A total of 43 pairs of CRC tissues and corresponding adjacent normal tissues were obtained form CRC patients treated with surgery at The Second Hospital of Shandong University (Jinan, China) between February 2011 and October 2015. None of these patients received either preoperative chemotherapy or radiotherapy before surgery. Tissues were snap-frozen in liquid nitrogen and stored at −80°C until further use. This study was approved by the Ethics Review of The Second Hospital of Shandong University and written informed consent was also obtained from each patient.

CRC cell lines SW620, SW480, HCT116, HT29, and LOVO were purchased form Cell bank of Chinese Academy of Sciences (Shanghai, China). Normal human colon epithelium cell line (FHC) and human embryonic epithelial HEK293T cell line were obtained from American Type Culture Collection (Manassas, VA, USA). All cells were grown in L-15 for W620 and SW480, McCoys 5A for HCT116 and HT29, Dulbecco's modified Eagle's medium (DMEM) for LOVO, FHC and HEK293T cell supplemented with 10% heat-inactivated fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA), 100 U/ml penicillin and 100 g/ml streptomycin (Gibco; Thermo Fisher Scientific, Inc.), at 37°C in a humidified incubator containing 5% CO₂.

TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA, USA) was used to extract total RNA from tissues and cells, according to the manufacturer's instructions. The purity and concentration of total RNA was determined with a NanoDrop^® ND-1,000 spectrophotometer. For the detection of miR-329 expression, cDNA was synthesized using Takara PrimeScript™ First Strand cDNA Synthesis kit (Takara Bio, Inc., Otsu, Japan) and TaqMan MicroRNA Assays kit (Applied Biosystems; Thermo Fisher Scientific, Inc.) was utilized to measure miR-329 expression. To quantification TGF-β1 mRNA expression, reversing transcribed RNA into cDNA was performed using M-MLV Reverse Transcription system (Promega Corporation, Madison, WI, USA). The levels of TGF-β1 mRNA expression were quantified with SYBR Premix Ex Taq (Takara Bio, Dalian, China). Relative expression of miRNA and mRNA were normalized against U6 and GAPDH, respectively. The primers sequences were as follows: miR-329 forward, 5′-AACACACCTGGTTAACCTCTTT-3′ and reverse, 5′-CAGTGCGTGTCGTGGAGT−3′; U6 forward, 5′-CTCGCTTCGGCAGCACATATACT-3′ and reverse, 5′-ACGCTTCACGAATTTGCGTGTC-3′; TGF-β1 forward, 5′-GGACACCAACTATTGCTTCAG-3′ and reverse, 5′-TCCAGGCTCCAAATGTAGG-3′; GAPDH forward, 5′-GGACCTGACCTGCCGTCTAG-3′ and reverse, 5′-GTAGCCCAGGATGCCCTTGA-3′. Normalization and fold changes were calculated using the 2^−ΔΔCq method (27).

Total protein was extracted from tissues and cells using cold radioimmunoprecipitation assay buffer (Beyotime Institute of Biotechnology, Shanghai, China) containing protease and phosphatase inhibitors. A bicinchoninic acid protein assay kit (Bio-Rad Laboratories, Inc., Shanghai, China) was used to detect the protein concentration. Equal amounts of protein were separated using 10% SDS-PAGE, transferred onto a PVDF membranes (EMD Millipore, Billerica, MA, USA) and then blocked with Tris-buffered saline containing 0.05% Tween-20 (TBST) containing 5% non-fat dry milk for 1 h at room temperature. Subsequently, the membranes were incubated with following primary antibodies at 4°C overnight: mouse anti-human monoclonal TGF-β1 antibody (1:1,000 dilution; cat. no. sc-130348; Santa Cruz Biotechnology, Inc., Dallas, TX, USA) and mouse anti-human monoclonal GAPDH antibody (1:1,000 dilution; cat. no. sc-51907; Santa Cruz Biotechnology, Inc.). The membranes were washed with TBST for three times at room temperature and probed with the goat anti-mouse horseradish peroxidase-conjugated secondary antibody (1:1,000 dilution; cat. no. sc-2005; Santa Cruz Biotechnology, Inc.) at room temperature for 2 h. The protein of interest was visualized using Pierce™ ECL Western Blotting Substrate (Pierce; Thermo Fisher Scientific, Inc.) and analyzed using Quantity One software version 4.62 (Bio-Rad Laboratories, Inc.).

