MicroRNA-431 inhibits migration and invasion of hepatocellular carcinoma cells by targeting the ZEB1-mediated epithelial-mensenchymal transition.

MicroRNA-431 (miR-431) has been recognized as an oncogenic miRNA, being implicated in the initiation and development of human cancers. Recently, deregulation of miR-431 has been reported in several tumors. However, the clinical significance of miR-431 and its underlying role in human hepatocellular carcinoma (HCC) are poorly explored. Herein, we found that miR-431 expression was reduced in HCC tissues compared to noncancerous tissues. Otherwise, down-regulation of miR-431 was observed in aggressive tumor tissues. The levels of miR-431 expression in HCC cell lines were significantly lower than that in a nontransformed hepatic cell line. Clinical association analyses disclosed that a low level of miR-431 was prominently associated with poor prognostic features of HCC including venous infiltration, high Edmondson-Steiner grading and advanced tumor-node-metastasis (TNM) tumor stage. Our in vitro studies showed that up-regulation of miR-431 expression reduced cell invasion and migration in HCCLM3 cells. In contrast, down-regulation of miR-431 expression promoted SMMC-7721 cell invasion and migration. We found that up-regulation of miR-431 expression decreased zinc finger E-box binding homeobox 1 (ZEB1) expression and inhibited the epithelial-mesenchymal transition (EMT) with increased E-cadherin expression and decreased vimentin expression in HCCLM3 cells. Otherwise, down-regulation of miR-431 expression increased ZEB1 expression and promoted EMT in SMMC-7721 cells. Significantly, ZEB1 was identified as a target of miR-431 in HCC. ZEB1 knockdown abrogated the effect of miR-431 silencing on EMT and cell mobility in SMMC-7721 cells. In conclusion, miR-431 inhibits migration and invasion of HCC cells by suppressing ZEB1-mediated EMT.

Introduction

Hepatocellular carcinoma (HCC) is one of the most prevalent cancers with the worldwide incidence of more than 700,000 new cases annually, and ranks as the third leading cause of cancer-related mortality rate in the world [1], [2], [3]. Prominent progresses have been made in the diagnosis and treatment of HCC, while the long-term prognosis of HCC patients remains unsatisfactory [4], [5]. Frequent intrahepatic or systemic metastasis is the culprit for the high rate of recurrence and poor survival of HCC patients [3]. Therefore, a comprehensive understanding of the underlying mechanisms of HCC metastasis is of great significance which will contribute to the identification of novel diagnosis and therapeutic targets.

MicroRNAs (miRNAs), which are a group of endogenous short non-coding single strand RNAs, act as important post-transcriptional regulator of gene expression by interacting with the 3′-UTR of the targeted mRNAs [6]. And they participate in various biological processes [7], [8], [9] including cell differentiation and development, cell proliferation and apoptosis, cell movements and stem cell renewal. It has been widely accepted that the deregulation and dysfunction of miRNAs play a critical role in the pathogenesis of human cancers [10], [11]. And emerging evidence has demonstrated that aberrant expression of miRNAs played a key role in the initiation and progression of HCC [12]. However, the exact mechanisms by which miRNAs regulate the metastasis of HCC remain poorly understood.

MicroRNA-431 (miR-431), which was initially recognized as nervous system specific miRNA [13], has been identified as a novel tumor-related microRNA. Elevated expression of miR-431 was confirmed in colorectal cancer (CRC) and increased miR-431 expression was able to differentiate healthy controls from CRC patients [14], suggesting the oncogenic role of miR-431 in CRC. Studies of medulloblastoma and glioblastoma confirmed that miR-431 mediated the inhibitory effect of human IFN-β on cell viability by modulating Janus-activated kinase–Signal transducers and activators of transcription (JAK–STAT) signaling pathway [15], indicating the inhibitory role of miR-431 in these cancers. Therefore, the exact role of miR-431 in human cancers is dependent on the type of human cancer. The study of HCC found that versican 3′-UTR could promoted the proliferation, migration, invasion, and colony formation of HCC cells by binding and arresting the functions of miR-431 [16]. However, the clinical significance and functional roles of miR-431 in HCC are still not fully understood.

