The HepG2 cell line was obtained from the American Type Culture Collection (Manassas, VA, USA) and passaged for a period of <6 months. The HepG2 cell line was originally assumed to be a hepatocellular carcinoma cell line, but was subsequently identified to originate from a hepatoblastoma, hence the emphasis of the present study on ‘liver cancer’ (21). The cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) containing 10% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.) and 1% penicillin and streptomycin, at 37°C in a humidified 5% CO₂ incubator.

CDDP resistant HepG2 cell line (CDDP-R) was derived from original parental cell line (CDDP-S) by continuous exposure to cisplatin (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) following initial dose-response studies of cisplatin (0, 1.25, 2.5, 5, 10 µM) over 72 h at 37°C from which half maximal inhibitory concentration (IC₅₀) values were obtained. Initially, HepG2 was treated with cisplatin (4.7 µM) (IC₅₀) for 72 h at 37°C. The media (DMEM + 10% FBS, Gibco; Thermo Fisher Scientific, Inc.) was removed and cells were allowed to recover for a further 72 h. This development period was performed for ~6 months. IC₅₀ concentrations were reassessed. Cells were then maintained continuously in the presence of cisplatin at the new IC₅₀ concentration (20.5 µM) at 37°C for a further 6 months.

pVax-based SIRT1 overexpression plasmid was purchased from Fulengen Bio Co., Ltd. (Guangzhou, China). The pVax empty plasmid was used as transfection control. All plasmids were transformed into DH5α cells (Genewiz, Inc., Suzhou, China) for amplification and DNA was extracted by the EndoFree Plasmid kit (Qiagen GmbH, Hilden, Germany), according to the manufacturer's protocol. The concentration was determined by measuring the A260/A280 ratio using a Thermo ND 2,000 spectrophotometer (Thermo Fisher Scientific, Inc.). The plasmid was stored at −20°C until further use. The plasmid transfection was conducted using 2 µg plasmid per well using Lipofectamine 3000 (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol.

The SIRT1-3′UTR containing the miR-29c binding site and the miR-29c mutant binding site were purchased from Genewiz, Inc., and extracted using the EndoFree Plasmid kit (Qiagen GmbH.). The plasmids (2 µg/well) were co-transfected with miR-29c into HepG2 cells using Lipofectamine 3000 (Invitrogen; Thermo Fisher Scientific, Inc.) according to the protocol previously described by Luo et al (22). A luciferase reporter assay (Promega Corporation, Madison, WI, USA) was purchased and used to measure luciferase activity at 4 h post-transfection, according to the manufacturer's protocol. The relative luciferase activity was normalized to the miR-NC group.

For proliferation assays, cells were seeded at 2×10³ cells per well in 96-well plates, as previously described (23). Cell Counting Kit-8 (Dojindo Molecular Technologies, Inc., Shanghai, China) was used and absorbance was measured at 450 nm for each well at different time points (0, 24, 48 and 72 h) using a microplate reader (Thermo Fisher Scientific, Inc.). For colony formation assays, cells were plated at 500–1,000 cells per well into 6-well plates and cultured for ~14 days, followed by crystal violet (0.5% w/v) staining for 30 min at room temperature, and counted using a light Stereomicroscope (×4).

miRNAs were obtained using the mirVana miRNA Isolation kit according to the manufacturer's protocol (Ambion; Thermo Fisher Scientific, Inc.). RT-qPCR was performed for miR-29c using miRNA primers obtained from Exiqon A/S (Vedbæk, Denmark). β-actin was used as a loading control. The primers of miR-29c and β-actin were not supplied according to the rules of the company (Exiqon A/S, Vedbæk, Denmark). First, cDNA was synthesized from all miRNA samples according to the manufacturer's protocol (Exiqon A/S). Synthesized cDNAs were used as templates for gene-expression analysis through RT-qPCR with SYBR Green (Applied Biosystems; Thermo Fisher Scientific, Inc.). The qPCR conditions were as follows: Denaturation at 94°C for 2 min, amplification for 30 cycles at 94°C for 0.5 min, annealing at 60°C for 0.5 min and extension at 72°C for 1 min, followed by a terminal elongation step at 72°C for 10 min. Data were analyzed with the 2^−ΔΔCq method (24).

Cells (2×10⁵) transfected with miRNA and treated by CDDP were harvested at 48 h post-transfection and stained with Annexin V-FITC/PI Apoptosis Detection kit I (BD Pharmingen; BD Biosciences, San Jose, CA, USA). Apoptotic cells were assessed in triplicate and the experiment was repeated three times independently by flow cytometry (FACS Calibur; BD Biosciences) with FACSComp software (version 5.1; BD Biosciences).

