MicroRNA-26a suppresses tumor growth and metastasis of human hepatocellular carcinoma by targeting interleukin-6-Stat3 pathway

Authors

  • Xin Yang,

    1. Liver Cancer Institute and Zhongshan Hospital, Institutes of Biomedical Science, Fudan University, Shanghai, China
    2. Key Laboratory of Carcinogenesis and Cancer Invasion, Ministry of Education, China
    Search for more papers by this author
  • Lei Liang,

    1. Liver Cancer Institute and Zhongshan Hospital, Institutes of Biomedical Science, Fudan University, Shanghai, China
    2. Key Laboratory of Carcinogenesis and Cancer Invasion, Ministry of Education, China
    Search for more papers by this author
  • Xiao-Fei Zhang,

    1. Liver Cancer Institute and Zhongshan Hospital, Institutes of Biomedical Science, Fudan University, Shanghai, China
    2. Key Laboratory of Carcinogenesis and Cancer Invasion, Ministry of Education, China
    Search for more papers by this author
  • Hu-Liang Jia,

    1. Liver Cancer Institute and Zhongshan Hospital, Institutes of Biomedical Science, Fudan University, Shanghai, China
    2. Key Laboratory of Carcinogenesis and Cancer Invasion, Ministry of Education, China
    Search for more papers by this author
  • Yi Qin,

    1. Liver Cancer Institute and Zhongshan Hospital, Institutes of Biomedical Science, Fudan University, Shanghai, China
    2. Key Laboratory of Carcinogenesis and Cancer Invasion, Ministry of Education, China
    Search for more papers by this author
  • Xu-Chao Zhu,

    1. Liver Cancer Institute and Zhongshan Hospital, Institutes of Biomedical Science, Fudan University, Shanghai, China
    2. Key Laboratory of Carcinogenesis and Cancer Invasion, Ministry of Education, China
    Search for more papers by this author
  • Xiao-Mei Gao,

    1. Liver Cancer Institute and Zhongshan Hospital, Institutes of Biomedical Science, Fudan University, Shanghai, China
    2. Key Laboratory of Carcinogenesis and Cancer Invasion, Ministry of Education, China
    Search for more papers by this author
  • Peng Qiao,

    1. Liver Cancer Institute and Zhongshan Hospital, Institutes of Biomedical Science, Fudan University, Shanghai, China
    2. Key Laboratory of Carcinogenesis and Cancer Invasion, Ministry of Education, China
    Search for more papers by this author
  • Yan Zheng,

    1. Liver Cancer Institute and Zhongshan Hospital, Institutes of Biomedical Science, Fudan University, Shanghai, China
    2. Key Laboratory of Carcinogenesis and Cancer Invasion, Ministry of Education, China
    Search for more papers by this author
  • Yuan-Yuan Sheng,

    1. Liver Cancer Institute and Zhongshan Hospital, Institutes of Biomedical Science, Fudan University, Shanghai, China
    2. Key Laboratory of Carcinogenesis and Cancer Invasion, Ministry of Education, China
    Search for more papers by this author
  • Jin-Wang Wei,

    1. Liver Cancer Institute and Zhongshan Hospital, Institutes of Biomedical Science, Fudan University, Shanghai, China
    2. Key Laboratory of Carcinogenesis and Cancer Invasion, Ministry of Education, China
    Search for more papers by this author
  • Hai-Jun Zhou,

    1. Liver Cancer Institute and Zhongshan Hospital, Institutes of Biomedical Science, Fudan University, Shanghai, China
    2. Key Laboratory of Carcinogenesis and Cancer Invasion, Ministry of Education, China
    Search for more papers by this author
  • Ning Ren,

    1. Liver Cancer Institute and Zhongshan Hospital, Institutes of Biomedical Science, Fudan University, Shanghai, China
    2. Key Laboratory of Carcinogenesis and Cancer Invasion, Ministry of Education, China
    Search for more papers by this author
  • Qing-Hai Ye,

    1. Liver Cancer Institute and Zhongshan Hospital, Institutes of Biomedical Science, Fudan University, Shanghai, China
    2. Key Laboratory of Carcinogenesis and Cancer Invasion, Ministry of Education, China
    Search for more papers by this author
  • Qiong-Zhu Dong,

    Corresponding author
    1. Liver Cancer Institute and Zhongshan Hospital, Institutes of Biomedical Science, Fudan University, Shanghai, China
    2. Key Laboratory of Carcinogenesis and Cancer Invasion, Ministry of Education, China
    • Liver Cancer Institute and Zhongshan Hospital, Fudan University, 180 Feng Lin Road, Shanghai 200032, China===

    Search for more papers by this author
    • fax: +86-21-5423-7960

  • Lun-Xiu Qin

    Corresponding author
    1. Liver Cancer Institute and Zhongshan Hospital, Fudan University, 180 Feng Lin Road, Shanghai 200032, China
    • Liver Cancer Institute and Zhongshan Hospital, Fudan University, 180 Feng Lin Road, Shanghai 200032, China===

    Search for more papers by this author
    • fax: +86-21-5423-7960


  • Potential conflict of interest: Nothing to report.

  • View this article online at wileyonlinelibrary.com.

