The CRC cell lines (SW620, SW480, DLD1, HCT116, LoVo, and HT-29) were purchased from the American Type Culture Collection (Manassas, VA, USA). The normal human colon mucosal epithelial cell line (NCM460) and the CRC cell line (KM12) were purchased from the BeNa Culture Collection (Beijing, China). The normal human colon mucosal epithelial cell line (NCM460) and seven CRC cell lines (SW620, SW480, DLD1, HCT116, KM12, LoVo, and HT-29) were grown in Dulbecco's modified Eagle's medium (Invitrogen; Thermo Fisher Scientific, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (Invitrogen; Thermo Fisher Scientific) and 100 units of penicillin-streptomycin.

This study was conducted on a total of 125 paraffin-embedded and archived CRC samples, which were diagnosed histopathologically at the Affiliated Shenzhen Sixth Hospital of Guangdong Medical University from 2003 to 2012. Informed patient consent and approval from the Institutional Research Ethics Committee of the Affiliated Shenzhen Sixth Hospital of Guangdong Medical University was obtained for use of these clinical materials for research purposes. Clinical information regarding the samples is summarized in Table I. Ten CRC tissue samples and their matched adjacent non-cancerous colorectal tissues were frozen and stored in liquid nitrogen until further use.

The 3′-UTR regions of human TRAF5 (from 1801 to 2211 nt, containing a predicted conserved miR-873 binding site) and TAB1 (from 3288 to 3700 nt, containing a predicted conserved miR-873 binding site) were generated by PCR and cloned into the modified pGL3-control luciferase reporter plasmid (Promega Corporation, Madison, WI, USA). Primer sequences were as follows: TRAF5-3′UTR sense, 5′-ACTCCGCGGATCCCAGATGATTAAATT-3′ and antisense, 5′-CTAACTGCAGTTCCTTGTTCTGGGATCAC-3′; TAB1-3′UTR sense, 5′-ACTCCGCGGCGGAGGTCCTGGCCCTCAG-3′ and antisense, 5′-CTAACTGCAGCCCATGGAGGAAACAACAGGGAG-3′. Point mutations in the putative miR-873-binding seed regions of the TRAF5-3′-UTR and TAB1-3′-UTR constructs were created using a Stratagene QuikChange Mutagenesis kit (Stratagene; Agilent Technologies Inc., La Jolla, CA, USA). A miR-873 mimic, miR-873 inhibitor, and the negative control (NC) were purchased from Guangzhou RiboBio Co., Ltd. (Guangzhou, China). The sequences were as follows: miR-873 mimic, 5′ GCAGGAACUUGUGAGUCU CCU-3′; miR-873-NC sense, 5′-UUUGUACUACACAAAAGUACUG-3′; miR-873 inhibitor, 5′-AGGAGACUCACAAGUUCCUGC-3′. The miR-873 mimic or miR-873 inhibitor (the miR-873 inhibitor is a locked nucleic acid (LNA)/O-Methyl oligo (OMe) modified antisense oligonucleotide designed specifically to bind to and inhibit the endogenous miR-873 molecule) was transfected into cells using the Lipofectamine 2000 reagent (Invitrogen; Thermo Fisher Scientific) according to the manufacturer's instructions.

Total RNA from the indicated tissues or cells was extracted using the TRIzol reagent (Life Technologies; Thermo Fisher Scientific, Carlsbad, CA, USA), according to the manufacturer's instructions. Complementary DNA (cDNA) was amplified and quantified on an ABI PRISM 7500 Sequence Detection system (Applied Biosystems; Thermo Fisher Scientific, Foster City, CA, USA) using SYBR Green I (Roche Diagnostics, Grenzach-Wyhlen, Germany). miRNA quantification was determined using Bulge-loop™ miRNA quantitative reverse transcription PCR (qRT-PCR) Primer Set (one RT primer and a pair of qPCR primers for each set) specific for U6 and miR-873 that were designed and synthesized by Guangzhou RiboBio Co., Ltd. The catalog numbers of these primers were as follows: miR-873 RT primer (ssD809230648), miR-873 forward primer (ssD090525061), miR-873 reverse primer (ssD089261711), U6 RT primer (ssD0904071008), U6 forward primer (ssD0904071006), and U6 reverse primer (ss D0904071007). The quantitative PCR (qPCR) conditions were as follows: incubation at 50°C for 2 min, 95°C for 10 min, followed by 40 cycles at 95°C for 15 sec, and 60°C for 1 min. The expression of miR-873 was defined based on the quantification cycle (Cq), and the relative fold changes between normal human colon mucosal epithelial cell line and the CRC cell lines, and between CRC tissues and their tumor-adjacent tissues (TATs), were calculated according to the formula 2^{−[(Cq of miR-873) - (Cq of U6)]} after normalization to the expression of U6 small nuclear RNA as a reference (30). REST-MCS beta software version 2 was used to further analyze the qPCR data.