CDDP-R cells stably expressing miR-NC or miR-29c (5×10⁶ cells in 100 µl DMEM) were injected subcutaneously into the flanks of Balb/c nude mice (4 mice/group) (5 weeks old, 18–20 g, male; Vital River Laboratories, Beijing, China). Mice were kept in a specific pathogen-free environment, on a 12 h light/12 h dark cycle at a room temperature of 22±2°C with ad libitum access to food and water. CDDP (20 nmol in 100 µl saline) was injected every 3 days (5 times). An equal volume of normal saline was injected as a negative control. Tumor volumes were measured every 5 days. Tumor weights were measured immediately after sacrificing the mice, and tumor samples were harvested for whole protein lysates and embedded in paraffin for sectioning, as previously described by Dai et al (25). Procedures involving animals conformed to the guidelines of the Institutional Animal Care and Use Committee of Sichuan Academy of Medical Sciences and Sichuan Provincial People's Hospital (Sichuan, Chengdu, China) and study approval was obtained.

Tumor tissue sections were examined for the presence of apoptotic cells using TUNEL assay, in which fragmented DNA from apoptotic cells is end-labeled with the fluorophore. The biopsy samples were fixed in 10% phosphate-buffered formalin (Thermo Fisher Scientific, Inc.) for 24 h at room temperature, processed and then embedded in paraffin. Serial 4-µm thick tissue sections were analyzed using the DeadEnd™ Fluorometric TUNEL system (Promega Corporation) according to the manufacturer's protocol. Following deparaffinization and rehydration, sections were fixed, permeabilized with proteinase K for 8–10 min, and repeatedly fixed. The sections were then covered with 50 ml terminal deoxynucleotidyl transferase mix for 1 h at 37°C in a humidified chamber. The coverslips were removed and the sections were immersed in 2X SSC buffer for 15 min, washed with PBS and mounted with medium that included DAPI (1.5 µg/ml) (Vectashield^®; Vector Laboratories Inc., Burlingame, CA, USA) at room temperature (22–25°C) for 15 min. Fluorescence images of three different fields of view were captured using a fluorescence microscope (Olympus Corporation, Tokyo, Japan).

The proteins were extracted with ice-cold lysis buffer containing 1 mM EDTA, 20 mM Tris-HCl (pH7.5), 1 mM dithiothreitol, 0.1 mM phenylmethylsulfonyl fluoride, 5 mM MgCl₂ and a protease inhibitor cocktail (1:100) (Pierce, Thermo Fisher Scientific Inc.), and then centrifuged at 12,000 × g for 20 min at 4°C. Protein concentration of the supernatant from the extract was measured with the Bicinchoninic Acid assay kit (Beyotime Institute of Biotechnology, Beijing, China). Equivalent amounts (30 µg) of proteins were loaded on 12% SDS-PAGE and transferred to polyvinylidene difluoride membranes (Merck KGaA). The membranes were blocked in Tris-buffered saline-Tween 20 and probed with anti-SIRT1 at 4°C overnight (dilution, 1:1,500; catalog no. ab32441; Abcam, Cambridge, UK). Following washing, the membranes were incubated with horseradish peroxidase-conjugated secondary antibodies (dilution, 1:10,000; cat no. ZB-5301; OriGene Technologies, Inc., Beijing, China) for 60 min at room temperature. All blots were probed with antibodies against β-actin at 4°C overnight (dilution 1:3,000; cat no. MABT825; Merck KGaA) as a loading control. Immobilon^® ECL Ultra Western HRP Substrate (cat no. WBULS0500; EMD Millipore, Billerica, MA, USA) was used for detection with X-ray film. The densitometry was measured by Image J software (Version 1.48; National Institutes of Health, Bethesda, MD, USA).

The potential target genes of miR-29c were predicted using three different online programs with databases of different algorithms, including TargetScan (http://www.targetscan.org/), MicroRNA.org (http://www.microrna.org/) and miRDB (http://mirdb.org/) using h-miR-29c as a keyword on the 8th November 2016. The predicted targets were listed.

Continuous normally distributed variables are represented graphically as the mean ± standard deviation. For statistical comparison of quantitative data between groups, analysis of variance (ANOVA) with Dunnett's multiple comparisons or Student's t-test was performed. All statistical analyses were performed using SPSS 22.0 statistical software (IBM Corp., Armonk, NY, USA). P<0.05 was considered to indicate a statistically significant difference.

To assess the expression of miR-29c in CDDP-R liver cancer, CDDP-R HepG2 cell lines were selected by gradually increasing CDDP concentration in the culture medium. The IC₅₀ value for CDDP in the parent HepG2 CDDP-sensitive (CDDP-S) line and the derived CDDP-resistant HepG2 (CDDP-R) cells was calculated. It was demonstrated that the IC₅₀ value in the CDDP-S cells (IC₅₀, 4.7±0.4 µM) (Fig. 1A) was significantly lower than that in the CDDP-R cells (IC₅₀, 82.2±3.1 µM) (P<0.05) (Fig. 1B). The colony formation abilities of the two cell lines was analyzed, and the CDDP-R cells were demonstrated to exhibit increased colony formation abilities (CDDP-S, 143±18 vs. CDDP-R, 264±25; P<0.01) (Fig. 1C-D). Notably, miR-29c expression levels, as detected by qPCR, were significantly lower in CDDP-R cells than in CDDP-S cells (CDDP-R, 0.21±0.05; P<0.001) (Fig. 1E). These results suggested that miR-29c is downregulated in CDDP-R cancer cells.

