Retracted: MiR-497-5p inhibits cell proliferation and metastasis in hepatocellular carcinoma by targeting insulin-like growth factor 1

Abstract

Background

MicroRNAs (miRNAs) play an important regulatory role in carcinogenesis and cancer progression. Aberrant expression of miR-497-5p has been reported in various human malignancies. However, the role of miR-497-5p in hepatocellular carcinoma (HCC) remains unclear.

Results

In this study, we found that miR-497-5p was downregulated in HCC tissues. The low level of miR-497-5p in HCC tumors was correlated with aggressive clinicopathological characteristics and predicted poor prognosis in HCC patients. The overexpression of miR-497-5p significantly inhibited HCC cell proliferation, colony formation, and metastasis in vitro and vivo. Bioinformatics analysis further identified insulin-like growth factor 1 (IGF1) as a novel target of miR-497-5p in HCC cells.

Conclusion

Our study suggested that miR-497-5p regulates HCC cell survival, partially through downregulation of IGF1. Therefore, the miR-497-5p/IGF1 axis might serve as a novel therapeutic target in patients with HCC.

Abbreviations
• AFP
• alpha fetoprotein
• BCLC
• Barcelona Clinic Liver Cancer
• HBs Ag
• hepatitis B surface antigen
• HCC
• hepatocellular carcinoma
• IGF1
• insulin-like growth factor 1
• IGF1R
• insulin-like growth factor 1 receptor
• MicroRNAs
• miRNAs
• OS
• overall survival
• RFS
• recurrence-free survival
• TNM
• tumor node metastasis
• UTR
• Untranslated Region

INTRODUCTION

Hepatocellular carcinoma (HCC) is the third leading cause of cancer-related mortality worldwide (Bosch, Ribes, Díaz, & Cléries, ; Forner, Llovet, & Bruix, ; Torre et al., ). Despite recent advances in cancer treatment, the majority of HCC patients showed a poor prognosis (Llovet, Fuster, & Bruix, ; Ryu, Jang, Kim, Lee, & Chung, ). Therefore, understanding the molecular mechanisms of HCC and identifying specific prognostic biomarkers are important for the development of therapeutic strategies for HCC.

MicroRNAs (miRNAs or miRs) are small noncoding RNA molecules of 21–25 nucleotides in length which recognize specific complementary sequences predominantly located in the 3′-untranslated region (UTR) of target mRNAs and function to either repress translation or degrade target mRNAs (Abba, Mudduluru, & Allgayer, ; Cho, ). miRNAs have been implicated in various forms of cancer by altering the expression of oncogenes or tumor suppressor genes (Abba et al., ; Cho, ; Fabbri, Calore, Paone, Galli, & Calin, ; Hernando, ; Nanasinkam & Croce, ; Osada & Takahashi, ; Setoyama, Ling, Natsugoe, & Calin, ). Many miRNAs have been reported to play an important role in HCC (Chen, Li, et al., ; Chen, Zhang, et al., ; Li, Deng, et al., ; Mirzaei et al., ; Takata et al., ; Yang et al., ). However, the characterization of miR-497-5p in HCC and its association with HCC progression remains unclear.

Previous studies have shown that miR-497-5p is of great interest in cancer therapies due to its association with various types of cancer (Chai et al., ; Chen et al., ; Sun et al., ). Here, we found that low miR-497-5p expression in HCC tissues was significantly correlated with poor prognosis of patients; thus, we focused on the roles and corresponding mechanisms of miR-497-5p in the progression of HCC.

MATERIALS AND METHODS Ethical compliance

All the clinical specimens were obtained with informed consent and approved by the Clinical Research Ethics Committee of Yanan Hospital. Informed consent was obtained from all patients involved in this study. All the experiments were performed in accordance with the approved guidelines of the Institutional Research Ethics Committee of Yunnan University of Chinese Traditional Medicine. The data do not contain any information that could identify the patients.