Abstract

Down-regulation of microRNA-26a (miR-26a) is associated with poor prognosis of hepatocellular carcinoma (HCC), but its functional mechanism in HCC remains unclear. In this study, we investigated the roles of miR-26a in tumor growth and metastasis of HCC and found that miR-26a was frequently down-regulated in HCC tissues. Down-regulation of miR-26a correlated with HCC recurrence and metastasis. Through gain- and loss-of-function studies, miR-26a was demonstrated to significantly inhibit in vitro cell proliferation, migration, and invasion. In addition, miR-26a induced G1 arrest and promoted apoptosis of HCC cells. Importantly, miR-26a suppressed in vivo tumor growth and metastasis in nude mice models bearing human HCC. Interleukin-6 (IL-6) was identified as a target of miR-26a. Knockdown of IL-6 induced effects on HCC cells similar to those induced by miR-26a. In contrast, IL-6 treatment abrogated the effects induced by miR-26a up-regulation. Moreover, miR-26a dramatically suppressed expression of signal transducer and activator of transcription 3 (Stat3) target genes, including Bcl-2, Mcl-1, cyclin D1, and MMP2. IL-6 treatment antagonized this effect, while knockdown of IL-6 by IL-6 short hairpin RNA (shIL-6) induced inhibitory effects on the expression of p-Stat3 and its main target genes, similar to miR-26a. The messenger RNA and protein levels of IL-6 inversely correlated with miR-26a in HCCs. Patients with high miR-26a or low IL-6 in HCC tissues had a better prognosis with longer overall survival (OS) and time to recurrence (TTR). In multivariate analysis, miR-26a, IL-6, and their combination were demonstrated to be independent prognostic indicators for OS and TTR of HCC patients. Conclusion: miR-26a could suppress tumor growth and metastasis of HCC through IL-6-Stat3 signaling and is a novel prognostic marker and therapeutic target for HCC. (HEPATOLOGY 2013)

MicroRNAs (miRNAs) are a class of highly conserved short RNAs that regulate diverse cellular processes by binding to the 3′ untranslated region (3′-UTR) of target messenger RNAs (mRNAs).1 Numerous reports have shown that miRNA dysfunction is involved in the development and progression of various human cancers.2-4 Many studies have demonstrated that some miRNAs are crucial for tumorigenicity,5 proliferation,6 apoptosis,7 angiogenesis, invasion, and metastasis8, 9 of hepatocellular carcinoma (HCC). Therefore, miRNAs could serve as therapeutic targets in HCC.

Emerging data show that miRNA-26a (miR-26a) is down-regulated and could serve as a potential tumor suppressor in several distinct cancer types including thyroid anaplastic carcinomas, nasopharyngeal carcinoma, breast cancer, and HCC.10-16 Restoration of miR-26a activity in HCC has been proposed as a novel treatment strategy. Systemic administration of miR-26a delivered by adeno-associated virus antagonized tumorigenesis in a mouse model of MYC-induced liver cancer.12 We have reported that patients whose tumors had low miR-26a expression had a shorter overall survival (OS) but a better response to adjuvant interferon therapy versus patients whose tumors had high expression of miR-26a.14 However, miR-26a has also been shown to exhibit oncogenic properties in glioma, lung cancer, and cholangiocarcinoma.17-19 Numerous studies have shown that miR-26a suppresses cell proliferation in nasopharyngeal carcinoma, breast cancer, and liver cancer cells,12, 15, 16 but promotes proliferation in glioma and cholangiocarcinoma.17, 18 Interestingly, miR-26a was highly expressed in lymph node metastatic tumors compared with primary tumors and enhanced lung cancer cell migration and invasion.19 The real reason for the dual effects of miR-26a is not clear yet, it might be in part due to organ-specific actions and the different cellular contexts of tumors. To date, experimental data suggest that miR-26a directly regulates several target mRNAs, including SMAD1, MTDH, CDK6, CCNE1, CCNE2, CCND2, EZH2, PTEN, RB1, MAP3K2, and GSK-3β.12, 16-24 However, the potential mechanisms by which miR-26a affects the malignant phenotype of HCC cells remain largely unknown.

In this study, we investigated the potential function of miR-26a in the development and progression of HCC. We found that miR-26a expression was down-regulated in the majority of HCC tissues. Down-regulation of miR-26a was significantly associated with tumor recurrence, metastasis, and poor prognosis in HCC patients. Moreover, in vitro and in vivo assays showed that miR-26a significantly suppressed tumor growth, invasion, and metastasis of HCC by targeting the interleukin-6 (IL-6)-signal transducer and activator of transcription 3 (Stat3) signaling pathway. These results provide a clearer understanding of the underlying mechanism by which miR-26a inhibits HCC.

Patients and Methods

Patients and Specimens.

A total of 130 patients were enrolled in the study. Patients received curative resection for HCC at Liver Cancer Institute and Zhongshan Hospital, Fudan University (Shanghai, China) between January 2005 and December 2006. Patients did not receive any preoperative cancer treatments. Patients were followed after surgical treatment until May 2011, with a median follow-up of 29 months (range, 2-73.2 months). During the follow-up, patients were monitored every 2 to 3 months as described.25 Computed tomography scanning or magnetic resonance imaging was performed when tumor recurrence was suspected. Ten out of 130 patients had extrahepatic metastasis, and 66 patients had tumor recurrence, including 57 intrahepatic recurrences. Recurrent tumors were treated as described.26 Among the 130 study cases, 62 plasma samples were collected; of these, three patients had tumor metastasis and 31 patients had tumor recurrence. Clinico-pathological characteristics of the patients are presented in Supporting Table 1. Clinical samples were collected from patients after obtaining informed consent in accordance with a protocol approved by the Ethics Committee of Fudan University (Shanghai, China).