In the present study, we find that reduced expression of miR-431 is observed in HCC with aggressive phenotype. The low-expression of miR-431 is correlated with adverse clinicopathological parameters of HCC. Furthermore, miR-431 inhibits migration and invasion of HCC cells in vitro. MiR-431 inversely regulates ZEB1 abundance and EMT in HCC cells. The effects of miR-431 down-regulation on EMT and cell mobility are abrogated by ZEB1 knockdown in SMMC-7721 cells. Our data uncover that miR-431 facilitates the invasive abilities of HCC cells by inhibiting ZEB1-mediated EMT.

Materials and methods

Clinical samples and cell lines

86 samples of HCC and matched tumor-adjacent tissues were obtained from the Department of General Surgery at the Taihe Hospital, Hubei University of Medicine during January 2010–December 2013. The demographic features and clinicopathologic date were shown in Table 1. All samples were used after obtaining informed consent from patients. Patients receiving preoperative chemotherapy or embolization were excluded. Intrahepatic spreading, venous infiltration or tumor invasion into bile ducts were observed in 18 patients. 41 patients suffered tumor recurrence. The Hubei University of Medicine Ethics Committee approved all protocols according to the Declaration of Helsinki (as revised in Tokyo 2004).

Table 1

Clinical correlation of miR-431 expression in HCC.

Clinicopathologic features		Total no. of patients, n = 86	No. of patients		P
Low miR-431	High miR-431
Age (y)	<50	36	15	21	0.190
⩾50	50	28	22
Sex	Male	65	34	31	0.451
Female	21	9	12
HBV	Absent	28	12	16	0.357
Present	58	31	27
Serum AFP level (ng/mL)	<400	32	14	18	0.372
⩾400	54	29	25
Tumor size (cm)	<5	32	13	19	0.181
⩾5	54	30	24
No. of tumor nodules	1	68	31	37	0.112
⩾2	18	12	6
Cirrhosis	Absent	36	15	21	0.190
Present	50	28	22
Venous infiltration	Absent	76	35	41	0.044⁎
Present	10	8	2
Edmondson–Steiner grading	I + II	65	27	38	0.006⁎
III + IV	21	16	5
TNM tumor stage	I + II	66	28	38	0.011⁎
III + IV	20	15	5

HCC, hepatocellular carcinoma; HBV, hepatitis B virus; AFP, alpha-fetoprotein; TNM, tumor-node-metastasis.

Statistically significant.

The human HCC cell lines (SMMC-7721, Hep3B, MHCC97H and HCCLM3) and a nontransformed hepatic cell line (LO2) (the Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences, Shanghai, China), were routinely cultured in complete Dulbecco’s modified Eagle medium (DMEM, Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS, Gibco) with 100 units/mL penicillin and 100 μg/mL streptomycin (Sigma, St-Louis, MO, USA) at 37 °C in a humidified containing of 5% CO2 incubator.

Real time quantitative reverse transcription-PCR (qRT-PCR)

qPCR primer against mature miRNA hsa-miR-431-5p (HmiRQP0499) and Homo sapiens snRNA U6 qPCR Primer (HmiRQP9001) were obtained from Genecopoeia (Guangzhou, China). Cell lines and frozen tumor specimens were subject to total RNA extraction using Trizol reagent (Invitrogen, California, USA). Complementary DNA synthesis was performed using TaqMan MicroRNA Reverse Transcription kit (Applied Biosystems, Foster City, CA, USA) according to the manufacturer’s instructions. Amplification and detection of miR-431 and U6 were performed using a TaqMan Human MiRNA Assay Kit (Applied Biosystems) and the ABI PRISM 7900 Sequence Detection System (Applied Biosystems). The relative expression of miR-431 was shown as fold difference relative to U6.

