DLD1 and HCT8 cells stored in our laboratory were maintained in RPMI-1640 medium, supplemented with 10% fetal bovine serum (FBS) (both from Gibco, Grand Island, NY, USA) at 37°C under 5% CO₂. We had entrusted Beijing Microread Genetics Co., Ltd., Beijing, China for the identification of the 2 cell lines. The authentication reports (data not shown) showed no cross-contamination of other human cell lines, and 100% matched cell lines are found in ATCC banks with the DLD1 and HCT8 cell lines. Precursor miRNA (pre-miR-375) and precursor negative control (pre-neg) (Applied Biosystems, Foster City, CA, USA) were transfected into the two cell lines using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. Additionally, following this method, the interfering RNA of sp1 messenger RNA (siSP1) and the negative control (si-neg) were transfected into the HCT8 cells.

Total RNA of the cells was extracted using mirVana isolation kit according to the manufacturer's instructions (Applied Biosystems). For validation of miRNA and target gene mRNA expression, quantitative reverse transcriptase-polymerase chain reaction (qRT-PCR) using TaqMan miRNA assays and quantitative RT-PCR using SYBR-Green PCR Master Mix kit (Applied Biosystems) were performed according to the manufacturer's instructions. The expression level of U47 small nuclear RNA was used for miRNA, and GAPDH was used for mRNA as the endogenous controls. All assays were carried out in triplicate.

For the colony formation assay, 1×10³ cells were plated in 6-well plates for each well in medium with 20% FBS for 72 h, and then fixed and stained with crystal violet. Then, the colonies were photographed and counted under a microscope. All experiments were performed in 3 replicates and were repeated 3 times, independently.

DLD1 (2×10³) and HCT8 cells (3×10³) were suspended in RPMI-1640 medium (100 μl) containing 10% FBS and cultured into 96-well plates overnight, and then transfected with pre-miR-375 and pre-neg. Cell proliferation was determined using CCK-8 assay at 0, 1, 2 and 3 days after transfection, respectively, and the absorbance of the samples was measured with a spectrophotometer reader at 450 nm.

DLD1 (2×10⁵) and HCT8 cells (2×10⁵) were cultured into 6-well plates, at 24 h after transfection of pre-miR-375 and pre-neg, as previously described. Migration and invasion assays were performed both using 24-well Transwell migration chambers with 8-μl pore size (Corning Costar, Inc., Corning, NY, USA). DLD1 (1×10⁵) and HCT8 cells (5×10⁴) suspended in 100 μl corresponding culture medium without FBS were loaded into the top chamber, and the bottom chamber contained 600 μl medium with 20% FBS. For the migration assay, the cells were allowed to migrate for 12 h. In addition, for the invasion assay, Transwell wells were pre-coated with Matrigel, and the cells were allowed to invade for 48 h. When both assays were stopped, the cells that migrated or invaded into the bottom chamber were fixed with methanol for 5 min, and then stained with crystal violet (0.05%) for 4 min. The cells that migrated or invaded were photographed and counted under a microscope. In addition, the migration and invasion rates were assessed by the formula: (motile cells transfected with pre-miR-375/motile cells transfected with pre-neg). All experiments were independently repeated in triplicate.

To elucidate the molecular mechanisms involved in the effects of miR-375 on CRC cells, putative miR-375 target genes were predicted through the gateway miRecords (http://mirecords.biolead.org/). To improve the accuracy of the prediction, the genes that were predicted by at least 4 of 11 databases (DIANA-microT, MicroInspector, miRanda, MirTarget2, miTarget, NBmiRTar, PicTar, PITA, RNA22, RNAhybrid and TargetScan) were selected as candidate targets. Wild and mutant putative targets of the Sp1 transcription factor (SP1) 3′-UTR (Fig. 3A) were cloned into pmiReport vector (Ambion, Carlsbad, CA, USA). 293T cells (2×10⁴) were co-transfected with 500 ng of wild or mutant constructs of the SP1 3′-UTR with pre-miR-375 or pre-neg. Each sample was cotransfected with 50 ng of pRL-TK plasmid expressing Renilla luciferase to monitor the transfection efficiency. A luciferase activity assay was performed 48 h after transfection with the dual-luciferase reporter assay system. The relative luciferase activity was normalized to Renilla luciferase activity.

DLD1 and/or HCT8 cells (1×10⁵) were incubated in 6-well plates for 72 h. Then, cell proteins were harvested and homogenized with lysis buffer (Tiangen, Shanghai, China). Proteins from the cells were resolved by 10% SDS-PAGE gel and transferred to a nitrocellulose membrane (Millipore, Billerica, MA, USA). The membrane was incubated with primary antibody at 4°C overnight, then the secondary antibody for 1 h at 37°C and finally visualized by enhanced chemiluminescence (ECL; Thermo Fisher Scientific, Inc., Waltham, MA, USA). The antibodies used were: SP1 (1:2,000; Abcam); matrix metalloproteinase 2 (MMP2) (1:1,000), E-cadherin (1:2,000), vimentin (1:500), snail (1:1,000), β-catenin (1:1,000) [all from Cell Signaling Technology (CST), Beverly, MA, USA]; β-actin (1:1,000) and GAPDH (1:2,000) (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA).