In Vitro Cell Proliferation, Cell Cycle, and Apoptosis Analyses.

Cell proliferation, cell cycle, and apoptosis analyses were performed as described in the Supporting Information.

In Vitro Migration and Invasion Assays.

Wound healing assays were used to examine migration of HCC cells. The invasive ability of cells was determined as described in the Supporting Information.

In Vivo Tumor Growth and Metastasis Assays.

All experimental procedures involving animals were approved by The Animal Care and Use Committee of Fudan University, China. Flank xenografts were established by subcutaneously injecting 2 × 106 HCC-LM3 or MHCC97-H cells, which were stably transfected with a miR-26a expression vector or control vector, into nude mice (BALB/c nu/nu) (SLRC, Shanghai, China) that were 4 weeks old. Subcutaneous tumors were removed and dissected into 1-mm3 sections, which were incubated into the left hepatic lobe of BALB/c nude mice to establish orthotopic implantation models. Tumor growth was monitored, and mice were sacrificed after 6 weeks. Tumors, livers, and lungs were removed, fixed in formalin, and embedded in paraffin. Consecutive sections were made for every lung tissue block and stained with hematoxylin and eosin. The number of lung metastases was calculated and evaluated independently by two pathologists. Based on the number of HCC cells in the maximal section of the metastatic lesion, lung metastases were classified into four grades. Grade I was defined as having ≤20 tumor cells; grade II had 20 to 50 tumor cells; grade III had 50 to 100 tumor cells; grade IV had >100 tumor cells.

Luciferase Reporter Assay.

HEK293T cells were seeded in a 96-well plate at 50% to 60% confluence. After 24 hours, cells were transfected with 120 ng of miR-26a expression vector, miR-26a inhibitor, control vector, or negative control. Cells were cotransfected with 30 ng of the wild-type or mutant 3′-UTR of IL-6 mRNA. Transfections were performed using 0.45 μL of Fugene (Promega, Madison, WI). Cells were collected 48 hours after transfection, and Renilla luciferase activity was measured using a dual-luciferase reporter system (Promega). Luciferase reporter assays were performed in duplicate and repeated in three independent experiments. Luciferase activity was detected using an Orion II microplate luminometer (Berthold Technologies, Germany).

Detection of Target Proteins.

Levels of IL-6 in serum or culture medium were detected via enzyme-linked immunosorbent assay (R&D Systems, Minneapolis, MN) according to the manufacturer's instructions. Protein levels of IL-6, p-Stat3, Bcl-2, Mcl-1, cyclin D1, and MMP2 were detected in HCC cells via western blotting and in paraffin-embedded, formalin-fixed tissues via immunohistochemistry. Detailed descriptions of the materials and methods can be found in the Supporting Information.

Statistical Analysis.

Statistical analysis was performed with SPSS version 15.0 (SPSS Inc., Chicago, IL). Values are expressed as the mean ± SD. The difference between groups was analyzed using a Student t test when comparing only two groups or one-way analysis of variance when comparing more than two groups. P < 0.05 was considered statistically significant.

Abbreviations

HCC, hepatocellular carcinoma; IL-6, interleukin-6; miR-26a, microRNA-26a; miRNA, microRNA; mRNA, messenger RNA; OS, overall survival; qRT-PCR, quantitative real-time polymerase chain reaction; shIL-6, IL-6 short hairpin RNA; Stat3, signal transducer and activator of transcription 3; TTR, time to recurrence; 3′-UTR, 3′ untranslated region.

Results

Down-Regulation of miR-26a Is Associated with Metastasis and Recurrence of HCC.

Expression of miR-26a was analyzed by quantitative real-time polymerase chain reaction (qRT-PCR) and normalized against an endogenous control (U6 RNA). The expression level of miR-26a was significantly down-regulated in HCC tissues versus adjacent nontumor liver tissues (P < 0.01) (Fig. 1A). Furthermore, in comparison to nonmetastatic HCC tissues, miR-26a levels were significantly lower (P < 0.01) in metastatic HCC tissues, which showed intrahepatic spreading or tumor invasion into blood vessels or bile ducts (Fig. 1B). Similarly, plasma miR-26a levels in patients with metastatic HCC were obviously lower compared with patients with nonmetastatic HCC (P < 0.05) (Supporting Fig. 1A). Moreover, miR-26a levels were decreased in both tumor tissues (P < 0.01) (Fig. 1C) and plasma (P < 0.05) (Supporting Fig. 1B) obtained from patients who showed intrahepatic tumor recurrence or extrahepatic metastasis in comparison with patients who did not have tumor recurrence. However, the difference between miR-26a levels in the group with intrahepatic recurrences and the group with extrahepatic metastases was not statistically significant (Supporting Fig. 2). These results indicate that significant down-regulation of miR-26a expression occurs in HCC and correlates with HCC relapse and metastasis.

Figure 1.