Tissue samples

A total of 166 pairs of snap-frozen HCC and peritumoral tissues were obtained from the Yanan Hospital. These tissues were used for quantitative real-time polymerase chain reaction (qRT-PCR) analysis. Clinical tissue samples were verified as tumor or nontumor using a histopathological examination and the Edmondson grading system. Micro-metastases were defined as tumors adjacent to the border of the main tumor as observed using a microscope. Tumor staging was defined based on the sixth edition of the Tumor Node Metastasis (TNM) classification system published by the International Union Against Cancer. The tissue samples were stored at −80°C until further use. Tumor differentiation was defined according to the R and Barcelona Clinic Liver Cancer (BCLC) staging systems.

Cell culture

Human hepatocellular cancer cell lines (Huh7, HCC-LM3, HepG2, Hep3B, and THLE-3) were purchased from the Shanghai Institute of Life Sciences Cell Resource Center in Shanghai, China. All cell lines were cultured in DMEM medium (Hyclone) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (Gibco). All cell cultures were maintained at 37°C in a humidified atmosphere with 5% CO₂.

Bioinformatics methods

The miRNA targets were predicted using a computer-aided algorithm from TargetScan (http://www.targetscan.org).

Quantitative real-time polymerase chain reaction (qRT-PCR)

Mature miR-497-5p expression was detected using a TaqMan miRNA-assay kit (Applied Biosystems, Foster City, CA, USA) according to the manufacturer's instructions. U6 gene was used as a normalization control. All experiments were performed in triplicate and repeated once. To verify the integrity of insulin-like growth factor 1 (IGF1) expression, the GAPDH gene was used as an internal control. PCR was run using the following conditions: 30 cycles consisting of denaturation at 94°C for 30 s, annealing at 56°C (58°C for GAPDH) for 30 s, and extension at 72°C for 30 s. The primer and probe sequences used in the qRT-PCR reactions are listed in Table S2.

Western blotting assay

Cell lysates were separated using 10% sodium dodecyl sulfate polyacrylamide gels, electrophoretically transferred to polyvinylidene difluoride membranes (Roche Diagnostics, Mannheim, Germany), and then detected using anti-IGF1 and anti-IGF1R antibodies. Protein loading was estimated using a mouse anti-GAPDH monoclonal antibody. Lab Works Image Acquisition and Analysis Software (UVP, Upland, CA, USA) were used to quantify the band intensities.

Luciferase activity assay

The 3’UTR of IGF1 was amplified and cloned downstream of the pGL3/Luciferase (Luc) vector. The mutant 3’UTR of IGF1 (several nucleotides within the binding sites were mutated) was amplified using the pGL3/Luc-IGF1 3’UTR as the template and then was cloned downstream of the pGL3/Luc vector. For the luciferase reporter assay, cells were co-transfected with either miR-497-5p mimics or control and the pGL3/Luc-IGF1 3’UTR or the mutant 3’UTR, together with the controls. At 48 hr after transfection, the cells were lysed using RIPA buffer, and luciferase intensity was measured using an F-4500 Fluorescence Spectrophotometer (HIT-ACHI).

Transfection

For stable transfection, the lentiviral expression vectors LV-Control and LV-IGF1 were obtained from Shanghai Gene Pharma Company (China). Lentiviruses were mixed with polybrene (5 mg/ml) and added Huh7 and HepG2 cells. Positive clones were selected in puromycin (5 mg/ml). Stable IGF1 transfectants were isolated after 2 weeks.