HCT8 cells (5×10⁴) were seeded into 6-well plates containing 13-mm collagen (Sigma, St. Louis, MO, USA)-coated coverslips. After 12 h, the cells were transfected with pre-miR-375 and pre-neg. Then, for another 48 h, the cells on the coverslips were washed using cooled PBS on ice twice with PBS for 5 min. Primary antibodies were diluted in 1% (w/v) BSA/PBS and applied to the coverslips and incubated on ice for 1 h. After 3 washes with PBS, coverslips were incubated with cy3-conjugated secondary antibody diluted in 1% (w/v) BSA/PBS on ice for 30 min. The cells were washed twice with PBS before coverslips were mounted onto glass slides using Gelvatol mounting medium. Slides were viewed using an Olympus IX71 microscope fitted with a U-RFL-T fluorescent lamp, and images were captured and analyzed using DP Controller software (Olympus Corporation, Tokyo, Japan).

Fifteen snap-frozen tumor tissues from 5 patients who were surgically resected at Shanghai Huashan Hospital in 2008 were obtained. Each group included primary cancer (T), lymphatic metastasis (N) and liver metastasis tissues (M) from the same patient. The present study was approved by the Institutional Review Board of Shanghai Medical College, Fudan University and informed consent was obtained from all patients. The RNA extraction and qRT-PCR protocols are referred to as above.

The data are expressed as mean ± SD. Student's t-test was used to compare test samples with the controls. Two-way analysis of variance was used to compare differences among three or more experimental groups. Statistical analyses were performed using SPSS 11.0 software. P<0.05 was considered to indicate a statistically significant result.

To further determine the role of miR-375 in CRC cells, expression of miR-375 was upregulated in DLD1 cells by 40.9-fold (P=0.0040) and in HCT8 cells by 89.2-fold (P=0.0052) (Fig. 1A) compared with the negative control. Overexpression of miR-375 repressed the proliferation of CRC cells as detected by CCK-8 and colony formation assays. CRC HCT8 cells transfected with pre-miR-375 were observed to grow more slowly by 18.8% (P=0.0096) on day 3 (Fig. 1B). The cell colony average area in the DLD1 cells transfected with pre-miR-375 was reduced by 25% (P=0.0201), and by 21.5% in the HCT8 cells (P=0.0136) (Fig. 1C and D).

To validate the involvement of miR-375 in metastasis, migration and invasion assays were performed in the DLD1 and HCT8 cells. In the migration assay, the numbers of migrating cells transfected with pre-miR-375 were significantly reduced, and the migration rate decreased by 33% in the DLD1 cells (P=0.0138), and 36% in the HCT8 cells (P=0.0140) compared with the pre-neg groups (Fig. 2A). In the invasion assay, the numbers of invading cells transfected with pre-miR-375 were significantly reduced, and the invasion rate decreased by 46% in the DLD1 cells (P=0.0479), and 52.3% in the HCT8 cells (P=0.0422) compared with the pre-neg groups (Fig. 2B).

SP1 was found to be one of the candidate target genes, which bears a miR-375 binding sites in its 3′-UTR. To determine whether miR-375 can regulate the expression of SP1, luciferase reporter assay was performed with a vector containing the wild-type (WT) or mutant (MUT) (Fig. 3A) putative SP1 3′-UTR target site downstream of the luciferase reporter pMIR-REPORT vector. When pre-miR-375 was co-transfected, the relative luciferase activity of the reporter containing WT-3′-UTR was significantly suppressed by 28.5% (P=0.0203) compared with that of the reporter containing NC-3′ UTR. In contrast, the luciferase activity of the reporter containing MUT-3′-UTR was unaffected by simultaneous transfection with NC-3′-UTR (Fig. 3B). Then, western blotting and qRT-PCR results showed that overexpression of miR-375 decreased the SP1 expression at the protein (Fig. 3D) but not the messenger RNA level (Fig. 3C).

To further explore the molecular mechanisms of how miR-375 regulates migration and invasion development, we tested whether miR-375 regulates the expression of MMP-2 and EMT-related genes through SP1. The data showed that overexpression of miR-375 in the HCT8 cells reduced SP1 and MMP2 protein (Fig. 4A and B). In addition, immunofluorescence assay showed that upregulation of miR-375 predominantly increased E-cadherin protein on the HCT8 cell membrane (Fig. 4D), even though the whole protein of E-cadherin did not increase obviously (Fig. 4C). Upregulation of miR-375 reduced N-cadherin, vimentin, snail and β-catenin proteins in the HCT8 cells (P<0.05, P<0.01) (Fig. 4E and F).

To further explore whether SP1 regulates MMP2 and EMT-associated genes, SP1 was interfered by siSP1 in the HCT8 cells. In addition, the results showed that interference with siSP1 reduced SP1 protein, and at the same time reduced expression of MMP2 (Fig. 5A and B) and EMT-associated genes (Fig. 5C and D) (P<0.05, P<0.01).

Expression of miR-375 in primary cancer (T), lymphatic metastasis (N) and liver metastasis tissues (M) were examined. In addition, miR-375 was downregulated in all the 5 lymphatic metastasis and liver metastasis tissues compared with primary cancer tissues (Fig. 6A). In addition, miR-375 was significantly downregulated in lymphatic metastasis by 65% (P=0.0163) and liver metastasis tissues by 94.7% (P=0.0000) compared with primary cancer (Fig. 6B).