Tissue specimens

All 76 HCC and control normal liver tissues (NLT) were obtained from Shengjing Hospital of China Medical University between Febraury 2010 and May 2011. The Ethics Committees of China Medical University approved this study, and handwritten permissions of patients were obtained. The clinical pathological data were confirmed by 2 pathologists according to guidelines for the standardised pathological diagnosis of primary liver cancer in China, and follow-up continued 5 years until April 2016.

All patients included 49 men and 27 women, whose ages ranged from 41 to 78 years. There were 37 patients with poorly differentiated and 39 patients with well or moderately differentiated tumour. Based on the seventh edition UICC/TNM classification, there were 44 patients in stage T1, 26 patients in T2, 6 patients in T3, 62 patients in N0, 14 patients in N1, 71 patients in M0, and 5 patients in M1.

Cell culture

Human liver HL7702 cell line and HCC HepG2 cell line were obtained from China Academy of Chinese Medical Sciences and cultured in DMEM medium with 10% foetal bovine serum (Gibco, USA) at 37° with 5% CO₂.

Quantitative real-time PCR

Total RNA was extracted with Trizol reagent and cDNA synthesised with a High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA). Then MiRNA-610 and HDGF expression level was detected with SYBR (Applied Biosystems, Foster City, CA, USA). The primers of HDGF were 5’-GAGGGTGACGGTGATAAGAA-3’ (forward) and 5’-GAAACATTGGTGGCTACAGG-3’ (reverse). U6 and GAPDH were selected to be endogenous reference genes. The relative expression was quantified with the relative quantitative method [13].

Vector construction and transfection

The MiRNA-610 agonist (agomir-610) and negative control (agomir-NC) were purchased from Thermo Fisher Scientific (Foster City, CA, USA). The expression vector (pc-HDGF) and negative control vector (pc-NC) were constructed by Genscript (Nanjing, China). 50 nM agomir-610 or agomir-NC and 1.5 μg pc-HDGF or pc-NC were transfected with Lipofectamine^TM 3000 Reagent (Invitrogen, Foster City, CA, USA) according to the operating manual. After 4 h, normal media was applied to cultured cells with 48 h.

Cell proliferation assay

The cells (2 × 10⁴/well) were seeded in a 96-well plate. The 96-well plate was given an additional 20 μl MTT solution (0.5 mg/ml, Sigma-Aldrich, USA) per well, and incubated at 37^oC. Four hours later, the MTT solution was abandoned, and then to each well was added 0.2 ml followed by incubation for 30 min. The 96-well plate was read on an enzyme-labelled instrument (Bio-Rad Laboratories, Inc., Hercules, CA, USA) at 570 nm wavelength, and the calculation of cell viability refers to the previous literature [13].

Apoptosis detection

Flow cytometry was used to examine the apoptosis rate with an Annexin V-FITC apoptosis test kit (Beyotime, Nantong, China) according to the operating manual. Next, the cells were read by flow cytometry (BD Biosciences, Franklin Lakes, NJ, USA) (Ex, 488 nm; Em, 635 nm), and the data were analysed through CELLQuest 3.0 software (BD, USA). Cells in the right lower quadrant were regarded as apoptosis.

Chemotherapy resistance assay

Cells were treated with cisplatin (DDP) at various concentrations (0.1 μg/ml, 1 μg/ml, 5 μg/ml, 10 μg/ml, 25 μg/ml, 50 μg/ml) [14]. The cell viability was detected by MTT assay 48 h later and a dose-response curve was plotted. The half maximal inhibitory concentration (IC₅₀) was calculated by probit regression model.

Luciferase reporter assay

The wild-type and mutant-type luciferase reporter vector (pmiR-HDGF-wt and pmiR-HDGF-mut) were constructed by the Genescript company (Nanjing, China) which included wild-type or mutant binding site of HDGF 3’UTR. The microRNA and luciferase reporter vector were cotransfected into HepG2 cells. The luciferase reporter assay was processed referring to the previous literature [13], the Dual-Glo luciferase assay (Promega Corporation, Madison, WI, USA) was applied to quantitate the relative luciferase activity, which was computed by the luciferase activity of Firefly to Renilla.

Western blot

Protein samples were harvested from cells, and were tested and treated with polyacrylamide gels electrophoresis, and then transferred to PVDF membrane. The membrane was hybridised first with Rabbit polyclonal HDGF antibody (ab176697, Abcam, USA) or Rabbit polyclonal GAPDH (ab9485, Abcam, USA) and then hybridised with secondary Goat Anti-Rabbit antibody (A0208, Beyotime, China) alternately. After that, the membrane was incubated with BeyoECL Plus reagent (Beyotime, Nanjing, China), and the images were obtained and analysed to quantify the expression of HDGF protein.

Statistical analysis

All experiments were repeated five times. The statistical analysis was completed by SPSS 23.0 software (IBM, USA). Differences were assessed by one-way analysis of variance (ANOVA) in nonparametric statistics and Student’s unpaired t-test. The correlation between the expression of MiRNA- 610 and HDGF was analysed by Spearman’s rank test. P < 0.05 means statistical difference.

The low-expression and clinical significance of MiRNA-610

As shown in Figure 1 A, the MiRNA-610 expression in HCC specimens was much lower than that in NLT specimens (p < 0.05). Moreover, the MiRNA-610 expression in HepG2 cells was down-regulated compared with HL7702 cells (Figure 1 B, p < 0.05). Also, further statistical analysis indicated that low-expression of MiRNA-610 was more common in HCC patients with poor differentiation and high T stage (p < 0.05), but the expression of MiRNA-610 was not related with sex, age, positive lymph node metastasis, and distal metastasis (Table I). Those findings preliminarily proved that MiRNA-610 might be involved in the genesis and progression of HCC, but not in metastasis.