Cell proliferation (MTT) assay and colony formation assay

The transfected cells were plated into 96-well plates at a density of 5,000 cells/well. At 48 hr after transfection, the cells were incubated with MTT (3-(4, 5-Dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide) for 4 hr at 37°C. The cells were then agitated with MTT solvent on an orbital shaker for 10 min while avoiding light. The absorbance was measured at 450 nm (OD450nm) using a spectrophotometer. The transfected cells were seeded into 12-well plates at a density of 200 cells/well. The medium was changed every 3 days. Approximately 10 days after seeding, most of the cell clones contained more than 50 cells. The clones were washed with 1 × PBS and stained with Crystal Violet for approximately 5 min. The clones were imaged and counted, and the colony formation rate was calculated using the following formula: (number of clones)/(number of seeded cells) × 100%.

Transwell assays

Milli cell 24-well culture insert plates (Millipore, USA) and polycarbonate membranes with a pore size of 8 μm were used in the transwell assays as follows. First, the insert plates were equilibrated with 0.5 ml of DMEM for 1 hr at 37°C with 5% CO₂. The medium was then replaced with 0.5 ml of DMEM supplemented with 10% FBS in the lower chambers. In all, 50,000 cells in 400 μl of serum-free DMEM were loaded into the upper chambers. After 24 hr of incubation, the insert plates were rinsed with PBS, and the upper surfaces of the membranes were scraped to remove the cells. The cells on the underside of the membrane were stained with Giemsa stain and counted using a microscope. The cells from each culture condition were examined in quadruplicate.

Wound healing assays

For the wound healing assays, monolayer of cells plated in 12-well plates were wounded by scraping the monolayer with a 200-μl plastic pipette tip. The cells were then rinsed several times with medium to remove any floating cells. The wound healing process was monitored with an inverted light microscope (Olympus).

IGF1 ELISA

IGF1 levels in normal culture medium collected after 48 hr from different cells were detected with the Human IGF1 ELISA Kit (ab100545; Abcam, CA, USA) according to the manufacturer's instructions.

Animal studies

The animal studies were approved by the Yunnan University of Chinese Traditional Medicine. To explore the effects of miR-497-5p on tumor growth in vivo, 1 × 10⁷ Huh7 cells, which stably overexpressed miR-497-5p, or control cells were subcutaneously implanted into the bilateral armpit of nine BALB/c nude mice. The tumor volume was measured each week following implantation (the volume, V = length × width × length × 1/2). All mice were sacrificed 5 weeks later. Eighteen 6-week-old male nude mice were randomly assigned into two groups; and cells (4 × 10⁵) were injected into the tail vein of the mice to establish a pulmonary metastatic model. The mice were sacrificed 10 weeks after the injection and microscopically examined for the development of lung metastatic foci via H&E staining. The animals were housed in cages under standard pathogen-free conditions, in accordance with the Yunnan University of Chinese Traditional Medicine.

Statistical analysis

All of the data are shown as the mean ± SD, and the experiments were run in triplicate. Statistically significant differences were determined using a two-tailed Student's t-test and p < .05 was considered statistically significant.

RESULTS miR-497-5p expression is downregulated in human HCC

To explore the role of miR-497-5p in HCC, we first examined the expression of miR-497-5p in HCC cell lines. As shown in Figure a, decreased miR-497-5p expression was observed in HCC cell lines compared to normal liver cells. Downregulation of miR-497-5p in HCC cell lines was correlated with the expression observed in HCC tumor tissues. Among 166 matched HCC tumor tissue samples, miR-497-5p was significantly downregulated in tumor tissues compared with the nontumor tissues (Figure b). Therefore, we speculated that downregulation of miR-497-5p expression might play an important role in HCC progression and development.