Cell transfection

All miRNA vectors were purchased from Genecopoeia (Guangzhou, China) including miR-431 expression vector (HmiR0213-MR04), the control vector for miR-431 (CmiR0001-MR04), miR-431 inhibitor (HmiR-AN0499-AM04) and the negative control for the miR-431 inhibitor (CmiR-AN0001-AM04). The targeted sequences for ZEB1 siRNA duplex (sense 5′-CAG UGA AAG AGA AGG GAA UUU-3′ and antisense 5′-AUU CCC UUC UCU UUC ACU GUU-3′) or a nonspecific duplex oligonucleotide as a negative control (sense 5′-GGA AAG ACG AUG ACG GAA AUU-3′ and antisense 5′-UUU CCG UCA UCG UCU UUC CUU-3′) were synthesized by Sangon Biotech (Shanghai) Co., Ltd. (Shanghai, China). The vectors mentioned above were transfected into HCC cells using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions.

Boyden chamber, transwell and MTT assays

The migration of HCC cells was detected by Boyden chamber assays (NeuroProbe, Gaithersburg, MD, USA) as previously described [17]. BioCoat Matrigel invasion chamber (Becton Dickinson Labware, Bedford, MA, USA) was used for Transwell cell invasion assays as previously mentioned [18]. 2 × 103 HCC cells were seed in 96-well plates and the 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyl tetrazolium bromide (MTT, Roche, USA) assay was employed to assess cell viability at 12, 24 and 36 h.

Western blot

RIPA buffer containing protease inhibitor was used for total protein extraction. Then, prepared protein samples were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS–PAGE) and transferred onto polyvinylidene fluoride (PVDF) membranes. The membrane was blocked with 5% skimmed milk in TBST and incubated with the appropriate antibody. The following primary antibodies were used: ZEB1 (D80D3, #3396; Cell Signaling, Beverly, MA, USA), E-cadherin (24E10, #3195; Cell Signaling, Beverly, MA, USA), vimentin (C-20, sc-7557; Santa Cruz, CA, USA) and GAPDH (sc-25778, Santa Cruz Biotechnology, Santa Cruz, CA, USA). Horseradish peroxidase (HRP)-conjugated secondary antibodies (Bio-Rad, Hercules, CA, USA) were used at an appropriate dilution and detected using enhanced chemiluminescence regents (Thermo Scientific, Waltham, MA, USA).

Luciferase reporter assay

Wild type (wt) ZEB1-3′-untranslated region (UTR) and mutant (mt) ZEB1-3′-UTR were constructed as previously described [19]. For the luciferase reporter assay, SMMC-7721 cells were respectively transfected with wt ZEB1-3′-UTR and mt ZEB1-3′-UTR and subsequently transduced with corresponding miRNA vectors. The cells were collected after 48 h. The dual-luciferase reporter assay system (Promega, Madison, WI, USA) was used to measure Renilla luciferase activity according to the manufacturer’s protocols.

Statistical analysis

The quantitative data are expressed as Mean ± SEM. The SPSS statistical package for Windows Version 13 (SPSS, Chicago, IL, USA) and GraphPad Prism 5 software (GraphPad Software, Inc, San Diego, CA, USA) were used for statistical analysis. A Pearson chi-squared test was applied to determine clinicopathological correlations. Comparison between two or more groups was subjected to a two-tailed Student’s t test or ANOVA when appropriate. Difference were considered significant when P < 0.05.

Results

Reduced expression of miR-431 is associated with aggressive HCCs

The mean level of miR-431 expression in HCC was obviously lower than that in matched tumor-adjacent tissues (P < 0.05, Fig. 1A). Aggressive HCCs were defined as cases with intrahepatic spreading, venous infiltration or tumor invasion into bile ducts. By contrast with nonaggressive HCCs, the expression of miR-431 was obviously reduced in aggressive cases (P < 0.05, Fig. 1B). Otherwise, the expression of miR-431 was significantly decreased in tumor tissues arising from patients suffered recurrence (P < 0.05, Fig. 1C). Taken together, reduced expression of miR-431 was associated with aggressive HCCs. Next, we found that miR-431 was obviously down-regulated in HCC cells compared with normal hepatic cells (P < 0.05, Fig. 1D). In addition, the expression of miR-431 in highly invasive HCC cell lines, MHCC97H and HCCLM3, were obviously lower than those in low invasive ones including SMMC-7721 and Hep3B (P < 0.05, Fig. 1D). These results demonstrate that reduced expression of miR-431 facilitates metastatic behavior of HCC cells.