miR-329 mimics and negative control miRNA mimics (miR-NC) were synthesized by GenePharma Co., Ltd., (Shanghai, China). pcDNA3.1-TGF-β1 and blank pcDNA3.1 vector were designed and purchased from Guangzhou RiboBio Co., Ltd. (Guangzhou, China). Cell transfection was performed using Lipofectamine 2,000 (Invitrogen) according to the manufacturer's protocol. After incubation at 37°C in the presence of 5% CO₂ for 6–8 h, the medium was replaced by culture medium containing 10% FBS.

Cell proliferation was determined with CCK-8 assay (Dojindo, Kumamoto, Japan). Cells were seeded into 96-well plates at a density of 3×10³ cells per well. Twenty-four h after being planted into 96-well plates, cells were transfected with miR-329 mimics, miR-NC, pcDNA3.1-TGF-β1 or pcDNA3,1. Cell proliferation was evaluated following 0, 24, 48, and 72 h of incubation at 37°C in a humidified incubator containing 5% CO₂. Briefly, 10 µl of CCK-8 reagent was added to each well and then incubated at 37°C for additional 2 h. The absorbance of each well at 450 nm was measured with the ELx800 Absorbance Microplate Reader (BioTek, Winooski, VT, USA).

Transwell chambers with polycarbonate filters with 8-µm pore size (BD Biosciences, Franklin Lakes, NJ, USA) were adopted to analyze the invasion abilities of CRC cells. After transfection 48 h, 5×10⁴ transfected cells in FBS-free culture medium were plated in the upper chamber coated with Matrigel (BD Biosciences, San Jose, CA, USA), and 600 µl of DMEM containing 20% FBS were added into the lower chamber as chemoattractant. Cells were then incubated at 37°C in a humidified incubator containing 5% CO₂ for 48 h, the non-invading cells were removed with a cotton tip. The invasive cells were fixed with 95% ethanol, stained with 0.5% crystal violet, washed, photographed and counted in five randomly selected fields for each well under an inverted microscope (CKX41; Olympus, Tokyo, Japan).

Potential target genes of miR-329 were analyzed using the miRNA target prediction program PicTar (http://pictar.mdc-berlin.de/) and TargetScan (www.targetscan.org).

Dual-luciferase reporter assay was performed in order to clarify whether TGF-β1 is a direct target gene of miR-329. The luciferase reporter vectors pGL3-TGF-β1-3′-UTR wild-type (WT) and pGL3-TGF-β1-3′-UTR mutant (MUT) were synthesized and purchased from GenePharma. HEK293T cells were cotransfected with miR-329 mimics or miR-NC, and pGL3-TGF-β1-3′-UTR WT or pGL3-TGF-β1-3′-UTR MUT using Lipofectamine^® 2000. The pRL-SV40 vector carrying the Renilla luciferase gene was used as an internal control. After incubation at 37°C in a humidified incubator containing 5% CO₂ for 48 h, the luciferase activity was detected using Dual-Luciferase Reporter Assay System (Promega GmbH, Mannheim, Germany).

The data was expressed as the mean ± standard error. Student's t-test or one-way ANOVA was used to analyze the significance of differences between groups using a windows-based SPSS 13.0 software (SPSS, Inc, Chicago, IL, USA). P<0.05 was considered to indicate a statistically significant difference.

In this study, we first performed RT-qPCR to measure miR-329 expression level in a total of 43 pairs of CRC tissues and corresponding adjacent normal tissues. As shown in Fig. 1A, miR-329 expression was significantly low in CRC tissues compared with that in corresponding adjacent normal tissues (P<0.05). We further investigate the association between miR-329 expression level and clinicopathological factors of patients with CRC. As indicated in Table I, miR-329 expression levels in CRC tissues were correlated with TNM stage (P=0.004) and lymph node metastasis (P=0.009), but not with age (P=0.977), gender (P=0.658), or tumor size (P=0.224).