Decreased miR-497-5p expression predicts aggressive clinicopathological characteristics and poor prognosis in HCC patients

To investigate the clinical significance of miR-497-5p in HCC, the cohort of 166 HCC patients was divided into two groups: a low miR-497-5p expression (below the median expression level, n = 83) group and a high miR-497-5p expression (above the median expression level, n = 83) group. We found that low expression of miR-497-5p was associated with a larger tumor size (≥5 cm) (p = .003), multiple tumors (p = .018) and positive hepatitis B virus (HBV) DNA levels (p = .007) (Table ). Kaplan-Meier survival analysis showed that the HCC patients in the low miR-497-5p expression group exhibited worse recurrence free survival (RFS) and overall survival (OS) than patients in the high miR-497-5p expression group (p < .001 and p < .001, respectively) (Figure c,d). The prognosis of some patients with early-stage HCC still turn out to be poor, suggesting that a supplementary prognostic predictor is required for these patients. Therefore, patients with early-stage HCC (TNM stage I) were stratified and subgroup analyses were performed. Notably, the prognosis-predictive value of low miR-497-5p in early-stage HCC (TNM stage I) was still proven (RFS for p < .001 and OS for p < .001, respectively) (Figure e,f). Similar results were also observed in patients with normal serum AFP level (<20 μg/L) (Figure S1a,b). A univariate analysis indicated that among the clinicopathological characteristics, miR-497-5p expression level, tumor number, vascular invasion, and positivity for AFP or HBV-DNA level were correlated with RFS and OS (Table S1). Furthermore, multivariate Cox regression analysis indicated that miR-497-5p expression level, tumor number, vascular invasion, and positivity for AFP were independent risk factors for RFS, and miR-497-5p expression level and vascular invasion were independent risk factors for OS in HCC patients (Table ). Taken together, these results indicate that miR-497-5p may represent a valuable prognostic biomarker for HCC.

Clinical characteristics of the 166 hepatocellular carcinoma patients based on the expression level of miR-497-5p

[Table omitted. See PDF]

Note:

Abbreviations: AFP, alpha fetoprotein; BCLC, Barcelona Clinic Liver Cancer; TNM, tumor node metastasis.

The median expression level was used as the cutoff value. Low miR-497-5p expression in each of the 83 patients was defined as a value below the 50th percentile. High miR-497-5p expression in each of the 83 patients was defined as a value above the 50th percentile. Pearson's Chi-square tests were used for the correlation analysis between the expression levels of miR-497-5p and the clinical features. Results were considered statistically significant at p < .05.

Multivariate analysis for recurrence free survival and overall survival

Variable	RFS			OS
	Hazard ratio	95% CI	p-value	Hazard ratio	95% CI	p-value
AFP, μg/L, ≥20 versus <20	1.636	1.142 – 2.342	.007
Microvascular invasion, present versus absent	1.689	1.093 – 2.609	.018	1.758	1.118 – 2.763	.014
Tumor number, multiple versus solitary	1.834	1.227–2.740	.003
miR−497−5p expression, low versus high	0.274	0.180 – 0.418	<.001	0.360	0.234 – 0.553	<.001

Abbreviations: AFP, alpha fetoprotein; HBsAg, hepatitis B surface antigen; OS, overall survival; RFS, recurrence free survival.

Overexpression of miR-497-5p inhibits HCC cell proliferation in vitro and vivo

To determine the effect of miR-497-5p on HCC cell behavior, we stably overexpressed miR-497-5p in the Huh7 and HepG2 cell lines using a lentivirus (Figure a). To confirm that miR-497-5p could function as a tumor suppressor, the effects of miR-497-5p overexpression on HCC cell proliferation were determined in vitro. As shown in Figure b,c, overexpression of miR-497-5p inhibited HCC cell proliferation. In parallel, Huh7 and HepG2 cells that stably overexpressed miR-497-5p formed fewer colonies compared with the control cells (Figure d,e). To determine the effects of miR-497-5p on tumorigenesis in vivo, we injected Huh7 cells that stably overexpressed miR-497-5p or miR-GFP subcutaneously into nude mice for xenograft transplantation. As shown in Figure f,g, mice injected with miR-497-5p-overexpressing cells showed significantly decreased tumor growth compared with mice injected with miR-GFP-transfected cells. Therefore, these results indicated that ectopic miR-497-5p expression inhibited HCC cell proliferation in vitro and vivo.

