5|. Hierarchical clustering was performed using the Agilent GeneSpring GX software (version 12.0; Agilent Technologies).

Total RNA was isolated using E.Z.N.A.^® Total RNA kit II (Omega Bio-Tek Inc., Norcross, GA, USA). miRNA was isolated using E.Z.N.A. PF miRNA Isolation kit (Omega Bio-Tek Inc.). Large RNA reverse transcription was performed using the RevertAid first-strand cDNA synthesis kit (Thermo Fisher Scientific Inc., Vilnius, Lithuania). miRNA reverse transcription was performed using an All-in-One™ miRNA qRT-PCR Detection kit (GeneCopoeia, Inc., Rockville, MD, USA). qPCR was performed using Real-Time Quantitative PCR SYBR Green detection reagent (Cowin Biotech Co., Ltd., Beijing, China). miRNA qRT-PCR was performed using All-in-One™ miRNA qRT-PCR Detection kit (GeneCopoeia, Inc.). Primers utilized for cDNA amplification are summarized in Table I. The following reaction conditions were used: An initial 10 min at 95°C; then 40 cycles of 20 sec at 94°C, 15 sec at 59°C and 30 sec at 72°C. The standard PCR reaction mixture then underwent a final extension for 5 min at 72°C, while the quantitative PCR mixture was subjected to melting curve analysis to validate the reaction product specificity. The relative expression of CTGF was normalized to that of GAPDH using the 2^−ΔΔCT-method. The relative expression of miR-133b was normalized to that of U6-snRNA using the 2^−ΔΔCT-method. Each sample was amplified in triplicate.

SW-620 and HT-29 cells were seeded in a 96-well plate at 2×10⁴ cells per well and incubated overnight. miR-133b was transfected into SW-620 and HT-29 cells using Lipofectamine 2000. At 24 h after transfection, cells were transfected with the pYr-MirTarget Renilla luciferase plasmid in combination with the CTGF 3′-UTR expression plasmid (Cignal Reporter Assay kit; Qiagen, Hilden, Germany). Cells were harvested at 48 h following transfection and assayed for luciferase activity using the Dual-Luciferase Reporter Assay System (Promega Corp., Madison, WI, USA). Luciferase activity was normalized against that of Renilla luciferase. Luminescence was quantified with a luminometer (Sirius single tube luminometer, Belthold, Wildbad, Germany). All experiments were performed in triplicate.

The human CRC tissues (histologically confirmed as adenocarcinoma of the colon) and adjacent non-cancerous tissue counterparts used for RT-qPCR and western blotting were obtained from patients who underwent surgical resection of CRC at the Third Affiliated Hospital of Central South University (Hunan, China) once written informed consent had been obtained. All specimens were snap frozen in liquid nitrogen for 24 h and then stored at −80°C prior to further processing. The study was approved by the ethics committee of the Third XiangYa Hospital of Central South University (Hunan, China). Patient consent was obtained from the patient as well as the patient’s family.

Total protein from tissues samples and the cultured cell lines was harvested in a radioimmunoprecipitation assay lysis buffer (Sigma-Aldrich Co, St. Louis, MO, USA). A volume of extract equivalent to 100 μg total protein was separated by 7.5% SDS-PAGE (Well Biological Co., Ltd, Changsha, China) and transferred onto polyvinylidene difluoride membranes (Invitrogen Life Technologies). The membranes were incubated with Tris-buffered saline and Tween-20 (TBST) (Well Biological Co., Ltd) containing 5% skimmed milk at 4°C overnight. Membranes were then incubated with the following primary antibodies: goat anti-CTGF polyclonal antibody (1:500) and GAPDH rabbit polyclonal antibody (1:2,000), at room temperature for 2 h. The membranes were then washed three times with TBST and incubated with HRP-conjugated rabbit anti-goat antibody (1:10,000) and HRP-conjugated goat anti-rabbit antibody (1:2,000) for 1 h. Membranes were then washed three times with TBST and treated with western blotting luminal reagent (Pierce ECL Western Blotting kit; Thermo Fisher Scientific Inc., Vilnius, Lithuania) for band visualization, the results were obtained using an ImageQuant LAS 350 (GE Healthcare) and quantified by densitometry. Blots were performed in triplicate.

All statistical analyses were conducted using the SPSS version 16.0 software package (SPSS, Inc., Chicago, IL, USA). Data are expressed as the mean ± standard deviation. Differences between groups were compared using analysis of variance and two-tailed t-tests. The χ² test was used to analyze the significance of the association between miR-133b expression and clinicopathological characteristics. Each experiment was performed independently at least twice and obtained similar results. P<0.05 was considered to indicate a statistically significant difference.

In order to investigate whether there was an effect of miR-133b on the expression of individual genes, cluster analysis was performed to organize genes into clusters based on the similarities of their expression levels. Statistical analysis using an unpaired t-test demonstrated that a total of 103 genes were differentially expressed (72 upregulated and 31 downregulated) between HT-29 and SW-620 cells (Fig. 1). The 20 genes downregulated to the greatest extent were differentially expressed in miR-133b-transfected HT-29 and SW-620 cells, compared with the corresponding type of untransfected cells, with the fold change calculated based on the mean value of the two cell groups (Table II). CTGF was identified by microarray analysis as a potential target candidate of miR-133b interaction (log FC=−4.41, FDR<0.05) and was selected for further study.

