miRIDIAN miRNA hairpin inhibitor negative control, miRIDIAN miRNA mimic negative control, miRIDIAN hairpin inhibitor or miRIDIAN miRNA mimic for human hsa-miR-122 (MIMAT0000421) were purchased from Thermo Scientific Dharmacon (Waltham, MA, USA). The miRIDIAN hairpin inhibitors and mimics were used at a concentration of 50 nM. Polyclonal anti-occludin and monoclonal anti-β-tubulin antibodies were purchased from Sigma (St. Louis, MO, USA). Polyclonal Anti-ACTIVE p38 (Thr¹⁸⁰/Tyr¹⁸²) and Anti-ACTIVE JNK1/2 (Thr¹⁸³/Tyr¹⁸⁵) antibodies were purchased from Promega (Madison, WI, USA). Monoclonal anti-phospho-Erk1/2 (Thr²⁰²/Tyr²⁰⁴), anti-Erk1/2, and anti-p38 antibodies were purchased from Cell Signaling Technology (Beverly, MA, USA). Monoclonal anti-JNK1/2 antibody was purchased from BD Transduction Laboratories (San Jose, CA, USA).

ccRCC specimens were obtained from patients while they underwent primary curative resection at the Osaka University Medical Hospital, Japan. Tumour-associated normal renal tissue was also obtained from a subset of these patients when possible. Prior written and informed consent was obtained from each patient, and the study was approved by the ethics review board of the Osaka University Medical Hospital and the methods were carried out in accordance with the approved guidelines.

Following excision, tissue samples were immediately immersed in RNAlater (Qiagen, Valencia, CA, USA) and stored at −20°C until RNA extraction. miRNAs were purified using the miRNeasy mini kit (Qiagen). Real-time PCR analysis was conducted to validate miR-122 expression in ccRCC using 80 tumour samples and 10 adjacent normal renal samples (for Fig. 1B–D) and another distinct 122 tumour tissue samples (for Fig. 1E and F) using the Mir-X miRNA first-strand synthesis kit (Takara, Shiga, Japan). Thermal cycling conditions included an initial step at 98°C for 30 sec, and 40 cycles at 95°C for 2 sec and 66°C for 5 sec using an miR-122 specific primer (5′-tggagtgtgacaatggtgtttg-3′), and 98°C for 30 sec, and 40 cycles at 95°C for 2 sec and at 63°C for 5 sec by using an U6 snRNA primer (Takara). Clinical and pathological data related to the clinical samples are presented in Table I.

Three human ccRCC cell lines (Caki-1, Caki-2 and, ACHN), obtained from the American Type Culture Collection (ATCC), were cultured in RPMI-1640 medium (Wako, Osaka, Japan) supplemented with 10% fetal bovine serum, 100 U/ml penicillin G, and 0.1 µg/ml streptomycin.

Protein samples were separated by sodium dodecyl sulphate (SDS)-polyacrylamide gel electrophoresis (PAGE) on a 7.5–15% SDS-polyacrylamide gel and then transferred to a polyvinylidene difluoride membrane using the Bio-Rad semi-dry transfer system (1 h, 12 V). Immunoreactive proteins reacting with the antibodies described above were visualized by treatment with a detection reagent (ECL Prime western blotting detection reagent; GE Healthcare, Lafayette, CO, USA). Densitometric analysis was performed using NIH ImageJ software.

A pmirGLO dual-luciferase miRNA target expression vector was used for the 3′-UTR luciferase reporter assay (Promega). The following oligonucleotides were used for the evaluation of efficacy of the miR-122 hairpin inhibitor on occludin 3′-UTR: 5′-CTA GCG GCC GCT AGT TCA ACT GGG CTG AAC ACT CCA G-3′ and 5′-TCG ACT GGA GTG TTC AGC CCA GTT GAA CTA GCG GCC GCT AGA GCT-3′. Luciferase activity was determined using a luminometer (Turner Biosystems 20/20 n luminometer, Promega).

Cell migration was examined by wound healing assay. In brief, Caki-2 cells transfected with the miRIDIAN hairpin inhibitor or the negative control inhibitor, and ACHN cells transfected with the miRIDIAN microRNA mimic or the negative control mimic were seeded in a 24-well plate (Caki-2 cells: 2.0×10⁴ cells/well, ACHN cells: 4.0×10⁴ cells/well) and incubated for 72 h. A wound was created in a monolayer of approximately 90% confluent Caki-2 or ACHN cells using a sterile 1 ml pipette tip. Cell migration pictures were recorded at 0 and 12 h after wound creation using an Olympus IX71 fluorescence microscope.