Overexpression of miR-497-5p inhibits HCC cell metastasis in vitro and vivo

Next, we determined the effects of miR-497-5p overexpression on HCC cell metastasis in vitro. A wound healing migration assay showed that miR-497-5p overexpression could inhibit cell migration in Huh7 and HepG2 cells (Figure a,b). Migration and matrigel invasion assays with these miR-497-5p overexpressing cell lines also showed that the exogenous expression of miR-497-5p dramatically decreased cell mobility compared with that of control cells (Figure c-f).To verify the in vivo consequences of miR-497-5p overexpression, we injected Huh7 cells that stably overexpressed miR-497-5p or miR-GFP cells into the lateral tail vein of nude mice and evaluated both the metastatic growth in the lungs and the survival of the mice. Ten weeks later, the mouse lungs were stained with H&E and lung micro metastases were microscopically evaluated (Figure g,h). Fewer and smaller metastatic foci were detected in the mice injected with the Huh7- miR-497-5p cells than in the control group. In addition, mice injected with Huh7-miR-497-5p cells had a significantly higher survival rate (Figure i).

miR-497-5p directly targets IGF1

The miRNA target prediction algorithm TargetScan was used to computationally screen for genes with miR-497-5p complementary sites in their 3’UTR. The results showed that IGF1 was a putative target of miR-497-5p. The miR-497-5p binding sequences in the 3’UTR of IGF1 mRNA (WT-3’UTR) or its mutant (IGF1-3’UTR-mut) were subcloned downstream of the luciferase reporter vector pGL3 (Figure a). As shown in Figure b,c, the relative luciferase activity of the reporter containing IGF1 WT-3’UTR was significantly decreased when miR-497-5p was co-transfected, whereas the luciferase activity of the IGF1-3’UTR-mut reporter was unaffected in Huh7 and HepG2 cells. These results suggest that miR-497-5p might suppress IGF1 expression through the putative binding site in its 3’UTR. Furthermore, real time PCR and Western blot assays were performed to determine whether miR-497-5p expression affects the expression of endogenous IGF1 at both the transcriptional and translational levels. Consistently, the mRNA level of IGF1 was decreased in miR-497-5p-overexpressing HCC cells (Figure d). Furthermore, Western blot analysis showed that the expression of IGF1 and its target gene, IGF1R, was significantly inhibited following overexpression of miR-497-5p (Figure e). To examine whether miR-497-5p decreases IGF1 secretion, we measured IGF1 levels in the supernatant from different cell clones. miR-497-5p overexpression decreased IGF1 levels in the supernatant (Figure S2). Moreover, the miR-497-5p level was negative correlated with the IGF1 transcript level in the tumor tissues (Figure f). These data indicated that miR-497-5p inhibited IGF1 expression in HCC cells by targeting its 3’UTR.

Rescue expression of IGF1 abolished the effects of miR-497-5p on phenotype of HCC cells

To determine if the IGF1 gene is required for themiR-497-5p's effects on HCC cell proliferation and metastasis, ectopic over-expression of IGF1 was performed to conduct functional studies in Huh7 and HepG2 cells (Figure a,b). In Figure c,d, overexpression of IGF1 abolished the effects of miR-497-5p on inhibiting of HCC cells proliferation. Furthermore, the capacities of migration and invasion in IGF1-overexpression HCC cells were significantly enhanced, while overexpression of IGF1 abolished the effects of miR-497-5p on inhibiting of HCC cells migration and invasion ability (Figure e,f). Therefore, overexpressed IGF1 abolished the effects of miR-497-5p on phenotype of HCC cells.

DISCUSSION

Although several studies have demonstrated the signaling pathways involved in the regulation of HCC cell proliferation and cell cycle, the molecular mechanism for HCC cell growth remains unclear (Bosch et al., ; Farazi & Depinho, ; Forner et al., ; ). MiRNAs offer a novel molecular approach to this question and have been reported to be involved in the pathogenesis of HCC (Toffanin et al., ).