The expression levels of miR-26a in HCC tissues and HCC cell lines. Comparing differences in the expression levels of miR-26a between (A) tumor and corresponding nontumor tissues, (B) metastatic and nonmetastatic HCC tissues, (C) tumor tissues from recurrent and nonrecurrent groups, and (D) HCC cell lines with different metastatic potentials and the nontransformed hepatic cell line L02. Expression levels of miR-26a were determined by qRT-PCR and normalized against an endogenous control (U6 RNA). *P < 0.05. **P < 0.01.

To further evaluate the association of miR-26a with HCC metastasis, we analyzed miR-26a levels in a panel of human HCC cell lines with different metastatic potentials.27 The miR-26a level was significantly decreased in seven established HCC cell lines relative to the nontransformed hepatic cell line L02 (Fig. 1D). Moreover, the expression levels of miR-26a in the highly metastatic HCC cell lines MHCC97-H and HCC-LM3 were much lower than those in the HCC cell lines that have low metastatic potential, including HepG2, PLC, SMMC-7402, SMMC-7721, and Hep3B (Fig. 1D). These results indicate that reduced expression of miR-26a correlates with increased metastatic potential of HCC cells.

Inhibitory Effects of miR-26a on Proliferation of HCC Cells.

To further explore the biological significance of miR-26a in HCC, we transfected an miR-26a expression plasmid or an anti-miR-26a vector into human HCC cell lines that have different endogenous expression levels of miR-26a. Expression of miR-26a was verified by qRT-PCR (Supporting Fig. 3A). Modulation of miR-26a function was confirmed by checking protein expression of EZH2, a well-established miR-26a target16, 22, 23 (Supporting Fig. 3B).

Up-regulation of miR-26a in HCC-LM3 and MHCC97-H cells, which have high metastatic potential and low endogenous miR-26a expression levels, resulted in significant suppression of cell proliferation, G1 arrest, and induction of apoptosis. Transfection of miR-26a for 48, 72, or 96 hours inhibited proliferation of HCC-LM3 cells by 56%, 52%, and 47%, respectively (P < 0.01) (Fig. 2A). The cell cycle was arrested in G1 phase, with 62.5% of miR-26a-transfected cells in G0/G1 versus 50% of control cells (P < 0.05) (Fig. 2B). In addition, transfection of miR-26a induced apoptosis (27.4% versus 4.3% in the control group; P < 0.01) (Fig. 2C). Similar effects of miR-26a were found in MHCC97-H cells (Supporting Fig. 4A-C).

Figure 2.

The effects of miR-26a on in vitro proliferation, apoptosis, migration, and invasion of HCC-LM3 cells. Up-regulation of miR-26a significantly (A) inhibited cell proliferation, (B) induced G1/S arrest, and (C) increased the apoptosis rate compared with controls. (D) The wound healing rate in HCC-LM3 cells transfected with miR-26a was significantly decreased. (E) The number of invaded cells in the miR-26a–transfected group was significantly decreased compared with control. *P < 0.05. **P < 0.01.

In contrast, miR-26a knockdown using anti–miR-26a in HepG2 and PLC cells, which have low metastatic potential and high endogenous miR-26a levels, significantly increased cell proliferation compared with the negative control (Supporting Fig. 5A). Cell cycle arrest (Supporting Fig. 5B) and apoptosis (Supporting Fig. 5C) were not observed. These results demonstrate that miR-26a regulates proliferation of HCC cells.

MiR-26a Reduces Migration and Invasion of HCC Cells.

Mobility of HCC-LM3 and MHCC97-H cells in wound healing assays was significantly decreased after transfection of miR-26a (Fig. 2D and Supporting Fig. S4D). In contrast, knockdown of miR-26a increased wound healing (Supporting Fig. 5D). Similarly, in Matrigel invasion assays, up-regulation of miR-26a significantly decreased migration of HCC-LM3 cells (78 versus 161; P < 0.01) (Fig. 2E) and MHCC97-H cells (73 versus 149; P < 0.01) (Supporting Fig. 4E). Silencing of miR-26a markedly increased invasion of HepG2 cells (169 versus 120; P < 0.01) and PLC cells (155 versus 103; P < 0.01) (Supporting Fig. 5E). These results suggest that miR-26a significantly inhibits in vitro invasion of HCC cells.

Effects of MiR-26a on In Vivo Tumor Growth and Metastasis of HCC.

HCC-LM3 or MHCC97-H cells stably expressing miR-26a or control vector were injected subcutaneously into nude mice. Palpable tumors formed within 1 week. Tumor volume was measured each week, and mice were sacrificed 6 weeks after tumor cell implantation. The average tumor volume of HCC-LM3 cells stably transfected with miR-26a was 1.12 ± 0.089 cm3, which was significantly smaller than tumors in the control group (1.72 ± 0.108 cm3; P < 0.01) (Fig. 3A).

Figure 3.

The effects of miR-26a on in vivo tumor growth and metastasis of HCC. miR-26a overexpression suppressed tumor growth and metastasis in subcutaneous and orthotopic implantation mouse models of HCC-LM3 cells. (A) Tumor volumes (left) and tumor growth curves (right) of subcutaneous implantation models of HCC are shown. (B) Tumor volumes in the orthotopic implantation models at week 6 are shown. (C) Metastatic lesions in the lungs of the orthotopic implantation models are shown at 6 weeks after implantation. The total numbers and grades of lung metastatic lesions in the miR-26a transfection groups were much less than those in controls (magnification ×200; scale bars, 100 μm). *P < 0.05. **P < 0.01.