Fig. 1

The expression levels of miR-431 in HCC tissues and cells. Comparing differences in the expression levels of miR-431 between (A) HCC (n = 86) and matched nontumor tissues (n = 86), (B) aggressive (n = 18) and nonaggressive tumor tissues (n = 68), (C) HCC tissues arising from recurrent (n = 41) and non-recurrent groups (n = 45), and (D) HCC cell lines with different metastatic potentials and the immortalized hepatic cell line LO2 (n = 6). ∗P < 0.05.

Clinical significance of miR-431 expression in HCC

HCC cases were divided into miR-431 low expression group (n = 43) and high expression group (n = 43) in accordance with the cutoff value, which was defined as the median level of miR-431 expression in this cohort. HCC patients with venous infiltration (P = 0.044), high Edmondson–Steiner grading (P = 0.006) and advanced TNM tumor stage (P = 0.011) showed a significant lower levels of miR-431 expression (Table 1). Herein, clinical association analyses indicate that low-expression of miR-431 is associated with adverse clinicopathologic features of HCC.

MiR-431 inhibits migration and invasion of HCC cells

We transduced HCC cell line, HCCLM3, with miR-431 expression vector and the control vector for miR-431. The expression of miR-431 was obviously overexpressed by miR-431 expression vector in HCCLM3 cells as assessed by qRT-PCR (P < 0.05, Fig. 2A). Cell migration tested by Boyden chamber assays showed that overexpression of miR-431 led to a significant reduction of migrated cell numbers in HCCLM3 cells (P < 0.05, Fig. 2B). Otherwise, Transwell assays indicated that the number of invaded HCCLM3 cells was obviously decreased after overexpression of miR-431 (P < 0.05, Fig. 2B). Next, miR-431 down-regulating SMMC-7721 cells was established and measured by qRT-PCR (P < 0.05, Fig. 2C). On the contrary, knockdown of miR-431 evidently promoted migration and invasion of SMMC-7721 cells (P < 0.05, respectively, Fig. 2D). However, MTT assays indicated that the cell viability was not significantly changed after manipulating miR-431 expression levels in both HCCLM3 and SMMC-7721 cells (P > 0.05, respectively, Fig. S1). Taken together, miR-431 inhibits migration and invasion of HCC cells.

Fig. 2

MiR-431 reduces migration and invasion of HCC cells. (A) HCCLM3 cells that were transfected with miR-control (control) and miR-431, respectively, were subjected to qRT-PCR for miR-431 expression. Quantitative analysis indicated that miR-431 was significantly up-regulated by miR-431 expression vectors in HCCLM3. n = 6, ∗P < 0.05. (B) Cell migration as measured by Boyden chamber assays was inhibited by up-regulation of miR-431 in HCCLM3 cells as compared with control cells. MiR-431 overexpressing HCCLM3 cells conferred a less number of invaded cells as compared with control cells. n = 3 repeats with similar results, ∗P < 0.05. (C) SMMC-7721 cells that were transfected with negative control (NC) and miR-431 inhibiotr (anti-miR-431), respectively, were subjected to qRT-PCR for miR-431 expression. n = 6, ∗P < 0.05. (D) Down-regulation of miR-431 promoted cell migration and invasion in SMMC-7721 cells. n = 3 repeats with similar results, ∗P < 0.05.