In a previous study, miR-497-5p was reported as an oncogene in renal cell carcinoma (Li, Chen, et al., ). However, the characterization of miR-497-5p in HCC and its association with disease progression and development remain unclear. In this study, we found that miR-497-5p was downregulated in HCC tumor tissues and HCC cell lines. MiR-497-5p downregulation correlated with a larger tumor size (≥5 cm) (p = .003), multiple tumors (p = .018) and positive HBV DNA levels (p = .007) and predicted poor prognosis. Currently, the HBV DNA levels level is closely related to the recurrence of HCC, and it is used to predict the risk of recurrence in HBsAg-positive HCC patients after hepatectomy. The relationship between the expression of the miR-497-5p and the serum AFP levels suggests that miR-497-5p may be a potential biomarker for diagnosing HCC. Furthermore, its close relationship with the HBV DNA levels and the degree of tumor size also indicates that miR-497-5p may be related to hepatocarcinogenesis. Furthermore, overexpression of miR-497-5p in Huh7 and HepG2 cells significantly inhibited cell proliferation and metastasis both in vitro and in vivo. These results strongly suggest thatmiR-497-5p has an inhibitory role in HCC.

MicroRNAs can function as tumor suppressors or oncogenes by targeting oncogenes or tumor suppressor genes. In this study, we explored the miR-497-5p targets that may contribute to its inhibition of cell proliferation and metastasis in HCC. Using TargetScan bioinformatics, we identified the IGF1 gene as a possible direct target for miR-497-5p. IGFs are composed of two subtypes, IGF1 and IGF2, and have been shown to regulate the proliferation and differentiation of muscle, cartilage, chondrocytes, and neurons (Joung et al., ; Kim et al., ; Patil, Sable, & Kothari, ; Sullivan, Kim, & Feldman, ; Wang et al., ). IGF1 is secreted primarily from the liver in response to human growth hormone (GH) and acts systemically (Perrini et al., ). IGF1 has been shown to play important roles in both normal and pathological biological processes including cancer (Brouwer-Visser & Huang, ; Bruchim, Attias, & Werner, ; Huat et al., ; Livingstone, ; Schlegel et al., ; Sekharam et al., ; Sjögren et al., ; Svensson et al., ; Tirapegui, ; Weroha & Haluska, ). Studies have also reported that human IGF1 pro-forms induce breast cancer cell proliferation through the IGF1 receptor. In this study, we performed luciferase reporter assay, real-time PCR, and western blotting to verify that miR-497-5p can directly target IGF1 by interacting with its 3’UTR. The expression level of the IGF1 downstream target gene, IGF1R, was also decreased in miR-497-5p-transfected cells. To the best of our knowledge, our study is the first to report that miR-497-5p acts as a tumor suppressor in HCC. Besides, we observed that miR-497-5p had a negative expression with IGF1 transcription in HCC, and. this implies that miR-497-5p might act as a tumor suppressor or have a similar role through decreasing IGF1 expression. Our results indicate that the miR-497-5p/IGF1 axis might serve as a novel therapeutic target in patients with HCC.

In summary, miR-497-5p expression is frequently decreased in HCC tumor tissues and may serve as a prognostic biomarker in patients with HCC. The overexpression of miR-497-5p inhibited HCC cell proliferation by directly suppressing the expression of IGF1, which not only sheds new light on HCC progression and metastasis, but also provides a potential target for cancer prevention and treatment.

ACKNOWLEDGMENTS

The study was funded by the National Institutes of Health grants NIDDK R01-DK46441 and NCI R01-CA095731 and Ann and Jerry Moss Foundation. The study was also funded by the National Science Fund for Young Scholars (3150090121) and by the project of scholar of Traditional Chinese Medicine University.

CONFLICT OF INTEREST

We certify that the authors have no actual or potential conflict of interest in relation to this article.