To further determine the effects of miR-26a on in vivo tumor growth and metastasis, HCC-LM3 or MHCC97-H tumor xenografts were isolated and implanted into the liver to establish orthotopic models. The average volume of HCC-LM3 orthotopic tumor in the miR-26a transfection group was noticeably smaller than the control group (P < 0.01) (Fig. 3B). In addition, the incidence of lung metastasis of HCC-LM3 orthotopic tumor in the miR-26a transfection group and the control group was 40% and 80%, respectively. The total number and grade of lung metastatic lesions in the miR-26a transfection group was much lower than the control (P < 0.01) (Fig. 3C). Similar inhibitory effects on in vivo tumor growth and metastasis induced by miR-26a were found in nude mice bearing MHCC97-H stable transfectants (Supporting Fig. 6).

IL-6 Is a Direct Downstream Target of miR-26a.

Next, we searched for candidate target genes of miR-26a using publicly available databases, including TargetScan (http://www.targetscan.org/), PicTar (http://pictar.mdc-berlin.de/), and miRanda (microrna.org and miRbase). IL-6 was selected for further experimental validation, because the complementary sequence of miR-26a was identified in the 3′-UTR of IL-6 mRNA by miRanda analysis (Fig. 4A). A statistically significant inverse correlation was revealed by Spearman's correlation analysis between mRNA levels of miR-26a and IL-6 (r = −0.910; P < 0.001) (Fig. 4B) and between miR-26a mRNA and IL-6 protein (r = −0.7571; P < 0.001) (Supporting Fig. 7). The inverse correlation between miR-26a and IL-6 mRNA levels (P = 0.014) and miR-26a mRNA and IL-6 protein level (P < 0.001), was further confirmed by Fisher's exact tests in 130 HCC tissues (Supporting Table 2). Moreover, overexpression of miR-26a significantly reduced the mRNA level of IL-6 in HCC-LM3 and MHCC97-H cells and reduced IL-6 protein levels in cell culture supernatant (P < 0.01) (Fig. 4C). In contrast, knockdown of miR-26a dramatically increased the mRNA and protein levels of IL-6 in HepG2 and PLC cells (P < 0.01) (Supporting Fig. 8).

Figure 4.

IL-6 is a direct target of miR-26a in HCC. (A) miR-26a and its putative binding sequence in the 3′-UTR of IL-6. The mutant miR-26a binding site was generated in the complementary site for the seed region of miR-26a (wt, wild type; mt, mutant type). (B) Spearman's correlation analysis demonstrated that miR-26a expression level was inversely correlated with IL-6 mRNA level in HCC tissues. (C) After transfection of miR-26a, both the mRNA and protein levels of IL-6 in HCC-LM3 and MHCC97-H cells were significantly decreased compared with controls. (D) In subcutaneous implantation models of HCC-LM3 or MHCC97-H cells, both the mRNA level in tumor tissues and the serum protein level of IL-6 were noticeably reduced when miR-26a was up-regulated. (E) miR-26a significantly suppressed the luciferase activity that carried wt but not mt 3′-UTR of IL-6. Anti–miR-26a led to a noticeable increase in luciferase activity of wt 3′-UTR of IL-6. **P < 0.01.

In subcutaneous models, the serum protein and tumor mRNA levels of IL-6 were found to be noticeably reduced when miR-26a was up-regulated (P < 0.01) (Fig. 4D). In addition, miR-26a significantly suppressed the luciferase activity of IL-6 containing a wild-type 3′-UTR but did not suppress activity of IL-6 with a mutant 3′-UTR (P < 0.01) (Fig. 4E). Down-regulation of miR-26a by anti–miR-26a increased the luciferase activity of IL-6 (P < 0.01) (Fig. 4E). Taken together, these results strongly suggest that IL-6 is a target of miR-26a in HCC.

Alterations of IL-6 Levels Influence the Effects of miR-26a on HCC Cells.

To further confirm that IL-6 is a functional target of miR-26a, we treated HCCLM3 miR-26a stable transfectants with IL-6. We found that cell proliferation was markedly increased after adding 25 ng/mL of IL-6 to HCC cell cultures (P < 0.01) (Fig. 5A, left). Moreover, IL-6 significantly rescued miR-26a–induced cell cycle arrest (P < 0.05) (Fig. 5B, left) and apoptosis (P < 0.01) (Fig. 5C, left). Similarly, the inhibitory effect of miR-26a on HCC cell invasion was also antagonized by IL-6 treatment (P < 0.01) (Fig. 5D, left). These data provide further support that IL-6 is a downstream functional mediator of miR-26a.

Figure 5.

The correlation between IL-6 alterations and miR-26a effects in HCC cells. (A) Cell proliferation of the miR-26a group was suppressed compared with controls. Proliferation was significantly increased after treatment with IL-6 (miR-26a+IL-6 group) compared with the miR-26a group (left). shIL-6 decreased HCC proliferation similar to miR-26a (right). The correlations between miR-26 effects and IL-6 treatment or knockdown by shIL-6 are shown in (B) cell cycle distribution, (C) apoptosis, and (D) invasive ability assays. IL-6 treatment abrogated the effects of miR-26a on HCC cells. Knockdown of IL-6 induced effects that were similar to those stimulated by miR-26a. *P < 0.05. **P < 0.01.