As miRNAs have been implicated in the repression of gene expression, it was hypothesized that miR-133b may target a negative regulator of the expression of CTGF. Using computational screening (TargetScan; Whitehead institute for Biomedical Research, www.targetscan.org), it was shown that the 3′UTR of CTGF contains a potential binding site for miR-133b (Fig. 2A). The transfection efficiency of synthetic miR-133b was evaluated in the human HT-29 and SW-620 CRC cell lines. As shown in Fig. 2B, RT-qPCR analysis demonstrated that miR-133b levels in miR-133b-transfected cells were significantly higher than those in untransfected or control-transfected cells (HT-29 cells, P<0.01; SW-620 cells, P<0.01).

In order to investigate whether miR-133b expression influenced CTGF expression, mRNA and protein levels of CTGF were examined in miR-133b-transfected cell lines (Fig. 2C–E). miR-133b transfection significantly reduced the levels of CTGF mRNA compared with those in untransfected or control-transfected cells (SW-620 cells, P<0.01; HT-29 cells, P<0.05; Fig. 2C). Western blot analysis showed that the protein levels of CTGF were also downregulated following overexpression of miR-133b in SW-620 and HT-29 cells, compared with those in the controls (Fig. 2D). Quantification of protein bands revealed a four-fold decrease in CTGF protein following overexpression of miR-133b (P<0.05; Fig. 2E).

In order to determine whether the observed reduction in CTGF expression is directly driven by miR-133b, reporter gene assays were conducted in HT-29 and SW-620 cells. The entire 3′UTR of CTGF was cloned into the pYr-MirTarget-control vector, creating a luciferase reporter gene containing the seed match for miR-133b. In addition, a mutated reporter construct was generated, in which the miR-133b seed match sequence was deleted from the 3′UTR of CTGF. Co-transfection with the CTGF wild-type-3′UTR construct and miR-133b yielded a significantly reduced relative luciferase activity (SW-620 cells, P<0.05; HT-29 cells, P<0.05; Fig. 2F). However, the luciferase activity of the reporter construct that had been mutated at the specific miR-133b target site was not reduced in comparison with that in untransfected cells, when simultaneously transfected with miR-133b (SW-620 cells, P<0.05; HT-29 cells, P<0.05; Fig. 2F). This suggested that the 3′-UTR of CTGF is a functional target site for miR-133b in HT-29 and SW-620 cancer cells.

In order to gain insight into the biological functions of miR-133b and CTGF in human colorectal carcinogenesis, the expression of miR-133b in human CRC tissues and paired healthy counterparts was analyzed using RT-qPCR. Following this, the expression levels of miR-133b and CTGF mRNA were measured using comparisons with the human CCD-18Co healthy colon cell line as a reference. The relative expression levels of miR-133b in tumor specimens compared with those in the CCD-18Co cells (0.55±0.41) were significantly lower than the relative expression levels in non-tumor specimens compared with those in CCD-18Co cells (1.32±0.24; P<0.001). The relative expression levels of CTGF mRNA in tumor specimens (1.46±0.36) were significantly higher than those (0.57±0.32) in the corresponding non-tumor specimens (P<0.001; Fig. 3A).

The effect of miR-133b expression on that of CTGF was assessed using data obtained from RT-qPCR. A significant negative correlation was observed between CTGF mRNA and miR-133b expression (n=71, r=−0.385, P<0.01, Pearson’s correlation; Fig. 3B). These data demonstrated a reciprocal regulation of miR-133b and CTGF, suggesting that downregulation of miR-133b may lead to CTGF overexpression in CRC tissues.

Furthermore, western blot analysis was conducted in order to determine whether protein expression of CTGF was consistent with mRNA expression. As hypothesized, the alterations in CTGF protein levels in tumor samples compared with those in non-tumor samples were in accordance with changes in mRNA levels in these groups. Protein levels of CTGF were verified in CRC tissues (P<0.05; Fig. 3C and D).

In order to evaluate the correlation between miR-133b and CTGF mRNA expression as well as clinicopathological factors relevant to CRC, the samples were divided into groups according to whether they exhibited low or high expression using the median values of miR-133b and CTGF mRNA levels. Patients with expression levels of miR-133b in tumor tissues below the median value of 0.47 were assigned to the low expression group (n=36), while patients with miR-133b expression levels in tumor tissue ≥0.47 were assigned to the high expression group (n=35). In addition, patients with tumor CTGF mRNA expression levels below the median value of 1.49 were assigned to the low expression group (n=36) and patients with tumor CTGF mRNA expression levels ≥1.49 were assigned to the high expression group (n=35). Current staging for colorectal cancer is based on the 7th edition of AJCC Cancer Staging Manual (16). There was a significant difference between the groups with low and high expression of miR-133b in terms of cell differentiation, lymph node metastasis and clinical stage, with a low expression being associated with poor differentiation, lymph node metastasis and a more advanced stage (P<0.05; Table III). However, no significant difference was found between the groups with low and high expression of miR-133b in terms of patient age, gender, tumor site or extramural vascular invasion status (P>0.05).

Furthermore, a high expression of CTGF was associated with advanced clinical stage (stage III or IV vs. I or II) and lymph node metastasis (P<0.05; Table III). No significant difference was detected between the groups with low or high expression of CTGF mRNA in terms of patient age, gender, tumor site, tumor differentiation or extramural vascular invasion status (P>0.05). The association between the levels of miR-133b and CTGF mRNA expression and the clinicopathological characteristics is summarized in Table III. In brief, the data suggest that miR-133b and CTGF may be candidate therapeutic targets in colorectal cancer, in particular in stage III and stage IV tumors.