Figure 1

MiRNA-610 was down-regulated in hepatocellular carcinoma (HCC). A, B – Relative expression of MiRNA-610 and hepatoma-derived growth factor (HDGF) in HCC tissues and cell lines. Compared to NLT or HL7702 cells, the MiRNA-610 expression was lower in HCC tissue and HepG2 cells, and HDGF showed significant downregulation. C – Kaplan-Meier analysis indicated that HCC patients with lower MiRNA-610 expression (n = 38) had a significantly worse prognosis than patients with higher MiRNA-610 expression (n = 38). *p < 0.05

Figure 1 C shows that patients with higher MiRNA-610 expression had a remarkably longer survival time than patients with lower MiRNA-610 expression. All patients were divided into higher and lower MiRNA-610 expression groups from the median value of relative MiRNA-610 expression. The overall 5-year survival rate of is 37.5% in patients with higher MiRNA-610 expression but 18.5% in patients with lower MiRNA-610 expression. Meanwhile, patients with lower expression of MiRNA-610 had a significantly worse prognosis by univariate Cox proportional hazards regression analysis of overall survival (HR = 0.6867, 95% CI: 0.085–1.288; p = 0.036). The outcomes verified that MiRNA-610 was a prognostic factor for HCC patients.

MiRNA-610 played a tumour-suppressing role in HepG2 cells

In order to validate the effect of MiRNA-610 on proliferation ability and apoptosis of HepG2 cells, the agomir-610 was transfected into HepG2 cells to up-regulate the MiRNA-610 expression.

In comparison to the respective NC groups, the cell viability of HepG2 cells with MiRNA-610 over-expression decreased significantly (Figure 2 A, p < 0.05). The results of flow cytometry revealed that MiRNA-610 over-expression promoted the early apoptosis from 2.89% to 8.67% (Figure 2 B, p < 0.05), but the late apoptosis (from 10.37% to 11.46%) and necrosis (from 0.64% to 0.83%) showed no significant changes (p > 0.05).

Figure 2

MiRNA-610 played tumour-suppressing roles in HepG2 cells. A – MiRNA-610 upregulation inhibited cell viability of HepG2 cells. B – MiRNA-610 upregulation promoted the cell apoptosis level of HepG2 cells. C – MiRNA-610 upregulation depressed the IC50 of DDP in HepG2 cells. *p < 0.05

Upregulation of MiRNA-610 inhibited DDP resistance in HepG2 cells

DDP is one of the first-line drugs in clinical chemotherapy for HCC patients. Figure 2 C shows how MiRNA-610 over-expression could depress IC₅₀ of DDP from 2.37 ±0.19 to 1.47 ±0.13 μg/ml in HepG2 cells (p < 0.05), which demonstrates that over-expression of MiRNA-610 inhibits DDP resistance in HCC cells.

HDGF was a target gene of MiRNA-610 in HepG2 cells

A conserved MiRNA-610 binding site was predicted in the 3’-UTRs of HDGF at 635-642 bp by several bioinformatics software packages, including TargetScan (Figure 3 A).

Figure 3

Hepatoma-derived growth factor (HDGF) was a target gene of MiRNA-610 in HepG2 cells. A – 3’-UTR region of HDGF mRNA is partially complementary to MiRNA-610. B – The effect of agomir-610 on the luciferase activity of pmiR-HDGF-3’UTR and pmiR-HDGF-3’UTR mutation. Firefly luciferase activity was normalised to Renilla luciferase. C – The influence of transfection with agomir-610 to HDGF protein expression in HepG2 cells. The relative quantitative analysis of HDGF protein normalised to GAPDH protein. *p < 0.05

The HDGF expression was up-regulated in HCC tissues compared with NLT tissues (Figure 1 B). Moreover, a negative correlation was identified by Spearman’s rank test between MiRNA-610 and HDGF.

Furthermore, luciferase reporter assay was used to investigate the above-mentioned prediction and the direct MiRNA610-HDGF interaction. Results showed that MiRNA-610 could inhibit the relative luciferase activity by combining with wild-type binding site, but not to mutant binding site (Figure 2 B, p < 0.05). However, the agomir-NC had no similar regulatory effects.

MiRNA-610 overexpression depressed HDGF protein expression in HepG2 cells as detected by western blot (Figure 3 C).

HDGF-mediated regulation induced by MiRNA-610 over-expression in HepG2 cells

MiRNA-610 depressed the proliferation and DDP resistance of HepG2 cells and the HDGF gene was the MiRNA-610’s target, so we assumed HDGF-mediated regulation induced by MiRNA-610 over-expression in HepG2 cells. The transfection with pc-HDGF could up-regulate the expression of HDGF protein silenced by MiRNA-610 enhancement (Figure 4 A). HDGF over-expression in HepG2 cells reversed the MiRNA-610’s regulation on cell proliferation, apoptosis, and DDP resistance (Figure 4 B–D). In view of the above, we identified that MiRNA-610 depressed the proliferation and DDP resistance of HepG2 cells through targeted silencing of HDGF.

Figure 4

Hepatoma-derived growth factor (HDGF)-mediated MiRNA-610-induced regulation in HepG2 cells. A – The influence of transfection with agomir-610 and pc-HDGF on HDGF protein expression in HepG2 cells. B–D – HDGF overexpression reversed MiRNA-610’s regulatory roles in proliferation (B), apoptosis (C), and chemotherapy resistance (D) of hepatocellular carcinoma cells. *p < 0.05