Cells (2×10³/well) were seeded into 96-well plates. Transfection was performed 12 h later. At the indicated time-points, 100 µl of sterile 3-(4,5-Dimethyl-2-thiazolyl)-2, 5-diphenyl-2H-tetrazoliumbromide (MTT) dye (0.5 mg/ml; Sigma-Aldrich, St. Louis, MI, USA) was added and incubated for 4 h at 37°C. The culture medium was subsequently removed, and 150 µl of dimethyl sulfoxide (DMSO; Sigma-Aldrich) was added. The absorbance was assessed at a wavelength of 490 nm. All experiments were performed in triplicate.

Cells (8×10²/well) were seeded into 6-well plates and cultured for 10 days. The colonies were stained with 1% crystal violet for 30 sec after incubation with 10% formaldehyde for 5 min. Colonies were counted only if they contained more than 50 cells, according to the established criteria for colony formation (31–33).

Cell lysates were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene fluoride (PVDF) membranes (EMD Millipore, Billerica, MA, USA). The membranes were probed with antibodies against TRAF5 (anti-TRAF5 antibody; 1:1,000; mouse, polyclonal; cat. no. SAB1409766) and TAB1 (anti-TAB1 antibody; 1:500; rabbit, polyclonal; cat. no. SAB4301002; both from Sigma-Aldrich) overnight at 4°C, followed by incubation with horseradish peroxidase-conjugated secondary antibodies (1:2,000; goat; cat. no. 7074 and 1:2,000; horse; cat. no. 7076; Cell Signaling Technology, Danvers, MA, USA) for 1 h at 20°C. The membranes were stripped and re-probed with an anti-α-tubulin antibody (Sigma-Aldrich) as a loading control.

Cells (2×10⁴/well) were seeded in triplicate in 48-well plates and allowed to settle for 24 h. Next, 100 ng of the luciferase reporter plasmids or the control plasmid, both with 1 ng of pRL-TK Renilla plasmid (Promega Corporation), were transfected into cells using the Lipofectamine 2000 reagent (Invitrogen; Thermo Fisher Scientific), according to the manufacturer's recommendation. Luciferase and Renilla signals were assessed 24 h after transfection using a Dual Luciferase Reporter assay kit (Promega Corporation), according to the manufacturer's instructions.

Histology was performed to quantify Ki67 expression in 10 paraffin-embedded human CRC samples. IHC was performed on sections using an anti-Ki67 antibody (1:1; mouse, monoclonal; cat. no. IR62661-2; Dako; Agilent Technologies, Inc., Glostrup, Denmark). H&E staining was performed using Mayer's hematoxylin solution. Immunostaining of the sections was quantified and scored independently by two observers based on both the proportion of positively-stained tumor cells and the intensity of staining. The proportion of tumor cells enriched for Ki67 was scored as follows: 0 (no positive tumor cells), 1 (<10% positive tumor cells), 2 (10–50% positive tumor cells), and 3 (>50% positive tumor cells). The intensity of staining was scored according to the following criteria: 0 (no staining), 1 (weak staining, light yellow), 2 (moderate staining, yellow brown), and 3 (strong staining, brown). The staining index (SI) was calculated as the proportion of positive tumor cells × the staining intensity score. Using this scoring method, we evaluated the expression of Ki67 by determining the SI, which was scored as 0, 1, 2, 3, 4, 6 and 9.

All values are presented as the means ± standard deviation (SD). Student's t-test was used to determine the statistical differences. The Chi-square test was used to analyze the relationship between miR-873 expression and clinicopathological characteristics. Multivariate statistical analysis was performed using a Cox regression model. Survival curves were plotted using the Kaplan-Meier method and compared using the log-rank test. Statistical analyses were performed using the SPSS 19.0 software (SPSS, Chicago, IL, USA). P≤0.05 was considered to indicate a statistically significant result.