We then detected miR-329 expression in CRC cell lines (SW620, SW480, HCT116, HT29, LOVO) and a normal human colon epithelium cell line (FHC) using RT-qPCR. The results showed that miR-329 was downregulated in all examined CRC cell lines, when compared with that in FHC (Fig. 1B, P<0.05). Especially, expression level of miR-329 in the SW480 and HCT116 cell lines was much lower than in other cell lines. Hence, SW480 and HCT116 cells were chosen for further experiment assays.

To investigate the biological roles of miR-329 in CRC, SW480 and HCT116 cells were transfected with miR-329 mimic or miR-NC, and then the effects of miR-329 on CRC cell proliferation and invasion were evaluated. RT-qPCR was used to evaluate the transfection efficiency, and confirmed that miR-329 was markedly upregulated in SW480 and HCT116 cells transfected with miR-329 mimics (Fig. 2A, P<0.05).

CCK-8 assay was performed, and revealed that increased expression of miR-329 suppressed proliferation of SW480 and HCT116 cells (Fig. 2B, P<0.05). Subsequently, Transwell invasion assay was utilized to test the role of miR-329 in cell invasion ability, and the results showed that upregulation of miR-329 decreased SW480 and HCT116 cells abilities (Fig. 2C, P<0.05). Thus, those findings suggested that miR-329 could inhibit CRC cell tumorigenicity and progression in vitro.

The molecular mechanism underlying the tumor suppressive roles of miR-329 was further investigated in CRC. Bioinformatics analysis was conducted to analyze the potential targets of miR-329. A great deal of different targets were predicted; of these genes, TGF-β1 was selected as a potential target of miR-329 since TGF-β1 was upregulated in CRC tissues and contributed to CRC formation and progression (28) (Fig. 3A). To test whether TGF-β1 expression is direct regulated by miR-329, the 3′-UTR luciferase reporter vectors containing wild type or mutant TGF-β1 3′-UTR were synthesized, and co-transfected into HEK293T cells with miR-329 mimics or miR-NC. The data of Dual-luciferase reporter assay showed that the luciferase activity of pGL3-TGF-β1-3′-UTR WT was obviously reduced in miR-329 mimics transfected cells (Fig. 3B, P<0.05); while, there was no significant difference in the luciferase activity when cells were transfected with pGL3-TGF-β1-3′-UTR Mut together with miR-329 mimics or miR-NC. Additionally, RT-qPCR and western blotting were utilized to examine TGF-β1 expression in SW480 and HCT116 cells following transfection with miR-329 mimics or miR-NC. As shown in Fig. 3C and D, both the mRNA and protein levels of TGF-β1 were negatively regulated by miR-329 (both P<0.05). Collectively, TGF-β1 was a direct target of miR-329 and was negatively regulated by miR-329 in CRC.

TGF-β1 was identified as a direct target of miR-329 in CRC; hence, we evaluated the association between TGF-β1 and miR-329 expression level in CRC. Firstly, the expression level of TGF-β1 in CRC tissues and corresponding adjacent normal tissues was detected. As shown in Fig. 4A, CRC tissues showed an increased TGF-β1 mRNA and protein expression compared with that in corresponding adjacent normal tissues (both P<0.05). Furthermore, the correlation between miR-329 and TGF-β1 mRNA in CRC tissues was evaluated with Spearman's correlation analysis. The analysis indicated that TGF-β1 mRNA was inverse related to miR-329 expression in CRC tissues (Fig. 4B; r=−0.5613, P<0.0001).

Next, a rescue experiment was performed to further confirm that TGF-β1 was the functional target of miR-329 in CRC. pcDNA3.1-TGF-β1 was adopted to increase TGF-β1 expression in SW480 and HCT116 cells confirmed by Western blot (Fig. 5A, P<0.05). CCK-8 assay revealed that TGF-β1 overexpression in SW480 and HCT116 cells could partially rescue the suppressive effect of miR-329 on cell proliferation (Fig. 5B, P<0.05). In addition, TGF-β1 restoration reversed the effect of miR-329 on the invasion capability of SW480 and HCT116 cells. These results suggested that miR-329 exerts tumor suppressive roles in CRC partially through inhibition of TGF-β1 expression.