Cell proliferation was examined by WST-1 assay. Caki-2 cells transfected with the miRIDIAN hairpin inhibitor or the negative control inhibitor, and ACHN cells transfected with the miRIDIAN miRNA mimic or the negative control mimic were seeded in a 96-well plate (0.3×10⁴ cells/well) and incubated for 72 h. After incubation for 2 h with the WST-1 reagent (Dojindo, Osaka, Japan) at 37°C and 5% CO₂, the optical density was read at a wavelength of 450/630 nm (Ex/Em).

Caki-2 or ACHN cells were seeded at 1.0×10⁴ cells/well in 6-well plates in RPMI-1640 medium (supplemented with 10% FBS and 0.3% agarose with 0.4% agarose underlay). Dishes were incubated at 37°C and 5% CO₂. After 14 days, colonies were stained with crystal violet (Wako) and counted.

The BioCoat tumour invasion system (Corning Inc., Corning, NY, USA) was used to perform the cell invasion assay. Caki-2 cells transfected with the miRIDIAN hairpin inhibitor or the negative control inhibitor were seeded in a 96-well plate (3×10⁴ cells/well). Following incubation for 12 h, the cells were labelled with calcein AM (4 µg/ml, Takara), and the fluorescence of the invaded cells was read at a wavelength of 494/517 nm (Ex/Em).

Cells plated on coverslips were washed with phosphate-buffered saline (PBS). The cells were fixed and permeabilized in 4% formaldehyde for 15 min at room temperature and then washed twice with PBS. After blocking with 2% bovine serum albumin in PBS for 1 h, the coverslips were incubated with the anti-occludin antibody (1:500) at 4°C. The coverslips were washed twice with PBS and incubated with the fluorochrome-conjugated secondary antibody for 1 h at room temperature. The cells were then examined under the fluorescence microscope, Bio-Zero (Keyence, Osaka, Japan).

The expression of occludin was determined by immunohistochemical staining of paraffin-embedded tissues of ccRCC samples with the highest and lowest levels of miR-122 expression samples (10 samples of each). Formalin-fixed paraffin-embedded sections (5 µm in thickness) were deparaffinized and rehydrated. After the slides were steamed for 20 min in 10 mM citrate buffer (pH 6.0) for antigen retrieval, endogenous peroxidase was blocked using 3% H₂O₂. Immunohistochemical staining for occludin was performed using anti-occludin (1:500) and the EnVision⁺ Detection System (Dako, Santa Clara, CA, USA) was used according to the manufacturer's instructions. Primary antibodies were incubated overnight at 4°C and counterstained with hematoxylin. The levels of occludin staining were classified into four groups according to the staining intensity of the positive cells (staining score, 0–3; Fig. 5F) and Quick score was calculated as follows; Quick score = (percentage of occludin positive cells) × (staining score). For each ccRCC sample, three points were calculated.

ACHN cells transfected with occludin siRNAs or miR-122 mimic for 48 h were reseeded (1.0×10⁵ cells/well) on a 0.4 µm pore size 24-well cell culture insert (Thermo Scientific Dharmacon) and incubated for 1 day. TEER was measured using a Millicell^® ERS-2 voltohmmeter (Millipore, Billerica, MA, USA). TEER was calculated as follows: (sample well Ω − empty well Ω) × (cultured area cm²) = TEER (Ω cm²).

The results are expressed as the mean ± standard deviation (SD). Differences between the values were statistically analysed using the Student's t-test or one-way analysis of variance (ANOVA) with Bonferroni post-hoc tests. Correlations between the values were analysed using Pearson correlation analysis (GraphPad Prism 6.0, GraphPad software, San Diego, CA, USA). A P-value of <0.05 was considered statistically significant.

In our previous microarray analysis identifying unique miRNA expression signatures (GEO accession number: GSE55138) (8), aberrantly high expression of miR-122 was detected in ccRCC tissues compared with adjacent normal renal tissues (Fig. 1A). To confirm the expression of miR-122 in ccRCC specimens we performed real-time PCR analysis. Although there was no correlation with pathological stages or grades (Fig. 1C and D), the expression of miR-122 was approximately 300-fold higher in the ccRCC tissues compared to the adjacent normal renal tissues (Fig. 1B) and was significantly high in ccRCC tissues accompanied by lymphatic invasion compared with those without lymphatic invasion (Fig. 1E). Noteworthy, ccRCC patients with high miR-122 expression had significantly poorer progression-free survival compared with those with low miR-122 expression (Fig. 1F).