MiR-431 suppresses ZEB1-induced EMT in HCC cells

To uncover the underlying mechanisms by which miR-431 exert its functional role in HCC cells, we searched two publicly available databases (TargetScan 6.2 and miRanDa) for the predicted target of miR-431. ZEB1, an important regulator of EMT in HCC, was found to be one of the predicted targets of miR-431. HCCLM3 cells that were transduced with miR-control or miR-431 were subjected to WB for ZEB1 and EMT markers. As shown in Fig. 3, overexpression of miR-431 significantly reduced the expression of ZEB1 protein and resulted in increase of E-cadherin level and decrease of vimentin level in HCCLM3 cells (P < 0.05, respectively, Fig. 3). In contrast, knockdown of miR-431 increased the level of ZEB1 protein and led to decrease of E-cadherin expression and increase of vimentin expression (P < 0.05, respectively). Herein, a dual-luciferase reporter assay was performed to disclose whether miR-431 directly binds to the 3′-UTR of ZEB1 mRNA. As expected, miR-431 obviously reduced the luciferase activity of ZEB1 containing a wt 3′-UTR (P < 0.05, Fig. 4A and B). While an increase in the luciferase activity of wt ZEB1 3′-UTR was observed after anti-miR-431 tranfection (P < 0.05, Fig. 4A and B). However, there were no significant changes of the luciferase activity of ZEB1 with an mt 3′-UTR after corresponding miRNA vectors transfection (Fig. 4A and B). Notably, ZEB1 knockdown by a specific siRNA abolished the effect of miR-431 down-regulation on EMT with elevated expression of E-cadherin and reduced expression of vimentin in SMMC-7721 cells (P < 0.05, respectively, Fig. 5A). And ZEB1 knockdown abrogated the effect of miR-431 silencing on SMMC-7721 cell mobility with reduced abilities of cell migration and invasion (P < 0.05, respectively, Fig. 5B). Thus, our results indicate that ZEB1 is a functional target of miR-431 in HCC.

Fig. 3

MiR-431 inversely regulates ZEB1 expression and EMT in HCC cells. Representative western blot indicated that up-regulation of miR-431 decreased ZEB1 expression and inhibited EMT with elevated expression of E-cadherin and reduced expression of vimentin in HCCLM3 cells. Meanwhile, down-regulation of miR-431 increased ZEB1 expression and promoted EMT in SMMC-7721 cells. n = 6, ∗P < 0.05.

Fig. 4

ZEB1 is identified as a functional target of miR-431 in HCC. (A) miR-431 and its putative binding sequence in the 3′-UTR of ZEB1. The mutant miR-431 binding site was generated in the complementary site for the seed region of miR-431 (wt, wild type; mt, mutant type). (B) MiR-431 significantly suppressed the luciferase activity that carried wt but not mt 3′-UTR of ZEB1. Anti-miR-431 led to a noticeable increase in luciferase activity of wt 3′-UTR of ZEB1. n = 3 repeats with similar results, ∗P < 0.05.

Fig. 5

MiR-431 inhibits EMT by targeting ZEB1 in HCC cells. MiR-431 silenced SMMC-7721 cells and control cells that were transfected with scrambled siRNA or ZEB1 siRNA were subjected to Western blot and Transwell assays. (A) ZEB1 knockdown led to up-regulation of E-cadherin and down-regulation of vimentin in negative control (NC) vectors transfected SMMC-7721 cells. Otherwise, ZEB1 deletion abolished the effect of miR-431 down-regulation on EMT with elevated expression of E-cadherin and reduced expression of vimentin in SMMC-7721 cells. n = 6; ∗P < 0.05. (B) ZEB1 knockdown reduced cell migration and invasion in control SMMC-7721 cells. Furthermore, ZEB1 knockdown abrogated the effect of miR-431 silencing on SMMC-7721 cell mobility with reduced abilities of cell migration and invasion. n = 3; ∗P < 0.05.

Discussion

Several studies have demonstrated that miRNAs regulate the expression of oncogene or tumor suppressor, suggesting a new mechanism involved in the initiation and development of HCC [20]. Aberrant expression of miR-431 in other human cancers promoted us to determine its expression status in HCC. Thus, 86 paired of HCC and matched tumor-adjacent tissues were subjected to qRT-PCR for miR-431 expression. Our data revealed that the expression of miR-431 in HCC was significantly lower than that in tumor-adjacent tissues. Moreover, the expression of miR-431 was significantly reduced in aggressive HCCs as compared with that in nonaggressive tumors. Otherwise, miR-431 was expressed at significant lower levels in HCC cell lines, especially MHCC97H and HCCLM3, as compared with L02. Importantly, our data showed that reduced expression of miR-431 was associated with adverse prognostic features of HCC. Notably, miR-431 has been considered as a potent biomarker of human HCC with high prognostic value [21]. Altogether, our results indicate that the status of miR-431 may be critical for prognosis determination in HCC patients.

