Colon cacer tissues and mathed adjacent normal tissues were collected from 17 colon cancer patients. All carcinoma samples were obtained at the time of operation and obtained with the informed consent of the patients. All human materials were used in accordance with the policies of the institutional review board at China-Japan Union Hospital of Jilin University, China.

The HT29 and SW480 cell lines were from American Type Culture Collection (ATCC) and cultured under conditions provided by the manufacturer. Twenty-four hours before transfection cells were seeded, then miR-31 antisense locked nucleic acid (LNA anti-miR31) and control oligonucleotides (LAN control) or RhoBTB1 siRNA and control siRNA were transfected with Lipofectamine 2000 (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions. Twenty-four or forty-eight hours later, cells were collected and subjected to further analysis. The human RhoBTB1 siRNA and control siRNA were purchased from Origene.

Total RNA was extracted from cells and tumor tissues by using TRIzol (Invitrogen). For miR-31 detection, stem-loop RT-PCR assay was performed. U6 snRNA was used as an internal control. The following stem-loop primers were used for miRNA-31: 5′-GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACAGCTAT-3′ and U6 snRNA: 5′-GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACAAAATATGGAAC-3′. For RhoBTB1 mRNA detection, M-MLV Reverse Transcriptase system (Promega) was used to reverse-transcribe RNA into cDNA. The qRT-PCR was carried out using the SYBR Green qPCR Master Mix (Tiangen) on ABI 7300 (Applied Biosystems, Foster, CA). Glyceraldehyde 3-phosphate dehydrogenase was used as an internal control. The primers used for detection of miR-31 and U6 were, forward: 5′-GGAGAGGCAAGATGCTGGCA-3′; U6-forward: 5′-CGCAAGGATGACACGCAAATTC-3′, and a universal downstream reverse primer, 5′-GTGCAGGGTCCGAGGT-3′. The primers used for detection of RhoBTB1, forward: 5′-GGAGTGAAGGAGCCTGTGAG-3′; reverse: 5′-TGCCAATGAACCCCTTACTC-3′. Internal control GAPDH primers for qRT-PCR, forward: 5′-ACGGATTTGGTCGTATTGGG-3′, reverse: 5′-TGATTTTGGAGGGATCTCGC-3′.

Twenty-four hours before transfection HT29 cells were seeded into a 24-well plate (2.5–3×10⁴/well), then the cells were cotransfected with Renilla luciferase and luciferase reporter plasmids containing miR-31 or vector control and wild-type or mutated target gene 3′UTR with Lipofectamine 2000 (Invitrogen). Forty-eight hours after transfection, luciferase activities were measured using dual-luciferase reporter assay system (Promega). Firefly luciferase activities were normalized to Renilla luciferase.

After transfection as indicated, cells were harvested and lysed in RIPA lysis buffer [150 mM NaCl, 10 mM Tris, pH 7.5, 1% NP40, 1% deoxycholate, 0.1% SDS, protease inhibitor cocktail (Sigma)] and total protein was measured using the Bradford protein assay (Bio-Rad, Hercules, CA). Equal amounts of protein sample was subjected to 10% SDS-PAGE and transferred to PVDF membrane (Millipore, Billerica, MA, USA), after blocking in 5% skimmed milk for 30 min at room temperature, the membrane was incubated in anti-RhoBTB1 (1:2000, Abcam) for 2 h at room temperature, followed by incubating with horseradish peroxidase-linked secondary antibody for 1 h at room temerature and visualized with ECL (Millipore). β-actin (1:5000, Abcam, Inc., Cambridge, MA) was used as loading control.

Tumor or matched normal tissues were fixed with 4% paraformaldehyde, then the samples were embeded in paraffin and sectioned. Followed incubated with RhoBTB1 antibody (1:500), the bound antibodies were detected with the biotin-streptavidinperoxidase system (Vector Labs, Burlingame, CA, USA) using diaminobenzidine (Sigma-Aldrich) as chromogen.

For in situ hybridization assay, DIG-labeled locked nucleic acid (LNA)-based probe specific for miR-31 (Exiqon, Vedbaek, Denmark) was introduced. The in situ hybridization signal was detected by incubation with horseradish peroxidase (HRP)-conjugated anti-digoxigenin (1:200, Abcam Inc., Cambridge, MA).

Cell proliferation rates were measured using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) reduction assay according to the manufacturer’s instructios. In brief, HT29 cells were seeded into 96-well plates 24 h before transfection and treated as described previously. Then MTT solution was added in each well, and the cells were cultured in incubator for another 4 h. The medium was discarded and 100 μl of DMSO was added to each cell, after shaking for 15 min the absorbance at 595 nm was measured with microplate reader (Bio-Rad).