To investigate the role of miR-122 in ccRCC, we first compared the expression levels of miR-122 in three ccRCC cell lines (Caki-1, Caki-2, and ACHN, Fig. 2A). Since the expression of miR-122 was highest in Caki-2 cells, we first analysed the function of miR-122 using Caki-2 cells. To investigate the function of miR-122, we examined the effect of a miR-122 inhibitor on growth, migration, and invasion in Caki-2 cells. Although the miR-122 inhibitor had no significant effect on the cell growth (Fig. 2B and C), migration and invasion activities were significantly reduced in Caki-2 cells (Fig. 2D and E). Of the cell lines examined, ACHN cells had the lowest miR-122 expression. Using a miR-122 mimic in ACHN cells, we observed upregulated proliferation (Fig. 3A and B), migration (Fig. 3C), and invasion activities (Fig. 3D). These results suggest that miR-122 is involved in regulating the expression of crucial molecule/s in ccRCC cells.

We used target prediction programs (miRbase, TargetScan, and miRanda) to identify the potential miR-122 target in ccRCC cells. We then focused on occludin as a potential target, because it is a component of tight junctions and functions as a tumour suppressor in several cancers (30,31). TargetScan predicts that the 3′-untranslated region (UTR) of occludin mRNA contains a complementary site for the seed region of miR-122 (Fig. 4A). To confirm whether or not occludin was a direct target of miR-122, the miR-122 inhibitor and a luciferase reporter plasmid containing the predicted miR-122 binding site (within the 3′-UTR of the human occludin gene) were co-transfection into Caki-2 cells. Luciferase activity confirmed that the reporter construct was working appropriately (Fig. 4B). The luciferase activity in cells co-transfected with the miR-122 inhibitor increased significantly compared to those co-transfected with the negative control inhibitor. Conversely, compared to the negative control mimic, the miR-122 mimic significantly decreased the luciferase activity (Fig. 4C). Moreover, the miR-122 inhibitor upregulated the expression of occludin protein in Caki-2 cells (Fig. 4D), and the miR-122 mimic reduced occludin protein expression in ACHN cells (Fig. 4E). These results indicated that occludin is a direct target of miR-122 in ccRCC cells.

To examine whether the malignant phenotypes upregulated by the miR-122 mimic in ccRCC cells were due to the decrease in occludin expression, we performed occludin knockdown experiments in ACHN cells (Fig. 5A and B). Occludin knockdown upregulated the migration activity of ACHN cells. The motility of ACHN cells was upregulated by occludin knockdown. Transepithelial electrical resistance (TEER) analysis showed that miR-122 mimic and occludin knockdown downregulated the TEER in the cells (Fig. 5C and D). TEER is a quantitative method to measure the integrity of tight junctions in cultured cells. These results suggest that miR-122 might upregulate the permeability of tight junctions via occludin reduction, leading to an increase in the motility of ACHN cells.

The miR-122 inhibitor upregulated the expression of occludin at the cell-cell adhesive surface in Caki-2 cells (Fig. 5E). To examine whether miR-122 expression and occludin protein expression are negatively correlated, immunohistochemistry analysis was performed on the ccRCC samples with the lowest and highest miR-122-expression (10 samples for each). ccRCC tissues with high levels of miR-122 expression had a significantly weak occludin quick score compared to those with low levels of miR-122 expression (Fig. 5F and G). Moreover, miR-122 and occludin protein expression showed a negative correlation in the ccRCC tissues (Fig. 5H).

Although the miR-122 mimic upregulated the cell proliferation, motility, and invasion activities, occludin knockdown only upregulated cell motility. Target prediction programs predict that putative miR-122 binding sequences exist in other genes involved in cell growth [e.g., p27 (36), PHD3 (37,38), and DUSP2 (39)] and invasion [e.g., CPEB1 (40), FOXP2 (41), and SOX17 (42)]. Occludin is known to mediate the MAPK signalling pathway via oncogenic proteins such as Raf (29). Therefore, we examined the effect of miR-122 on the MAPK pathway by miR-122 overexpression and loss of function experiments. We found that miR-122 affected the phosphorylation status of Erk1/2 and p38 (Fig. 6A). Likewise, occludin knockdown upregulated the phosphorylation status of Erk1/2 (Fig. 6B). Therefore, miR-122 might also mediate ccRCC cell growth and invasion via occludin and other target molecules including MAPK signalling components.