IL-6 short hairpin RNA (shIL-6) knocked down IL-6 expression in HCC-LM3 cells (Supporting Fig. 9A,B) and significantly inhibited cell proliferation (P < 0.01) (Fig. 5A, right), induced G1 arrest (P < 0.05) (Fig. 5B, right), increased apoptosis (P < 0.01) (Fig. 5C, right), and inhibited invasion (P < 0.01) (Fig. 5D, right). These results were consistent with the effects of miR-26a overexpression and provide further evidence supporting IL-6 as a downstream mediator of miR-26a.

MiR-26a Inhibits the IL-6-Stat3 Signaling Pathway.

To explore whether miR-26a exerts its functions through the IL-6-Stat3 pathway, which contributes to cancer development and progression, we examined a number of the main IL-6-Stat3 target genes, including Bcl-2, Mcl-1, cyclin D1, and MMP2. Expression of p-Stat3, Bcl-2, Mcl-1, cyclin D1, and MMP2 were decreased in HCC-LM3 and MHCC97-H cells that stably overexpress miR-26a (Fig. 6A, left, lane 2). In contrast, expression of these genes was significantly up-regulated after 48 hours of IL-6 stimulation (Fig. 6A, left, lane 3). Furthermore, shIL-6 had effects on p-Stat3 and its target genes similar to the effects induced by miR-26a (Fig. 6A, right). Staining intensities for IL-6, p-Stat3, and the main target genes were greatly decreased in HCC tissues from subcutaneous implantation models of HCCLM3 stably transfected with miR-26a compared with controls (P < 0.01) (Fig. 6B and Supporting Fig. 10). These data indicate that miR-26a inhibits IL-6-Stat3 signaling in HCC.

Figure 6.

miR-26a exerts its functions by inhibiting the IL-6-Stat3 signaling pathway. (A) In HCC-LM3 and MHCC97-H cells with miR-26a overexpression, the expression levels of p-Stat3, Mcl-1, Bcl-2, cyclin D1, and MMP2 (left, lane 2) were significantly decreased compared with the control. IL-6 treatment abrogated the decreased expression of these genes induced by miR-26a in HCC cells (left, lane 3). Knockdown of IL-6 by shIL-6 inhibited expression of p-Stat3 and its main target genes, similar to miR-26a (right). (B) Immunohistochemistry staining for IL-6-Stat3 pathway-related proteins in tumor tissues from mice with subcutaneous HCC implantation. Compared with controls, much lower staining intensity and fewer positive cells for IL-6, p-Stat3, and its main target genes, including cyclin D1, Bcl-2, Mcl-1, and MMP2, were demonstrated in the HCC tissues from the miR-26a overexpression group. (magnification ×400; scale bars, 50 μm).

The Prognostic Values of miR-26a and IL-6 for HCC Patients.

Although significant correlations were not found between miR-26a and the clinico-pathological features of HCCs (Supporting Table 3), the 1-, 3-, and 5-year OS rates of patients with high miR-26a (91.4%, 68.5%, and 38.5%, respectively) were significantly higher than those with low miR-26a (66.8%, 39.6%, and 19.2%, respectively; P < 0.001) (Fig. 7A). Further, the 1-, 3-, and 5-year tumor recurrence rates for the high miR-26a group (27.6%, 56.0%, and 63.4%, respectively) were much lower than those of the low miR-26a group (51.6%, 71.7%, and 78.5%, respectively; P = 0.025) (Fig. 7B). Similarly, the patients with higher plasma miR-26a levels had better overall survival and lower possibilities of tumor recurrence (Supporting Fig. 11).

Figure 7.

The prognostic significance of miR-26a and IL-6 for HCC patients assessed via Kaplan-Meier analysis. The patients with higher miR-26a had (A) better overall survival (OS) and (B) lower possibility for tumor recurrence. Patients with higher IL-6 levels in HCC tissues had (C) poorer OS and (D) higher probability of tumor recurrence. Patients in subgroup I had the (E) longest OS and (F) lowest possibility of tumor recurrence among the four subgroups, which were divided according to combinations of miR-26a and IL-6 (subgroup I, high miR-26a/low IL-6; subgroup II, low miR-26a/low IL-6; subgroup III, high miR-26a/high IL-6; subgroup IV, low miR-26a/high IL-6). For each cohort, different subgroups were plotted according to the cutoff values of miR-26a and IL-6, which were defined as the median of the cohort.

The IL-6 level in tumor tissues was significantly associated with capsule status (P = 0.034), vascular invasion (P = 0.007), and tumor-node-metastasis stage (P = 0.004) of HCC (Supporting Table 3). The 1-, 3-, and 5-year OS rates of HCC patients with high IL-6 were 65.8%, 49.3%, and 18.2%, respectively, which were lower than those for patients with low IL-6 (92.0%, 81.9%, and 67.1%, respectively; P < 0.001) (Fig. 7C). The 1-, 3-, and 5-year tumor recurrence rates for patients in the high IL-6 group (54.3%, 85.8%, and 85.8%, respectively) were higher than those for patients with low IL-6 (21.9%, 43.4%, and 58.5%, respectively; P < 0.001) (Fig. 7D).