Pre-warmed 1 ml of 2X RPMI-1640 medium with 20% fetal calf serum (FBS) and 1 ml of 1.2% Bacto-agar (Difco) solution were mixed and transferred onto a 3-cm dish, and then incubated at 4°C for 30 min to allow the bottom agar layer to solidify. Next, mixed 0.5 ml melted 0.6% agar at 50°C with equal volume of 2X medium and stored in 40°C water bath, then 1×10³ cells were added and mixed quickly and poured onto the bottom agar. Cells were left to grown for 2 weeks in the incubator, then cell colonies were observed under a microscope, and colonies of >50 cells were counted.

Data are presented as mean ± SEM. Student’s t-test (two-tailed) was used to compare two groups (P<0.05 was considered significant) unless otherwise indicated (χ² test).

It has been reported by many studies that miR-31 was dysregulated in various cancers, so we examined miR-31 expression level in human colon cancer tissues by using qRT-PCR. As showed in Fig. 1A, compared with the normal colon tissues miR-31 was significantly increased in tumor tissues. ISH assay results further confirmed that miR-31 was induced in colon cancer tissues (Fig. 1B). Moreover, we detected miR-31 expression levels in colon cancer cell lines HT29 and SW480. qRT-PCR results showed that compared with normal colon tissues, miR-31 was also increased in HT29 and SW480 (Fig. 1C), this was in accordance with the previous studies (19–21). These results suggested that miR-31 might be involved in colon tumor progression.

To examine the role of miR-31 in colon cancer development, we inhibited miR-31 in HT29 cells by using locked nucleic acid-modified anti-miR31 antisense (LNA anti-miR31). As showed in Fig. 2A (right), after tranfected with LNA anti-miR31, miR31 was effectively inhibited in HT29 cells. MTT assays showed that inhibition of miR-31 reduced viability of HT29 cells (Fig. 2A, left), which indicates that upregulation of miR-31 might attribute to colon tumorigenesis.

Anchorage-independence of cell growth abitity is a property of cancer cells. To verify the involvement of miR-31 in colon tumorigenesis we next performed soft agarose assays. The results showed, after knockdown of miR-31 colony formation in soft agarose was obviously supressed (Fig. 2B). Collectively the results demostrated that repression of miR-31 inhibited human colon cancer cell viability and anchorage-independent colony formation, which further confirmed that miR-31 plays an important role in colon cancer development.

Since miRNAs regulate translation and degraduation of target mRNAs by direct interactions with 3′UTR of the target mRNAs, we investigated the targets of miR-31 to further uncover the mechanism underlying its tumor promoting effect on colon tumorigenesis. By using online miRNA target prediction databases (http://www.targetscan.org), we hypothesized that RhoBTB1 is a target of miR-31 (Fig. 3A). To investigate whether RhoBTB1 was directly targeted by miR-31, we performed luciferase reporter gene assay in HT29 cells.

Firstly, we constructed luciferase reporter plasmids containing RhoBTB1 3′UTR sequence (wt-RhoBTB1) or with deleted putative miR-31 target sites (mut-RhoBTB1). As shown in Fig. 3B, after co-tranfected with wt-RhoBTB1 plasmid, the luciferase activity of miR-31 transfected cells was significantly reduced compared to control miR tranfected cells, while co-tranfected with mut-RhoBTB1 plasmid did not change the luciferase activity. Taken together, these results indicated that miR-31 directly regulated RhoBTB1 by targeting its 3′UTR.

RhoBTB family belongs to the Rho family of small GTPases, which comprises three members, of which RhoBTB1 and RhoBTB2 have been proposed as tumor suppressors. Here we found that RhoBTB1 was a target of miR-31 in the colon cancer cell line HT29, then we investigated the effect of miR-31 on RhoBTB1. Compared with HT29 transfected with control RNA, RhoBTB1 mRNA level was significantly downregulated in miR-31 mimic-tansfected HT29 cells (Fig. 4A), and western blotting results showed RhoBTB1 protein expression level was also downregulated with miR-31 overexpression (Fig. 4B). These observations suggested that miR-31 upregulation in colon tumor might depress RhoBTB1.

As mentioned above, RhoTBT1 was proposed as a tumor supressor (32), and we also found that upregulation of miR-31 in colon cancer cell line HT29 could depressed RhoBTB1, then we performed ICH and western blotting experiments to examine the expression level of RhoBTB1 in colon cancer tissues. As showed in Fig. 5, RhoBTB1 was clearly decreased compared to normal colon tissues. These results suggested that the reduction of RhoBTB1 might attribute to colon tumorigenesis.

To further confirm that miR-31-induced colon tumor progression is mediated by RhoBTB1, we knockdown RhoBTB1 in HT29 cells by RNAi. As shown in Fig. 6B, RhoBTB1 mRNA was effectively inhibited after RhoBTB1 siRNA transfected, and MTT assay results showed HT29 proliferation was induced with suppression of RhoBTB1. Soft agarose colony forming assay results showed inhibition of RhoBTB1 promoted colon cancer cell clonal growth. These data proved that the miR-31 promoted tumor development at least partly through suppressing tumor supressor RhoBTB1.