As miR-29c expression was reduced following acquisition of resistance to CDDP, the present study investigated the role of miR-29c in chemoresistance. A plasmid expressing miR-29c was introduced and it was validated that the plasmid effectively enhanced miR-29c levels in the CDDP-R cells (miR-29c vs. miR-negative control (NC), 14.2±3.2 vs. 1.0±0.02) (Fig. 2A). Overexpression of miR-29c along with CDDP treatment (20 µM) in CDDP-R cells resulted in significantly reduced viability compared with that in the untreated group and the miR-NC plus CDDP-treated group (decreased 45.0%; miR-29c vs. miR-NC, 0.82±0.13 vs. 1.49±0.21) (Fig. 2B). The colony formation assay results corroborated these results; treatment with miR-29c and CDDP (20 µM) resulted in reduced colony numbers in CDDP-R cells (decreased 72.9%; miR-29c vs. miR-NC, 67±15 vs. 247±27 mm³) (Fig. 2C and D). Furthermore, an improved rate of apoptosis in CDDP-R cells was observed following treatment with miR-29c and CDDP (increased 8.0 times; miR-29c vs. miR-NC, 21.83±4.25 vs. 2.73±0.72%) (Fig. 2E and F). These results indicate that overexpression of miR-29c could restore the CDDP sensitivity of CDDP-R cells in vitro.

To validate the observed phenomenon in vivo, a xenograft tumor model in nude mice was established, using the selected CDDP-resistant HepG2 cell line with CDDP treatment and stable expression of either miR-NC or miR-29c. The present study demonstrated that restoring miR-29c in CDDP-R cells markedly reduced xenograft tumor growth, including tumor volume (decreased 55.5%; miR-29c vs. miR-NC, 428.4±59.2 vs. 963.2±102.3 mm³) and tumor weight (decreased 53.4%; miR-29c vs. miR-NC, 0.48±0.09 vs. 1.03±0.14 g), whereas there was no significant difference between the CDDP + miR-NC and normal saline control groups (Fig. 3A and B). RT-qPCR analysis demonstrated that the expression of miR-29c in the CDDP + miR-29c-treated tumors was 9.42 times higher than that in the control groups (Fig. 3C). TUNEL staining revealed a markedly increased number of apoptotic cells upon CDDP + miR-29c treatment (increased 3.45 times; miR-29c vs. miR-NC: 76.3±14.3 vs. 22.1±6.2) (Fig. 3D and E). These results indicated that overexpression of miR-29c may sensitize CDDP-resistant liver cancer to CDDP in vivo.

To investigate the possible mechanisms by which miR-29c restores liver cancer sensitivity to CDDP, bioinformatics-based prediction was performed using TargetScan and miRDB (http://mirdb.org), and it was demonstrated that miR-29c may potentially target the SIRT1 3′-UTR. To investigate if this was a possible mechanism for miR-29c-mediated restoration of liver cancer sensitivity to CDDP, the protein level of SIRT1 was determined and substantially increased SIRT1 protein levels were observed upon acquisition of CDDP resistance (Fig. 4A). A luciferase assay indicated that luciferase expression in SIRT1-3′UTR constructs was significantly affected by miR-29c, whereas no significant reduction was observed in SIRT1-3′UTR mutant constructs (Fig. 4B). A SIRT1 expression vector was ectopically expressed and verified that SIRT1 levels were restored (Fig. 4C). The present study assayed for cell viability and colony formation ability and identified that overexpression of SIRT1 relieves the effect of miR-29c on cell proliferation (CDDP + miR-NC, 1.74±0,18; CDDP + miR-29c, 0.89±0.11; CDDP + miR-29c + SIRT1, 1.82±0.13) and colony numbers (CDDP + miR-NC, 213±25; CDDP + miR-29c, 42±8; CDDP + miR-29c + SIRT1, 243±34) (Fig. 4D-F). Additionally, overexpression of SIRT1 restores the effect of miR-29c, which promotes apoptosis in CDDP-R cell lines (CDDP + miR-NC, 3.38±0.62%; CDDP + miR-29c, 23.52±4.21%; CDDP+miR-29c+SIRT1, 3.04±0.76%) (Fig. 4G-H). Collectively, these findings revealed that miR-29c can target and suppress SIRT1, and restore sensitivity to CDDP in CDDP-R liver cancer cell lines.