Univariate analysis showed that miR-26a, IL-6 level, tumor size, tumor encapsulation, vascular invasion, and tumor-node-metastasis stage were significantly associated with OS and time to recurrence (TTR) in HCC patients (Table 1). However, sex, age, liver cirrhosis, hepatitis B surface antigen, and tumor differentiation did not significantly associate with OS or TTR. Multivariate analysis showed that miR-26a, IL-6 level, and tumor size were independent prognostic indicators for both OS and TTR (Table 1).

Table 1. Univariate and Multivariate Analysis of Factors Associated With Overall Survival and Time To Recurrence in HCC Patients
Clinical VariablesOverall SurvivalTime to Recurrence
HR (95%CI)PHR (95%CI)P
  • Analyses were conducted using univariate analysis or multivariate Cox proportional hazards regression. Combination of miR-26a and IL-6: subgroup I, high miR-26a and low IL-6; subgroup II, low miR-26a and low IL-6; subgroup III, high miR-26a and high IL-6; subgroup IV, low miR-26a and high IL-6. The low or high of IL-6 and miR-26a level was defined according to their cutoff values, which were defined as the median values of the cohort of patients tested. Boldface values are significant.

  • *

    Edmondson grade.

  • Multivariate analysis of miR-26a, IL-6, tumor size, tumor capsule, tumor thrombus, and TNM stage.

  • Multivariate analysis of the combination of miR-26a and IL-6, tumor size, tumor capsule, tumor thrombus, and TNM stage.  Abbreviations: AFP, alpha-fetoprotein; ALT, alanine aminotransferase; CI, confidence interval; HBsAg, hepatitis B surface antigen; HR, hazard ratio; TNM, tumor-node-metastasis.

Univariate analysis
 miR-26a (high versus low)0.34 (0.19-0.63)<0.0010.61 (0.39-0.94)0.025
 IL-6 (high versus low)6.92 (3.81-12.58)<0.0013.43 (2.13-5.52)<0.001
 Sex (female versus male)1.37 (0.59-3.19)0.4540.78 (0.42-1.48)0.455
 Age (>50 years versus ≤50 years)0.85 (0.51-1.40)0.5230.87 (0.56-1.36)0.552
 HBsAg (positive versus negative)0.93 (0.40-2.21)0.8690.98 (0.46-2.16)0.996
 Cirrhosis (yes versus no)1.01 (0.56-1.80)0.9641.05 (0.63-1.75)0.851
 ALT (≥75 versus <75 U/L)1.46 (0.79-2.69)0.2231.07 (0.60-1.90)0.812
 AFP (≥20 versus <20 ng/mL)1.55 (0.87-2.76)0.1371.46 (0.88-2.42)0.140
 Tumor size (≥5 cm versus <5 cm)3.60 (2.04-6.35)<0.0012.57 (1.61-4.08)<0.001
 Tumor number (multiple versus single)1.76 (0.97-3.19)0.0631.68 (1.00-2.85)0.051
 Tumor capsule (complete versus none)2.19 (1.30-3.71)0.0031.92 (1.22-3.01)0.005
 Tumor thrombus (yes versus no)4.04 (2.36-6.93)<0.0013.53 (2.16-5.78)<0.001
 Tumor differentiation* (>II versus I-II)1.24 (0.70-2.19)0.4501.13 (0.68-1.89)0.639
 TNM stage (III/IV versus I/II)3.50 (2.10-5.84)<0.0012.88 (1.82-4.54)<0.001
 Combination of miR-26a and IL-6
  I versus IV0.06 (0.02-0.14)<0.0010.23 (0.13-0.43)<0.001
  II versus IV0.21 (0.10-0.44)<0.0010.32 (0.16-0.63)0.001
  III versus IV0.51(0.28-0.94)0.0300.79 (0.45-1.38)0.403
Multivariate analysis
  miR-26a (high versus low)0.394 (0.229-0.676)0.0010.65 (0.41-1.01)0.043
  IL-6 (high versus Low)5.639 (2.939-10.820)<0.0012.82 (1.71-4.62)<0.001
  Tumor size (≥5 cm versus <5 cm)3.278 (1.811-5.934)<0.0012.11(1.29-3.46)0.003
Multivariate analysis
 Combination of miR-26a and IL-6
  I versus IV0.06 (0.02-0.14)<0.0010.24 (0.13-0.44)<0.001
  II versus IV0.25 (0.11-0.55)0.0010.39 (0.20-0.79)0.009
  III versus IV0.48 (0.25-0.89)0.0210.68 (0.39-1.21)0.196
  Tumor size (≥5 cm versus <5 cm)3.06 (1.67-5.62)<0.0012.13 (1.30-3.50)0.003

Next, we divided patients into four groups based on miR-26a and IL-6 expression levels. HCC patients with high miR-26a and low IL-6 had the best OS, lowest TTR, and best prognosis. In contrast, those with low miR-26a and high IL-6 had the poorest prognosis with the lowest OS and highest probability of tumor recurrence (P < 0.001) (Fig. 7E,F). The combination of miR-26a and IL-6 was an independent prognostic indicator for OS (P < 0.001) and TTR (P < 0.001) and was even better than miR-26a or IL-6 alone (Table 1).