Our gain- and loss-of-function experiments demonstrated that overexpression of miR-431 inhibited migration and invasion of HCCLM3 cells. While knockdown of miR-431 led to increase of migrated and invaded SMMC-7721 cells. However, the HCC cell viability was not significantly changed after manipulating miR-431 expression levels. These results indicate that miR-431 truly inhibits migration and invasion of HCC cells. EMT is a crucial mechanism hijacked by hypoxic cancer cells for metastasis [22], [23]. The term EMT has been coined to define the conversion process of epithelial cells into migratory and invasive fibroblastoid cells, mostly characterized by loss of epithelial cell properties, acquisition of mesenchymal cell phenotype, and enhancement of HCC cell migration and invasion capability [24], [25]. Next, we determined the effect of miR-431 alteration on EMT makers, E-cadherin and vimentin in HCC cells. Western blot analyses found that overexpression of miR-431 suppressed EMT with increase of E-cadherin protein and decrease of vimentin protein in HCCLM3 cells. On the contrary, down-regulation of miR-431 facilitated EMT in SMMC-7721 cells. In summary, miR-431 may inhibit cell migration and invasion by suppressing EMT in HCC. Recent studies reported that down-regulation of miR-431 increase cell invasion and migration in HepG2 cells [16]. ZEB1 has been considered as a crucial regulator of EMT in HCC [26]. Accordingly, we investigated whether miR-431 plays a regulatory role in ZEB1 expression in HCC. Our data indicated that ZEB1 abundance was inversely associated with miR-431 alteration in HCC cells. Herein, we validated ZEB1 as a direct target of miR-431 in HCC. Importantly, the regulatory effects of miR-431 knockdown on EMT progression, cell migration and invasion were partially abrogated by ZEB1 knockdown in HCC cells. Thus, we propose that miR-431 inhibits HCC cell migration and invasion by suppressing ZEB1 induced-EMT.

In summary, we find that down-regulation of miR-431 is observed in HCC tissues and cells, especially in aggressive HCCs and highly metastatic cell lines. In addition, the low-expression of miR-431 is associated with poor prognostic features of HCC. We prove that miR-431 inhibits migration and invasion of HCC cells by suppressing ZEB1-induced EMT. In conclusion, we suggest that miR-431 functions as a tumor suppressive miRNA, and inhibits the invasive behavior of HCC by inhibiting ZEB1.

Conclusions

In summary, our study discloses that the expression of miR-431 is down-regulated in HCC tissues as compared matched tumor-adjacent tissues. Furthermore, miR-431 is expressed at significant higher levels in aggressive and recurrent HCCs. The relative expression of miR-431 in HCC cell lines is significantly lower than that in the normal hepatic cell line. In vitro studies demonstrate that miR-431 inhibits migration and invasion of HCC cells. Importantly, our data indicate that miR-431 inversely regulates the abundance of ZEB1 and EMT progression in HCC cells. Herein, ZEB1 is identified as a direct downstream target of miR-431. Notably, the effects of miR-431 down-regulation on EMT, cell migration and invasion are abrogated by ZEB1 knockdown in SMMC-7721 cells. This study reveals that miR-431 may play a critical role in the tumor metastasis of HCC and may be a potential therapeutic target for HCC.

Competing interest

The authors declare that they have no competing interests.

Author contribution statement

Kexin Sun, Tiancai Zeng, Dong Huang, Zizhong Liu and Shang Huang carried out the cell biology and molecular biology experiments, participated in the sequence alignment and drafted the manuscript. Kexin Sun, Tiancai Zeng, Jiong Liu and Zhenfan Qu participated in the design of the study and performed the statistical analysis. Zhenfan Qu conceived of the study, and participated in its design and coordination and helped to draft the manuscript. All authors read and approved the final manuscript.