To examine the expression levels of miR-873 expression in CRC, we conducted qPCR in one normal human colon mucosal epithelial cell line, seven CRC cell lines, and ten pairs of CRC tissues and their TATs. The results revealed that miR-873 was markedly decreased in all seven CRC cell lines (SW620, SW480, DLD1, HCT116, KM12, LoVo, and HT-29) compared with the normal human colon mucosal epithelial cell line NCM460 (Fig. 1A). Consistent with the results obtained from the cell lines, miR-873 expression in the ten CRC tissue samples was significantly lower compared with that in their TATs (Fig. 1B), indicating that the expression of miR-873 was downregulated in CRC.

To investigate the biological significance of miR-873 downregulation during the progression of CRC, we transfected the miR-873 mimic into HCT116 and SW480 cell lines. MTT and colony formation assays revealed that overexpression of miR-873 markedly decreased the growth rates of both CRC lines compared with the negative control (NC) (Fig. 2A-C). Furthermore, transfection of the miR-873 inhibitor significantly increased the growth rates of both CRC lines compared with that of the NC (Fig. 2D-F). However, neither transfection of miR-873 nor the miR-873 inhibitor altered the apoptotic rates of CRC cells (data not shown). Therefore, these results revealed that miR-873 suppressed the proliferation of CRC cells.

To identify the direct targets of miR-873 regulation, we searched publicly available databases (TargetScan, Pictar, and miRANDA) and found that TRAF5 and TAB1, which encode components of the NF-κB pathway, may be potential targets (Fig. 3A). As predicted, western blot analysis revealed that ectopic expression of miR-873 in HCT116 and SW480 cells decreased the levels of the TRAF5 and TAB1 proteins, whereas ectopic expression of the miR-873 inhibitor increased their levels (Fig. 3B-D). However, we determined that the mRNA expression levels of TRAF5 and TAB1 did not exhibit evident alterations in the miR-873 dysregulated cells (data not shown), suggesting that miR-873 negatively regulated the expression of these proteins at the translation level. To further test this, we subcloned the 3′-UTRs of TRAF5 and TAB1 into the pGL3 luciferase reporter. Transfection of miR-873 consistently attenuated the luciferase activity of the TRAF5-3′-UTR and TAB1-3′-UTR luciferase reporter in both HCT116 and SW480 cells, whereas transfection of the miR-873 inhibitor rescued luciferase suppression. However, dysregulation of miR-873 did not result in the alteration of the reporter activities driven by the 3′-UTRs of TRAF5 and TAB1 mutated within the miR-873-binding seed regions (Fig. 3E). Collectively, these results further supported the view that TRAF5 and TAB1 are genuine targets of miR-873.

Furthermore, the luciferase assay revealed that ectopic expression of miR-873 in HCT116 and SW480 cells significantly reduced NF-κB luciferase activity, while ectopic expression of the miR-873 inhibitor enhanced NF-κB luciferase activity (Fig. 3F). This suggested an important role for NF-κB signaling in the CRC cell proliferation induced by miR-873 downregulation.

To further evaluate whether miR-873 downregulation was associated with the clinical features or prognosis of CRC, we examined the expression of miR-873 in a large cohort of clinical CRC samples using qPCR and performed a correlation analysis between the clinicopathological features and the expression of miR-873. The data revealed that miR-873 expression was inversely correlated with Dukes' stage (P=0.005), clinical stage (P=0.002), tumor-node-metastasis (TNM) classification (T, P=0.036; N, P=0.001; M, P=0.010), and histological differentiation (P=0.014) (Table II). Additionally, Kaplan-Meier survival analysis revealed that patients with CRC with lower miR-873 expression levels had shorter overall survival (Fig. 4). Moreover, univariate and multivariate analysis indicated that miR-873 expression levels were an independent prognostic factor for CRC (Table III). Collectively, these results indicated a possible link between miR-873 downregulation and CRC progression.

Finally, we examined whether the reduction in miR-873 expression induced cell proliferation in CRC samples and whether this was clinically relevant. IHC analysis of ten CRC specimens revealed that low miR-873-expressed specimens had a higher proportion of cells expressing the proliferation marker Ki67. In contrast, high miR-873-expressed specimens displayed a small proportion of Ki67-positive cells among the ten CRC specimens (Fig. 5A). Correlation analysis revealed that miR-873 expression levels were inversely correlated with Ki67 in these CRC samples (Fig. 5B). Collectively, these results established that miR-873 suppressed CRC cell proliferation via inhibition of TRAF5 and TAB1, which are key components of the NF-κB signaling pathway.