Collectively, these data suggest that miR-26a and IL-6 are independent prognostic indicators for HCC patients. The combination of miR-26a and IL-6 shows the strongest prognostic value.

Discussion

In previous studies, miR-26a has been reported to regulate hepatocyte proliferation during liver regeneration,28 block G1/S transition,29 and suppress in vivo tumor growth.30 The molecular mechanisms by which miR-26a inhibits HCC remain unclear. In the present study, we found that miR-26a expression was markedly decreased in HCC tissues versus adjacent nontumor liver tissues. For the first time, we illuminate the role of miR-26a deregulation in tumor growth and metastasis using in vitro and in vivo assays. We found that miR-26a overexpression suppressed in vitro cell proliferation and invasion, induced cell cycle arrest, promoted apoptosis, and restrained in vivo tumor growth and metastasis. Conversely, down-regulation of miR-26a inhibited apoptosis and promoted proliferation, invasion, and cell cycle transition to S phase. These results indicate that miR-26a might be a novel tumor suppressor that plays important roles in the regulation of tumor growth and metastasis of HCC.

We found the following evidence that miR-26a inhibits tumor growth and metastasis in part by suppressing IL-6. (A) The mRNA levels of miR-26a inversely correlated with IL-6 levels in HCC tissues. (B) Up-regulation of miR-26a significantly reduced IL-6 levels in HCC cells, whereas down-regulation of miR-26a increased IL-6 levels. (C) Overexpression of miR-26a decreased the luciferase reporter activity of wild-type 3′-UTR but not mutant 3′-UTR of IL-6. (D) More importantly, the effects of miR-26a modulation on cell proliferation, apoptosis, and invasion of HCC cells were accompanied by changes in IL-6 levels and activities. Introduction of IL-6 abrogated the effects induced by miR-26a. These data support IL-6 as a downstream mediator of miR-26a function in HCCs.

Autocrine production of IL-6 by tumor cells has long been linked with malignancy.31 IL-6 promotes tumor growth and invasion in many types of human cancer,32-35 and high IL-6 levels are associated with the development of HCC.36 IL-6 stimulated Stat3 phosphorylation and promoted cell survival. Further, inhibition of Stat3 signaling has been shown to block the antiapoptotic activity of IL-6 in human liver cancer cells.37 Stat3 has been found to be constitutively activated by IL-6 in many types of human malignancies.38 Activation of STAT3 induces transcription of target genes, including Bcl-2, Mcl-1, cyclin D1, and MMP2.38 We found that miR-26a significantly decreased expression of p-Stat3, Mcl-1, Bcl-2, cyclin D1, and MMP2, whereas IL-6 treatment abrogated this effect. Knockdown of IL-6 inhibited phosphorylation of Stat3 and expression of Stat3 target genes, similar to miR-26a. These results indicate that miR-26a inhibits IL-6-Stat3 signaling.

Cyclin D1 is a crucial mediator of G1 to S progression. Up-regulation of cyclin Dl results in rapid growth of a subset of HCCs.39 Thus, down-regulation of cyclin D1 could be a mechanism by which miR-26a suppresses cell proliferation and promotes cell cycle arrest at the G1 phase. Bcl-2 and Mcl-1 are overexpressed in some HCC tissues and play important roles in regulating apoptosis of HCC.40, 41 In this study, we found that miR-26a significantly down-regulated Bcl-2 and Mcl-1 via inhibition of the IL-6-Stat3 pathway, which led to apoptosis of HCC cells. MMPs are a family of enzymes that proteolytically degrade various components of the extracellular matrix. High levels of certain MMPs are closely correlated with the invasive and metastatic potential of tumors.42 Specifically, activated Stat3 regulates tumor invasion of melanoma cells by regulating gene transcription of MMP2. In contrast, blocking Stat3 activity significantly suppresses MMP2 expression, tumor invasion, and brain and lung metastasis.43, 44 Therefore, down-regulation of MMP2 is one important mechanism by which miR-26a inhibits HCC invasion and metastasis.

The fundamental role of miR-26a in tumor cell proliferation, invasion, and metastasis suggests its potential application as a prognostic predictor and cancer therapeutic target. Our previous study reported that low expression of miR-26a was an independent predictor of survival in a large cohort of patients from China who had hepatitis B virus–associated HCC.14 In the present study, we confirm that miR-26a, IL-6, and the combination of miR-26a and IL-6 are independent prognostic indicators for OS and TTR in HCC patients. The prognostic effect of the combination of miR-26a and IL-6 is much better than that of miR-26a or IL-6 alone.

In conclusion, our results show that miR-26a is down-regulated in HCC and plasma, and inversely related to tumor recurrence and metastasis. We demonstrated that miR-26a is an independent prognostic factor for HCC, particularly when combined with IL-6 expression level. Using in vitro and in vivo studies, miR-26a was confirmed as a novel inhibitor of tumor growth, invasion, and metastasis of HCC. The multiple tumor-suppressive functions of miR-26a were mediated by the IL-6-Stat3 pathway. These data suggest that miR-26a deregulation may play an important role in tumor growth and metastasis and may be a novel prognostic marker and potential therapeutic target in HCC.

Ancillary