From the GEO database (https://www.ncbi.nlm.nih.gov/geo/), the microarray data of miRNA GSE6857 and GSE30297, which represent miRNA expression profile data from 496 liver cancer patients and 35 normal controls, respectively, were obtained. The data sets were based on the platforms GPL4700 and GPL8786, respectively. All raw data, including the original CEL, GPR and SOFT files, were obtained for further analysis.

Normalization of the raw miRNA data was performed in R platform (version 3.1.3; https://www.r-project.org/) using the Robust Multi-array Analysis (RMA) method (13). The final log2-transformed RMA expression values were then stored for further analysis. Three independent tests, including Fisher's exact test, t-test and Wilcoxon test, were used to identify significantly differentially expressed miRNAs between liver cancer and normal control samples. The miRNAs that were supported by the above three tests were considered to be confidently involved in the carcinogenesis of liver cancer.

Prediction of target genes for the differentially expressed miRNAs was performed using current available methods, including DIANA (10), miRanda (11) and TargetScan (12). In order to reduce the false-positive rate of target prediction, the predicted targets supported by at least two independent methods were selected as reliable target genes of miRNAs. In addition, miRNAs target genes with wet experimental support in the TarBase 6.0 database (14) were also included in the final pathway analysis.

The most significantly differentially expressed miRNAs identified in the previous steps were selected for IPA. The Ingenuity Knowledge database (http://www.ingenuity.com/products/ipa) and IPA tools were used to identify the enriched roles of miRNAs and their target genes in cellular functions, pathways and diseases.

The human liver cancer cell line HepG2 was obtained from the Chinese Center for Type Culture Collection (Beijing, China). The HepG2 cell line was first cultured in RPMI-1640 medium (Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 10% fetal bovine serum (Thermo Fisher Scientific, Inc.). Cells were maintained in a humidified atmosphere with 5% CO₂ at 37°C. HepG2 cells were seeded in 24-well plates at 6×10⁵ cells/well and incubated overnight. Transfection of the miR-1297 mimics, anti-miR-1297, inactive control cel-mir-67 or pMIR-REPORT vector (all Thermo Fisher Scientific, Inc.) was performed using Lipofectamine 2000 transfection reagent (Invitrogen; Thermo Fisher Scientific, Inc.) with 300 nmol miRNA or 1 µg/ml DNA plasmid, respectively. Total proteins of HepG2 cells were isolated at 48 h post-transfection using M-PER Reagent (Thermo Fisher Scientific, Inc.).

Proteins were separated by 12% SDS-PAGE and then transferred onto nitrocellulose membranes (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Membranes were blocked with 5% non-fat milk and incubated with anti-retinoblastoma (RB)1 antibody (cat. no. ab181616; Abcam, Shanghai, China) or anti-β-actin antibody (cat. no. ab8227; Abcam) at 4°C overnight. Following extensive washes with TBST, a secondary antibody (cat. no. ab150077; Abcam) was added to the system. Finally, enhanced chemiluminescence (Abcam) was used to detect the immunoreactive protein bands.

Cell proliferation assay was performed with Cell Counting Kit-8 (CCK-8; Dojindo Molecular Technologies, Inc., Kumamoto, Japan). HepG2 liver cancer cells were plated at 6×10⁵ cells/well in 24-well plates. Then, cells were incubated in 10% CCK-8 at 37°C for color conversion. Proliferation rates were detected at 24, 48 and 72 h post-transfection.

HepG2 cells were seeded at 6×10⁵ cells/well in 24-well plates and incubated for 24 h. Subsequently, the cells were co-transfected with 0.8 µg pGL3-RB1-3′UTR or pGL3-RB1-3′UTR Mut plasmid or with 0.08 ng phRL-SV40 control vector (all Promega Corporation, Madison, WI, USA), and with 100 nM miR-1297 or inactive control RNA, using Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific, Inc.). The Renilla luciferase and firefly luciferase activities were detected using a dual luciferase assay (Promega Corporation) at 24 h post-transfection.

Statistical analyses were performed using R (version 3.1.3; https://www.r-project.org/). Values were expressed as means ± standard deviation. Differences between groups were estimated with the Student's t-test. P<0.05 was considered to indicate a statistically significant difference.

Based on the combined results of the three independent tests, five miRNAs were identified to be significantly differentially expressed in liver cancer (Fig. 1 and Table I), including three upregulated and two downregulated miRNAs. Among these miRNAs, miR-1297 had the most significant deregulation.

Since miRNAs serve their functions by targeting mRNAs, the predicted target genes of the differentially expressed miRNAs were retrieved. The most significant deregulated miRNA, miR-1297, attracted our attention because one of its target genes supported by multiple evidences is the tumor-suppressor gene RB1 (Fig. 2), which is involved in the regulation of the cell cycle and in human cancer pathways (hsa04110 and hsa05200 Kyoto Encyclopedia of Genes and Genomes pathways) (15–17).

The predicted target genes of miR-1297 were collected and imported into the IPA system to investigate their biological functions in liver cancer. Table II contains the top five most significant networks identified by IPA. Among these networks, cell death and survival was the most frequent function, with a significant score of 43 (Table II). IPA also indicated that miR-1297 target genes were involved in various biological functions, including cell cycle and cellular development (Table III). Cell death and survival as well as glutamate receptor signaling were the most significant pathways enriched in target genes of miR-1297 (Fig. 3).

The potential impact of miR-1297 on liver cancer cell proliferation was assessed in the HepG2 cell line. HepG2 cells were transfected with miR-1297 mimics or miR-1297 inhibitor, or with the inactive control cel-mir-67. CCK-8 proliferation assay indicated that cell proliferation was significantly promoted in miR-1297-mimics-transfected HepG2 cells compared with that in inactive control cel-mir-67-transfected cells (Fig. 4A). Conversely, miR-1297 inhibitor significantly inhibited the proliferation of HepG2 cells (Fig. 4A).

miR-1297 mimics significantly reduced the protein levels of RB1 in liver cancer cells (Fig. 4B). Conversely, miR-1297 inhibitor significantly increased the protein levels of RB1 in liver cancer cells (Fig. 4B). As predicted by bioinformatics analysis, there was complementarity between hsa-miR-1297 and the 3′UTR of RB1. The effect of miR-1297 on the translation of RB1 mRNA into protein was then determined using a luciferase reporter assay. miR-1297 mimics significantly reduced the luciferase activity of the reporter gene with the wild-type construct but not with the mutant RB1 3′UTR construct (Fig. 4C). The inhibitor of miR-1297 significantly enhanced the luciferase activity of the reporter gene with the wild-type construct but not with the mutant RB1 3′UTR construct (Fig. 4D). These evidences indicate that miR-1297 directly binds to the 3′UTR region of RB1. In general, miR-1297 targets and negatively regulates RB1 in liver cancer cells.